Cycles The Science of Prediction

This volume belongs to set nun1ber _C/dt.
e of I 000, of
CYCLES CLASSIC LIBRARY COLLECTJC)N .
The Foundation for the Study of Cycles
124 S. Highland Ave.
Pittsburgh, PA 15206
(412) 441-1666
JUNE 19!17
CYCLES
1' 1-IE SC I El\C; E OF PREDICTION
f>.y EO\VARO R. DF.,VEY
o.11tl ED\VIN F. J) AKIN

Contents
Charts a11d Diagrams
Concerning Eco11omic Preclictiori 1\n 1ntro<l1lction J<l
I l,Vhy Trenrl.< A re Important 1
II P1ttterr1s i11 Growth of Orgariisms I()
111 The Growth Tr~rtd iu Our Basic Industries

IV Trends in Some Other JntltL<tries 42
V Some Rhythmir Cycles in .Vatural Phe11t)me11a 51
\'I The 54-Year Rhythm
vu The 9"Yettr Rhythm
Vlll The 3~-Year Rhythm 102
IX The z 8· Year Rhyr.hm

x Causes, Corre/otifl11s, Co11;ectu.re.r 135
XI Anal)•sis a11d Syntltesis 160
XII 1"iming a Br~si11ess 172

XII I Avoidi11g Some Economic l/lusi-011s 191
XIV fVar and I ts Dislocations 200
XV Postwar Trends
XVI Postwar Rhythms
Al'l'ENOIX I
APPENDIX IT
APPl!l\OJX II I
Charts aml Diagrams
(l f•AP TER t
F J C. J>AGE
1. Trct1tl ()/ tt 1l)•[Jotlteti.:al1Ju.1i11t<,\f 0 1gu1rizfltiQn 3
2. Tlie 'J'tdt•n ;,, the Tlnited Slatet ?
4. Index of Sponis!. T 1atk 9
CHA r TER II
t.Growth in nody l¥eigltt of o Afnle IYliite Rat
f, Grvu:tl1 i11, lVt>igJtt of a Pu1nplt.i11
3. Growllt of 11 Populatio1' of Ycost Cells
.."
'3
4. Growth of n Populatio·1• of Fr1,it f.'li1:s 13
5· Growt}i of tJ1e POfJU/JJ.tio11 of Swetltn 14
6. Growl!. of tlit Pop1.ta1ion of tlie U11it<d States 15
7. G111DJth of 1J1c Population of Front, 16
8. G111R1tl1 of the Indigenous .l\'atiw: Pn/1ulation of Algeria 17
9. Growl}1 of the Pnj1ulatiot1 of Ct:"'"""'Y 18
C llA .. ' 1 t : R Ill
1. Population of Co>1ti11e11tal U. ~'., 1$;0-1940

2. U .S. 1\f11t1 ufartttres, 1831~19.fS
S· U.S. Mercharulise Exporll, '8J<,_,9-H
4. U.S. l\l<rchttndise lmpurts, 1830-1915
5- U.S. Iron and Steel Ptotluction
6. Stft(l."1 H.o;'lw°'js
1· r..l . s. ~·1ti/1br1ildin.~, 183t~l fJ15
a. U. S.1lt1to11)obile Produrtiot1, 1900-191,t
9. JlrJf~,6 of u . ~·.11or.o,·e·Dra111n Vel•itlt: Prorltl(tio11, ISJ!J-'939
io. Gatti~ ()r;
U. 5'. Fartr-u1 } ll1•uary 1li4f1-r944
• 1. U.S. Corn Production, '8J!J-'!NJ
t t.U.S. Cotton Production, 18J0--19/J
•5- U.S. Wool f>rodw:tion,,84<>-'90
14. U.S. IV/ital Production, 183<J-194)
1;. Malt Liquor fo the United Siat<t
rvi l
CHARTS AND DIACRA M S
F 1 C. P:\CE
16. U.S. L11mber Produt:tion, 18)1)-19/5 39
17. U.S. Couon Spi11dla in Operation, 18;1)-I!JIJ ~9
18. U.S. Cf>OI Produclion, 184~191; 40
'!>- U.S. S111eller Copper Productinn, 1R5t>-191; 40
to. U . •~. n~fi1ied Prit,rary Lead Prnductir'"''' 1S;t>-19-13 41
r.r1ArTEll rv
1. 1! • •~. ll<l)'Otl Ynrn C<Jtl.ftJtnfJtio·ri, 1911 19/S
:i. ll. S. Aircraft Pro<luc,if>n, 1914-1 914
!· U. S'. Elt'ctric £1ier{f)' Prod1J.ct ir>n, l9n7-19,15
4. J!<1lttt; of U. S. l\ras-urol (;as Prodt1ctitJ1J, 1!1!1>19<1J
!i· U. ~'.Petroleum P11Jtl11r:lion, 1E>'_;g-194;
G. CJ.:). Paper and 1¥oo<l Prllp Ptodu.ctio,., 18;9- 1914
7. U. .~ . T obocco l'rod11ctio1i, t8)!r-' 91S
8. Trends of J1o riOttf U.S. J11dustries
CllAPTKk V
1.A /Jundonu of Atlantic Salmnn 55

•· Abundance of L ynx 54
S· A/)undnnce of Te11t Caterpillars 55
4. O:one nt l .ondon (K~w) a1td PariJ ti6
5. The 9~·Yea1 Rhythm in Insects. Pi.ih ~ ·\·fnr1•111als. 1\.1011, and Ozf>11e 57
G. A butJdant·c of ,it ice 5B
7. llol<S and Foxes 58
8 . .~ol11r and T e rrestrial Dala 59
~)· Variatiof1, i11Hu111<1ri £ 11iotio11 fi 1
IQ. nr.athJ fro11iP11eum,inia. 62
11. Death~ frorn l11flue:11za 62
1 t. 1'ree Grnrutl;. 63
13. Flowrr Crowth 6j
14. Ab11,1drLnceof Sunspots 64
15. //4riatiu,• i11 Te1restrial .~agnetir1'• 64
1G. S)•ntl&.esisa/ T wu C)-cles 65
17. Syrrth~sis of To.'O Cycks 65
18. Syrrthniso/ FourCy<ks 66
C11APTEI. ' ' I

I. u..(j. A verage l·Vholesale P·ricu
2. 1..1.S. AT1r-ruge JVl;olesale Prices,Exccplinf{ f.Var Peaks
8· 6So Year:s ,,, ii:heat Prices i·n England
4. The $4 ·Year Rhythrr1, f.11 JVlletJt J',ltesi,1 1:.'1igland
5. Long Wa ves itt World tVho/esale Prices
[vii 1
CHARTS AJ\'D DIACRA~!S
FIC. PACE
6. lJ'orl(l f.nten-clationships: Pn'ces and E."(ports 77
7. Lottg J-Vaves i11 J.J'ages So
8. Q1.lO/(llions of J,tlr.rr.st-Ocari11g Ser.un'ties 80
9. The Lc111g vi:of1e in Co(ll g,
1o. Pig 1TfJrt an(l Ltad Prod1lction i11 Eugla11d 81
11. Lot1g JVa:ues i·11 Prices 82
CH,\1'1''l=; R Vil
1. Th e 9-Ycar Rhythn1 i11 P·rices 89
2. U.S. Com.n1012 Stock Prices 90
8· Tile 9--l'car Rltylh'l'lt in Co1nmon Stock P·ri,·cs !)l
4. U.S. l~'hnles(Jlr. Pn'ces, 1830-z9.1 5 93
5. The 9-Year Rhythm.i1i J>i'ltolesale Prices 93
6. Ari. lndividua.l Bu.sirtess 94
7. RcttJf>ilr.1/atiori, the 9-Year Rh)'lllrti 95
8. ."i)'nth.esis 96
9. U.S. Wholesale Prices 97
1o. Gold P·rices a11d Currenry 98
CH,\J.>'rF.R. \'ITJ
1. Tl1e JV!·Year llh)•lhrn. i·n lnrt11.s.t"Tial Prodttction 101-1 ()5
~. Rate of Chartgc in Inclustrial Production 104- 105
3. TJ~e J~'t·Year Rl1)1t}11n in Pt'g Jro11 Produ,c1io1t J 06--1<ry
4. Ila le of c;Jiat1gt: i11. Pig lrori Producfio11 1ofr-107
r,. Tlte JY! Year Rhyllt111- l'tt Coru.1non .~tu1:k Pri,;e:s
4 lltr-111
6. Rate oj CJ1ange i1i Com1rion Stock Prices 110- lJl
CHAl''[Elt. I X
1. The r8}~·Year Rhythm in Real Estate Activity 116-117
2. Th e I 8 M-Year Rltytltrn in Btt.i ltl.irig Acti-vity 118-119
3. The 18 U l'1:ar Rltythm. iri iifa·rriag1:s
4
'20
4- The r8)1i·Ycor Rhythm in fVheat Acreage 120
5. B·rit:k J~clu.c.~tlu11, J :t l
6. B11iilrli11g Con.1tr·1i.ctiori i11 J{a111f>r1.r g J :.?:!

7. J\~t:t•' York .~k)·scrapers
"3
8. Chicago Skyscrapers 123
9. Pig 11'ori 129
10. Tlie I8J.i·Year Rh)1thn1 in Loa·n s and Discounts 130
t t. The 18~~ -Year Rh)•lli111 in Railr(>tJd .~tot:lt Prices 131
12. The 18%·Y<~tlr R.h)•lh·1n in Corn1>11>n-Slock P.ri<.·es 132
13. "/ }~e 18~ -Year Rh)•t}trn iti Comrtion Stock J>rices
4
133
14. Thr, 18~ ·Year Rh)•th111in the S<ll'.:s uf an Industrial Company 133
[viii l
CHARTS At"l> f>IAGKA1'f~
CllAPTJ:R "'
r tc. PAC.,~
1. Rrsi1ltnlial Building 136

2. 1'extiles 136
3. .Sugar l'rices 1 37
4. CtJttle Prirr.J J !Ji'
!i· ]Jepper Prices 13K
6. Sales u/ a1' !1ttlu1lrifll Cotnpa'1'1 138-139
i· S11!1;pots a1ul P 1odur11011 t4:t
8. Solor Plvnnmt'no a11d Btuin~ss .1ftivi1y 143
g. Solar 1Cod1011tn1111111 .'tl«A r rit<' 1'14
JO. Sur1spclsa11d ~''""l P11 rt:l 14r,
11. lt'(l'J(.' .';J,et:lftlrtl 15<r-lf,t
J 2. Sun.spot Cycles 1 53
J .3· Sht:l,:to1rited Repr1·sf'rl '"' i1Jrt <If 1lte I 8 >3. r enr f~}tylh ,,. i 1t Stt11spot
i\'utnbrrJ·
14. Ot/ter Sunsfrul Cyclt:S tJ11d Prices
~llArTER. XI
1. Calton's Qui1111111.'< 16•
2. /'·requerteJ T>islTtbutiCJtt 16~
3. Bearing Ball> 163
4. Sound Sy-rrlht5i\ - ~rwo c:o1nf1n11r.nt..t 16.t
5. .~ound .(jyntht:-si1 - "I'wtlv~ (.'onifr<Jrtt:r.ts 165
6. Tide Preilict1t1r1 167
ct-rA .. 'l'l'.R. XII

1. A Sale.t Ch111t l i?S
2. A Rate ''f (";ron:t/1 Chart 174
3. Tu."O Regular C-.•t.lrt 1 75
4. Synt hai.s 176
5-- The Tinid L;-ne 176
G. Tre-ntl l.i11e ond C'Y'ltJ • 77
7. Projectio1l nnd Cor11/Mti.io11 t78
8. /f11 Etlla1ge1ntt1t 1 79
9. Sale:,· of a11 I 1idtcsr.1·itTl Com/><1rf'y. 1871- r9,10 1S•

10. Analysis 183
1 1. S)•t1tl1es1:s t8:t
J 2.Soles, I']'J7-19"11, uf •..f110,Jrcr l11.dusl1ial Co111pnny 181
13. tfnQ!)·sis 184
14. S)·ntllrsis IB)
13. Compa.ri.s,1n 18.;
[ ix)
CHARTS ANO DIAGRAMS
FIG. PAGE
16. Projectio1t 186
17. Firiat Sy11lltr.si':i· 157
18. Diagramnintic Ref)reserttation of the 51·Year, 18M·Year) 9 ·Year)
a11d41°;l1011t/1 J'-J1yth11is 18~189
i. Illt1st'oti 197
CHAPTER XIV
i-•.\.f'it-st Cltitiese E/>ocJt (800 ,,ears: 211 u.c.- A.D. 588). 202
1-R. •f)r.cond Chint:Sr. F.-/,ot·/1 (7R'' )'t:tlrS: 58frA.D. 1 J67) • 24>2
' ·C. 1'hird C.:hi11ese Epoch (A.o. 1;68 to prese.11t). 2 ()~
i. J¥ar~1·i111e Distor1io11 205

3. I'et.siste11ce of J.J.-aue /1~{ltu:1ice 106
4. lnte.-rest·flt:tin'11g [Jebt <>/ tlie Unite.:lStotcs 213
C HAPT ER X'/
1. U.S. Pate111> lssued,z8;7-z941
CHAPT.t'..R XVI
i. Profits i11 t/1e Steer Jr1dif.Stry, 1901-191)
APPENDIX I
t. Eq11al R r1te of Grott:tJi - A~·ithmetic Scale •12
2. E<111al natt: of (;'f ()U..'lll -Ra.l io Scale
3. Eqt1al A1,iott1it of (.;rot11tli- Ratio Scale
•4•
243
4. Equal At;no111its of GrottJtli - Arith•11etic Scale 244
5. Trend of a .Y yf1othelical Bu.sin.ess Orga1iization •44
6. The 1·re11d and The Cycle 245
7. Growth of l n1luSlTial l'rodu,ctioti it~ tl~e U11itedStates 246
8. L1td€X of .Spartisli Tr11dc •47
:\PP£KDIX II
1. Effect of ~fol1ing Auerage, uf n;flere1•1Letigtlts1tpotl a1i ideal
9-Ye11r Rhythm
2. Devt'ati01is of a1i ldt·al 9·Y1:ar Rh)•lhrn f rorti 11·1 ovitlg Avtrages
of Different Le11gths
[x]
Cortcernz"ng· Ecor1,omic P1·ediction
AN I NT R ODUC'f l ON
·r 15 1hc business of science co predict.. An exact science like

I astronomy can usually make vccy accurate predictions indeed.
A chemi>t make.s a preci.e prediction e•'<'T}' time he writes a
formula. The nuclear physicist advertised to the world, in the
atomic bomb, how man can deal with en tities so small that they
arc ro111pletely beyond the realm of sense perception. yet n1ake
predicrions astonishing in their accuracy and sign ificance. Eco-
nomics is now reaching a point where it can hope abo to make
rarher accurate predictions, within li1nits which this study will
explain.
Jn these pages we shall be primaril y concerned with a new ap·
proach to economics and tl1e problem of econornic forecast, with
the near·term future of the United Star.cs particularly in mind.
T his approach mo\'CS partly through some avenues tllat in tile
past have been the prov ince of other sciences as vaTious a.• biol-
ogy, psychology, and mathematics.
The study here falls into two ))arts. First, it shows that rhythm
and periodicity exist in Lhe natural world, and that our economic
world, analyzed with similar sta:istic;al tools, also duplays cur·
vilincar form$ and distinct rhythm s. Second. it deals with some
of the ideas which underlie these 13cts, sugi,>"Csts im plications which
seem safely implicit in them, and indicates son1e meanings which
such facts hold for all of us.
[ xi]
AN INTR.ODUCTION
The debt of the authors to those whose names, cqu:it.ions and

graphs line the pages of this hook - and to many others unnamcd-
is without end . Theirs is the pioneering that is moving economics
ouL of the blind alley where it stood for many years, so Lhat it can
take its rank as a true science.
There are those who, admitting that cr.onom ics has not been
an <:xact science, also insist that it cannot be, in the sense of pre-
dicting outcomes in hmnan affairs. There are even some who
f.'lltlsi1ler }:'1Tet1it·tion regarcfing l1t1ma11 life ~ a kir1d of irrtpiety -
or fakery, at best.
That prediction regarding human affairs so often stands in
ill repute with sober tnen (regardless of whel11er it " comes true "
or not) often stems fron1 techniques usecl in formulating it. It
is less the forc(ast than the questionable methodology that often
lacks scientific credibility.
The reader will be introduced to a met.hod of thinking about
l11e future which - new though it may be to him - seems deli-
nitdy t<> have proved of value. It is this method which is of funda-
mental importance- an impori.ance greater llian any specific
conclusions Lo which it may leacl. For on its validi.ty depends the
whole value of llle conclusions.
\Ve shall find, as we go forward, that in this approach to economic
phenomena we are abandoning the dassical approach based on end-
.less argument over cause and effect. It is hoped t.hat the reader's
reward will be the diS<~overy that in economics, as in other sciences,
we are apparently dealing with laws regarding rhythmic human
response to certain stimuli that give a remarkable working tool
to any man who is i·esponsibly concemecl with future outcomes -
whel11er he he businessnwn, community leader, or statesman.
Law in nature, of course, is not the kind of law that is handed
down by an authority. It is merely a summary of our observations
concerning what has consistently happened, and what we may l11ere-
fore expect to continue in a consistent universe. Such law permics us,
in effect, to assay certain probabilities. The ability to calculate
[ xii]
AN INTROOlJCTION
probahilities is a vital pan of all our rnodcrn scien tific progress. As

F.ddington has p<>inced out, in spc-•k.i ng of his own field 0£ physics,
"The c1uant11Jn physicist docs not fill the aLOm with gadgets for
directing iu future behavior, as rl:e classical physicist would have
done; he fills it with gadgets determining tlie odds on its future
behavior. I le studies the art of the l>ookmakcr, not of tJ1e trainer." •
Probabilities have to do with averages. ' '\'hen the physicist pre-
dicts that a given group of atoms will act in a certain way, he is
rel ring-•• physicisu now believe-on a knnwleclge only of what
the average atom will do. Quite similarly, when an insurance com-
pany more or less acc urately predicts the number of dc:u,hs for next
year, it is relying on statistical averages of a like kind.
"fhe discovery that the law of averages applies to humanity-
that cenain activities of people, viewed en nta.'l.~e, fall into definite
patterns, some of whid1 repeat th emselves with periodic rhythm -
promises LO be of great aid in making economics function as a true
science.
These patterns will not tell us what any g iven i11dividual will
do - any more than laws in physits will tell what a panicular atom
will do, or life insurance statistics will re,•c-.tl whether a given indi-
vidual will die. But the patterns do reveal how masses are lik ely
to act at gi ven times. And to that extent they are a fot·mulation
of law.
This discovery has had to wait upon the development of the
science of statistics, the invention of inclcx numb,-rs, and the com-
pilation o f statistical series over long periods. Today t11e work is
only beginning.
Adequate statistks have not been kept for more than a few gen-
erations. ' [ hus we are like doctors following l'asteur, who had in·
controvertible evidence l.hat germs exist, but in only a few instances
had isola1cd tbe genns involved in particular human states.
To assume that from our present limited store of k nowledge we
• A. S. £cl<Hngton, The Nature of the Pl1ysical Wurl<! (p. 301), The Mac-
millan Company.
[xiii J
AN JNTRODllCTI O N
ca n henceforth predict the course of life for humanity would be

childish. But we already have diKO''eries of importance. As the
community becomes beuer infom1ed of their value it will doubtless
help us 10 extend them . l'vfost important, to whatever extent the
limited knowledge we do have can he put to pracrical use, it would
seem that now is the ti me of times to bring it forth.
To be completely exposed to surprise by event.~ - in the com·
plcx age we live in - is a fair route to the insane asylum. Experi·
rncnrs at Yale University, on a pig 1.hey called " T he Broker," have
demonsLraLed tl1at a nervous brea kdown can be instituted even in a
porker, if he is b.;ven shocks of surprise repeated often enough.
A people must plan in order to live. Even an in~talln1cnt purchase
assumes a plan for the future. Presidenti:tl predictions that " pros·
perity "was just around the corner in 1m10 were well intended, hut
badly scrvt.>cf an entire nation. And esteemed government econo-
mists, who predicted unemployment in this country of at least
8,000,000 fnr tl1e spring of 1946, similarly performed a national
disservice. A whole government program was la id out on the basis
of a fnre<:ast which events proved erroneous.
\\lhen a people linds 1.hat predictions of many financial advisers,
srntesmen, historians, and other proclaimed ex pens are seldom bet·
tcr than the predictions of the ;is\,.ologers, o ur social scicn<:es have
demons1.rahly not been earning their way. lt is time for action.
Tl1is study is an attempt to show that something is indeed being
clone. The scientists who arc busy at the problem seldon1 report
their prllb'TCSS in the langu;ige of the average citizen. So their work
often escapes his knowledge. The pioneering scientistS will hardly
he satisfied with this aU~'tnpt to restate, in a simple way, the out·
come of some of their researches. Some average readers, conversely,
may feel equally dissatisfied, on 1.be ground 1.hat 1.hc subject still
seems abstru.~e and the language used is 01.hcrworldly, regardless
of all efforts ro avoid scientific jargon.
To b-Oth sources of justified criticism th e authors apologire in
advance. Every book, like every ho use plan, represents compromise;
[xiv]
and a cornplc:tdy satisfactory compromise is a c.:ontraclict.ion in
'"·or<"L'>.
This book is an au.empt LO show, in an elementary way for the
reader unfamiliar with this form of research, how some of the
inept arguments over economic outlooks ean be avoided by using
a few facts that should now he familiar to all. This applic<ition of a
new method to a study of economic activity, while relatively young,
sec111s ne\.·ertl1elcss 1r1c)re 1>rorr1ising, i11 <>ffering· resl1lts, tha11 tra~
d i tior1:1l CC<>rio111ic Ll)eor.ies tl1at fili textl.10(.,lzs "\•.;itl1 ''l'ir1inns and
arg·ume11ts O\'Cr \\'l1<.:tl1t~r a gi,•cn cause is i·eall}' an effcc.:t, ;.1rlfl ·vice
''ersa.
Here are traced trends evi<lcnr.c<l in various parts of the Ameri-
can economy- existing trends that can be calculatecl, measured,
;incl clc:monstratcd beyond reasonable 1!011ht. Th.: overwhelming
evidence for distin.ct rhythm or periodicity in the cycles that ac-
company theS<~ trends is set forth. How such information may be
used to assay future probabilities is then suggested.
Before rear.bing the end of the book, the reader will have at·
tained, it. is hopecl, some new insight for gauging t11e probable
future that faces America in the ye.ars following the most disrupting
w;ir in all world history.
The student of periodic rhytluns in human affairs has a tool
which tht~ law of averages itsel( puts into his hands. Jf trends have
continued for dcca<ks, or if the osdllations o[ cycles around th<:
trend h;ive repeated themselves so many times ancl so regularly that
the rhythm cannot. reasonably he the result of chance. it is unwise
to ignore the probability that these behaviors will continue.
'rite result is not prc<liction, in the sense in which the word is
ordinarily unclcrstoocl. If th<~ reader nevertheless wishes to regard
essenti11l parts of this book as prediction. then ic should be <~mpha
sized that the " forecasts ·· are written by the data themselves. They
emerge as tendend<:s in the organisms being studied. They do not
rest on the: opinion of any wan, or men. They are, in effect, the
pro/J11bilities of tomorrow.
[xv]
I
Why Trends Are Important
ut> FACI1i of growt.h are common knowledge to most mothers.

T who are ern'.ouraged by doctors to keep a weigh t rhart around
the nursery, for reference al weighing time. All healthy bahics, like
other ltealthy young organisn1s, show large initial rates of growth -
over 1oo per cent for babies the first year. As they get ol<lcr, the
rate of growth gradually falls off. At the approach of maturity, the
rate of i,'Towth finally reaches zero.
\ Vhy it;,, that an organism stops i,rrowin g we do not really know.
\\'c believe that biological organisms-whether dogs or babies or
other anill)als - have theit· growth controlled by inhibiting secre-
tions of glands. \Ve arc not so sure what it is t hat controls the size
of difrcreut kinds of trees, or, say, of a Jimsonweed.
But knowledge that such an inhibiting fact0r docs exist is im-
port.ant to us, e,·en when we cannot explain it. \Ve find it rCdSOll·
able that such a factor should be at work in organisms li ke a bahynr a
tree, just bcc;iuse we are used to obsel'\'ing it in actio n. Rut it also
works in other kinds of organisms, such as human institutions and
business organii;ttions.
Few executives are used to thinking of a business enterprise as
an" organism." But it clocs have a rate of growth that can be shown
by a tl'end line. For Lhe weight figures, which we might l:onsult in
establishing the trend for a baby, we ca n readily take tl1e \Jusincss
output, as revealed in successive sales figures.
( I)
CYCLF.S
Suppose we draw sue.It a trend for a hypothet.ical business organ-

ilat.ion whid1 shows average annual sales of $20,000 during its first
i•ear of business, and proceeds as follows:
1n>iual Sales Fit•e·Year Groulth
1905 Company f(Ju11<ftcl $ 20,(M)()
1910 Five.: years after founding 38.000 $ t8,ooo
1915 ..f'en years after fouucli11g 68,0(IQ 30,00<>
t920 Fifteen )'Ctlr:> after {ot111ding 116,000 48,000
1925 T'\'enty years alter founcli11g 18fi,<1<"1() 70,000
1930 --1 ·,"'e1lt)'··fi\•t: yc;~{rs t•ftcr fr.t~J11c;ling .a79,ooo !)!).000
1935 Thirty years a(ter fou11di11g 391,000 11 .2,000
I !),JO Thir<y·fi\'e }·ears alLt?I' fow1di11g 50H,ooo l 17.000
1945 1:orL)' yeal':S afLer f<n.11u.Jing 6<.HJ,000 101,000
Offh<ind this looks like a business that has been expanding

rapidly, with a satisfactory forward thrust every five years. But if we
analyze the figures, we find them showing signs of what, on the
contrary, is a" dying" business. Years may pass heforc the business
really goes under. nut it has long been approaching maturity.
\Vhat the figures show us is a steady decline in the r<ile of growth.
The declining rate, by five-year periods, is as follows:
RATE OF GROWTH
1go:;-19lo, 9•>%~ t)f a.clual s..1les it1 1905
19l0-l!)J5, 80% " " •• " l f~ IC)
1915-1920, 703 " .. 1915
1920-1925, 6o% .. .. .. .. l!J2f)
1925-193<J, !JO~~ "
.. " 1925
I 93<~193.!J, {Oo/:;• " "
.. 1930
'!}55- 19,..0, .;so% "
1940-1945. 20% .. ..
"
"
It is obvjous from chc table that during each five-year period the
rate of growth of t.his hypothetical bu.1iness has decreased 10 per
cent and that, if these tendencies continue, the rate of i;•rowth in
the future will be:
1945-1950, io% of actual sales in 1945
195<r I 95!;. 0% " " " " t 950
(2]
\\r11y TR l!.NDS ARV. l~it-OR'!ANT
In other words, we see that by 1955 the momentum will cease

enrircl)'. fly then the org:111ization will beco111e another one t~I'
those many whkh follow a groove in a marnrc and conservative
way - p1obahly entering into a m1>derate du line until ci1her ag-
gressi\'C cornpc1i1 ion shoves it asitlc entirel y 01· .. new blood .. comes
iuto the pictu re to give the aging institution a new stan-olf. A drnl't
•howing the sales for ll1is h)'polhetical businw would look like the
solid line in fig. 1.
,,,,.,,---
;
600
TREND
flr.. 1. T• ..~.._D OF A HYwrrttETtCAL Rusna:" Oac.,~17-"-TlOf'I.

Data '!)O!j-19,15, \\till\ a projccti()J'I to 19.'>.?· T11e ru·cJjeccion is liased on tf1e
&su1np1ion c1£ a co1lrir1uation of tlu; c..:otuLnnt clc~:lill e of Lhe rate 4)f gro,,•ch. :1~
cfi~ in the Leit.. (For the same cJa•a pJut1A'1 vcJ '.i.tJo .Kale, .Kc Al'J>Cll.dix I,
fig. 5.)
This line can easily be projected. \Ve have assumed chat sales itt
1950 will be 10 per cent greater than 1915. The table shows that
sales in 19·15 were $609,000. Adding 10 per cent gives $669,900,
projc:ctcd sales for 1950. A"su1ning that 1955 sales will show n o
growtlt o ver 1950 gives a s.'les tottl of $669.900 for this year also.
TI1e projected figures are shown by a broken line. Projections ot
(3)
CYCLES
tl1is son, based on rate of wowtl1 pattern and showing the approach
to what. we may ca ll " maturiry," are imporrnrn tools for all stu-
dents of practical economics.
Jn our hypothetical picture here, the design is that of a very regu-
larly dedi11ing rate of growth - a 1·eg11lariLy which is hardly typical
o( a n)' of t.hc institutions we shall study. But it is useful for illus-
trating a fact t.hat both husines.~men and investors usualli• overlook:
The rate of growtl1 in an organism is a sound index to its vitality.
It has become the c ustom in recent times, for instance. for in-
vesrors to look well on organizations that have paid regular divi-
dends over a long periocl of yea is without a gap. Sud1 a record docs
bespeak sound, conservative ~agement. !Sut it docs not suffice,
taken alone, to show tllat an organization has in it the vitality w
continue successfully if dynamic new competition rises to face it.
In an era when entirely new sources of energy are being discov-
ered, and when oul' ir1dustrfal chemists are creating comrnercially
n ew 1nat<.-ria Is that never even existed before; when we ensla vc new
breeds of li \ ing organisms and put them to work for us in vats; when
any counll y with raw materials c:an m:tkc machines and an)' nation
with machines can make raw materials• - in such a day new fonns
of competition can arise rapidly. Therefore Ille trend line in estah-
lishcd industries becomes a more important economi~ study than
ever befor·c. Just as the figures for any one business may be analyzed
to rc,•cal tbe trend, so may the fii,.,nes for a group of businesses, such
as those that comprise a manufacturing industry.
In tbe United Stares we have a number of great industries -
often called .. basic .. industries - which arc fundamental to the
support of our esta blished patt<.'rn of living. The tren(l' in these
industries are of significance to all of our pCQple, and not merely
to the executives and workers and stockholders in those particular
fields. Such trends clo n.ot have isolated meanings. They indicate
plainly tile state of given organs in the eu>nomic body we call the
• See Virgil Jordan's Manifesto for the Atomic Age (Rucge" Press, 1946),
p. a1 ff.
[41
WHY TRt:l'l>S AR£ T'.\frORTA:<T
nation. If a similar tTend ~hows up simultaneously in a number of

tl1e vital organs, we have conclusions that are in many ways ap·
plicable to our n;irion as a whole.
As we shall sec shortly, the I.rends existing in a number of our
great industries show 1kfini1.ely that we have been reaching- a period
of basic" n1a1.urity " in our whole economic development. ~·his is
a fact of enormous implications Lhat reach in many directions. The
implications arc so great, indeed, that many people (as usual ·with
l1l1111a1Uty) fin d it ea,.,it.r to deny the fact-. i11 l1etll~ nrgt1mc r1t, t.l13Jl
face it hont-stly and tl1en proceed to deal with it. The argument has
reached into the realm of politics- where the only justifiable
debate should be concco1c<i 1101. witl1 1.he rc11lity of the fact, but
with what t.o do about ii.. 1t has entered the life of 1.hc average man;
it has enterccl 1.hc lives of businc" and social institutions; il is Te-
llectcd alike in natio nal <lcfcnse problems ;111cl litera1·y patterns.
Hete the fact will concern us in just one funcbnicntal way: its
effect on business q·dcs. llusiness cycles - we shall later define the
tem1 more exactly - arc not up and down departures fro1n a hori.
zontal line. Tt is more useful to think of cydc• as waves moving
around a curve<l axis, that axis being the t.rcnd . (lf you like to tl1ink
in terms of pictures, throw away the one of eye ks seen as the jagged
peaks of a level picket fence. Vi.~ualize instca<I the peaks of a fence
goi ng up and clown hill, or a coil-spring which is stretched over a
bent poker a.~ !lie core.)
A more or less 1.ypiClll trcnd line, around which a simple cycle is
mndng. lonks approximately like Fig. 2. H this trend line repre-
sented ilie growth of your own businL'SS, and tl1e cycle represented
ups and downs in sales, one thing would be clear: you would not feel
dips in the cycle nearly w seriously when you were growing rapidly
as you would when the trend had leveled out. for when I.he trend
is shooting up rapidly, tl1e hottom of one cycle is a " dc:pression"
only by comparison with its own peak. The bouom may he actually
at a higher level than the peak of activity in the preceding cycle.
This fact applies to every single busincs.s, and it applies to a
[5]
CYCLES
whole economy. Jn the yea,.,; between the Civil 'Var and 'Vol'ld
\Var 1, when l11e American econorny was expanding at a very rapid
race, the bottom of. one clcprcssion oft.en represented a level of
economic activity not much lower than preceding peaks of pros·
perity. Hence many a businessman of that extraordinary era knew
j u.\t. two kinds of " times" - til:ies when business was good, and
aoo~-~--~-~----~-~--~-~--~-~
I
I
I\ I
(
I \
\
I
TREND & CYCLE I
I \
;'I('
.,..--~
r
I " \ I
600 I \ I
I I
1/
,,
I '
I
I
l
/} _j
I
200 - I.
_,
1915 1925 1935 1945 1955
1:1c. 2. TtlE TRF.'.\1> :\'.' 7HT: Cvr.tl!

Tre:oll •>f a f1yp<>tllt:ti«:tJl l>ui.inc.ss <lrg;1nizati<•n as slM>\'in ir1 Fi~. 1 '''ith a
rc:gul:tf 9-Yeti r c:yc;le of ~c"• 1>er c:tr)l t:tnl1)1 ir.ucl<~ SUf>Crilll}H)St:(l. (F«>r the .s:inic drt ta
l'lot•.<.xl o r• r.:iti-0 .sc-:~le, .s<.:c 1\ppen<lix I, Fig. 6.)
it \vas better. nusitlt~.ss, (:\'er\ 111 <lcpressions, '"'as

otl1e1· ti111es \.vl'1e11
guod: although it was under the level of recent peaks. it was still
equal or pos.sibly even superior in volum<~ to lc:vels cstahlished at
other peaks in the fairly recent past.
Figure 3, for in.1t.an1x:, is a di;1rt which shows roughly the ups and
downs in our industrial production hetwecn 1884 and 19.~7, with a
superimposed trend line. The chart ha.~ purposely been split into
[ 6]
WHY TR1'Nl>S ARE l~tPORTANT
two pans. Jn the first part, up to about 19un, the trend was rising
rapidly. Jn the seconcl p:ut, after 1920, the trend has pretty well
leveled our. In the forJ11cr, the eye can c;isily note that tile depres-
s ion lows a, b, an<l 1: wtt'e still practically as high as eat'lier peaks
al A, B, and C. H c.>wcvt'r bad business may have s<:crned at ;uch
lows, it was still at a level that had represented an all-time peak not
-·
too long before. Hence it still ~ccmed preny good.
120 ~~~~~-,.~......~~~~~~~
INDUSTRIAL PRODUCTION
100
60
20
1920 1930 1940
Frc.!· G'RO\'/T11 nr -t~ousnuAL t•ioouCT•OK t~ Tl~ l1si-rw STATF.S

Data-1884-1937, toi;tlber with Tr<nd (alter O>vi>). 'J be chart bas been
•1>lil into °'"-o parts to e111pbasiic the changing charJctu of the trend. (For thC'
'\an1c daLa pJocted on ra1ic' scale; ~c :\p_pendix I. Fig. 7.)
Rut now note what happens following 1920, when the growth
uend begins to level vut. From chat point on, deprc~ion point.~
really mean somctliing for the economy to experience. They are
depression point• n<>L mel'ely by comparison witli recently experi-
t·nced peaks. They are dcprcs-ion; that become al.>svlutc in charac·
ter. There is no lift in the trend line to alleviate them.
T here are factors present in our economy to indicate that hCTe
f7j
CY CLl!.S
aher our major depressions will !:ave this character. These factors
suggest scriou.~ fulu1·c problems for the nalion and for the indi·
viclual businessman. \.Yi thout enjoying henceforth the ad vantages
of a rapid rare of growlh, we may find depressio11 periods harder
to bear. Also chcre is no g round for belief, as we shall note later,
that a cornplct.cly" managed "ttono my i.~ more likely to avoid wide
economic swings than is a free o ne, granting no difference in the
rate of growth.
TI1c busineuman ic going to ha,•e to ancicir:1r1• tlr1~rP$.._~ion~ ancl
plan for them wisCI), or suffer unprecedented penalties. Jn lhe
years when our economy was expanding r.ipidly, a businessman
might make very faulty j udgments and yet be saved from disasler,
or even thrust into s11cc;css, by the sheer momentum of the national
growth. If he failed to sense the coming o[ a depress io n, C)r to time
it accurately, why worry? Even the depression borwm wa.~ higher
than the peak of a few years before. Th is siluation is no longer
true. l t has not been true since 1918. A greater prctnium than ever
before is imp<"cd on the ability :o time the turn> in the business
<.-ycle correctly.
l\.fany of o ur businessmen have looked on the depths that were
readied in 1932- 1 9~3 us a development. c.ompletcly ac:c:idc:nt.al, and
one unlikely to recur. But one glance at the "":o ncl pa rt of Fig. 3
shows why the recurrence of a fall of magnitude seems probable if
and when our economy slides into a depression ag-.ii n. This chart
shows how the trend line in our economy has leveled out, so t hat
no rising trend is present to compensate {or a fall in the business
cycle.
The phenomenon of increasing intensity or depression - as an
econon1y approaches the upp<.'T level, or asymplote, of its trend -
is shown suggestively in th e history of the Spanish Empire, one of
the earliest social organ isms for whi<:h we have any useful statistical
record. I .. J. IIamillon, in his study A rnericar1 Treasure a11.d the
Price Revolution in Spa.in, bas provided us wilh elaboratel y de·
tailed estimates or Spain's total imports, valued in standard pesos,
(8]
\\'HY TRF..NDS ARE l?\f PORTJ\NT
from 1503 to 1660. On the basis of this data, Harolc! T. Davis has
constructed an index of Spanish 1.rade for this period which is
charted in Fig. 4, 1.oge1.her with a trend line which has hecn added
by the authors. The period includes most of the great. growth and
expansion of the Spanish Empire. 0£ particular interest is the
SPANISH TRADE
zoo
ISO
l.00
so
O'----'~-'-~..L~L---'~...l...~-'-~L--1.~...l...~-'---'~-1.~-'
15@ 1560 1580 1600 1620 1640 1660
frc. 4. I~vli:X. OF SPA.NlSH TR.-\t1r.

Data - 153~1650 (after Da\'is an(l H~milton) togetl)er '"itb Trend. (For
the same data pJ,,tt.e d Oft ratio scale. see Appendix I, Fig. 8.)
series of deep depressions that begin to appear after tne trend line
begi1Js to rt~ach 1raatt1rit}r.
111 very recent years we have discovered a great deal of new infor-
mat.ion on this su bj&t, and a new application of scientific methods
in its use. llut before we reach the problem of the cycle we
must know 1nore about trends, and how they are determined and
estinlated.
[gj
I I
Patter1is in G1·owth <!f Organisn?S
CROWTll TRl!.ND amounts to a pattern; and the pattern fa similar

A f(n- almost ;ill organisms, whetlier a group of cells in a pumpkin
or a group of human beini,'S in a nation. The one has a ra11ge of life
measured in months, the other in <:cnturics. But the general pat-
tern of growth is much the same. Even more surprising, a sim ilar
pauem may he discovered in the growth of human institutions like
corporations and industries, as we shal I sec.
lVIuch of our current knowledge of the Jaws of growth has devel-
oped from the pioneering work of Lowell J. Recd and Raymond
Pearl, of Johns Hopkins University. Pearl has briefly described the
outcome of their initial research, begun in 1920, as follows:
As a result of applying certain biological r easoning to the problem,
we hit upon an equation to describe t.he growth of popularions, which
SL1bse<1l1e11t '"'ot·k has clearly cietnonsr.rate<l to be a first. apfJroxitn::.tion
to the required Jaw. As we were in process of pl1blishing the first dis·
cussion <>f tl1c 11'1attcr, \\'E: fot1r1d tl1at a. Belgian 1natl1ematician, P. Jo"',
Verhulst, had as early as 1838 used this same rurve, which he called t.he
••logistic cttr\;e," a~ the expression of the l<.t\V of population gro\\•th. •
• The BioloKJ of Pofmlatian Growth (p. 3-4) , copyright, 1925, Alfred A.
Knopf. Inc.
''\111ile tftel'e a1·e scie111iscs \\•ho 11a"·e shO\\TI a disposition to question some
oC l11e l'earl n1cth0<ls-ort the grot1r1tl, for instance, tl1at 11~ did not cletermine
the probable erro~ of cc>11stants- at1d ,.,.hile alinost limitless \\'Otk in tl1ls field
is still to be done, it remains that Pearl's contribution is of extraordt11ary use-
ful11css.
PAT"rllRNS IN CROWTH OP ORCANISMS
The study which Pearl published in 1925 under the title The
Biology of Population G1'owth is still, some n»enty years later, a
readable, elegantly simple statement of a profound truth which has
since been put to worl:. in many uelds. 'l'he curves Pca.1·! and Reed
explored permit city planners to forecast future city populations
wilhin a small margin of error; enable great utility companies to
21
240
WH ITE RAT
·210
•
~
ill
180
.£
2: 150
"'~
i tzo
m
90
60
80
0 30 60 90 120 ISO 160 210 240 270 300 850 360 390 ~zo
Ai• in DU'S
F10. 1. GkO\~"TH t:io1 l\ODY \•VEICHT OP A 1'-fALK \Vttmt RAT (Affer Pearl)
l:.now ...,-jth fair accuracy where facilities will be in greatest demand

at a given future time. The Biology of Populatio11 Growth is still
required reading for any executive concerned with a scientific ap·
proach to tlie future; a few of its many charts are reproduced here
with the publisher's permission.
As Pearl points ou t, e,•cry living being starts as a single cell; the
cell divides and is multiplied; the process goes on at different r.ilcs
but withour cessation until complete adult development is reached .
Counting the cells is impossible after the very earliest growt11
(II J
CYCLES
stages; but periodically repeated weighingi; give a rough yet suf·

ficiently accur.11e index o f the increase in their number.
" The result~ of such periodic weighings gi,•c rise, when plotted
upon co-ordinate paper, to ;L cun•e of peculiarly charac1.eristic
sh;1pe," P earl showed. It is something like the shape of an elongated
italic f . The cun•e is similar, for insi.ance, for lhe brrow1h in body
weight of a white rat (Fig. 1) ; for the gTowth in weight of a pump-
5"400
qm)
-
4200
PUMPKIN
•~
a
.s
)000
i 24-00
lj:
1800
l'200
600
0 2 • &
FIG. 2. GRO\\>JII r~ 'i\'t:.1c11·1· OP A Po~tPKIN (Afttt-r Pearl)
kin (Fig. 2); and for the growth of a population of yc:t5t cells in a
test tube (Fig. 3) and of fruit Hies, or Drosophila, living in a
bottle (Fig. <t) .
This mud1 seems extremely logical. A population of yea.st cells
or fruit fiics, living in a dosed environment, could be expected to
reach some upp<.T limit 0£ balance between the number of cells and
the living space available. To find a consistent curve in approaching
such a balance does not seem surprising. But when rcarl (urned to
[12]
PATTERNS I N GROWTH 01' ORGANISMS
Asyqlit.OCl•'65.0
YEAST CELLS
150
0 z 4 6 8 10 14 10 1• zo
l>oJl
Fie. , . CIC..O\Y'lli OY A POPULATION' OF \'u~T ()u.1.s (After Pearl)
zso~~~~~~~~~~~~~~~~~~~~~~~
ZZ5
zoo
175 - FRUIT FLIES
a w u u ~ u zo ~ u n n ~ z ' & a w u u u
Aot1l MIY
FIG. 4· GRO\\'TH OF A POPUlATIO:i-1 OY }"t.lllT FLIES

1N A BO'rnz. (After Pearl)
CYCLES
the study of hun1an populations, he found the same law of growth

operating.
This demonstration clocs n ot conform so read ily with 1.hc proc-
esses of our human logic. For lnunan popula.1.i<>ns are subject to
plagues and wars, to emigration and immii-,craiion, to birth-<:ontrol
movements and counter bil'th·control movements. 'fhcy grow
9..--..---.--.-..--r-.--.--..--.---.--.--.-.-...---,.--...---.--.
8 ____ _!.ll!*l"!lof P_Ja!Sl~ 7.:'!.!:...~- --- ----
---------
7
SWEDEN
---- --~--
~6
=
"' 5
.e
.g
:0 4 -
"'
le-
2
.---=------------------------
Lowe< Uait of Pr""'I trdt l.SlS,000
1
o ._.,._.,,,,___.~.._......-.,,___.,.,._._.,...___._*'"_,.,,..,.~-,f:;'_.,=-.,.,_-=='="
1700 25 50 75 1800 2$ SO 75 1900 25 SO 75 2000 2S 50 75 i100
Year
F1c. 5. C¥O\\.TJI OF n1E PoPL'"L1\TJOI\' OP S\\•t.Oi<N (After Pearl)
under conditions t.hat seem far rcrnon:d from the controlled en-
vironment of DrosflphilC1 in a test Lube. And yet, as l'c•rl showed:
5,..,·eden has grol\'n i1t :t rn:1nncr l\'hich, in it3 ql1ar1titative relatiorlS at
least, is es.<enrially like the manner in which a !>Ol'Ulation of yeast cells
grows. . . . Except for the amount of time covert'll l>y rhc obscn'ations,
this curve for the Unit<'<! States is strikingly like rhat of Swc-den.. , .
For France .•. the b'ToWth has e•i<lently followccl, during the epoch
or tyde in which it now is, the same ba•ic law as 1ha1 of Sweden and
the United States. The s:une thing is true of the k11m~n population
gro10.·th of 1\tlSlt'.ia, llcJgi11m, Den1nark, El1gla.11d a11tl \·\'all's, Hungary,
llaly, Scotland, Serbia, Japan, Java, Philippine Island~. the world as a
whole, and Ballimoi·e dty. •
• Ibid., pp. 13- 17.
PATTERNS IN GROWTR OF ORGANTSMS
The similar·ities Pearl n1cntions show up graphically in his charts

for Sweden, the United States, anct France, reproduced in Figs.
5, (),and 7.
Now comes a vital and significantly useful fact iVhe11 a consistent
pattern of growiil exisls, we have sound gtounds for Tnaking fJte·
dicticms. However qualified those predictions may be, they have the
200
175
-c
150
.g 125
:i
UNITED STATES
.E
c
100
~, 75
1
so
25
0
1700
- - 1800 2000 2100
FlG. 6. GRO\\'l'H: O.F TH& POl'UL1\.TION OF Tf·(l!.

U1Sn1!1> STA,1!S (.~/ter Pearl)
probabilities on their side. Thus, when we have enough census

records-as for the United States-to establish a number of
fK)ints on the chart of its population growth, we may observe the
pattern of growth that exists; and knowing what the pattern is, we
may project it into the future for purposes of prediction.
\Ve may similarly project it int.o the past. France, for t'.Jrnmple, is
an old nation; but we lack reliable census data for France before
1800. Data av;iilable since then, however, are sufficient to create the
segment of a curve which may be projected backward as wel l as
forward. The curve for Fn1nce's present cydc of growth had prac-
tically reached its upper limit in prewar days, and suggested decline.
[ 15]
CYCLES
The period of time required by a given nation to go UU'ough the

familiar pattern of growtll varies greatly. "Vith France, the period
seems to have been something over four centuries for its current
cyde. llut records for Algeria convinced Pearl that this one small
country had lived through almost a complete cycle of growth, from
youth to rnat1n-ity, in less than a hundred years - I.hat cycle com-
50,--,-,--.,--,-,--,--,--,--,--,--,--,-...,.....,.....,.....,..-.,.-.,.-.,.-,-,
FRANCE
/
,/
,,
,,,''
10
~--~:-
_...........-· -·
_______________ .. _____ _
L~·er limit ot frcscnt Cyclo 6,604,000
I I
0 1500 1600 1100 1800 2000
1900
Fie. 7. GROV."'Tf-f O.F TJIE P0.1!'0J.A7TOK· OF FltANCF. (.-1/ter Pearl)
prising the period of open French cont.rol following conquest,

which seelns to have been corttpleted around 1 Br;o (see Fig. 8) .
This leads to anothc.r striking observation. Algeria was an old
country, with an indigenous population, before it was brought
under French domination. The introduction of French control :ip-
parently launched a new cycle in the nation's life. Alterations in
th<: environment were sufficiently radical to permit a new (,-ycle of
population growth to superimpose itself on whatever old one had
existed - one for which we have no census records. ' ·\lhen the new
trend hcg-an, it followed exactly the pattern of expansion we are
now familiar with in other countries.
\Ve may thus reach an interesting observation. New cycles of
growth 1nay be superimposed upon old ones. "\Ve cannot predict
[ 16]
when this will happen. >!or can we, as regards humanity, predict
the conditions that permit it to happen. llut a new cycle can appeat·
when some fundamental change occurs.
If we put our colony of Drosophila into a larger bottle, with
greater food supplies, a new r.ydc of h'TOwth would begin, but with
a pattern <1uite similar in shape to the old. Similarly, a nation can
5 0
ALGERIA
0 0
~_:___.::~.-- o Lower Asymp101~ ~2,238,000
- - - o - - - -- - -
1856 1866 I876 1386 1$96 l!l-06 1916 1926

Y~<tr
Fie. 8. GRov.·n1 OF 1'11£ J:-.:01c.i::.-,.;ous NA'fl\'J::. Po1•t•u·110N

OF Atr.•kJA (After Pearl)
apparently effect, through a reorganization of its resources, a ch<ingc

in environmental essentials which accomplishes n1uch clte same
end. It seems to be established, for instance, that the introclu<:tiori
of an inclustrial economy permits a new rate of population growth
substantially in excess of t.hat which wa.~ sustained by c11e com·
munity when it was dominantly agricultural.
The world has striking evidence of this in the case of Russia,
which following 1917- 1918 entered· into a new growc11 cycle. In
terms of chis new cycle, Russia stands coday as the youngest of the
b'Teat nation.< in the worlcl. The United St.ates is comparatively an
old nation, in the sense of c11e dynamic forces that are reflected in a
rate of growth, biologically and industrially.
[ •7]
CYCLES
Pearl was convinced that a similar new cycle was once evidenced
in Germany, where census records go back for enough to permit
him to find a rnodem cycle of growth superimposed on an old one
In the first half of the nineteenth century, Gcmiany's population
was reaching the upper level of growth in a cycle t11at had prevailed
UD
--·. . - ·----- -
Upper Limit of Upper (Pte$llnt) C1cle 116,531..000
100
GERMANY
90
0 ~1~10~0,..-~25,....~s~o,..-~1~s-1~aoo=---=2S=---==so,.--1~s-~19~00=-~25=---==so..---1~s,--2~0~00~
Year
Frc. 9. CRO\\'TII oF i.1;11; Port:LATION oF GERJ1.fANV
This cl1art (after l'carl) makes a1>parcnt l1V"O cycles o( gtoyotJt '"}\i('Jl l1a\'e
overlapped dLu-ing the period of census history.
since t11e 16oo's. For years the nation had been predominantly
agTic:11lr11r;:1l. T•1e al><>rti"\'P. rle.mocra1ir rcvol11tic~n of 1848 '"as c.·oi11-
cident with the beginning of a 11ew industrialization, conduct.ed
under Prussian leadership, and new political forrns. In the twenty
years tl1at followed 1850, this industrialization reached a develop-
ment that permitted the swift defeat o[ France in 1870. The further
unification of Germany that Tlismarck then imposed seems merely
1.0 have added impetus to tlle new growth in strength ah·cady
then established. These events were almost immediately refkctcd
in a rapid increase in t11e population's rate of growth. 'l'he up-
[ 18]
PATTF'R~S IN GRO\YT!J OF ORC:A:-.llS~lS
surge was dearly in evidence l>y 1870, and suhscciuent census

figures demonstrated that a new cycle of growth had begun, as
shown in Fig. 9.
Germany's new pattern of i,rt·owth, following 1850, de1nonstrates
an interesting and perhaps significa11t point, when her population
chart is compared with Lhose for £ngland, France, and the U nited
States: Germany in 1914 was in one sense the youngest of these
four industriali1.cd nations. This may help account for the ex-
traordinary vitality e'•iclenced by the German VC01>le in their rapid
rise from their 1918 defeat- one which hardly disturbed their
social integration. \\'hat it may mean hereafter is wholly problem·
atical, since coday's Germans, living in the world·s greatest mass of
ruins, seem only individual remnants or survivors of a population
that n<l lo11g·er is a t1nltecl r1<ttion.
For our own purposes here, the past pattern of Germany's popu·
lation gi-owth is worth noting only because it leads to a conclusion
which Pearl phr.iscs suocinctly:
" ·'hen human social evolution does manage to put a kink in lhe curve
of population growth, it has not done it by alcering any biological law.
I~ merely shifts, by a greate1· or smaller amounr, the ;ibsolutc base from
wh.id1 the law opera tcs. Theo the process goes on as before. •
It is possible to clraw another conclusion from which Pearl ab·

stains, but wl1ich seems in herent in the data avai lable to us. The
renaissance in the pattern of growth apparently does not occur,
except coucuJTc1nly with revolutionary changes in the environ·
ment. Jn \.'telving che United States, we l1a,·c 11v g1tJunds for
believing that the prevailing trend in our population growth will
undergo any material d 1ange, unless we simultaneously have
grounds for believing that a J.,'Teat revolu tion in American aJlairs is
due- and a revol\ltion. at that, complete enough LO permeate our
political, economic, philosophic, and social insLicuLions.
This is excec<li11gly important to understand, for it has furthe1
• 111ut. P· 20.
CY CT.ES
implications. 'Ve are going to discover that patterns of growth in

human organis1ns - an organism like a group of workers asso-
ciated in an industry - look and act very much like the pallerns
or trends in population growth. And, in further similarity - these
patterns show what we might call an "urge to fulfillment.'' That
is, they are consistent; they do not change capriciously; and there
seems no reason to expect any change unless we also imjJly a revo·
lution in the constitu.tion of the org<misni, or in the environment
where the pattern is l1eing worked out.
lt may well be that our discovery of a new source of energy, such
as has been revcakd to us by nuclear physics, or some of our reVO·
lutionary new advances in chemistry, which in effect give us new
sources o( raw material, do herald a fun<lamental revolution in
our society. For instance, Dr. Hornell Han, Professor of Sociology
at Duke University, who believes many of !lie activities of the
human race can he fitted to the Pea1·l logistic curves, suggests that
harnessing of atomic energy may mean a steep upward swing in
some of our familiar graphs.• Others may fin<l themselves inclined
Lo agree with him. Some hope, si1nilarl)', that the United Nations
is a precursor of a new form of world governtnent that could re-
lease wholly n<~W energies in our nation. But it would he some tirne
before we could be statistically sure of any such change. fVhen a
new trend line is superimposed on an ol,l one, the 1·ise will seem
to immediale observers like a mere rise in the cycle that rides on
the 1.1·e·nd. Some years would have to pass before we could be cer-
tain chac an}' st.tel) rise is not j1Jst c1rclic i11 cl1aracter, but a111our\ts
t.o a rise in the trend line itself - the imposing of a new, rapidly
rising trend line on the old.
This much is worth noting: changes that accompany a new rate
of growth in any given society are so profound, and so all·rcachi11g,
that old established institution• undergo a severt: shakc·ttp. This
shake·up will he in cvi<knce Ion!{ before we get the statistical data
to show what is really happening. It is not inipossible that the eco·
• Americatl ]ounta.l of .Sc1tiolog)•, Jon<-'. 1946.
: 20]
P ..\TTER~S TN GRO'\VTH OF ORGA:--lJS~IS
nomic and social upheavals being experkncecl hy our own genera-

tion are the forcsha<lows, as it were, 0£ revolur.ionary changes that
ultimately will give us an entirely new curve in our population
b'TOWth - to mention just one graph. Simultaneously, many new
industries would be born; some older industries wou ld mke on a
new lease of life; some wou ld doubtless die.
For the present, we shall avoid speculation on mattcrn of tl1is
kind, whith arc still far beyond the realm of statistkal projection.
Our concern here is not with gu~cs concern ing possible new
trends in our economy as t hey might appear in the future, but with
stucly of the ~tablished trends as t l.ey can be statist.kally tra~cd
right now.
III
Tlze Growtli Trerul,

in Our Basic Industries
o lll\.•w the I.rend lines that Pearl uses in such populat ion
T <.:harts as were reproduced in the foregoing pages, we need to
deal with equations, of which che smooth curve is the graph.•
Such a method is too complicated for the average businessman who
wants to explore trend;,. It is not only tliat he may lac:k the
training in 1nathematics. Even more important, he will probably
not recognize the fact t hat he is dealing with a logistic ~urve when
he puL~ figures clown on a chart.
A man who cha rted the g rowth of his i nfant son in pounds, every
week fro m tbc clay the boy w;is horn, would ordinaril)' know that
the rapidly r ising line woukl even tu ally level out. (Ile can get
weight cbaru that will tell him just when the " lx:nd-over " in the
Jine is dl1e.) nut ,\·hen he dlarts his r.orl""lr.tfr- .ul~ hr .'i:elrlorn
realizes - as he sc<'s the line go steeply up - that sometl1ing like
a logiscic curve is beir1g worked ouc here, wo, and that the line will
eventuall y bend over suddenly. Further, when it doe~ eventually
bend, he will usually regard chis as a tempornry • lump that will be
remedied b y a resum ption of progress i n due course.
• T hus the ec1uatil)n ol the logistic curve for fr:tncc ¥-·hich Pearl e1nplo)•ed
. . ..
THE CROWTll TREN D IN OUR RASIC IND USTR I ES
It is possible to avoid the danger. leading from such fallaci()us

jud1,m1cnts by charting figures on semilogarithmic., 0 1· ratio, scale.
That is, instead of using the Pearl t yf)<: of graph, we can use a ratio
scale, then get a curve which, h y it~ changing shape. auto1natically
reveals the falling off in the rate of growth. \Vith such a scale the
clong:u.cd italic .r ol' ]>earl usu<illy becomes a smooth elongated
italic parenthesis ( ·
Let us Lum back for a moment c:o the Pearl chart showing tl1e
P''PL1l~tiu11 b'TOwtl1 of the United St.ates,, on ~gr 15. Now if l Ye
stood at t!)<O o n this chart, and no dotted line indicating the future
were there to warn us, we might well assume that the upwa1·d sweep
of tlic line would coruinue indefinitely. \\le might assume this, at
least, if o ur knowledge were limitecl ro that of the "''erage Ameri-
can, and if our emotions were geared to the prc,•ailing American
sentiment that progress has always existed in thi s nation and
tl1ercforc always must - i( not at rhc same rate as in the p~st, r11en
at a still better one. Onl y specializctl knowledge an cl training could
warn us t11at this chart (as of 19!0) could not juscify such ex-
pectarion.
Fig. 1 on the next page is a c.hart of our population's growtll on
1·a1.io, not arithmetic, scale. Standing at 1 920 on this chart, we could
have looked back and hnve been readily aware that we were al111ost
at the Lop of a Jong slope that was already slowly leveling o ut. The
steady decline in the rate of growth at this point is dearI)' evident,
ancl the approximate point at which growth will peter out entirely
is cle;irly indicated by the nature of rhis simple curve icsel L Taking
our pla~c on this curve at 1920, we cou ld have projected it into
today wir.hout vast margi n o f error, even if we had 111erc:ly worked
freehand wit11 a pend!. This kind of chart, which shows us the
rate of !,'Towth at a quick glance, is preferable for our purposes on
many grounds.
l'carl has taught us something very valuable - that a law of
growth exists, and that it operates consistently. In using hereafter
a form of chart somewhat different from Pearl's, the authors are
[ 23 J
CYCLES
merely charting the operation of t:hat law in a manne.r they believe
t11e average reader will find n10re re;idily nseful ~1nrl applicable to
his immccliatc problems.
It is in this form that the growth of America's basic industries will
he graphed. The reader who ll1en wishes to apply the method to
__ ._,,,..-
POPULATION
100
80
l-"1c. 1. PortJLATION OF Co~l"J~.t:N·r..\.L t r. S., 1R3c>- 19.1<.>

D<~ce 11nial data.:\ L1'erl<l is shf>\'lll, f)l'c>jected t<.:t\t:i.tive1)' t-0 lgfi1). Rl.ilio s':ale.
(For a fttll explanation oi ratio sc.1-1~. sec A1>1>t11clix 1.)
his own business may easily come up wich a cotnparahle cha rt. All
he n(:ed do is to chart his own figures on a similar grid. H <: can then
knO'\\' jr1·11r1cc:ll1Jtcly \vl1crc l1is bttsiness stanc.fs, jn c<>tllJ'aristltl \•;ith
other broader national develorrnenr.~. without: having recourse to
any special mathematical knowledge.
The charts show11 in connect.ion with this chapter are rather
striking in the uniformity of the story tltcy tell. Consider first our
tot.al 1nanufacturcs, which arc shown in Fig. 2 in terms of their
growth since 1830, in dollars compensated for changing purd1a.sing
power." The rate of growth of out· manufactures has been steadily
• All doJlal' cJ1arlS jo Ll)is cl1apter ar1d the one next follo,,ri11g are so com·
pens.aced. That is, the val1rt:s for (:-"At:h yc.:i1· l1a,•e been. clivided b)1 1J1e i11dex o~
the ;rvcrage lV"ftolcsale prices for that y<:~1r, 19~ 6 bciug coosiderc(l as ltnity.
[ 24)
THF. GROWTH TREND IN OUI<. BASIC INDUSTRIES
ded<ning since around 1900. Hy 1920, as can be seen from the

cb<trt, it had become: a fraction of the rate prevailing in t11e late
1 Soo's. In at least one sense we may completely ignore the peak
established during \·Vorld \\tar IL Noc.ice Lhat it reaches far above

the trend line, and surpasses all other peaks attained previously.
200 ,-----,~---,~-r~-.~-,-~--.~-.~-.~-,-~..-~,-~.------.
100
MANUFACTURES ----
50
30
3
2
.5
.3L----'-~:-!:;;,~.l-----'~-'-~-'--7.!:::---'-~-'-~'----'---,~--'
18.SO 1900 1550
FlG... u. s. MA~Uf'A(;llJRF.<, 183<>-1945

Data cJtct:nnial 1830-1899; quinqt1ennial 1899-1919 \\tit.h c,,'$titnates for· v.·ar
yen.rs; biennial l9'~'fl39; ant1l1a1 e.sti1oates tl\ereafter. 1'he d:lta ha,•c 1>ccn
adjusted {or the fUlr(:J1asi11g Jl<)v1:cr c1f tl\e d·:>Jlar, 1926 = 100.
1\ U'eJ\d is sb.ov;n proJCCtc 1~)60. · J·r.e ,,rar )'ears, as tl1e text
CXJ>laix1i,, l1<tvt! het>n igll()r~cl i11 <lt!LCl'fllining trerid in this anti i11 all olht:r cl1;li:ts
porl1a)'il1g growth. l<atio sca/r..
Unfortunat<:ly - for the purposes of our sodety - it 1·cpresents

production for purpo~c~ of dest1'uctio11. Being a wanime phenome-
non, it tells us no1hin!1,' beyond the fact that under the centralized
compulsion of a war economy we had an enormous capacity to
produce.
[ 2,r, l
CYCLT>:S
The underlying trend line tells us ruuch more. vVhat that line
says is this: Under the econom:c system prevailing before the war
-to which we are presumably retun1ing-wc were closely ap·
proaching the uppe1· limit of our abilil)' tO f>rod·uce arid 11istribute
u•ithin the frame of that system.. Following iL~ warcime peak, our
20
10 EXPORTS
~
5
3 ,/
-----
~
Q 2 v
/
ltl
~
0 I
.2
ii
.5
/
.3
.l
IBSO 1900 1950
Fm. 3. U. S. MERCJ!AND!SE llxro•n 1830-1945

D.ata <lecenr1ir1I rKgo-191<i; tt••nu••l r91 o-•!J15· Dat;;i are :t(ijttsted for tJ\e pttr·
chasing pol\·cr of the dollar. 1926 = ioo. :\ t.retld is sho\v11 proje<:t<.:<l tcntativ('ly
to 1960. Ilatioscale.
\:ult1J.Jtt: uf 1tl;·111taf~cttJJt:!i i.) Wttltll tu fall L<lt.:k. tuv1.. arcl tJrc trc.:rt<l
line, and perhaps well below it, hcfore again stahiliT.ing itself
around the trc1HI. Rather than a matter fo1· philosophical argument,
this assmnption seems supported by the evidence of the pattern
that. bas been csr.ablished in our economy since the very begi11nin~s
of our nation - a pattern that can he ;ipproximatcly detennined
and, consequently, approximately pro.jeered.
\\Then we CoTJJpare the chart for nrnnufacturcs with tl1e charts
that show rare of growtl1 in ow· exports and imports (see Figs. 3
[26]
THE CROWTH TREND Ir< OUR BASIC INDUSTRIF.S
and 4). we meet a consisrency that we should now be prepared to

expect. Both export., and imports arc measured here in clollars that
are cou1pensatcd for varying purchasing powel', like the value
of 11i;111ufactu red goods charted previously. I .CL us look Jirst at f'ig. 3,
showing exports. The basic pattern established by the trend is un-
10
5 IMPORTS
,
ez
~
mI
-".s
.
j
.s
-
.2
.1
1850 1900 1960
Fie. 1· IJ. S. MucnANDISf IM.t'o•TS, •83<>-1945

D(lr.t decennial 1fl3~1910; a1111t1al 1910--.19115. na1,a nre adjt1sted for chl· p1,r-
cl\asi11g poY•cr (11 Lhe clollar, 19:.i-O = 100. ,,\ treu<l i.1;slto\lt'n, prujt..:cted tentati\•ely
to 1960. Ratio sctJlt.
deniable. and it tells us something for the future. The export levels
we established during \"\/orld ' \'al' II, by giving away almost un-
limited volumes of goods via lcn(l·lease, mean nothing - except
that we did give these huge volumes of material arvay. In peacetime.
our export volume should tend to return toward lhe true trend
line - and perhaps sink below it for a while - unl ess we continue
giving goods away in much the same fashion. Or unl ess a new trend
is to be established, one we can not now calculnLe.
Devices like the nrcuon \\foods Economic agreement, or Treas-
uri• grants made as political " loans" to foreign nations. may tend
for temporary periods to mask this truism - much •s the working
[ 27]
CYCLE~
of the Federal R eserve Ar.t tends to r.onceal from the general public
the far.t that we print dollars to meet government deficits. But our
cl1art says forthrightly, in statistical language, that on the bll.)is of
the long-establ ished 1.rencl. and in terms of foreign trade handled
at a rea l profi t, we should not be too optimistic iu looking for sud1
trade to be increased over the prcwa1· volume.
In other words, th e pattern here shows what we have seen in
other charts: ' Ve are approaching an upper li1nit of action, within
the Lrame of our economy as it has long existL-d. If we now continue
givi ng our goocls away as in '\1orld \\i:ir 11- either 11s f(Oods or in
the forrn of unr.ollc<:tible loans exchangeable for our goods - the n
of course we rnight presumably maintain any chosen voluinc of
exports. But tltat method of cxponing. if used in peacetime trade
over an)• long per iod of time, will itself be a kind of re,olution.
All our chart r;;1n tell us is tha:. failing a revoluLion of some sort,
the pattern as e,1,ahlished will rrcsnmably prevail.
M ercharidise impons, similarly charred for raLc of growth in com·
pen11<1ted dollars, 1cll us wuch the same story. Ile1c we can see 1ha1
(ba1Ting the period of \\lurid War II) t11e peak was reached in the
decade that followed 1920. There seems no evidence to suggest that
in o ur tirnc, or in th e frame of our economy as we know it, the pl:ak
of the twenties will ever be impo rrn ntly exceeded. Certainly in a
world where economic facilities outside the United States have
been so extensively wrt,-cked, and where the llnitecl States itself has
an unprecede nted volume of production facilities of iu own, we
]l :lr<lly need a cbJ 1 L 1.u ti.:11 lt.S U1t1l levels <.,f ust:ful i111i:>or1s i11 tt1e
future will r.crtninly not exr.ccd, for :111y considerable period, those
already estahlished in the past. Not, that is, within the frame of
familiar economic relationships.
Probably no industry so accurately mirrors progress in our in·
dustri;1 lizcd economy - and, in Lum, is minored in that progress
- as does our iron and steel industry. Iron and steel are truly basic
prmluccs in the machine age that has hcen with us for th e last <~en·
tury and a quaner. They are tools in almost every .. civilized "
[ 28 J
TUE C RO W T tl TRF.N D 1'1 OUR BAS I C DIDUSrRIES
activity. That is why iron and steel produ('tion reflects both the
ph}sica l ancl the psychological drives of a nation like the United
States.
fortunately for statistical purposes, the records for iron and steel
production go back further, and are more nearly exact, than those
I OOlr:~~~.-~,----,,----,~--r~-r~-r~-.-~-.~,--,---,-----,
&O
IRON & STEEL ------
20
Ftc. 5. U. S. IRON A!\U STVJ.L Pa.0DuC1'10N

Sho,,rn in ~rt't>S:S to1u o( 2240 lbs. each. Pig iron production J830--191~ (dc-
cen111"'-l 11S,u-1ts,:,o, ai1nual 1854-1915). Steel lr1g•1l f'roductloo 1914-1945 . .<\
ttcnd is Wo~·~ projected tcntati\•cly co 19'60. Ratio l'tnle.
for most of the other great Americin industri<·s. Jly using pig iron
production before 1913, ancl measuring sted output by steel ingot
production after that date, we i-.an trace the l1 istory of the industry's
activity hack almost. to its beginning.
The steel industry has been an old ind ustry since 1914. Note
the curve showing its l:,'l·owt.h, in Fig. :,. T he relapse it had around
[29]
CYCLES
1914 was nonnal in extent, as co1npared to all previous depres-

sions t hat followed peaks in activity. But Lhe next depression
that came along, following \Vorld \Var I, c.a.rried steel production
to lower levels than were reached in the 1914 relapse. The depres-
sion of the early thirties carried it to still lower levels. Such a series
of sinking spells, each more serious than cJ1c last., woul<l indicate
in a human organism a slowly declining vitality. In the steel indus-
try we may see, over a long period, a slowly declining rate of profit,
and a rapid decline in the rate of growth, until as of 1939 that rate
for the underlying growth trend was probably near zero.
The peak in steel production that occnrred during \Vorld \'Var II
may be ignored, {or purposes of significance in a creative and solvent
economy. That fantastic peak did not represent creative construc-
tion to serve the purpose of man as an economic creature, but
rather represented an explosion like that of aerial bombs which
destroy themselves when they reach their target.
Steel production in whatever peace we enjoy hereafter promises
ultimately to rt:t.urn to so111ewhere around the trend levels already
established in the long history of the steel industry; it may even
sink temporarily well below these levels - once it has supplied
whatever pent-up needs consumers feel as they emerge frocn the war
years, and assuming we avoid bo:.!1 armament races and gifts of free
steel to other nations. Failing the introduction of a new series of
economic or social forces in our national life - forces cJ1at would
ovcrtum its old patterns of economic growcJ1 - an average annual
prod\1ction of ap.,l.l)Xi111a1.cly ,.l o rriil1it)fl g1ot1~ t.v1n; is tlti:'. 111ea11
about which future fluctuations will probably oscillate. (A gross
ton is 2240 pounds; in term.~ of net tons of 2000 pounds the figure
would be 45 mill ions.)
?.fany of our other great industries have afao reached their ma-
turity. Note, in Fig. 6, the chan for the growth of steam railroads
(miles of road operated). The maximum was reached between 1910
and 1914; since then there ha.~ been an actual decline. The railroad
industry as such will have no more real expansion under economic
[30}
THF. CR0\\'1'11 'l R.£NO IN OUR BASIC JNL>USTRl£S
conditions a.s we know tl1em. just as significant a.s the maturity

reacbcd in mileage operated is the fact that the tons of freight
originated have ;ictually shown a decline on tl1 e per capita of popu-
lation basis, even during the years of \'\!orld 'Var II.
..
100
STEAM RAILWAYS
---------
"°° ~-,---,.----.--..--..--...--.---.--....---.----.---.,---.
$00
30
J 10
e :
&s
0
i 3
7
''
l 1
I
I
~ .s
.3
.1
.os
.03
.OIL--'--,8L"°_ _,__ _.__......_..__ 19LOO-...J...--'--..L.--'--19,_50_ _.
F10. G. S'J:t.AN RAtL\\l'AYS

Miles of road opcra1ecl (fu•t track) in oontinental llniled States, 1831>-1940.
Dcc;er111t.al data. A trend iJ s11ow11$ projttted tenlali\·c1y to l ~. Rotio scak.
Much wa.s made of the unprecedented volume of freight traffic

carried by the raili·oads in the recent ·wartime period, and of the
record of freight-miles built up: Such figures latgely reflect long
hauls and the fac.t that the usual unit of measurement is the ton-
mile. But by 1943, when our war elfon was in full swing, total
tonnage originated was still only about 9 per cent more than it had
been in 1929. On the per capita basis it was actually less, because
the population had increased. General Leonard P. Ayres of the
f 31 l
Cle veland Trust <.:om pan)', one of the country's foremost statisti-
cian.,, could only call it " astonishing" that our rai lroads in 1943
actually originated IClill freight per «apita of the popularion than
thcr 1li1l in most of the years Crom 19 11 through 1929, a nd far less
than in tl1c war yea rs o f l!Jti-1 9 18.
100. 000~-~-~-~-~-~-~-~-~-~-~-~-~-~
SHIPBUILDINC';
10.000
100
- •.M/.~ - ---- ----
10.__ _.__ _,__ _,__...__ _,__ _,__....__....__.....__ _._.._....._,.,.,,,..-.....

18$0 1900 19S.O
Fie:. 7. L'. S. SttT1•11v 111l1N<:, 1830-1945

J·'<tr •fl!•1- r939 U\!! 1;ross ronnage i!' ~htt\\'J'I o f 111crcl•ant ve.;s<;I$ launched ( 1uo
l<n1 $ r•r t)vcr). l'rit>r to 1894 tltc; •:h:i.rt s11ov.·:\ tonnage c>f vc.;sscls built, a<.ljusted
to C01\£t>nn to the a\t1~gr tonnage l.tunched lroro 1894- 1907 :11ld set back •)he
)C'... r . .. rom 193y-1915 lhe gTcm tcN1n.1gt is ~ho"'-n of 111cr<'h:an1 \'f:SSel.$ built
( 1ooo tons or O\•c:r) •
'l 'v.() trr.ucls ;1T(' i n 1lt1<;111·rl, the present nno proje<:te~t to 19Co, Ratio sclll.e.
Tlw shipbuildi11g industry, which rcac:ned fanLastic peaks of

co11s tr11cc ion und~r the war impact, " ill dou btless go righ t bacl:
to the level> it cam e from (s<:e Fig. 7) . J::ven con t inned ~uhsidies
to o pcr:itors can h:ird ly save it from such deflat ion; these existed,
tha11ks lO 1.hc l\·t erchant l'vlarine J\ Ct, long before the war began.
Tl11·rc is n0Lhi11g in the long-time growLh pat1e111 of the shippin,i:
i11<lu stry to sug_!leSt that ~onlinued su hsid ies, of whatever probable
[ 32 J
TllF, GllOWTH Tll£ND IN OUR BASIC ll<DIJSTRIES
volume, will change the pattcl"ll to any apprc«iable degree. \Vorld

\Var I raised shipbuilding twcrn~[old - from ahout 200,000 10
11lmut 4 million h'TO~ tons a year. Thereafter the trend prevailing
since arou nd l!)07 , when a decl ine had :>c:t in, w~s resu111ecl. This
peacetime 1.rcnd will probably be resu med once 1r1orc as the axis
around which yearly shipbuilding Yohune will llucLuaLe.
10,000 i:--.----,,--.--,.--..--.---.--..,....-..--....--~-..---,
AUTOMOBILES
1.000
10
1 ~-.._..,.,..,,,..__.
1850
_ _.._..... _~_,,.,.,_...__...__,__..__.._
1900 1950
.....
I'IG. 8. u' s. AU1'0Jtf011JLE l'RO l>l'C'l10~. l ooo )94.4
J•aclory• sales of mrJt(>r vt:l1i(les. \~.ith a tr<.:nd pr<'1je(••~d teoLati,·ely lO igOO.
Ratio scale.
·rhe aucomobilc industry in early 1946 hacl a huge volume of

new urders, but thl'TC "'"" no evidence that these wou Id r.rcatc a new
trend line. The inclusll)'S rate of growth was declining sharply well
before 1940, and was clo>cly approaching maturity, as sh<)wn in
Fig. 8. \ Vhatever tc111porary postwar splurg-c the indusLry may enjoy,
the trends revealed in 1.he chart Slli,'"g"C.sl.ecl ii. had a lready attained ,
prewa r, the bulk of the uiature maximutn of output. A Fo·rtune
cstimaLe of rnicl-19.11. for instance, c<Jncluded that a procluction <Jf
4~ million cars annually from 1nid-1945 on would completely sat·
[33]
CYCLES
urate· t he market by 1950; a production of 6t million cars would

saturate it b)' the end of 1947. Actually, of course, all early fo.-c-
casts of automobile production for the rnonths imn1ediately follow-
ing the war were vitiated by tl1e occurrence of a strike wave; even
1946 product.ion fdl much helow the manufacturers' earlier hopes.
\Vhile failure to produce more cars in 1!J16 came from inability to
manufacture, and not from the lack of a market, subsequent years
soo~~~~~~~~~~~~~~~~~~~~~~~~~~~
soo .........................,_.........................
---"""::.:..
HORSE·DRAWN
VEH ICLES
18SO
FrG. 9. \lALUE OF lJ. S. f·loRSE-ORA\\'N \'E11rcLE PRo.oucuo.N. 1839-1939

Data dece1111ial 183~1899, quinquc·!11 t1ia1 189!r1.919, l>icnni.:tl thcrcaft<,'T.
Dala ai:e adjusl.ed for 1J1c 1>urcJ1asing p( >Wt...'1' of the <lolJar, 1926= 100. A trcn(I is
sl\<>Y•fl, Jlrr·,jec:tec1 ju tlasllcs tcnt:~tiYcly to J960. Ratio ~·ca le.
\·\ re hai.:e here a11 example of ~J1l inclusLry lvl1icl1_, e•;e1\ before rcacbjng ma-
' >'
turity, \\•a:-. .'iupeTSetl(,'tl :t11t)thc..T. '·\1itl10\1t tl1c a<lvc:nt of the dyna1n.ic automcr
bile in<lt.rstry, '\•hidi <lcstroyc<l it, the horse·drawo \'Chicle: industry \'IOuld doubt-
less hal'e reached n1alurit~' at tl1e levels ir1dicated by tJ1e dotLed li11c at the top.
should be watched for an answer to an apposite question: \'\Till e\'en
"deferred demand," so called, enable the automobile industry
hcrc;1fter to gc..:t very• f~ir nl>o"·c its pt'C'\"(aT avcrngc pro<1u<.:tion lcvcl.s?
\Vhal happened to the earlier carriage industry is shown in Fig. g.
Grouped together on pages 36--11 the reader will find charts
showing the trend in a number o( basic activities as vario11s as cattle
on farms, corn production, cotton production, wool production,
wheat production, malt liquor consumption, lumber production,
cotton spindles in operation, c~1l production, copper production,
and lead production. These are shown not because any one of the
series c;harted is particularly significant in itself, but because as a
[ 34]
'!'UE GROWTH TREND IN OUR JlASTC INDUST l<TF-5
group the charts tel I a consistent story, and - most. important -

bccause Jong series of figures ate avail;i hle.
l hey indicate tliat maturity has been or is being reached
throughout our economic fahric as a whole. In che light of such
charts, any talk of the unlimited frontiers that lie before us is a
pure expression of faith and hope. "fhe statistical evidence does
not support it. Like goldfish nibbling at the glass of t11eir bowl,
we ha,,c demonstrably been reaching the circumference of our eco-
nomic wo--rld.a.s-it-is.
l ' his is not a trag;c fact; it is merely a fact. \ Ve may adjust our·
selves to it well or badly. Still more, we may decide as a society that
we refu~c to Jive in a world of such an established circumference,
and try to break whatever I.muds have been holding us the re (if we
can di1<.v•cr what lhcy arc, of what !hey are made, and how to break
tb.en1), and then create a world with far "itkr horizons for our·
selves. Some may suspecLthat this course will be the one that A1neri-
cans will ultiinately try to follow.
T here are some newer indusrrics in tlie nation which still have
:nuch of their growth ahead of them. \ Ve shall survey a few of these
briefly in the following chapter. They should be o[ interest tO
invcsLors who are concerned with the long-term 011Llook, and Lo
young people who arc inLclligently choosing industrial <'arecrs.
There is no reason, however, t.o assume that the activity of any
one of them, or of all of tbe1n put toi,ri'ther, will change the near-
term national outlook sui,-gcsted by the chans discussed in this
chapter.
CYCJ.J::S
CATILE
50
20
,
/
·~
1450 1900
FIG. JC), C:A1·1i.,J!, ON u. s....AR,,1S, JA~lJAR):'

1840--1911
Data clccennlal 1840-1910. annual tllercafttr. :\ trend is sl10\\·n. projcr.red
lt"tltati~ely to 1!)Go. RaJiq sCtJk.
3 CORN
...• 2
-
"m
w
~
..
~
.s
FIG. ll. u . ."i.

Cok..'' 11ROOVC'l'JOI'\", •889-'9'f5
Data <kccnnial 1R,!)-1870, quinquennial 187~18g5. annual ·~•9'1S- A
lrEnd i.11sho1m,1,rojccled L1,:nr.ative1y to 1960. Ratio scale.
[ 36]
TH£ GROWTH TREND IN OUR BAS IC INDUSTRIES
COTION
1•' 1G. 1~. U. S. Co·1-ro="i 1'ROOuc1'!or-.', 1830-1945

Da.1.adecen11ial 1830--1850. a1111ual thereafter. ·1·v.·o distit1ct trends are sho,m.
one prior to and tl1e OLhcr foll<.>,,•ing tJ1e c;ivil \'\'ar. ~1~11e 1aLcer !us bee11 pro-
jecled centaLjt•ely to 1900. Ratio sea.le.
-----
200
WOOL
--
.'!
c
~
0
A.
c
100
,.~
50
/
/
'° 1850 1900 1950
Fw. 13. U. S. '·l'oor. P•O'> VCTJON, 184<>-1943

Daca decenni<1l 1840-1 B6o, a1mual thcrc;t{tcr. T w<• <lisli11ct lt~t)d.:; ai·e
sl1ow11, one prior LO an.d <>ftc; follo\.\·irlg Llte 'l'arilf Act of 1924. The latter has
\1een projcctc;cl tcncativcJy lO J 960. R(Jtio scale..
[ 37]
CYCLES
2DDO
WHEAT
1000
·---- ...-
~ 500
£
•c
~
J
~
100
$D
185D l9DD 1950
FJ(;, 1.j . t:. s. J8!9-l 91!j

\ •\:Im.AT PROflUUl'!ON,
Data decennial 1889- 1870. qtljnqtJtr·1r1iaJ 1870-1895. annual 1B9S-19't5· A
t rend is sJ1o'''n, projected te11<ali,·e ly to 1960. Ratio scale.
MALT LIQUORS
l,DDD
100
Fie. 15. MAt,T L1Qt;OR lN ·r1·rJ! t Jr-:11'nO ST,, ., ,~,;

Consun:11)tion 188!J- ' !J:.tO. ProducLion. i921 - 1944. l)..1 1a tli·1·~·n 11ia l 1 83 9~185 9.
l'lt\r111al 1863- 1944. 1."h e cJTcct o( j)roblbitic)ft is t learl)' irulic~ted. _.\ Lrend is
shown, projecLed LtnlaLivc:ly to 196o. Ratio srale.
THE OROWTll TREND IN OUR BASIC 1:-IDUSTRIES
50
LUMBER
40
~ 30
1!
a:• 20
.."
.~
10
1850 1900 1950
FIG-. 16. u. s. Lt:).fDER }lROPCCT10N, 1830-1945

Data decennial .1839- 18-99 ( 1 8~~J rni~l!ling) , annual 1904- 194.;. Fiist t'll'>
poi111.S are esli1nt1ted tro1u ' 'altte daca. ...\ trcn<.l is slJO'l\'r1, 1>r<1jected tentati,·ely to
1960. J<atio scale.
COTION SPINDLES
20
~
E 10 -
:i;
,,
5
/
/
//
/
2
1850 1900 l9SO
Fie. 17. U. S. Corro:-.i Svr:-:nus If\ <>rERAl'JON. 1839- 1945

Data decennial 1839-1880, at1r1t1al 188tr19.15...\ treud is slio,.,rn, proje<:tc<J
tentatively to i96o. Ratio scale.
[ 39 J
CYCl,£ S
1,000
-----
COAL
100
•e
0
~
t:
~ 10
~
~
~
I
1850 1900 1950
Fi<>. 18. U. S. CoAL PRovucnoN, 184.,...1943

Daca dec:e11nial 1840-1850, quinq11enf'ti;tl 1850-1870, an1)ua1 1.l\c;reaftcr. A
trencl is sl1ov.·n, J>rojecced tentacj\rely to 19fio. Ratic1 seal<:.
1.000
COPPER
-- ------------
100
-- .....
.t ~-~.....,=-~-~-~-~-:-:'=-:-~-~-~-~--::::::--
1850 1900 1950
Fu:;. l!). u. s. s~t£LTER c:oP'PF.R P1tooucr10N. i85e>--tf,·1~

])ala qt1i11qtte1l 11iaf tB:)<~1R70, arJAt1al tl1ereafter. t\ Ill"\\' cycle of i:,.,.o,,•th.
coi11ciding '''iLh 1Jtt:: birtJ1 rJf. th.1,: clcccrj:~tl lndttsuy, began i11 1880. ,-.,,,o trends
ar-e shf>\Vtl, the present trend projecLed tt:tltativeJ)' to 1960. Ratio Jco.le.
[ 40 l
THJ:: GROWTH TR.F.ND IN OU R BASIC 11'DUSTR1£S
1..000, .---.----.---.--~-~--.---....---,--~-~--.---.--,
LEAD ---- -
10
1950
·~
1850
Fr<:. ~o . lJ. S. ltl!.flKt.1> 1,lllf\i:\R\' LEAD PRoot1CT10N, 1R.~<~1943
Data <lece11oial 1830-18:,0, <1uinc111cnr1 i.rJ I 1850-1870, af11111al Lhtrea.Ctel'. Jlere

again t\\'< ) c:ycles of E'il'O\Ylll are i11 cvidt•na:. Tlu:: l><.:gjn11ir1g <>f th~ see<>nd (:ycle
<:ojt)ridi....i '"iLh I.be junccio11 of the Ux•ion P~lcific an<.I (~111.ral Pacific tr".tcks ac
Utah in 1869, \vl1e11 '''eSl fl'tl. Jead ores ,~·ere made availal>le co rnark<::t.. T-.v<>
1.re11ds are sho,o;n, tJ1c 1-.resent trend pl'ojecLed te11tali\•el)' to 1960. R''tioscale.
IV
Trends in Some Other Industries
URING THF. ·nuRTUlS, when panaceas were sought on every hand

D for the then prevailing depression, hopes were expressed time
and again thaL a new industry would spring up to absorb the un-
employment slacl<.. " Somethi ng lil<.e automobiles" was the com-
parison frequently heard. l\o industry capable of fulfilling the
e.xpressed hope appeared. Nor does it seem probable that one will
appear on today's postwar scene with (:nough <lrive to take up
whatever slack in employment may develop in the late forties or
early fifties.
It. is 1.rne 1.ha1. we have a number of relatively new industries. It is
equally true Lhat a number of them hold p1·01nisc of great national
service, and eventual large expansion. But from what we know of
trends, we arc now able to under.tand that industries follow an or·
dcrl)· paltc1n of gro,,·th, jt1st as d..) all otl1cr c.>rg-<lni.'STl'l.,. TflC::)' <to 11-0t
spring full-blown upon the world. Any great industry that can func-
tion to create impressive employinent in tl1c innnedia te postwar
ye<1rs is with us right now. It is already on the .scene, with <in estab·
lished rate of gTOwl.h subject t.oourcxarnination. f urthermore, while
these indusLries offer interesting careers, they are forging ahead
wit11 rneLhods that usually produce more material witl1 less hunwn
effort. than do the processes of older c01npetitors.
\'\lnat about the various chemical ind11stries, for irutance? These
[ 42 J
TRENDS IN SOME OTHER INDUSTRIES
arc a group of in<lusu ies bailed as cl:e prohablc savior of the nation's
future, a new frontier that will indefinitely cxlend the nation's
progress to ever higher standards of living. Figure 1 provides an
answer for one such industry- rayon. R ayon mny he viewed as per-
h<lps typical of the young chemical industries. Ignore its wartime
distortions, and observe the basic trend. It has a gratifying and in-
s.0001.--.,--,---,,...--,---.---.--r--r---r--r-..---. - -
l,000
RAYON
•
1100 ~
10 •
F1<:. 1. U. S. RAYON \'.'\kN CoNSUlfPTtON, i911 - l!J45

:\ trcn<.f issltO\\ln, 1>1·oje<:ted tentatively ~o 19<io. Ratio scale.
teresting coun;c of growth lying ahead of it. ILs owners wi ll doubtless

make money. But itsccrns doubtfu: if they will <'reate any phenome-
nal an1ount of extra employment. 'l'hc fact is t11at practically none
of our chemical industrk'1! m.akes for a volume o( direct employment
as compared to industries ot analOb'()Usdollar sales volumes founded
in an older era. There arc technological reasons involved.
Suppose we consider aircraft. Aircraft was a baby indu;try before
the war; war gave it lite status .of a munitions industry, and air-
planes 1vere turned o ut as Aying platforms for hombs, in nurubers
limited only hy our capacity to bu ilcl and fly them. To appraise the
prospects of the aircraft indusLry in postwar da)'S we need only
glance at the trend line in Fig. 2. Allow for a 100 per cent margin
{43)
CYCLES
of error on the side of underestimate, and we would still have to

c-ond ude that following the war the aviation industry would revert
to very youthful dimensions. T hat is just wh:ot began happening in
late 1945. The two million people it was able LO employ in v.rartime
were reduced Lo a 1.cnth (or less) of tbeir number by Lhc middle of
•!H6. for the purposes of building peacetime planes.
100.000 ~-~~-~-~-~-~-~-~-~-~-~-~~
AIRPLANES
J0,000
Ii 1,000
I
:
100
Frc:-. 2. U. S. AmcR,\FT l)at.OJJCCJJON, 19 14-1911

A trend is sh0,\'11. projected tentatively tO 194:;0. R"tiosca{e.
Charts showing the trend lines for some of our other younger
industries suggest a similarly orderly icrowth- allowing for due
cyclic interruptions of the kind we shall later discuss.
Some industries, like radio 111a11ufacturing and eleCtTical refriger-
ation, are not so young in point of trend as some might suppose.
They were already slowing markedly in their rate of growth before
the war; rcplacemenL~ had become even then an important factor
in maintaining ''olume; and, after a postwar spun, the volume of
sales will probably not be sufhcicnt to brcnk established growt11
p<itterns.
[ 44)
TR£:<1JS lN SOME OTH ER IN DUSTR IES
ELECTRIC ENERGY
j 100 •
!i 10
I
I
Fir.. -'· lJ. S. Eu;cJlUC [ '.'\'F.Rf.,. PRODL'C110N, 1907- 1945

D.tt.a lor l!J4>(. 191.:.0, 1917. 1919; annt1al tJu::n:.Ulcr. A ti·end is shO\\'TI, pro-
jt:<.ttd ttotaci"·cly tc-J ·~· &110 scale.
----
2000•.-~..-~~~~~~~~ .....~......~.....-~.......~~~~~..-~
1000
NATURAL GAS
100
,
/
I
I
!
I
fl(: , 4· \ ' A.Lt:E OF u. s. NATIJRAL GAS PR.OD\.'C:l't(JN, 188~'-l!M-3

naui arc adju:stccl for the J)Urchasi11g poll/er of cite ,h,llar, igsr6 = ioo. A
ltcnd is sl'o'Yn, projected tentatively co 19€>0. Jtotlo scale.
[45]
CYCLES
S,000 ~-~-~---~-~-~-~-~-~-~-~-~-~
1,000 .
-----
PETROLEUM
100
.1
.01
:
.001,___..__..__..__ _,__ _,__ _,__ _,__ _.__ _.__ _.__ _.__ _.__ _.
1850 1900 1950
F1c;. 5. U. S. I•.1::r.KOLf..t:M l•R(>l>t.1c·r10N, 18.59-1945

Dat~t f\.IJ. at~
.!iltu,,·11.
1659, J.8(io, 1$(j5. 1870: i'.u11uctl ll1t:1<:<1(c.c..·1" ' l ·\.,U t1c.:ucJ!)
t.J\e i>rese1lt one 1>r<>j<~Ct1: 19fi<>. Jta:io .ltalt:. \·\:f)al '"e
rcal'y have
Jlt:f't~ is firsl Llu~ illurniu<itirig-<>il incJttStry upcut l"Vl1ic.:J1 \•.:as su1>crirnr><1:>t:cl Ll1t:
ttt(lt(n··ftu::l ln<.l\1stry.
On the other hand, natural expansion still lies ahead in the pro.
duction of ekctr-ic cncrg)'. natural gas, p<:trolenm , ;ind paper and
wood pulp. And industries like dectronics and certain branches of
chemicals are still almost in the:r infancy.
[ 46 ]
TRENDS I N SOM£ OTHE R INIHIST R IES
The chart on page 4!1 is adopted fro1n an idea ori!Pn;illy pro

posed by Roger Babson, lO indicate only suggeslively- and nor
with any degree of measured accuracy - the relative po~itions of a
large number of our industries, in terms of their comparable age,
or present raLe of growth.
PAPER & WOOD PULP
ID
flc. 6. t J. S. PA~na. A.,.,'D \\'ooP P1;r.P PRoour.·noN. 18.sg-1944

Daca c-lcrennial 1859 l!)Ot•. quinquennial l!)f>!r•9 •9· bien1lial 1919-1929.
annual thereafter. Dala for 18s9-•899 eui:nale<I fntm v.aluc Ggures. A trend is
a.bO\'lll. projected ltl"atatJVtly tv 1g6o. Jtutw ~t-ult:.
Tt provides occasion to reiterate that organisms cliff~.,. greatly in

the time they rct1uirc for reaching maturity. There may be as much
variation between two indu;tries, or two <~orporations in the same
industry, in tl1e time ~pan they require for growing from infancy
to maturity, as t11cre is IJctween different kinds of animals and dif·
fcrent populations.
[ 47 ]
CYCLES
'fhe younger industries shown on 1.his cotnposite chart have, in

the rapid raLc of growth that cl1aractcrizes }'Ollt.h, a resiliency that
may allow t11em lo face Lhe downsweep of posLwar cyc:lcs with
great.et· aplomh than son1e of the industries 1.hat have reached ma.
curity- for reasons indicatccl in Chapter I.
2
TOBACCO
I
.2
Fie. 7. U. S. TosAcco l-.Ronuc11or-.:, JS39-191.;

Data decerinial 183~1870, quir1q1it:rtnial 1870--18!)5; annual 1898-1945. A
trco<.I is shrJ\!iD, projectc<J tent;.itivcly to 1960. R<itio s<:ale.
On the other hand, many mature industries, like .~tee] , have man-
aged tc1 l)uild ll J> larg·e firtar1c;i;1l res<.:rvcs to <~11'r)' t.l1roug·f1 l(:arl cla}'S,
011 a scale , ...ihl<.:h 1•ot1ng indt1sLries 11ave 110L managed to eJnt1late.
Building of comparabk reserves has been difficult for today's
ronngel- industries like aircraft, which have had their first taste of
prosperity in a period of heavy wartime taxation.
Now that. we have clcalt with trends, we are ready to inquire into
the crdes that acc.ompany trends.
To distinguish accurately a trend as it is manifested in the life of
the nation's industry, or business, we must first know the cycles.
[48]
1·R t:: i"D~ JN ~O~ff: OTll E R lNDUSTRIES
Suppose, for instance, 1hat a corporJtion is enjoying a period ol

relatively high sales volume, compared to its experience nf live years
previous. Is this vol u111 e the result nf its norrnal rate of growth -
a growc.h proceeding at. a n :lat.ively rapid rate, beca use the corpora·
L00.000,,....- - - - -- - - - - -- - -- - -- - - - . - --...,.
~
~
!:
•••
o-
-
8~
10.000 ••
~Q
oO
1,000 :-
iif.
100
F1G. H. ·1~1t"~NO$ OF V ..\RlOl}~ u. s. lND UiTR.1~$

Diagranlffialic re1>rCS("rlt~11jon or a growth ct1.rve on ~·hieh there ha\·e beeo l<:r
c:itcd '"'itlt 0t11y approximate ac_<;uracy, variOltS Arueric-Jrt iodtutries. acct>nJing
K> tht'.ir J>r(:$(::f'lt rate of growth. Ratic1scal1:..
\\'hc:"n ••• il1Ju:::io y 4 ~ou•1K anti growtng (:;ut., it appa.n here at tl1c lei•. where
the a1rve is ris.ing rapidly. As il gets older and more 1oa1urc;, it b located near
die 1op of the cur"e, or even beyond. (Chart afcer an i<lca by Roser Babson.)
Lion is young? Or is it rather tlic r esult of an upsweep that is

cyclic- a temporary gain that will be heavily compensated for
later? To what extent may the rise be botl1? So far as the rise is
cycl ic, when must we be prepared to face a declinl·? ln so far as it
is a trend, may it be ~xpc:.><:tcd more or !ess to continue?
[49)
CYCL ES
A knowleclgc of prevailing trends provides L" with the first key

r eq uirecl for answering such questions -questions that may be ap-
plied alike to a whvlc economy or to an indiv idual corporation.
\Vhen we Ii.now what the cycle is, we are then in a position to isolate
its e fft:c.t from chat of 1.ltc trend, and study trends as separate en-
tities. Studies of tl1c trend witnou t prior study of the cycle must
therefore he p1ll'cly tentative.
'\'c shall quickly discover that the so-called business or economic
cycle is in reality n composite of many different cycles. \Ve shall he
particularly inwrcstctl in studying thc1n for evidences of rh)·thm.
!:'o r if a nd .v hcn we find rhytlr n we can, to the extent jusiifiable,
predic t recu rrence; a ncl we shall have p red ictab ility o n the basi~
of much 111orc 1.han finite logic and gllc.~~.
It has Jong been assumed that if we could only isolate the cause$
of economic c:yde5 W(' «ould 1.lten effectual I)' prevent them, anrl so
eliminate the downsweeps mat plague high ly organi1ed societies.
;\fanv• ,.oJumes have been written to discuss 1hc causes. Here we
~hall be far less concerned with the cause ihan "ith the timing.
L ike the weatherman, we are conc.erned initially nor with alterin~
t he weat.lte.1·, bu t wi t h the problem of predicting it with fair ac-
curacy, so 1hat those d ue oo be out in it can be prope rly prepared.
v
S(rrrte Rl1)itfttnic Cy cles 1
itz Natural Pftl'1tO'nZeJZa
YCLE, as a word from th<~ Greek meaning circle, implies a

C coming around to lhe place of bc;,..; nuing. It. is used by most peo-
ple in a very general sense, and we sl:all employ it here in that sa me
general way. T wo 01her words which hold more exa~t weaning are
"rhythm" and "periodicity."
C)•cle does not necessarily imply any regularity in time interv;ils.
Thus one b usiness cycle, as measured from the trough of one de·
pression to the trough of the next, coulcl he four years in length,
and Lhe next seven years, "ncl Lhe -next thirteen. And one <'.yt.k o[
growt.h in a hun1an population, as measured from on<~ period of
basic stabil ity to another, might he 100 years or a thousand.
Rhythm, or rhythmic cycle, will onlina1·ily be used to denote a
~ycle which repeats itself at rather u11iform time intervals. Rhyth rn
implies a kind of heat. Thus the beating of the l1 cart is rhyt11mic,
though it mar be slightly irregular.
Periodicity, periodic cycle, or regul11r cycle, will be used to dt:note
a <:ydc which repeats k~elf at n1aLhematically <.:xa<:t intet"vals. True
periodicity, like a true straight line or a uue circle, docs not exist
in na tul"C; but we have dose approximations.
[.:11 J
CYCLES
'fhe idea that cycles in human affairs 111ay be rhythmic or peri·

odic h as stimulated re.~carch for manv. v<:ars.
'
•
Toda)' hundreds of scientists are engaged in some phase of re-
sea rch in tl1e sr.udy of cydcs. To the 1nan in the street some of. their
research mav, see1n spent on trivial suhr'.cos. Cvdes
' in the incidence
of field mice, tent caterpillars, and lynx may appear a far cry from
the question of whether or net there are rhythmic cycles in steel
product.ion an<I <~ornrnoclicy prices. llut just as great advances in
ooedicine, psychology, and other sciences have resulted from the
otudy of lower founs of life, so recent progress in studying rhythms
in human affairs owes much to knowledge ga ined in other fields.
• Tltc list of those lO ,,1 J101n tJ1e n'oc1,:rr1 student of business cycles rnust l>c
i11debted is Jong. T . c:. I .. de Sistn<>ndi ,,•as amo11g the firs.t i>i<u1(•r·~. \\•ith 1-..is
lv~otlutou,x Prir1cJ·pes d'Economic 1-'o!itl'q1.1r. (181~}). CJ(mcr•t Juglar l)t\blisfled
his Des Cf'i.ses C:lt1nmr.r<:i(1./es et de leu:· 1·esot1r pbiodiqur. io 1860. :\. ,,:\,, C:'Bttrnot,
----J.· .
'
pJ iii..,, .· !i·:j ..-$~.'J..._~ -:;J.ID . r;:I;~·:. sur f,:! t1rint·i;1;.,,· fi,1rafh•=m11.'it/ '''· · ~·c la 1 .
·,~ch·e1ses·m 1838. is t.reclilCcl \\oiffi 1nspinng Jc...ons. rcon \\.'atrus frB31-19~
F. Y. Edg.,wonh (18.15-192f). V. Pareto (184~1923), aod A. Marshall (1
··i·' .-.,.~.-
>• 1914). \V. S ..Jevon:;, to;ho ptlblished On tlte Ssu,dy of Period.i<.· c;o7n1nr.rr.ifll '
~- t u alt'ot1.J i11 J86;c, h:i.s bctll (llled lhe father of irldex t1,J1nhcrs. Sir Francis
fi~ tc)rt ( 1R2:r-19 11) int'enled correlatic•t• a11alr.i:is, a.nd Kar] Pcarso11 (1857-1~
6) de..·cloped il as a lOf>l fc>r 11t;.ti!Stic<tl analysts . .J. H. J'oyntir'g i11 1.Rltf a11d R~
~· llo•>ker in 1~> 1 contribt1te:I 10 v.·ork 011 tl1e prol>Jc.·m of the secttla.r t~
lL H. I .. i\foore in 1914 de\'elopt:d f1arn1f>r1ic an;ilysis and correlations i11 l1is b
O· no11iic (.:ycJcs: Thr.ir J."it' and Cattse. \\tarren 1'.1. J•ersc.uls, \,\:}10 m:i.dc the farsl
his b1111i11css barometers jn 1915~ l>e.~g}l n f1is '\'Ork on btlsiness ryclc.".S <tt T·l'J.I\''
ii\ t9 17 . .Josc·plt 1\. S<httmpeter, J•ro(essor or
Et;(,ll( Hfl i l'.$ at H;1r\!ard, creaLe1
a lll(,llUnlC.ril v.·jth his pttblicatior\ in. J939 i)f nu.\·i1it:.\•s Cy(:/es: A Theoretical,
fo'Tical and Statistical Anal)•.Sis of the C41pitali.st P·rocess. 1"he l'.'Ol'k <>£Georg
\·\:a11-e1l and 1:ra11k /\. Pearson of Cornell is '''ell l110i\'J). ffarf>lcf T. D<1\'iS
~
1e
The. An"l)'sis of Econo-tnic Ti,rie .~cries, l'ublishc:cl as f\.fon.ograpl1 No. G
Co1\•lcs Comm.issiort for R est-arcl• in Ec;c>r1omic;, 11:.l.s made a11 itnpol'tant 001~
by j
.•. bution. Irving Fii.her'5 Tlte li!aking of lndex ,,'\'urt1bet-s~ publisltc;c1 ira l 92.?, ~
a landr1\:1rk. N. D. l\.ondratieff, ,,•he> y>ul>li!"hc;cf The Lo11g J.l..at't'S l1t Ec011oi
Life i o t92fi, ha.11 lllt">wn much light Oil the 54-)'ea,. t":y<:fe. \Vt;!!.lcy C. ~·!itc~
wlu• J>U(tli$hccl [fusi1ieS.l' C)•C~c.\·~ Tl1e. rr,th/c,,, nrt(l its S'ctting in 1935. is •'.>Jle
wl1om t>\'Cf'}' sut,i>eq11c.:.ot 1\•orkcr in tl1c field nlust be indebLl·c-1. :\. F. Bt1rns
Jti!:i Prl>tl11ctio1i C..'tin.•es ;,,_ ll1r. [,'nitr.d Stntes sit1ce I870J is required re..din).I. S
a l>rie{ lisljng i:a11 <•nly skirt the frj1}i:,rt: <>f th(~C to \\'l1om the "ttthors of '
book are det:1-.ly obligated. Still oll1c.·r11 :ire mcn1ioncd in footnotes elsev.·he
•. [52 ]
SO'.\!E RHYTHMIC CYCLES IN NATURAi. l'll EN O~I EN A
Kot on!)' is huma11 life complex; it is also evitlc11t.ly subject. to a

whole complex series of rl•)'thms, as we shall see. And Lhese various
rhyt.hms supplement ;ln<l interfere with each other so thaL they arc
often extremely difficult to isohttc and trace.
SALMON
~2 ,,
g,
pl\
I \
~
8 '
1-11 ~--
,. ' ,
\/
'
~~88~0;-->1t.2~x--.-1a~s~o~7.,~~s~7.1s~o•o~~,.~os,.--1~0~10:--~1~•~1s,---~1~•20!,,,......,,~.2~s,.....~1~0Jo
Yeaf
F1c. '· AacNOANCE. OP :\TL:\KT·f<: $A1.\tON
CattJ1 r(~T TO(f per day on ll1e ll<:SLi~•()Uthe River, c::tn:l(f~t. 18So-1929.
sn1u<>the:cl by :l. tl1r<:C·}'C:l.T n10,,·ing a'•er·age (;ifter PJu·~Jps :-tncl R~ 1'1ing), toge tl1cr
\\·jth <t rcgtrl~tr 9%·ycar cycle. (1\1>1>end ix 11 tell~ :1h<.n1t movi11g averages.)
Tn tl1e various lower forms of life, which are evidently subject

to simpler laws, it is usually possible to find rhyth1ns 1nore clearly
revealed Lhan in human affairs. Note, for inst.all<;C, Fig. 1, which
shows 1.ltc "bunda1u:c of salmon on the Restigouche River in east-
ern Canada, from 1880 LO 1929. This study was made by Profrssor
Earl H. Phelps of Columbia Cniversity, in co-operation with
Professor !)avid L. ll<:l<li11g of the Hoston U niversity Sd100J of
1\·lcdidnc. The statistk.s were available because an exclusive fish·
ing club along tl1e river had for years insisted on full records re-
garding the number of ~sh c;aught by its members, and the time
spent in fishing. The cat.ch of lish per rod per day shows five dear
maxima between 1880 and 1930. \Vhcn a 1.rend and a periodicity
( 53 l
CYCLF.S
of 9·~· years are added to the <:hart, it is interesting to note that -

even though at sorne points the rhythms are clisLorted - they later
snap back into the pattern.
A rhytlun of around 9l years has likewise been fonnd in the Lives
of lyn.x. Note the graph in Fig. 2 . !Ylany years ago, when great
fluctuations in the annt1al catch of pelts by the Indians troubled
• 60
LYNX
• ..' ..... ,\
. ,\ ,I f,
·' •'~
~ 40 • A
•
~ " ' ! ' '" ' '' " I"I I'' ''I
'''')
.;20 ''
i
'' '\ ' ' I ' 'I ' I
.. '' 'I I'' ' I' '
J
0 ' ' I
'' '
I I
.:; 10 I ' '' ''' "\I '' I I' 'II ' 'I
' '' .
' '''
'..' ''I'
~ 8 ':
g 6 .. '' ' I f
"" '..
'
''
·l' .
....
t " "v .'
~
~ 4
3
'
0
~
~
0
~
~
- - -
0
!;; fi!
0
~
Fie;. 2 . ABUf'DA:'llCL OF LYNX

-
0
0
~
~
~
0
~
N
- *...
0
~
~
0
~
el
0
~
~
Offerings of skil)S to the J Jud!lon·!i ltay Company, 1844 - 1933. togeLher wilh
a r<'gi.1lar 9~·1' ·year C)'Cle. Jla.l z'.o !1cale. Si111ilar bt:lla.,.i<Jr is <.:vi<lcnccd back LO J735·
(Data tro1:r1 Cle1vt:tl a11d IT<irnilt«>t'1.)
the Hudson's llay Company, an inquisitive worker in the organ-

iwtion started to plot the annual catch. Since tl1e Lrappers would
work at least as l>ard in had years as in times when the catch was
good- and probably harder, for a good catch always meant plenly
of moru:y for firewater - the plotting may be considered a fair
record of the ntunber of animal' at large·:. Ir. apparently v•ri•n
gTeatly. The chart shows fluctuations in offerings from as low as
4,000 skins i11 a poor )'Car tc.> <>vcr 7f>,c><JO i11 a rc:all;• goo<l one. The
plou.ing reveals a distinct 9~-year 1·hythm.
A .,imilar (>-10 yea< rhythm has been found by other investiga-
tors in the abundant<; of gn>usc. Curiously enough, it has also been
found in the abundance of tent (:aterpill ars. Figure 3 charts the
nu111her of letters dispatched in the years 1!)13-•939 to the N-::w
Jersey F-xµeriment Station to complain of the pest. Records of
( 54]
SOME RHYTllMIC CYCLES lJS NATURAL PHEl<OMENA
rhn:c rhythmic cycles arc not enough to permit an aa;ur.itc esti-

mate of the true length; but -so far as th e evidence in Lhis par·
titular chan is conccn1cd- the ltngth is dearly in the general
neighborhood of 9 to 10 years. A survey of ch inch bug incidence
covering 120 years, fro m 1820 to 1940, made by V. E. Shelford and
\·\I. P. Flint, also shows some evidence of a rhythm with an average
lndf"t
TENT CATERPILLARS
40
Fie. 5. Aat.IHOA~CE OF 'Ji:t\T CJ'\ l¥kt'll.LAa.~

' .(:ttCT.S of c-Ontpla.int acrcrrc.'1$e<l lO -r·. J. Headlee, N1:v: Jcr:iey State J:.ntomol·
ogist, 191 3- 1!J3fJ·
Ieng-Li>. of about 9.G years. The last three maxima in Illinois oc-
curred in 1914, 1923, and 19~4. these elates being each one year in
advance for flie maxima in tent caterpillars in ::-.Jew Jersey.
Similar rhythms of a little less than 10 years" average length have
been t!i,;cover ed in the lives of such other ani111als as Canadian
marten, fisher, lllink. muskrat, ancl snowshoe rablJit, <Jtherwise
lno''-"11 as the ' ''ar~ing hare.''
1
Jntcres1.ingly enough, :1 rhyl11mic cycle of 9 to 10 years-per-

haps uj rears - has aIs<> been found hy Profc.o;sor El lswonh Hunt-
ington, of Yale, in the rate of death from heart disease in north-
eastern United States.
Even more curious!)'. Huntington has traced a r hythm of this
same approximate length in the ''ariations in aunospberic ozone.
[55 ]
CYCf . !;S
fhe only records for 01one covering an appreciable length of time

arc 1.hosc n1aintained by tl1e Greenwich Ob,crv;nory in London
from 1877 to 19u). ancl those by the :Vlontsouris Observatory at
Paris for about the same period. (See Fig. 4.) The existence of
a relationship bctw~cn the death rate fron1 bean disease and the
vo lume of ozone in the air is of course not claimed b~ Profrs.sor
Huntington to he proved, and needs further investigatiort, but ilie
150
1880 1884 1888 1892 1896 1000
Fu::. 4. OtoNE AT Lo"OON ( KL'"") AND PAius

1\\"('r.:tged 18;7-1 9'-'7· Sc:tM>nal {yearly) p.-.t(C-m :tnd u:cular (long t.c;nra)
tn:rt<Irliminatcd, 1111<K•11u<il by 1nearu t)f a .st'\t·n·mo11c)\ mo,· i 11~ a\•cra~"t!. :.111(1
a rt-gular 9~·ycar cycJe <tddcd. (:\ftcr Jtu11ti11g1011).
coincider1ce is prO\'OCative. Prcs11n1at>I}' tl1e :t1n<>1.t11 t of OZ<>Ile in

1he air depends largely on ultraviolet light.
I-Iuntingtnn <:oridudes that the 95-year cycle, whatever its cause
or correlations, is liasically a biological cycle. It i' primarily evi-
denced both in ;mimals and in men on the "animal," or health,
kvrl. (5"c Fig. !>·) \\lhdtc-.:r the conditions gh•in~ rise to il, it
seems in hiologica1 U!rn1s to be expresser! basicall)' through ani-
n1al \.'ig:or.
'rhere is ;inother rhytflTn- not to be confused with the rhytJun
of 9S years -- of about !J years' average length, or a little over,
whic:h ;is a physica l rhythm shows up in electrical phenomena in
the form of v;iriation.s iii cl1e electric ch;irge in the terrestrial atmos-
phere. Ellsworth IIuntingLOn states that thi.\ !)·year rhytlun - as
we •hall call it - i> revealed distinctly in the growth of the Sequoias
[56 J
SOME l\HY'l'HMIC CYCl.ES l:-1 NATURAL PHENOMENA
in California, as shown by tree rings. \Ve shall noce later that a

pronounced rhythm of 9 years or so also exists in important eco-
nomic phenomena.
60
Tent Caterpillars
30 '
'\
0
125 Salmon I
I
100 \, '
'
75
Lynx
so
I
25 I
I
0
105
Heart Disease
100 (3-Year M~lns Avet.t8ts)
95
+ 200
+100 Ozone I \
I
I ''
0
\,
-100
-200
1840 1850 1860 1870 1880 18'l0 1900 1910 1!120 1930 1940
f .u;. 5. 'T ttv. 9%-\'£AR RttYTJJ).f IN INSECTS; F1s11. !\'fAllZ..iALS, A1.~N. AND
0zON:N
The 07.<'ule cw"\•e shoi'is L11e amottnc br \\'hich tl1c rJ1.c)11e <>f any given year
fliffr;no from tJ1aL of tl\e fifth precedi11g year., after .5ecular U'etl.ds have been
Pli1ninarcr1 (afler iluT1liogl<1n).
"-'Tany other cycles have been noted at the biological level. One
of approx imately 4 years shows up in the rnunber, migrations, and
epiclernic.s oE Jemmings, lield mice. <1nd the foxes that prey on them
in regions a.~ far separated as Non~ay, Newfoundland, and Kew
York. (See Fi6'S· 6-7.)
Evidence t.hat a rhythm of 41 months exists in the solar constant,
a.~ measured by the Smithsonian lnstitution, has been found by
[57)
CYCLES
200
MICE
~
0
... 150
~
·~
~too
•E
:ii
0
... ...
"• ••c ~ •c ~ •c •c
- a" ,."•
- a"
J
:a • • .'! • Ii •" •
~
• • • .'!
~
•
1933
~
1934
~ 0 :E
1935
~
"'1936 ~
Frc::. 6. AuuN11A."1Ct'.: o.i:· ~fltt

f\:fous(~ population in a .:z:<.racrc field in central Ne"'· York. JV33-•936 (after
Hamjft-0r1}. ( :l•arc shO'\'S sc•~sonal patccrra a11\Jtlcl:1 r1cc of \toJc$; C•Jnsigt1mc11ts of ant.i-·vole matc:ri~ls sent out in
second half o{ each year by the Agrikt1ltttr Bota11ische A11stalt o{ P..turtich, 1905-
191(> (af1t:r t•:Jacnl), t<>gcllttr \Vi tit a regular 3 5/6-y<:ar cycle. Ratio scale.
B. :\btt1lda11ce of !'ox.es: Nun1ber (>I f1)xes brougltL i11 f(1r bo111\ty iri Nor·,\'ay,
18Rtr-r~J~ 1 (after F.ltort). tc.•gcthcr \\rlth a rcguliir .3 5,/6-ycar cycle.-. Note that the
fox cycle t'cachcs its crest a year after the crcsc of the vole cycle. Ratio Jcale.
[ 58]
SO:.l:E RHYTH)llC C YC L ES I:< NATURAT. PHENOMEN A
T. E. Sterne of Harvard lJniversity Observatory. An almost identi-

cal rhythrn has been shown by Ht.:ntingt.on to exist in the daily
varial>ility in atmospheric electricity (see Fig. 8) .
\Ve shall note later rlaal. a l'h)'thm o[ see111i11gly identical length
is evident in an cx1.raordinary number of economic phenomena.
1912 1914 1916 1918 1920 1922 192t 1926 t9ift 1930 l!J32 l9M 1936
1:;-10. 8. Su1....lR Ar-:o TERR.ESTRtAL DA'l'A

Abo,·c. rhc Solar c:or1~IOl\l, He lOl\I, \ l:triallilit>' ,-,f t\lrll()l>Jlheric Elcctrjcit)' al
J.on<l•*'' (Ke\'' and Esktlalcmuir). Regular forty-one mv11Ll1 cycles ha,•e been
;id(!('(). TJ1e e:lect.r'ical tun·c is ba$Ccl or1 pt2'ttnt.ages of a ni11e-ycar ulO\.'ing al'er~
"'8C: "";'h seasc>oaJ tr~nd.s cli1ninat('(l (aJlc:r Jl u11tinglon) .
Evidence of rite existence of other rhythmic cycles in solar phe-

nomena has likewise been accumulated in recent years. Jn addition
to sunspot rhythms or approxim;itdy 11 and 22 years in duration,
investi~ators have isolated or.her cycles of about (.h 10, and 18 years.
·rite solar constant, or 1hc amount of heat from the sun received
on earth, pnrsuc.s a course different from that of sunspots, but also
is suhjc<·t to rhythmic variations. Our record of solar phenomena,
however, is still too shorr to give cs any complete certainty about
the exan length of these rhythms.
Some students have laboriously sought to find correlations be-
t\veen such solar rl1ythms (Incl \\'eather, and :1lso t11Jtt~an econontic
activity. Some interesting evidence of correlatio n has been ad-
duced:• The reader might here refer again to Fig. 8. l3ut correla-
., See, for in$ta11ce, the Quarterly ]1)t1rnal of Econo111irs, Novei:nber. 1934:
"Solar and Econ<.lmir Rel:i1ion!l1i1>s. a J>rc:Jimi11ary R<":J')(U'l '' by Carlos Garcia·
Mata and Felix Shaffner: also Sunspots •M lh<ir Effe<ts (\\lltittlesey House,
1937), by H. T. Ster.son.
(59]
CYCLES
t ion does not prove causality, and data available are still too few
w speak of in Let relationships in terms of compl ete scientific certi-
tude. l'\or docs th e problen1 o[ interrelationship concern us at this
poinL. One do<'s not need to know the cause of' night to know that
morning will follow.
One of the earliest known rhythms whi ch has interested men is
the so-<:alled 13riickner q·cle in weather. Sir Francis Bacon was one
of the first to remark it; he wrote over three centuries ago:
They say it is obscrv<'<I in 1he Low Countries (T know not in what
part) that in eve1 y five and thirty years the same kind and •11it of years
and l\~eathers c:o1nc a~ )ur, again; as great frosts, gre~t \\'Ct, great droug-!1t,
\Var1l1 v:i11tcrs, su11ur1(•r;1i v.•jll'l liLtlc l1t:at1 a 11d tl1e like, a11(l tltey call it
the pri111e; it is a thi11g I clo tl\c ratl1cr m enli0 1l bec:ause, co1npt1tir1g
Uack\i.:ards, I 11ave fot111d S<->tr1e concurrence. •
An Austrian, F'.• llriickner, gave his na1nc to this 35-year cycle in

189 1 b y writing a book about it. sho,.;ng ckarly thaL the weather
of Europe varies in C}'clcs wiLh an a\'crage leng1h of abouL 35 years.
]'\ umcrous othe-r weather investigators have found c\'idcncc of a
rhytlunic crdc of the same length. Recent evi1lcnce has been ac-
curnulated for an(>C,hcr weather rhythm of something more than
22 years in avcr;ige duration.
David Brunt, an l:;nglishman, published in 1927 an exhaustive
study tracing some !M different weather rhythms, with durations
o[ from Jess than two ycnrs up. H. P. Gillette, a.s a result of a study
of droughts, has concluded that one of the basic climatic rhythms
is 747 years in length - a figure agreeing closely wi1h a climatic
rhythm of 7.50 yea1·s isolated by llrunt, and of 7.42 years as figured
by Beveridge. Gilleu e multiplies his estimate of 7.47 by three to
!\'Cta rhythm of 22.4 years, which agrees with a rhythm of that length
isolatccl hy 13runl in European "'eather, and is practically the
dominant cycle of 2•.3 years, :is found by C. N. Anclerson in sun-
spots with alternate cycles reversed.
• J\s quote,."() 1Jy f.11.f\~Orth Huntingtoo it• ;\f ainsprings of CiviliUJlion (John
\Viley 8: Son•, Inc.) , p. 455-
[6o]
SOME RHYTHMIC CYCLES TN NATURAi, rREl'OMENA
H . H. Clayton ha\ found a rhythmic cycle of 7·!'>4 years in baro-

metric pl'e,surcs prevailing all the way Crom latitudes above Go in 0
Sihe1·ia and Iceland to the tl'opics at Calcutta. Gillette has con-

cluded that other cydcs could occur in both an upward and a
downward series where Lhrec is either the n1ultiplier or the divisor
of 7·17· The mulLiple of three applied against 22.4 agrees rather
... . . • ' I
w........,..
j • ..,.
H
-~i--~- HUMAN~~~~~~c...~o.t::i.A..o.~"°"~A.-l
-• EMOTION WorUr ...r"
+•
+~ ~---------~"'7Ll,<'.:..<::..1~~'4,C::::..l,t::::...\..c:~ \ /·T-<
-z -
-4
Woftlc.r .. 0 ..
••+2~-~~~~~~~~t::::c.>J..:.~~::,v-,.<.:.:i-..e:1<::::1..,....Lo..~aj
-~
-4
+4 WOfliecr ..K'•
+~ V::.':::i::'.:\,C::.:::.C:::i.i:::._ _ _ __ _ _-.:i
:1 Afl#. M.,-Ae..Jal ALq...$ep.Qd.ftov
' .Dec.Jn. feb.Mat1 ~.M«tJun,
•••
J.a. A111.S.,.OcU1cn.Dec.J-. Feb.
to
1929 1900 1931

l 'IG. 9· , , AKIATI():-. 1::"1 HUMAN EMono,:i.i
"Ups" a11d ••do,,~," in the ~ m,>ti<>•l$ of f1>ur ruale croploycci of tJ1e Penn·
&yl,,·;trtia Railrc>ad C.:01r1p:tny (nftcr Hcn;(:y). Rr.print~d from ~·Vu-rkcn' Erno·
tio11s i11 !J"hop and Ho1nr., l1y kex.Jord B. Hers.e)'. CoJ'}'right, 1931, by the Uni·
vcr&ity <.•f P~r• u sy l ,:ania J•rcss.
well with a rliyt11m of nil years isolated by Beveridge, and approxi-

m;ttcs a rbyLhm of 6G years which C. E. P. Brooks found evident in
floods of the ::-Jik. But it should he pcinted out that it is t.>asy w
ro11iance over intl"rrl"'l~1itlrl~hip 1hait ougftt to <!xist. and to attach
importance to seeming coincidences that more careful measure-
ments may prove to b(' entirely disparate phenotn<'l1a.
Rhythms have even been found in the emotional li[c o[ human
beings. Professor Rex ll. Hersey of the University o[ l'ennsylvania
found chat every intlivitlual tends to have his own emotional ups
and downs in rhythmic 1:yclcs which may vary in periods, de-
pending on t11c individual, from 2t to 9~ weeks. Further studies
conducted in both Europe and America have confirmed Professor
[61]
CYCLE S
Hersey's original findings, some of which are charted in Fig. 9.

Rhythms have also been found to exist in the incidence of dis·
ease (Figs. 10- 11). They also exist in growth (Figs. 12-13). They
exist in sohir pl1cnomcna (Fig. 14) '""' in 1en estrial magnetism
(Fig. 15). The purpose of a brief >lltnmary of this kind '"rt only
be to show hricfly, for readers unfamiliar with tit<: subject, the
scope of some of the research work being devoted to the suhj~-ct
o1 rl1yrl1111s in f>11r cnviroomenL
PNEUMONIA
Fu:. 1o. O EATtts FRO~t ll:o:,..u":\10~1A
?\.1orlaljl)' per 100,000. '!Jll"-1y28 (afLer ~·tctropo licnn Life J11.s11r.ancc.: Co111.
pa11y). CJi;u·t sl)Oi\'sseosonal pattern a.11<1 three-)'ear tll}'Lhm.
INFLUENZA
••
3
2 Sc1lo of \Veek'
1 •
1 10 20 30 40 so 60 70 80 90 100 110 120 130 140 150
1923 & 1925 192• ~ 1927 1925 & 1928
J.'10. 11. D£ATliS t-"llO~t lNVf l 'P.H7.A

t.f onaliry pc·r 1()(),000, 19t3- 1928 (~frcr i\fc=1n·,[W)lita11 LiCc Insurance Com·
pany). Chan shows SMIJt>nal pouern and three-year rhythm.
(62]
SOP.f £ RJ! Y J Hf\f JC C YCLES I N NATURAL l'li l!NOM.EN.-\
5.6 L8
5.5
PINE
s.• •c
Js.a 1.7 p:
,; !
~Js.2 ,,,.
Eij
• s.J
·go ~Q
.fis.o 1.6~i
"'•.9 REDWOOD
<>
u
.,
I WO '
1500 '
HOO !.JOO 1800 !SW
1.5
Fie. 11. 1'1tr.1: Cao\,,Ttt

Long-period fluctt1:11inn." ln growtl1 of l\ri1.<J11a red\\'ood (Sc<.Luoia) ftot11
'~90 to 19 10, atl<l <•( yellO\I.' Jl inc (Pious-) i11 Calaveras C(•11nty ft'Oill 1620 co
1920. One.h11ndrt<.l·)'(;U r 111t1\•i11g ti.v<.·ragc.s of t1·ee-ri11g ~· id ths ~\rt.: slu1,,•11, ,,.tL11
tJ1e St:quoia curve conrctt'fl fen• U'eil.d. A C)'Clt ,,( 150 ycnrs, pr<.:sumal,ly ir1
v•catflcr, i~ d isclosed. (Afccr Thut111·n1vu.)
0.006
CROCUS
g0.004
0.002
•
- 40
Fie. 13. I• to\VEll GRowtlf

SI1ort·period fll1ctuatio11s ir1 gi·o"·Lh o( che crocus. 1\ crocus grov;!I b)' little
jerk~. racf1 "\'.' it l1 a11 an1pli1udc of aboltt o.c.io:: 1rt1n .. ei.•try IWC! l\LY scco1u.I~ fir so,
each incrcntcut 1Jci11g follo,vcd by a partial recoil. (,'\fter .Bose.· a11d ~1 ·11ompw11 .)
CYCLES
120 SUN SPOTS

80
4'l '
O''--,..,L""'--..,,.,.,_-""-~:--l<..,.!-,---=>.l!,-,-:o;.l:..___::io.<:..__----"..,;_--'-"!r.:...._J ''
184'! 1850 1860 1810 1880 1890 1910 1920 1930
!G. 14. ADUNDA~CF. 011' S1.:r\St>(Yf$

0
Sunspot ntlmbers. J835-1930. (C~an~egic Tr1!5,titution o( '\·\1ashington.) A

regttlar t 1.4·)·car c..;·<:1.e l)a.o; l>ccn a<.f<.fod. Taken over a longer t.ir11e-s1>atl, lf)e
J>Crirxl averages so1ne,,•hat less.
µ
1.8
1.2
''
0.8 '
0.4
0
18-40 1850 1860 18?0 1860 1890 1900 1910 1920 1930.
Frc. 1 &· VARJ,,l'ION 111> TERJU:.•rtRtAL hfAGl"E1·1s"t
Jvtagnctic activit)', t.835-1930. (C;a:'r:i.e~ie lnstitutior1 of '\'ashingto11.) ..:\

rcgtilar 11.4-year cycle l1as been addecl. Note tl)c: close <:<>rrc::;1>-0r1<lcncc 'vith che
curve of su11spot numbers.
The study of rhythm is of rdativdy recent origin; it has had to

wait not only upon the cornpilation of statistics, but- even more
important- tlie development of adequate methods Cor dealing
with the stacistics, and the funds with which to finance the work.
It has been obvious almost from tl1e beginning of research in this
fiel<J that, in tracing Lltc actit)fJ t..)( C.}'l.Jc.~:> i11 vu1 ~vu.1.l<.I, \Vt:: arc tle~tl·
ing t1ot \vitl1 a few si1TipJe rll}'t.hms, but rallter ''titl1 a \ ast ltr1knO\.\rn 1
ntunbcr which interlock and interweave and yield themselves to

discovery only through elaborate analysis.
Figures 16 and 17 from \Varren and Pearson illustrate the com-
plexity of this problem by showing how two simple rhythms may
combine to create a rhythm unlike either of iu componen ts. Fig·
ure 18, also from \\larren and Pearson, shows similarly how several
such rhythms may combine into a synthesis. Later on, in Chapter
[ 64]
SOME RllYTl-l:.HC CYCl. BS IN NATURAL PHENO'.\IE'.'IA
Ttuee~ear cycle
•
~ 0
.s
-12
+12-
~
.s
-12
0 2 4
• 8 12
y,., 14 15 18 20
F•G. 16. SYf\"T'HESIS OF T\..'O CY<..:L.l:':S
Regular ,3•}'ear an(l 4-ycar ty<:lcs, <11ld 1.he two c.ombinccl. The sult). of the t\\'O
repc;\t.s every 12 yea.rs. (After \V:itrcn auC Pearson.)
SiX·)'!3t C)'CIO
~
~
E
-20
+20
••
~ 0
E
-20
...... _ SiX and Elgllt·Ya•r c'.fl;lci combined
+20
•
.lj 0
.s
-20 .
-·10 z.;
0 5 10 15 20 30 •o 45 50
Yt:at
fJv. 17. SvNTII.£s1s OE T\o\•o CYctns
Rcg11J;ir fi-year and 8·}'C<•r c:ycJ(·..s, and I.he two t'1rnblned. T11e sum tJf 1.l1e
t\\'O repeats cvcry 24 years. (1\ftcr \\tarrt:1l ar'd f'earson.)
[ 65]
CYCLES
+20
-20 Three-year Cycle

+20
-20
Five-year Cycle
+20
-20
Eight-year Cycle
+20
-20
Seventoon-year Cycle
+20
Four Combined
-20
0 5 10 15 20 25 30 35 40 45
Year
Ftc. 18. SYN·1·m.;.s1s 01" FocR Cl:'CJ..l!S

Regular 3-. 5·. 8·, and 17-year cycles and tJle fottr combined. The suul of the
(ot1r repeats e\'ery 2040 years. ( ..\.cter \<\'<lrren and Pears.on.)
[66]
SO~tE RllYTfl>-flC C YCLES JN :-.IATUJt.AT, PHENOf\fENA
XI, 1nore ,.;11 he said about synd1csis and analysis in clealing with
rhythmic c~cles. Here it is suJiicient to nor.c in passing that the
cydc~ with which we must deal in economic li[e are usually very
1:omplex inclcccl, being secm.ingly the synthesis of various rhythms
of different lengths - Wgcther, of course, with numerous acci-
dental variations.
The work of economisl.S in analping such complexities is some-
what comparable to that o( the astronomers in tbeir task of deter·
111Uting the rl1}•t.l1m.s in the l1ca,·e11$
with the cliffcrence that the
astronomers have h<:en dealing with much simpler behavior . As
Ilarold T. J)avis has pointed out:
Historically the investigation of time series b~nn with the astron·
nnlers. , .. Tile astr<>r101ucrs, llO\Vever, ,,:ere n1t1cl11)lOrc fOl't\111ate than
tllc ecOtlomisrs i11 011<: ''C.:ry llupOl'lant matter. --J'hc structt1re of their
series as it applied 10 planetary motion was determined by one or two
dominating causes. 1 he motions of the planets w~re influenced mainly
by the excessive mass or th~ sun and .econdarily by the ma.s of Jupiter.
. . . Yet, in spite or d1is unusual dominance of the sun, one mathemati-
cal equation i1l 1he set tl1ar tletenu.i11es the motio11 of 1!1e moon rf'..ar.hes
the incredible leng1h or 170 pages. 'fhe economists may learn patience
from t11<~ astro1101.t1ers, '"°ho ha,'e r1eeclecl tl1r~e centltrics Lo attain tl1e
co11trol ,.,.·hicJ1 1he}'11<>,1o l1 ;lvc over the elerr1c11ts of t11eir t.i1n<! series . .•.
1
lt is w.JJ known 1.ha1. 1.hc problem o[ three uodies - tho< is to say, the
clctcnninatiOtl or 1J1e m<>LiUtlS of three hodics - tJlOvitlg 11nder their OV.'Il
gravitational innuenccs has never been completely solved. Hence, tl1e
general probktn of four, live or more bodies is almost hopelessly dilli-
c..uJ1. llul .."J.1c11 u11c dominoating in.Bucncc c.-xists, such ne the dominance
of the mass or the sun O•cr the ma.ses o( 1he planets, lhen the approxi-
matio11 to a cnmplere solL1tioi1 is rclati\•cly acet1rate. •
As we move forward into examining some major rhythms which

research has revc11Jccl at work in 1he economic life of our society,
we sha ll do well to remember that we are adventurers in a field of
new knowledge. \Ve must be scientifically aware tbat all our con·
• The JJ.nalys.is of Economic Time Serie.s,, pp. 1-2.
[67]
CYCLt:S
clusions are to be qualified by the knowledge that they are tcnta

tive. This is still no valid reason why we should not marshal them.
and use them in daily life if they can he made applicable. Had we
waited to apply the tentative findings oC any oC our n1ajor sciences
unli t tJ1ey had been proved beyond all possible question, we should
doubtless still have the biological staLus of cave men. I ndeed, all
our scienc.es have been developed only through having their
hypo1heses tested in practical appl ication. \.\/hen the hypothesis
worked, it was thereafter regarded as proved. Until it was actually
put to work, no amount of argument was adequate to prove it.
As n avis has poin ted out, much of our prOb'TCSS in isolating
rhythms depends on the use of elaborate sLitistical methods which
even yet are in the process of evolution. Jn the words of Harold
Hotelling: .. If only our t)T.ulnical sun were smaller, the family of
planets would enjoy some of the chaos of democratic societies, and
tJ1e astronomer would he closer to the statistician. &ience would
have arisen later and statistics earlier."
The d oaos that seems to prevail in the society we live in is
doubtless one outcome of our own chaotic perceptions. As we have
learned what to look Cor in nature, we have discovered a remark·
ahl c dci,,..,.ce of existing order. But we have been plc;ised to insist
t hat the activities of man are largely excepted from this rhyth mic
orucr that we note throughout all other man ifestations of the uni-
\'Crs<•. Sud o a c.r cdo is as pleasant for our ego as was the long·
enduring geocentric astronomy of our ancestors. Bu t the reader is
Invited to suspend it - however :cntatively- anti to consider with
open mind some data which may lead to other conclusions.
Jn the pages that immediately follow, we shall consider four
major rhythms that show up prominently in human activity, and
discuss tl1eir impl ications in our economy.
[ 68)
VI
The 54-Year Rhythm
PARENT in many aspects of the economic life of the United

K States, since recorded statistic..s have been available to mark it,
fa a rhythmic cycle of gre<1t importance. Its <iverage length seems to
be almost exactly 54 rears. That makes it longer than the economic
lifetime of most individuals. Thus it readily eso1pcs t ltc not.ice of
the community and of 1.he average businessman whose fonunes.
110 11c tile less, rna)' i11 tnat\}r \\•ay·s reflect lt.
It is a rlt ythm that is sltown statistic.ally, and most clearly, in
wholesale price swings. Interestingly enough, there has been iu
modern times no pronounced long-time trend in commodity prices
- contrary to general belief. The Uniled States, for instance, is
old enough as now organized to have experi(:ncecl three of these
54-year rhythms; during each one, average commodity prices began
at a low level. nwved (with interruptions) up to a peak, and moved
down again (with interruptions) to approximately the level they
ha,J spr 111lg fr( ltll.
Figure 1 shows the average price le\'cl charted for the years 1790-
194,~. with an ideal 54-year rhythm incUc.ated in dotted lines. One
thing is i1nmediatcly dear. Altho1lgh the three of the four great
wartime price peaks shown on 1.his chart fall fairly well in place,
the ideal pattern as it stands could not at any time have foretold,
in advance, either the ti,ning or the amplitude of tltesc war peaks.
\•Vars provide a definite distortion. \·Ve shall come back to this
matter.
CYCLES
t.'Ieanwhilc, look at Fig. 2, which graphs the price average hut

omits the wartime peaks, leaving merely a gap wh~rc they have been
eliminated. ' l'hc acu1al price wave (ii,moring Lhe g<tps) now is
clcal"ly more in harmony with the ideal wave. It is obvious here,
140
WHOLESALE PRICES
l!O
100
FJC.•• u. s. A\"F.R..(;£ WROLIJSALC l'lolCIS
Th.rce-year moving :.'·c-r..xc or an index. 17!)0-191,;~ 1926 ~ 100, togethtr

,.,.itlt a regular 54·yca r C:)'C.IC". Tllc highs of lhe rc:gt11ar C'Yc:Je fall it\ 181j, 1871,
i925. •97!J, ett.., iu IO'"-'S !all in 1790, t841. 1898, 1951, aoo6, t:tc:. Tl1e J"egular
cycle has been cxter)ded i1llO the future t'> indicnte apprvximately \vf1at will
happ~:11 irt this series, i( tltt rltyll11n continues.
however, that some other force or forces are present Lo distort the
actual wave so th:it it is not in perfect conformity with the icleal
pattern. \Ve may cc:irrcctly suspc,;t that other rhythms should be
wughL Vve shall shortly look for them. :\feanwhile, we mwt first
make as sure as we can that the !Yl-year rhythm is real.
In a re1narkably comprehcnsh·e study that filled se,·en volumes,
II History of Agriculture a11d Prices in Engla11d, J. f .. ' J". Rogers
assembled invaluable data on wheat prices over several centuries
-data wh ich, howev1;r approxin1ate, are u~cful enough for aver-
ages and cover the longest range for any price dac;1 we have avail-
able. Because wheat in its average value is closely :1ssociated with
the price of commoditii;,; in general, it provides us with a rather
useful index.
THF, 5 4 - YEAR RHYTHM
The Rogers wheac-price data, pcesented in 10-year aver.t!,<es 1.c

smooth out annual Auctuacioru and converted by N. C. l\furray to
equivalent rc:ncs-per-bushd on the basis of equiv;ikut values of
gold and silver, are graphed in Fig. 3. (Shown as pcl"centages of a
140 -
WHOLESALE PRICES
120
20
0 '--...L..-'---'~'---±---'--±~'---±---'--±~'--,.__.__..~.__-±--_,_-±~'---o!
~ ; ! ~ ; ~ § ; ; ~ ~
Fie. :'.!'. U. S. 1\\'.FJtAGt, \\.•HvL~ PktCD, EXCEM'IJliC \\1.\k Pr-"-KS
l"hree-yea.r mo\ing avn"llgr ctf i1u.lex. 17~1945: omitring period~ co"ering
the \\Tar of 1812, the Ci"il \\'ar, \·\ torJd \\"ar I. an<I \ Vorl(l l\.,.r ti . Cooformiry
to the regular 5-1·yc;:ir cycle iJ 1uoce e-.ide11t \\'ith these omUsiutut..
50-ycar moving average, they ;ire graphed in Fig. 4. Superimpo.~ed

in dotted lines is an ideal 54-year panern.) Severn I fac1.:> are striking.
Note, first, tliat from the early part of t11c thirteenth century until
around 1520, then: was no important rising " trend" in prices.
Then began a lung and almost steady upward sweep lhat was TJ)
la.t tlHough the next rhrcc centuries, as world exploration was
begun ancl brought to ftuilion, and England deYC!OJ>lld her colonial
system and hcc11t1<: a great empire.
The first part of th e rise, from, say, 1520 to 1576, was rather
gr<idual; from 1576 to a crest around 1660, it was meteoric.. Then
followed about So years n( mild clcdine, to a double botto1n
arou11d 1740. 111 t11e next 70 years there 4\\'as :.tgai11 a great 1111!r1ovecp,
which hardly halted in 1790. and the11 c.on1in11c1I i1s rneleoric night
up to an unprecedent~ ~ak coinciding with the Napoleonic ' ·Vars.
[ 71 J
CYCLES
c.....
111r Bll"sllel
31)0,~--.~ .....~~~..--~..---.~~~..--~..---.~.....~...-~..---.~--.
WHEAT PRICES
200
100
ol=~::C:::::::;::~:;;;'.~...,.+.-:----L--:-:1:,,........L~,,.........L----,-,l~L_J.
1300 1400 lSOD 1600 1700 JSOO 1900 2000
Fic~. .3· 680 Y &.\RS or \'.r11EAT PR.1cns IN ENC.:l .A!'\O

1 -.en-year averages ti:Go ~ 19;10 (after Roge1·.s a11cl Oavis), "rhc 5.·i ~year tl1yth1n
c:.u ltl' <li.sccrned. but i~ nbsrurcc.1 l>y the trcn<l. Sec Fig. 4 ior tJ1e sao1e figure$,
smooched, '"illl tJ1c.: t rcrtd c limi11ated.
,,, eeni
WHEAT PRICES II '

120 OEV1A flONS FROM TREND I
I
110
"' ," ,1
, t '
...r.•,
I
I
''
I
100 I
' '' , I I
: '' 1/ \ '
'' '. ' i t ' ' I I
',., \ !,
90
'
•/' i
"' , I '
v v "'-'I '''
I
I I
I
I
80
1250 1300 1400 1500 16Q) 1700 1800 1900 2000
111c. 4.. T11 e 54· Y&A1t RtrYTJn.t I~ '"' Hr.A's· P JUCR.• 1N ENGLAND
'To r<.'1Ilovc
, the trend, LJ')e fl1·ircs shm\'ll i1l 1:ig. Ji have been cx1>rcssed as per·
ccntages o( Lheia· r..•>-)l'ar J11ovir1g average; the Cltrvc h:t ~ been sn)oothe<.l lJy
avc:ragiug t<Jnsccutive pertc·(11.agt-s. A regula1· 54-}'c:ar cycle has hccn added. A
rllythm Lhn.L 1tn...~ re1)(~:111 ·cl i1,;el( :-is many ti111t:i (1 3Y2) o.ntl as regularly as tl1i:1
one ca1t11ot easily be the result ()f 1)11r~ cl1ance.
Tne 54- VF.AR llHYTllM
Then followed almost 100 yearsofdei:linc. In the periorl from 1845

Ll1ere \\1as a f)ricf rc<:c>\'Cl'):; t11en decJir1e <:<>f1ti11Ltc:d [ronl ll1e '"''ar
peak to a hot.tr.mi around 1898-t~JOO. This bou.om, reached almost
a century a[Ler the peak, w;is lower tl1a11 any previous kvd since
the lace 15oo's. The t'C:covcry that set in thcrcaft<~r with 1900, and
continued up through ' ·Vorld \Var I, was almost canceled out by the
193o's.
Approximate as tl1e figures on which these averages are based
lllU~t lu:, tllC:)' ~liJl r1.1tr1i::.li Lt:) V\• iLfl C:vidCll.C(' of a !)0-6v~yc,;ar \•ia'\'C ill
price moH:ments extending well back into history. It becomes i111-
mecliately apparent ii' '"' correct for die trend dcscrihcrl ahrwc, and
then chart the remainder as in Fig. 4. The rhythm, when studied
more carefully. proves to have an average length o[ very dose to
54 years.
One of the difficult pieces of work. in the history of pel'iodogram
analysis was accc.unplislted by Sir 'Villiam H. Bev<:rirlgc, t<.> de-
tcrrltit1e ,,:Jletl1er c>r rt(>t t~1cre \.V<~rc pcr1na11e11L C}'Cles itl l1i:J1e;.1:t
prices in Eastern and \Vcstcrn F.nrope. In view of the fact that Eng-
latl(l \1.:;1s fcJr c·c11tt1ries a free-trade oot1ntr}r, its \\•l1c;tt y)riccs C.<)Ulcl
he <:xpected to run in consonance with those of the Continent. Sir
'INillian1 usec.I inc.lex 11u1nhers so constructed that the upward leap
i111) riccs cl111·ing t11c t.ransiLion from mcdie\:;1) to r11ocl<:r11 t.iJ11es \Vas
rc:11lc>vecl, st1bjecting cflem to },a1·111c,11ic a1ial}'Sis <.)\!er a range o[
approximate.ly 300 years. Ile found several true p<~riods that gi1vc
stro11g· cvidc.:n<:e elf coxreJacion '"'ith kno,vn '"'<:att1cr t·l1ytl1u1s ancl
st>nspot ph1·r•rlt11t-r'l::t . An11t.
such con'C:lation) he also found a rchnivdy significant concentra-
tic>11 c>f enet~gy.
This 54·year wave as we h;1vc experienced it in tlte Uniterl Stat.<~s
registers itself in commodity price charts ralhcr dearly. The hl'<t
wave, from low to low, dales itst:lf •7!j0- 18.11.; ~!~ :;.". .cr.<)nd runs 1f\4.1-
1Ar18:. a11cl thC' rfli1·cl, i11 t1'·l1i<~J1 tll~ V11ite ~,·c>rfcl '"'ars. f~tll.s bet\...-eeti a le.>\,. i11 18~)8 ;111<'!, if rlic rliytl 1t11
COllliTlllCS, (.1JI C' al'{)llll<l ltt.=)2.
[ 73)
CYCLES
Because of the incidence of three great wars in American history,

each of which occurred around Lhe peak of the ideal 54-year wave
and coincided roughly with a peak in commodity prices, some ob-
servers have callc<l this 5•l·yeat· rhythm a recurrent " war cycle."
Harold T. Davis has gone so far as to Sllh'!(r.st that. the average in·
ter\'al of about 50 years between c:onfficu in the Punic \Vars of
Rome and Carthage might lend force to such a theory, and points
out that those wars undoubtedly created in ancient times a degree
of inflation in commodity pric~'$ such as has aucnded our 1nodetn
war periods.
However in teresting that thought may be as historica l spec.uJa.
tion, of more importance to us is a fact which can be ascertain ed
on 0<11' pauerns by simple observation: in the Uniter! Stat.c:s price<
have always 1·isen through the upward sweep of the r,4-ycar wa"c to
a major pL»1k, and I.lie)' hav<; always falkn during its downward
turn to a major bouom. Further, as we may observe in the Euro·
pean price panerns, the E uropean price..' always tended to make
a recovery around the peak o[ t'le 54-ycar wave, 1:ven if Ll1ey were
steadily falling in a long-tcn11 price trend; ;ind practically always
they have tended to halt their rise, even during a steadily l'ising long·
term price trend, when the ~1·ycar wave reached a Lrough.
Since 1925 the cour~c of the 54-year rhythm in th e United States
has been generally downward, reaching toward a trough due, as
said above, in 1952 (see Fig. 1) . It is sii,rnificant 1.ha1- despite the
recurrence of war in the 194o's - a\•erngc n·rail and wholesale
price levels in this rountry up to m1d· 1941i dad not reach anywhere
near the peak of I!)'!) 1920, even t hough foo :u a group did
l'i;c to 1920 levels, as shown in indices lik e rh l' Dun and Bradstreet
index of thiny-onc fooduuffs.
\'le shall learn, when we study the lesser g·year rhythm which
also affects prices, Lhat another rhythm ic Fon.<· was present after
1941 to carry prices upwarcl to moder;.,1.e degrees. Tlut as a lc~~er
1·bylhm, it has not sctvL'Ci LO give price.< Lhc enormous rhythmic
support they bad during the period surrounding \Vorld \\far I,
[74]
ru.e 54-YF.llR l!HY1RM
when both the 9-year and 51-year rhythms were u1oving upward in
concert.
Further interpretation of our prke dat.~ will corne more conven-
iently in tl1c next section, when the reader meets the !)·yca1· cycle
in diagram and can sec it combined with the 54-year rhy1.hm.
For the present, having ;cen romc evidence for the statistical
l'eality of this 5.1-year wave, we shall find it usdul to consid1:r some
challenging ideas regarding- iL' imporo:;n1cc in our organized econ·
omy. lt was I'\. D. Kon1halieff who first succt·cdcd, through his rc-
sear<J1c:s, irl l1npressl r1g man}' (:t·<)11omiscs '"'it.It C\•idenc·c; tllal a long
wave of npprvximatcl y 1.his durntion charat:tcrized the tlcvclop-
rnent of our capitalist economy. Joseph A. Srhumpeter, Professor
of Economics at Han·ard University, approaching the problettl
fr<>rn ''ario11s filstc,rica l angles, h".s ncJt onl)' <.:ot1fir111ed th e r11ajor
conclusions or Kondra1 ieff, but has gone fa1· to proviclc the basis
for a rlC\\.' ttn(ierst;i11ding of che \\'ay our ccoJlOmic orga11ism '''orks.
It shoul<l uc empha.i1.t:<l . in justice to Profcs."ir Schumpeter, that
he docs not accept 1hc ueed for e'Xact timing or eX<rCI. e~tinmtes ol
dur.ition for rlois rhythm , wloid1 he regards as the outcome of ne.v
efforts in tbe society to»ai·d p•<Jb'TCM. For doc purpose o[ bis rc-
se;ord1es, '"l1id1 he reporter! in the two volumes of his 1Ju,ti11es!
Cycles, he d:or.cs 1.he lirst long cycle in our inod~-rn America n histm;
- measured from low LO low - from the eighties of the eighteenth
n!ntury to 18.j2 - noting that "we have seen reasons to believe
d.at this lont; wave was not tl1e first of its kind ." The follo» ing long
w:n•c as he saw iL ran h<:tween 1842 and 1897. And the thircl is our
currc11t one, reaching from 1898 en. Perhapi. to avmd implying
that he consicle1ed the darings completely exact, l' rofessor St:humpt'-
1.cr has called these periods" Kondrn1.ielf" cycle;, rather than cycles
of any spcci!ic length.
In his admirable study, which survered almost two centuries of
econornic history in England. Germany. and America, he ha> shown
condusivcly that every Kondraticff cycle has been uniC'Ju ely ~h:ir
acterizcd by rome particularly important economic or industrial
[ 75 l
CYCLE!;
inno•ation. T he first Kvn<lrJtidJ <:)de of the last century was the

wa•c in which we moved Corward t11rough the process that ha> been
calkd rlw iulustrial revolution, and the ahsorption of that revolu-
tion ·s e tkns. ·1 he next Kondratidf "'as the age of steam an<l steel -
a11cl o l rail rt acl deve lopment. The K.ondratidf tha is now drawing
0
to a close ha.; been the ag-c of chemi,u y, clcc-1.ric ity, and rnotors.
WHOLESALE PRICES
1,40 1)00 1600 l810 1820 1830 1840 1850 1860 1870 1880 1890 )900 1910
Ttr.. 5· Lf)NI; \'\t..\\.t:.S 1:0.' \ •\ :fJRl.I) \\' 1-10 1.1<..SALF. l 'RTC'l!.C:

Data for Gern1any. (;real l~r·i1.ai11, a11cl the Uriited St.'\tf.~. Xin<:-ycar mo\•ing
%11VtrnAe.'i ol t\>hc·•f(:S.:.Je price jndiCC'.S nl'e 1,IOllfCL on rtJlio scale. (.-\(ter
'11.f 1UlllJ)(•l1:r.)
Figure 5. horrowcd from Schumpcrcrs study, answers the read·

c·r·s n:tlt1ral =i•1estion as L•> l1(J\\.. f-Jr ~11c;h a 1Y''1\'C 111a) l)t• a11 <:xcllisi,•ely
1\me1 it...""";tJl 1,!1e11,,menon. Here \\'~sec tr-.:trC<l. in t'le 9·>·ear 1110\·i11g
a•·erageo[ uie price indkcs for the three countric<>, ;t (;orrespondence
in "''"T rno•·ement t11at is vhihh• wirhvut analysi1. As Sd111mpeter
point> out:
Neu only do cycles in dillcrenc counrrirs systematically offect each
Nher, <o thot chc history of horrll)' all)' one of tht:m c.in be written
(7G J
THE 54-YEAR RHYTHM
'vitllout rcfcrt11ce to si111t1lt~111co11s cyc:lic:al pl1asc·s i11 otl1cr c<>u11tries;

but cycles rl:ally ar<!, CSJ>tcially as regards the greac i1111ovac.io11s tl\at
produce the Kondraiielfs, international phe11omena. That is to say,
suc:l1 a pro<:Css a.s tl1c r:1ilroatlizatic>r• or tl1e clcctrific:~tti<>rt <>f the ,,.•orld
transccn<ls rl1e hc>11racJ;Jries of ira<li\•idu~I c:c>1111tries i11 suc:l1 a \'ia}' as to be
more 1n1ly desCJ'ibed as one world-wide proce$S chan as the sum of
disti11c1 11atio11al 011cs;t:
WHOLESALE EXPORTS
100 PRICES /F 100
,.1"'-..,_\!:K.
~.s.
, - - -.:::-' ..../ . ...........G
50
·-·-........
··"
/ ,/
I
50
F1c:;. 6. \\.'or.r.r •.> lrf'J'v.RJU;t.A'OO~Stt1rs: PR..1('£S ANO Ext•oR'ts

:\. Jndices of l\•holcsalc prices, 19~9 = 100; B. l11diccs o( total exports,
19.:.-9 = tc)(->. ( •.\Cter Davis.)
Davis makes the same point about the unity that prevails in our
econon1ic atrr1(>spl1erc :
, ,\..,.as it acr:idcnca.J, for example, that. the 11nrelatcd inflatiorls of t.he
South Sea Ruhble in F.ngland and the f>fississippi Bubhle in France
should have developed ronr.urren<ly and should have collapsed almost
\Vitl1i11 tlle sa111e 1nc>nch ? [172'Jl Anir1<:itlenc:e tl1:ic
the collapse of the American securities market. in Or.toher 1929 should
have been accomplished without appreciable lag in rhe decline of
prices and the commerce of all the other nations in the world? t
Davis also shows the two charts which al'e reproduced in Fig. 6
to illustrate price levels for Germany, frJ.nce, the United KinC!·
• R11sim:.<1 Cycles (J>. G6G). McGraw-H;IJ llook Co., Inc., 1939.
t 1·11e Atifll)'sis of Ecoti<>1nic T1'1ne Serie.s, p. 558.
[ 77]
C\'CLJ>.S
dom, the United States, and J ap an in the decade from 1930 w 1910,
togeLher with exports for four ol the live nations; and he says:
1'l1c asto11isl1i11g llti11g to be o l)ser\'ecl in t.llese figttre!> is tl1at v.•hole·
sale pri<:cs a11<l total ex1)or1s for ail co\1111ries fo1 J o,~· rl1e san1e 1)attern.
Could this have been occasioned by Lht collapse of the American stock
market. or is ic e''idence or son1eth ing ,, l1ic:l1 '''as tJ1e cause, a.u1oog 0Ll1er
1
1.hings, of the American deflation iL<df? •
\Ve 1ni3,l1t ask, e\•er1 rr1or-t.-: poit1tc<ll}'• v...J-1}' tl1c r~co\•erie~ sl1o''"'Tl
after the clcclinc sh ould show such para llels in a group of nations
as di\'erse as these c.:t>1Jrl1·r ies i 11 tile eco11011lic t11.ecl1a11isn1s the,, . h<~(l
prc\:a ili11g· after 1933 . In Lllat )~ear, for irtsta11(:e, the!\<;\\;' Dea) l>eg·ar1
in the l:nited Swtes and Hitler wok over in Germany. The United
Stat<.:s l'>eg·:111 f)l'On10Ling recipro('al tr;1<le treaties, art([ (~er111a11 y
triecl her hand at forced exports, back<:d b y subsidies and 1.he
Schacll! invent.ion of several dilierent kinds of marks. Ilrirnin
moved toward prefcre11tial tat'iffs wir.hin the Empire, and France
movc::d toward t.lte llltun concepts of socia lism. "Vith all such divcr-
gencies of mcdmnism, th e nations concerned went through the era
'"'itl1 a co11sistent an(l concin11c<.1 similarity i11 eco11omic C>tttcomes.
F.ven more to the poi nt, the various nations, through all their
depressions and ensuing pol itical disturbance' and n:coverics of tlac
193o's, displayed in their economies the rhythmic patterns that one
could have anticipated fron1 observing thdr past p;1tten1s of re·
sponse. Says Professor Schurnpet<~r con<:crning the world-wide de·
pression itself: "Ve not only know that all the essential features
H
of the post-war period up to the world crisis, hut. also Lhose o[ the
world crisis itsdf, answer perfecLly to expt:ctarion from our moclcl,
i.e., past experience." 'I' Of the e1.1s11ing economic recovery, he said,
after tracing its outcome in EngJ;md: ":"Jeither new policies nor the
numerous disnarbancc.~ to which England's economy was exposed,
effaced the cyclical con1.our linc.s which, taking account of sp<:·
cilica ll y F.nglisl1 condi1.ions, we should expect from our sch<:me." !
• Ibid. t Bwit1ess Cydr:s, I" 984. t Ibid., p. 966.
[ 78]
T RE 54-Y E AR R R Y TR M
As to Germany's burst o( activity after 1933, he underlines the fact

that in a go,•ernmen t-managed economy the r hy1hms of the previous
capitalist economy are still pt'rpetuatcd, hy saying: .. 'f his is ex-
ac1.Iy the kind of performance o u r model wo uld have le<! us t.O
expect from unfeue rcd capiwlism. Very obvi ously, however, capi·
ml ism was not un fc1tcrcd.'" • Of the rise in A111crican prosperity
that was attributed after 1933 to the n1casu res o[ I.he New Deal -
but which arrived strictly as might ha\'e been an Lid pated by anyone
projecting the familiar ~-conomic waves of the past- Schumpcter
declares shortly: " ll would be contrary to all ex p<·ricncc and co m-
mon sense, Lhough of course no logical impos.sibility, th at a process
which can be str icLly proved to have been run ning its course since
at least the sixLcenth century a nd right to the en(! or 1932. should
have come to a sr.op suddenly on J\hn:h 1· ' 93:J·" t
\Ve shall not labor this point here, noting it ch idly as a matter
of background infomi:1tion. T housands of words could be " Tittcn
in the conventional attempt to trace all the working.. of unlike
causes which produced such like effects, and in the rntl no on e
would be much wiser. Fo1· our purposes it is suflicicn t to no te a
simi larity in the patterns of the phenomena that prevail over such
a wide geographic arc11; those patterns <1rc very closely related, be-
yond rlou bt.
Fig ures 7-11, borrowed from Ko ndratidf, sho w some of t he
work ings of cl1e long wave in diverse phenomena.
\ Vhile the major rhy1hms we a1e dealing wi th seem Lo prevail
more or less universally tluoughout the world's industrialued econ·
omies, some show up more prominently in one nation, others more
marked! ~· elsewhere. It is as if d ifferen t organisms responded
with d ifferent d egrees of receptivity Lo sti muli in 1l1e environ-
ment. \Ve shall lind this part.icularly 1n 1e when we note lat.er how
even different ind ustr ies and d ifferent p rod ucts niay be more re-
sponsive to one particular rhyt hm than 1.0 anoLhcr.
T h us we need not find it unusual if the 54-ycar rhyth m usually
• I bid., p. 971. t Ibid., p. !)85.
[ 79]
CYCLF.S
sho\\S up in one nation in wider swini,-s, more dearly marked,

Lhan in ot11er naLions. 'fhe g-year rhythm, which we are shortly to
meet, appears more clearly in Germany and America Lhan in Eng
land. ·niere is also a 41 -month rhythm which similarly stands out
more clearly in America than in F.urope. On the other hand, there
is a rhythm of around 5i years in the Continencal economies whicli
shows up there almost as clearly as the .p .montn rhythn1 in IJ1e
United States. H erc we do find a rhythm ol' 5! years in corn and
C.:OttOJl }J Tjccs, i11 SCC\l rily priccti, J l \ll i 11 (l l ll~ 1111Jcr u( U L}1cr Series
SC•I• of f Scale of ti
HO..--..---~--.---,..--..----r---.----.--..----r---.---,+15.0
+20 +10.0
+ to +s.o
-10 -s.o
1790 1800 1810 1820 1830 1840 1&50 1860 1870 1880 1890 1900 1910
FIG. i· Le>i.._C \\1A\ll{'i IN '\VAGYS
\ ·\ 'n:-.At:.4l o[ aJ:,<Ticultura.l laborers :,111d <extile v.·orkers in E11gl:11)d. n eviatior1

(t'Orn tl'e 11d CHI :-1 g"(>l{l b asis, Sffif)Othcd 1))' JilCilflS Of l'l 9·)'eftr Jl\Ovit)g a\•tragt:
{aC1c:r Kc111!lraticfl). A regular 54·yc••r cycle l1as beetl add c:cl.
-20 • - -20
,
-10 •• I~, -10
Ellrffsh Con.:lols f
0 0
+10 10
+ 20
Ll~820~-'--1~
840L.,--L-18'0..l-.-"-J~880,.,.__.__,,l90CJ
~-~).9~1.0"'
Fu;. 8. QuOTAnoNs oP lNTBRL~T-n!.A•r,.,c Su:vam•.s

Dtvir.1 1.ioos !rom creod o( Fn;uc;h 1cntc :.nd English o>n&<>ls, srnf>Othcct hy
mcn ns of :t 9·year n1oving average {;i ftt:r Kon<lra~icff). The curve.: h:Js been
in\'Cl'lCd and a tegular !)4· )·et.1r t:yt1c has l>c;cn atl<lc<I.
(80 l
Seafe for f'1~ noe Scale lor £nit1tnd
t40 +200
,-,
I I
+20
·-
,
I '
\
+100
,,- I
I 0
0' , ,/
J
I I
I
I -100
I I
I I
' , II I/
Coel productlon \ }v... - -200
El'l~l.and"'"' I
\1"'
- •o'.L .,--'--''-..!...-J.-'---''--'---1--'--'--''--'--1--'--'--L-..J -soo
1830 1840 1850 18&0 1870 ltW 1890 1900 1910
1:10. 9. TJJE I .oNc \\l.\ vr: 1.~ Co .~i.

C•>:1) producliOfl j11 l!:t1gla11d and (.<~iii c:c•1•:.u111ptifn1 1n F rance (afler 1\.01)·
d~ ·3LieJJ). :\ r<•gular ryt· ycar <:y<:ll' \l"S hc<.:n added.
Sc:ile lur Pig Iron Seale for Lead

~ ~
- 20 -
-30 '---'---'--'--.l~-'-_:.-.1-_.;._..__._~'--'----'~-l--.l~.L.-1-450
1840 1 ~0 186-0 1870 l880 1890 1900 J910 1920
t·1c. 1<•. P1<; lR<.lN ..\.:->D LEAD l'r..onucnol'\ I:.i Ez.;cr.ANn

l)e"·i.-.Lioo Jr<11Tl trend, smooLhed by 9--)'ear 1tlo1;·ing avetages (after K«)11tJr·a•
tit .ff) . r\ rcgul:-it ~1 ·yeai· cycle })as b-tx11 acl<lt-:d.
[ 81 J
CY(;l . RS
of industrial importance; but witl1 us the 41-month rhythm tends
to overshadow it.
Thu.~ far we have srokcn of che 54-ycar wave primari ly as a price
phenomenon. And indeed. bcc.~usc it has regisi~rcd itself so dearly
in prices, the chnrti11g of those prices is the dearest. ;ind simplest
way of recogniiing Ll1c phenomenon hist0rically, and traciug it.
380
360
' ' • ' I
....
340
PRICES ..!
f~ I~
••"
3ClO
280
260 "I•"
••
240 '
220
200
180
160
140
lZD
-1 ·
-
"' /
'
/
/
/
',
'
,, -
v Ul'l1ttd Sttlirs
100
80
60
<O
20
1~80 1800 1820 1860 1880 1900 1920
L"10. 11. LoNc ,.,t1\v~_;:; '~ PR.recs

Commodity 1>rice$ in 1:.ngland, France, a11d Lhe Uniled Slntes. The index h"s
l>een rcdttccc.1 1.0 a ~Id bnsls. (After Kondrat.ieff.) A regul~a· !)4-year cycle has
bccxt added.
But to understand what we are really tracing, let us remember

that prices arc al'"':\)'-S only S}'TI'llWJ)oe; TJ,,-. 51 ·y,..:.r 11rirt> '°'""a\.·P. ~n
only emanate from a 51-ycar wa,•e in human acti,<ity, and it is
doubtless this activity which is the ha<ic reality.
In the bcginnin)( of the wave's upward sweep, as Professor
Schumpetcr anrl others see it, tl1c economy is poised for expansion.
The economy is ju.• t ending a lime of depression when, as in winter,
few things grew. Storms have stripped clown 1hc decayed trees and
weak limbs that coul<l 1101. starnl 1.hc testing of tempests. In such
an economy many debts have been wiped out and others gycatl)•
rs21
THE 54- YRAR. RHYTHM
reduced; excesses of many sores have been paid for; surpluses have
been largely distributed; people finally begin lo sense in the air a
kind of burgeoning opportunity, like the faint s111ell of spdng when
t.herc is no <~hangc yet visible to the eye.
That was the air in our own nation, for instance, around 1790;
around 1845; around 1898. lt is a tin1e when venture capit<tl st;irts
coming fouh - wheu iudustries that got start.ed in a small way
toward the <iutumn of the last cycle now begin to find growing
opporluniLies - \vllen t~\'Ctl tf1t~ ttl<>st <:asual sc:attcri11gs c)f seed rna)1
spring up into a lush growth. It is in chis early phase of the cycle,
as Schumpeler demonstrated, that the innovations which char-
acterize each new cycle o( this kind begin co spring up in full force.
Just. as the railn><id was on the scene in primitive fonn before 1815,
but waited until afterward for its real development - just as elec-
tricity <:amc into being before 1898, but bccarne a great economic
force only in the 190o's - the great industrial innovations that
rise itl stL<:l1 a sprir1gtin1e , ..~ere 11st1all}' pla11r.t-:(I i11 t~1<~ prC\'ic.>t1s
cyde. Their really great and characterizing work - a work that
is to end in t1·a11Sfor111i11g· tl1e economic scene - beg'ins in tl1is 11e,.,
spring, as tJ1c 1lC\\.' i11<lustrlcs 1·isc to 1najest.i<~ a<lult J>l'<>portio11s.
A multifarious number of other contributary industries rise
simultaneously, sonic gxeac, some small; and as they thrive, so do
older industries that dot the :><;enc. Employment gTows; money
stans flowir1g; credit expands. \'Vhcn rni<lsummer has arrivc<l,
when the peak of the growing cydc is reached, there seein a 1nillion
evidences around of a new era, a florid age of growth beyond t.he
mcmorr of living man. Speculation regarding the boundkss future
is rife. That was 1929, in our own memory.
llut now the summer stonn gathers; suddenly from nowhere
douds appear, the haromcter snddenly falls, and in the quick hail
and rains and winds there is a scurrying in from the fid<ls. \\!hen the
sun c;ornc:s out again, the humidit)' is drained froin Lhe air, and
in its clear light there is already a hint of auttimn. That was, say,
1880, and our own 1933.
CYCl.I!'.S
This late s1mm1cr and coming a11tumn is the p<~rio<I of the great
harvest when the products of inclustry usllally are distributed
more wideJr than ever before. Prices have con1e down from their
peak . And, though they will fonn Oc(:asional tables, they now
usually continue to drift generally downward-while wages, on
the whole, stay up. Tl1is mean.~ growing purchasing power for the
workers who have jobs. It makes a spc:dal kine! of proletarian pros·
peril}', \ot•J1ere tl1e be11efits of tl1e c.~c:c>1l<>1n~· arc tlt>\'>' C\'C11 more
widely sh11recl than in the days when industrial expansion was so
rapid.
'!'here is, 11nfortuJ1atcly, more unemployment. For, as prices
decline, manufacturers naturallr look for ways to diminat" extra
rmm-hours, to turn out t!tc:ir producL for less. And competition
g1·<J\\'S a1n(>tlg· tf1e 111anufactt1rers. Neti.•con1ers s1)rir1g up, .attr;1(:ted
b)' the 1ecords of lush profits made by rhosc: now long in the field.
(:f.>t11l>i11ar.ior1s l>Ct\.veen 11lanttfactt1rers, som<;limt<;s 0111~· i11 the fc>n1l
of sile11t ',;orking· <-tgreen:1er1ts, c.'.on1e alo11g 11ar.•1r;11 l)'. as tl1c)SC 111osc
expose<! to competition try r.o protc<:t tJ1cru,elves. And some of Lhese
cou1bi11atio11s 111ay, llaJlg O\'er i11to the n<:xt era - ;1s t}1(:\,.' flld lO
plai;HC the first Rooscvcl t - and give: Ii im pol itica I opponuni Ly.
People l>q;i11 to t.ry t.o hold on Lo what they have. Some of the
old en1erprise has lost irs fervor. There: arc; hursL< of 'J><:t.ulation
a11cJ l>11si11C.'» ex 1,a 11sic>tl fl f>\\f a11d tl"1e11 - '·\le s11all see Lhat tl1ese come
alc>tlg; :1t [airly regt1lar 9-}1,•ar intc-!n•als, sorr1el10\\r . .Bt1t cl1e great
pro;\n:s.~ of du: c:ra li<i.< now been 1nade. i\s it s1owly diaws LO an
end, J>t~opJe look backl·var<l to rhe pas(. 1rti<lst11r1111<·r <l;t}'S '"'lien brrcat
fort1Jn<'S c4>•Jlcl st>rn<:rirrtcs \)<; ff>U11cl al1nost b)' Lt1rni11g· C>\·er a slc)tle,
if ;:ot1 \\'ere Jttc.:k~·; ar1l>!rc.:c<',!C'11tt:cl
j11cl11stria1 a11<l (·<.<>1u >rll ic acf1ieve1nents Vi.·el'e lra11sforn1i11g· tl1c.: la11d-
sc.ape. And 1l1e people begin 10 wonder if progres~ has really
sLop1)e<l..511<:11 ti.•as tf1t; 111<><>,) ir1 tl1c r1i11eti<":s. for inst<tr1ce, \.;hen
nry;111 \V;tS to11ri11g tl1C f'Olllltry I.e.• 1eJ I rar>t :lttdien<:C:S tJ1{~ l);tt.ic.>O
h<: <'.rucilie<i on the cross of gold. \'\'e ma)' sec itc.ome again
mn., L 1101
as the n1ood of the late forcics or early fifties.
[84 J
TH£ 51-YEAR RHYTHM
This, in very general terms, is the outline o[ the 54-year rhytl1m

in the economy as it has heen outlined hy l:'rofessor SchumpeLer
and 01hcr ohscrvers. F.ach cycle star~~ out anew - no! where the
previous one began, but where t11e previous one ended. Each leaves
.
tlle environrucnt v;ist.lv. d ifferent from its state a hal( centurv before.
The recorcl of the rl1yt.l11n that appears in the price charts is only
a vcr)' skdctonizecl record, but it is at leasL a record useful for sta-
tistical purposes.
Care is needed in interpreting and using tlle record . just because
it is such a skel<:trnl. Jn \·icwing the ;iutumn period characterized by
price declines, for instance. we nlust not assume that the <kdining
prices necessarily mean business dcpressior1. Unckr certain condi-
tions, our economy actually works bet.t.Cr and smoother when prices
are slowly receding. It has been pointed out. a thousand times over,
that our chief econo1nic security lies in distributing t.hc output of
consumers' goocls t:ver more witlcly. A slowly falling price le.vel
may be the sign of such an accomplislunent. Such facts s<:rvc: to wat'n
us that a strong 11pward sweep in the long-term price chart is not
neces.;;1rily to l!!: 1.ra1i.latcd ;L< pro~ptrity; and a downward sweep
<:e rtainiy does not necessarily mean depression .
If our 61-ycar rhythm, which shows up so clearly in the long-
cerm chart of wholesale prices, does not of itself signify times of
prosperity ancl srahi lity and clepression, then what does it do? I t
acts, in effect, as a clock. In association with three other rhythms
we are lo observe shortly, it gives us a way to know what time it is
it1 (>llT economy.
Suppose a young man has some capital and is going to start a
business. Bcc<1usc he is too young to have much personal nlemory of
the great economic tides in the world, anti hccau.5e school t:tlu<:arion
is u.;uall)' so deficient in such rm1ttr.rs, he is like: an airplane traveler
who travels for around the world without a watch, and then lands
irl a f<>g. , ,\:l1al li111e (>f da}r is jt? HO\\' lllU(:)1 tj111e r·•as he l>cfore
evening? Is he sure even of the clay of the month? It might be t11e
begirming of a new week, or it might be the en<l of an old one -
[85)
CYCLES
depending on thr inlernational rlatc line, and where be is in

reference to iL. UnleS5 he knows, or can quickly find ou l, he may
manage himself vt:ry badly .
•>\nyo1ic who Slarls o ut in hu.•iness without k11owing what time it
is on t.he economic dock will obviouslv owe 11111«11 lo luck if he
'
succeeds. And rrrn11y men who have eng:•gl:rl s11c-cessrully in busin"ss
for a long time owe much to thei r sheer good fortune in having
sta.ru·d <>Ut 1\'l\e11 tile economic· .;u11 \\'as risi11g.
T oday the grc:11 5 l·}eat rhythm is seuing: it has been on the
dcdine since th" middle twenties. In the middle 195<>'s it should
be rising- "!f"in, if lhc pasl rh~thmic palt<'f'lh l'epeal themselves.
Urrkss past precedent fails, we may look forward to that economic
tin1e as far111crs l<><>k fc>r\\•ard LCJ sprir1g after (I lr>r1g \vinter- a
spring when seeds long quiet in Llie ground slio11lhooting up
almost mir;iculously, witl1 a new spiri1. of c11tcrprise 1evealing itself
everywhere.
J ust as the million,s saw c;lectricity and the mot0r completely
transform 011r cnvir,•11n1c:11t a11<l ,.~..ays of Ji, ing fC'll l<J\\'i 11g 1898, tllose
1
who live into the third quarter of our •·entur)' may well see the
growth of new forces just us dramatic, impelling, an<I 11niver~al in
their etfect - forn:.< of which atomic energy may he o nly on<~.
1' hc start of a new trend in our econom y ramiol be forecast, as
we have seen. But even without expecting a11y lift in the tr(:n<l
lines, Ll1is new long-term cycle wulcl he read in advance to provicle
hope for a corning long tenn c:1a of prosperi1y anrl progr~ in our
Jl3Li011, regardless Of :lll}. dcprt.-s:;it»JI tl1al 111•1y iJ1le1 \ .C:J:JC:.
f 86]
\'I I
r< CHAPTER v we briefly met a rhythm that has 1.>ccn cakulated to

I have a length of very close to 9 years. As previously point<:d out,
this p<': riod icity should uot he confuse<( with the 9~·yea,. ,.hythm,
that appe;irs in biological phenomena. 'Ve shall now 1neet this
r1l}'lh111 again i11 COlllf)all)' \\:iLl1 ot1r 5,1·)'ea1· "'ra\·e, and sJ1all refer
to it hereafter as the !)·year rhyth1n, for purposes of sirnplicity. It
is the ,.hythm which Schumpe1.er, without delining its length too
cl(>s<:ly, lik.<-:s I.<> t:all tl1c Juglar <:}·<~le, i11 l1•>t)(>r <>f tJ1at pic>tl<,:<:rir1g
Frcnduna n:
His great lllC'ric is ch;it he pushed the crisis into the background and
th:tt lie <lis<:OvC'rC<-f Uc;Jo,-..· it ar1othcr, 111ttc:l1 111orc ft•J'1c.la1nt1ttal, pfle-
nome11ort. . . . Ile11cc:fortJ-1, alrf1oug}1 it t<>ok <l<.:ca<le.s fc>r rl1is nci.i.· \·ie\\i'
to r>revt·ti.1, tl1<.: wave (Jttstc-d tltc crisis frc>nl the role of pr<>tag<>Jtist or tl1e
play. . ..l ..I-1e f\t·oUl(~Ut l1:ts agai11 dt:.tr1gt:·1.l its co111plt:xi<>11. It is r10
longer tltc probk:rn of the wave. It. is the problem 0£ idcnrifying and,
if possible:, isolating the many waves and of smdying their interference
one \\•itl1 each <.t11ochcr. •
• B11sit1ess c:·ycles, p. af>)l.

Only a re\v yrar'> affcr Jttglar's '"ork appe;trcd in 186-o, an 1\mcriC':tn, Sam111;I
Ilt-noe:r, Jlr<ldt1cccl a v;ork cili11g regular cycles in the aver~tgc yearly })Tice- o( pjg
iron - crclcs tha1 averaged nioe yt';•r:s iJt length. 8e11ner. \\•110 called 11i1l1sel{
"a11 OJ1io fa1·1~1er," p11lllish(':cf in 1875 a curi<.l\1<:>ly itttcrcstiug l)<HJk (:allc"-d Rr-·n ·
nr.rs Prophecies of F11t!4rc Up.s and Dou.•1!.s of Prt'.ces, '"hich l.;cnt thro\1gh
[ 87}
CYCL~S
Sd1un1petcr finds his " .Juglars" clear!)' ind icated in the eco.
nornic life of Germa ny, F.ngla1u', ;111cl the Unit<'cl States; by charc-
ir1g· J>C:rc:er>tag·c c1<'\ iario11s of 1)ricc i11dice-:. frc>111 Ll1c•ir !)·}'t·ar 111cl\: ir1g
1
;1v<."r~tgcs i11 Li1ese Ll1rcc tf>1.t11tri<~s. lle i)rocll•C~S a cJ1art '""flere tl1e
.JLtglars are aj)parerll C.\'Cn tc• (111 Ltr1cli.!)c:c:·r11ir1g eye (see Fig. 1).
In <>Ur stt>(:k rnarket. as '~.rel l a!' ir1 c:c_>1rt111c>clit}' pricc.:s, '"·e find e\.·i~
<lence of an appioxiurntc 9-year rhylhm clea1I~ \f;Ht'cl (see fig. 2).
Figure 3 shows the ckviaiions from trend d1an<'d wi1h an ideal
!J·year pauern. • R.ising upon the: 54-yc.1r rhythm, the shorter "ave
o ften serves by its u ·oughs aucl pc.1ks to maik importaut C\'ClllS
in the economic Row of things. The peak in t l1is 9-year rhytbin in
lq•f), c:o111i11g- at a Li1nc \;,•) 1e1 1 tl1e fJ1·)'Car rll)' tl1r11 :tl!if> \Va.~ reatf1ing
tc>\\•a rcl a crest, cc>it1<:ic l<:<l ,vicl1 tl1at ye~\r's l'ea k i 11 tl1c 3·}'ear mo\•i11g
average of cornmorliry prices, highesL in this century (compare
Figs. u ancl 3, with ,, and 5).
It w:is in 1928. again, tha t this 9-ycar rhy1hmic panern in the
stock market on (e moic rcachc..-.1 its crest. The top of t11e 54-year
nuntl'N>·us ('c1itio11ot. Tn additio11 le) 1l1c !J·)'{«tt' cycle i.Jt 1>ig i1e.n prin;s, h e <ited
c:yr·Jc·s a\·cra~in.~ :,~lz yc:;,,1r. i1l ('Or11 y>rict·s. t t ye::11s in c~u11111 i1rict:s, and 1$ ye:.t~
bet,,•reu fii> <tncial J'anic~. T-lc no ted tit.it the: v:1ri:i1ir,n$ rro1n a !il:·1ud<ircl ~)·yrar
f)rttltr11 i11 iro11 r->rj c~~" tc·1ulc<l to re 1)e:.t1 thr.tnsc·lvr':) evc:1}. tl1ird t:yCI('. T irnc lt<ts
11n;:i\·cd l1i1n ri~f11 tv the ex1cn 1 1f1:11 , if yott hacl pttl'1 ha.'lt'cl :''"' sc>ld pig irot1 011
tl1e b:isl$ 4lf his prOjt'< ll'c·rioc(
'""liJd hn.,..e J,t"r11 t,,.<.·r\'t li11,e-1 »<Htf los1.
There 1.. a lw clt::.tr Cl> itlc11cr flf rlt<' cc>'ltin11cJ t"A b1t·111 c of rhc 5~~·)"ear rh~tl1111
he di~:.1\rn,.'rirt""- On the o.h~r 11.an<I~ the 1 1 )·(':lr rhytll1n fit: fcn1nrl
i11 aKtcm prices 1... ,. f;1ilrci 10 001'l1i1lttt."" tcl ,f,>minate. J Ji.. 11li.'4:t"'\-a1icH1$ as to the
.-~iscet1ce tJl a11 1ft.·}'C:lr lh) 1J11n - '''i{h Q r i1<ivn1 (>f 1u1L 111,tre 111:in lt\'O }eats ir'
c·iihet dl1 e-cti1>11- frt \t'lf \\'t·ll v.'i1l-. tl:~ facL~ abcn11 1111· 1A1,:i .,-car rl1ytl~1ll v.·e
sha ll cli'l<:ll"S in C:h:'\JJtc--r IX ; and che pa11ir·" c>r 1P.o!J· J907. :-i11d 19:.-~) '''(·re all
\vii hi11 t'\liO year:" - 0 1tc '"n) en· a 111.ifu...T - of a. t ixicl 1H·}·t'.t r (>:11 rcrn based 011
1873.
• 0\.'Cl" U1~ 1>c.·rio<l {or \ Vhi1J1 t\11" h:tt'e clala 91 ~ yc.11~ fir~ s.cocl:. rn:irl:ct be·
J1a\ ior J>c·111·r charl a 1h ~· 1J1rn of cxa<.ll~· 9 )'ear,,. iu lcu gth. C.i\.'c1 t l11: sau1c 1u·ri11(I,
VJG years b~tt1•r tito lhe S\\•ifl ~l> in v.:hQfcsale vriC~:i. l•W• : hut C)\.CT chc longer
l'eri1xl £c•r t<ibich '"l1r,1~ ·~1le prlc(! data al'e ;J\ailal>1t·, rhe J\:Of;lh is see11 tt> be
n)<we rly 9
11c;.. ~t:Otn. Sc."\: Fig. 5.
[88]
THE 9-YEAR RHYTHM
rhytltm had alre.1dy been reached :ilx1ut •!1~5· J n the late twenties,
so long as the 9·year rhythm continued rising, th{· momenturn of
the market was maintained. 13ut when 1.hi, rhyr..hm reached irs
Pt:r C.llt
+20
WHOLESALE PRICES Germtny
+10
+20
Greet ''111ln
-10
120 -
+ID
} \ ,,,
\
I
,
I
0
-10 ..'
I
\
' ,I
'
-20 .
-30
1830 1860 1900 1910
FJC. J. TtfE 9-\1F.AR. Rrr\'T'U>.C lN rRJcES

Oat.a for Cernlany, Gre."lt Brit:-1j11, a11d tltr tJuittd States. Percentage cte ... ia.
1i<ll1~c;f JJrit..c indices trom thc·1r 9-yc-;,r 1r11>\'111~ a"erages. i830-1910 (after
Schum1K'tcr) . Rq,.rular g-year C)"Clo ha\'C been :.tl(lt-<L
crest in 1928 and then turned a[tcr che 5•1·year rhytlun, which had
already 1,...... going down [or chree years, 1hc beginning of the !,'Teat
fall was do><: at hand (compare fib"'· •and 3}.
Inver·s('Jy, it was after thi> rhythm h~d re:whecl a trough in tg2~-
1924 chat the economy started rising C<pansivdy to the peak
rcuchcd in 1928. And it was afr.(:r its u-ough in 1!)3<- 1933 that re·
[ 89]
C Y CLES
200
COMMON STOCK PRICES t

100
t •/
Fie. • . u. s. Go!ntOH Sl'OCK PRl<'.l!S

Axc-H<X1ghtf1t1 Tr1dtx.. 18.)4- 19'15· $D}()()fht!d h)· 11~ns 0£ a 3·)·ear moving
average, togc1l1r.r ~·lth a \Vtighrcd 9·ycar 1Jl()\•ir1,g average, extrapolated to 1944.
Shaded areas sh<.)\Y' 1 i111~· and txteot of relative st.rcngth. ArtO\\'S u1ark highs of
regular 9 '& ·y<,;ir cycle$. (Sec [ootnote on page 88.)
covery began from chc 19u8 fall in commocli ty prkcs and from the
subsequent 19•!) crash in stock prices. The well-k nown peak in 1937
follo wed. The next peak of the ideal 9-year wave was in mid- 1946.
Sud1 "calling of the turns·· may be traced to the action of this
rhythm, time and again, for scores o f years in American ttonomic
ltiSt(>ry. T he signal is cl caT~t, as a tc.tlc, \.vl1c11 ••t:-i• ·by turns come
also in the other rhythms we are obscn•ing. The concurrent move-
ment of two major rhythm~ may often suffice to indicate such a turn
in the symbols graphed on our charts; C<)ncun·en t movement of
three major r lty1.l11ns indicate i t. almost unerring ly. As Schumpeter
has said:
It is clear that 1hc coincidence at any time of corresponding phases
or all t.hrtt cycles will always produce phenomena of unusual intensity,
especially if tllc phase$ that coincide arc ll1ose o r prosperity or depres-
[ go l
TllF: 9-YEAR RHYTHM
P6r Cent
150
1'5 I
I
I
,,
100
"' /\
'
I '
,'
" '' \
'
\ I
I
'
i
' "
'
75
COMMON STOCK PRICES
DEVIATIONS FROM TREND
50
1860 J870 1880 1890 1900 1910 1920 1930 1940 1$150
FJG. 3. Trn: 9 -YCA.R RHYTH'.\.r TN CoM~toN STOCK PR1c:ES
Perce21tagc <l<'Viacior1 of tl1e 3·ycar moving a'•erage sbo'ffl in 1:1g. 2 from
the 9-)t'ar 1no\•ing a\·cra,gc abt1gi'"en in tl:c a;1mc: ch~ together " -ilh a regular
g-ycar cycle.
1 ·ht high:; of the rf·gu1:.r cycle; Call :u [ollc,ws! L8!)6.o, 1865.0, ~f1' ;:(if~ 5': "1~ti3.n.
I A9::.<1. I !Jlll .0, 1~1 10.0. l 0.!J.I), 192R.1), J 937.0. '1t-I b.(t, 19.)J·<>· I !>')1.11. 1973.0.
lf)k:t.c>. r1c. ·rhi:: lo,,·~ J,,IJ h:df\\':1~· l)t:l'11:cen thcs.c cl;1tc.'i. ('l l \t exact yt~ar iud l·
c:11r·\ 1h1· 111id-poin Lo( th.11 yt'c1r. Tl1us" 19:.11:'..u" 1uea1".ls Ju11c !$(J, 1928:" 1!>·16.n •·
u1t.ln~ .Jun('.~<•, l!H6.)
\ rh\·l.11111 that h.-\ ttrc·J1ed itself J'li '11.111,- limc1: (•••) :ind a.s rcgtibrly i•
Lhi~ <>tic· c.J111u•t easily t't(' rhr rc::-.lilt o( pu1'<." ch.i111'.e.
'l lu: tC~Ltl<"tT <·yr·Jr: It.a~ IJccn <:Xt<·ruk·ll into th<.: rutUJ't• (() in\lit:af<.' <l}•})fOXi-
1\'l;lf<'I )· 1111: 1,ell:-1vj1)f tl•:tt 111ay be c-xpc.. 11·<J i11 Lile dc-vi:1ti<u1 ~ 1>1 tlte 3 y<·ar 1nt1\••
iJ\g :l\('r;1i;l' frt>11\ ti><' !I y<·;,r 111lJ\·i11g ,,,·c:r~gc, ii' this rbytllru 1:.>11li11ue$.
sion. Th• thrt-e <11·•1•'>1 an<l long•11 "<lcp~ion~" within the epoch
,o ... <·tPlay that <.ll:ir:trt11•i"tir •
And furtlaC'r, re11:ariling attempu to look back whil e the r<:c::orrl is

still Jn,],, and fincl ll1e "causes" o( the 1929 collaps1·:
' ' a 11larl ma}· staff<:r from m<:tn' ill' and ''<'l for nn i11dcfinire time
Iea<.l •l \tigc>rou:, Jirf" ,,·ict1ot.1t l)eing :11cott1r·eni<: ncc<l 1Jy tt1c111 Lt11til, ,,rl1rn
t1i'i g<.:1lPr:i l \'itali 1y h:i:, cbUctl ;rt'1a~·. tl1usc ill$ or ar1}' tlllt: of tl1c.:w 1n ay
suddenly ~cquirc: what ro the sper.ialist's eye will seem parau1ount or
• Busa:rn:.ss Cyclt11 p. 1 ;3.
CYCLP.S
eve11 of fatal importan<'C; so the economic organism always does bleed

from many wounds whid1 it be;;rs light!~ in three out of the four
C::)'<:lical pltas<..-s, a11ci \vliid1 spcJJ discon1for1. '\\•l')tri Ollc t)•c.:le, distress
whcu two, catastrophe whcu all tltc <:yclc:R arc in th e depression phase.
No doubt, cxter1lal irtjltries v.•ere o[ lJl1llSl1ally grcac i111portance in
this case, bttt explnnatirm car1,.01 be dcTived from them . . . . The crisis
was nowhere else anything lil<.e so sCYcrc as in the United Stales, 1he
country most nearly free from injur)' by external factors.•
This rhythm of !l yc-ars, which is of such itnponanc.c iu our CC<»
nomic statistics, is like every other rhythm we rliscuss here - it is
the registering of a wave of activity in tl1e cco11ornic organism.
Tf, when we see lhe wave rising, we t hink only of statistics ris-
i11g, 'l\'f I1a·vc a to<> li1nitc<l ''it"\\". For t11c st...'l.tistics, c1f <:<>UTSC, ;1re
only a reflection ol' human movement; t11ey arc merely one way of
seeing it, and o( meaSUl'ir1g it. Schurnpctcr bas shown, for instance,
that what this 9-ycar wave reallj' reflects is this: Repeatedly- at
something like the 9-ycar inte1val- the industries creaLing t11e
great economic tltrusts·fnrward in the 54-year C)clc have new surges
of activity. There arc new bursts of enthusia•m, of energy, of specu-
btion, of product.ion, of progress. The rise of 1.lais rhythm regis·
ters this almost uneningly. It. is not surprising that the rise in .~ud1
activity is accompanied by a rise or a "recovery" in wholesale
prices.
If the reader will glance again al the chart for whnlc<alc prices
in the preYious chapter, on page 70. he will he hcitcr prepared 10
luvl dt Fig. 4, wl1id1 is uae laner pare o( this .same chan u~th a
9·year moving avc·rage crend adced. Figure .i shows 1he deviations
from !l'Cnd chaned with an ideal 9-yea1 pauern. It is dear at first
glance that they fit with astonishing regularity. The 9-year wave
hel'e is extremely evident. It shows up stri kingly in maray individ-
ual businesses (sec Fig. 6) as we:! a.~ in wholcsak prices and pric~s
nf comrnon stocks. (Sec Fig. 7 for a summary o( the "ave.)
Now we arc read)' for another step: in Fig. 8 >CC 1he experiment
• Ibid., p. 9•" lt>lia suppLied.
TUE 9-YF.AR l\HYTHM
l•du
160
WHOLESALE PRICES I
1'0
I
uo
100
eo I
60
~-
20
0
llilO 1850 1870 1910 1030 1850 1970
Fie. 4. u. s. \VnolJOiAOF. P•1cn, 1830-1945

(1916-=- 100. A 5·)"CaT m<•ving a\eragt. rogcthcr \\·itl1 a ''-eigl•ted 9-Jell
1TH)\·i1~g a\•t.rage. ex1.rapobccd to 1914.) Sl\:kded areas show time and extc:n1 of
rt:latl,·e suength.•~o~·s mark l1iglu of ~u.lar {)*yc:ar cyc;lc.'5.
P•r Cent
150
,
.. /\
v ,,
75
"
§
- - --
~ ~
...
0
:i
0
~
~ ~
Q
!!! ~ 2·
:!:
-
~
5• 1-.HB g--)' EAR R.ttVTk\f IN \V1tOl,IM.LE PR.ICES
J.41c.
~ ~ -
:il
~
s~
-
Q
~
~
Percent.age JC\.·iati<•OS of tlie 3-year DlO\·ing a\'tnge fn>m 1.lae g-year mov·
i1>s; a~tragc. 1830 -1944., together 1tTich a rcgi1l:-..r 9-)tar cycle.
A t~ubr !r)ta.t cycle has been added and proj<"Ct•:d lnto the future ro sl'tOl\'
\vh;et will t1;1ppc;n if tl1i$ rhyrl1m Q'H1tinues.
Tf1e date~ of tlte Lu..1\i11g points of the regut:.r cyc;lc att the same as th06C
given ft,r th4.: regular c;yclc of t<ltn1no111:,ilCt:ls iu: .Fig. 5.
[ 93 J
CYCLF.S
of creating a composite pattern liy combining the 54-year pattern,

which we examined in I.he previous chapter, with a pat.tern show-
ing 9-year peaks. Against this " ideal " pattern, in Fig. 9, is now
imposed a graph o[ An1erican wholesale prices, smooth(:<! hy a
3·)'ea1· mo\··ing a\:erage, as \\~e first sa\'' it on 1>ag·e 70. Tl1is c.i111t~ 1.l1t~
pe;iks are not cli roin;ited, a.s 011 page 71, when the graph was first
applied t.o a .~itnple ,54-year pauern. Ilere the wartime peaks are
allowed to remain. And what do we find?
Per cent
ls• ~~~~~~~~~~~~~~~~~~~~~~~~~~~
120
,• I
I '
100 , ''
I
,;
7S
A MANUFACTURING COMPANY
lll80 18!10 1900 )910 1920
6. A.'1 J~o"1DUAL BuSlNl!.SS

Frc.
The 9·ycar rhythm in the sales of comp~ n)' A. 1>crcenlag-e deviati<,ni> <lf tlu;
3-year moviog ~''ctagc from tl1c 9-YCar moviog a'\·crage. A t'egular 9-yc:ar <:yt1e
has been added.
Strikingly enougl1, we find that the peaks in the actual record

fit over the peaks of 1.hc ide;il pattern laid out, with a most drama1.ic
conespondcncc in timing. In panicular, we sec that three great war-
time price pea ks, occurring near the crest of each of the 5+-year
price rhythms, came at 1.hc top of t.hc !)-year wave which rides t11e
longer rhyth.m. Further, it was a 9-year peak occurring on the up-
side of the 54-year wave. The <:<Jrnhination of. three [ot·ces - t.he
upsweep in the 54-year wave, the upsweep in t11e 9-year wave, and
the incalculable force of war-provided together a leverage of
J;"l·eat intensity. The height of the t.hn:c !,'Teat inflation peaks here
reflects that.
Now <~om pare these three price pc;1ks, as shown on the chart, with
the fourth and smaller peak toward which vur price l ine was reach-
[ 94]
TllE 9-YF.A R RHYTHM
ing in 1945. From 1939 on into 1915 the1·e was also a war to aet as
"" i11Aationary force 0 11 the price level. Tlut - in lhi.s '93!rl915
J>Cl'iod, the 54-ycar w~vc was declii:;iug. \\!o uld uie reader have said
100
A ." '
'A
•' ' ''
•'
' ' I
"''
'
'~' ''
'
I
' '' J
''
\
' '
18W lllSO 1470 1440 1890
1•1G. 7. JlP.CAt>J'l'IJT.AT1or.;, um!rYl£AR R •4V'l'H'-'

Pcrcc:r1ta.1o,tt: deviations of chc ~-yea r mr)\'ing a\•erage$ from tl1c 9·ycar 1rt(1vi11g
•"•rages in (A) S•I'"' ol (;ompauy A; (B) S..les of Conlpauy B (onotlu" com-
p.i.n y in a totally different iodu5try) ; (C} Pritt:S 0£ Common Sttx:ks; a11d
(D) Wholesale PTicco.
Regular ,-yc-.ar- t)tle~ ha\'e bttn added to all con'cs. The tint, thjrd, and
(ow·th 11at'e been $m0<1th<'tl 11}' .a s nine·)•ear-scction rru1vi11g average.
in advance that prices during \Vorltl \Var II would rise as high as

in \Vorld \Var T- knowir1g that the 54-year rhythm was no longer
moving up. and LhM 011ly tl1e rising 9-year wave was lhere, working
in conjunction wit.Ii wartime forc"5, to hold up prices?
On the hasis of t hCM' paLterns, he could have f;UC$Sed, when the
[ 95 J
CY Cl.ES
war con1menced - and before i1e knew anything about the gov-
ernmen t prke controls that woul<l be imposed - that wholesale
price levels during \Vorld \Var II would probably not reach the
levels auained during or shortly after '·Vorlcl \Var I. .Nor did they .
.Kot in England, o,. the United State~, or Canada, or Australia, or
New Zealand, or numerous other countries.
Nino Year Cycte
1840 t860 1900 1920 1940 19GO 1980 2000
Fto. R. S\':-.lT11r.srs
:\ 9·}'ear <:yt.:Jc, a 51·ycar c;yc:lt:, t1t1(} t.111~ I.\\'() co1r1l>i11c<l. (St.~c al~<> J,.i~. 16,
17 ;.i11d 18 or1 p~gcs 65 and 66 ir1 Cl1apt<..'T 5.)
The real price inflation 1..hat occurred in connection with \Vorld

,.\:<~• 11 t:.1t..l •t.:t..I it.s .111<1~.i11111111 JlOl. '"'it-It 1..Ju:: c::11t1 of lilt' \\•a1, l11JL .1lJv1.1t.
a year and a half later. There have been indications in the rhythms
. rjt:t.:s 111igl1t sir11i1ar1}' 1r1c'>t1r'1t for a
cJf tile .;.\111cri<.:;1r1 ct:c)Tl(>111y tJ1at. P
year or so following "INol'ld ' ·\lar JI - say. int.o 194 7. flut whether all
aveno,;e wholesale prices would reach the high levels of 1919 and
1920, before rc<:cssion, c<:>uld be doubted by those aware of tit(: tim-
ing of the r,4-year and 9-year cycles.
The man in the street could have told you that government" con-
trol " was Lhe " rea.~on " why \\lorld \Var II prices stayed below
( 96]
TH F. 9 - YEAR RH YTllM
vVorld vVar I kvels. But our charts an; not concerned with .. rea-
sons why" - they reflect d1ythms.
I.ook at the d111rt >ig-ain , t:ircrt 1933. Do you see the ()·year wave
rising there? That is the year in which Amcrk;m <.:(m11nodity prices
started to rise again from (kpression bottoms of the 193o's. It is
WHOLESALE PRICES
,, /" f, I
/•\
I v I ' ' \, ,.."'\
,, ' ,, I
'
' \
,,'\ 'r \
\,/
\ 1'-'
,,1
o.___,,..,.._,__,____,__...__.._.i..__,_ _.___,..,.._.__,____,__..__.._ _.___.

1840 1860 16!!0 1900 1920 1940 IS60 1980
Fl(,_;, 9· u. s. ' 'ltlOLliMLE l~RICJ::S

Tf)t'Ce·year lUO\'ing ~\'Cr3g<:, 1926 - J(J(), togetl)er,,rilh the 9•yc;ir :lncl s1·y¢:a.r
cycles i11 OOO)bina1ion (see Fig. 7 above). The 1)1·ojectio11 s.110\1ld be tlun1gl1t
of i11 cerms of gof<f. •!)37· lt slu>V.'$ a1)proximaccly '\'}tat \'ii1l J1a1>1>eo (of course
lia11·i11g '\',:ar <li.st<.>rti <.>n~) if tl1c:K: L\\'<> rl1:1lh1os continue.
also the year in which t.hc United Statt:s al>andoned che gold stand-
ard. :\>Jany reasons hav(; been cited for the economic upturn which
bei:ran aL th;1t time. Rut if we record the fac.L• only as seen in our
rhythms, we: may note that wlH:n tJ1(: Jong-prevailing 9-year rhythm
signakcl for a rise, the rise came. After Lhe count.ry went off gold,
prices in terms o f gold still kept on going clown. But prices in terms
of tbe dollar - " real •· prices so far as t.he i\mtric.an public was
conn:rn ed - now went up, just as a projection of the 9-year rhythm
could have forecast (see Fig. 10, borrowed from \Var.-cn and
Pearson) .
It has of course been argued that Llie rise in prices was" caused "
[ 9i l
CYCLES
by the departure from gold; and in so far as our price level is in-
flucuccd by the price of imported commodities, that thesis is de-
[ensible. But whether departure frmn t11c gold standard then ac;tco
causally for the whole subsequent uptrend in prices is exceedingly
problematical.
Index;--,..-,--,--,--,--,--,--.--.--.---..---.-.-,--,--,--,--,-...,.....,..-,..-,..-,
290
240 -
Currency
190
l40
90
Cotd_>- _,...---
oT I I I I I I 1 I I l I I I I f I ! I I I I I I T
1913 1915 1917 1919 1921 1923 J92S 1927 1929 1931 1933 1935 1937
l'IG. 10. Gow I•Juc.~~ ,,.-.i1> Cu1<1<E.'JCY
!'rices of Corty basic co1nmoditics in tJu:: Uuilc(l ."itatcs in o irrcr1c}' ar1(l in

goJd, 1913- 1936 (after '\'\'arrcn anti Pcarsvo). 191t.r-191.1 = 100.
This rise of prices fro1n 1933 on, in harmony with the indi<:at<~i1
9-yca1· pattern (the rise can possibly be dated from an earlier turr.
in 1932), was doubtless the ev(:nt that inspired Professor SchUTnpe-
ler's observation that a process running its course since at least the
sixteenth century could hardly have <:ome to a stop on l\1arch 4,
· ~~3·
If tl1is \.'t-rges 011 a cleter1l1i1li.s111 di.'itl1rbi1lg L(> so111e i eaclers, ic is
wonh recalling-parenthetically- that all of us are born into
a highly determined fonn of socicr.y in which everyone adjusts
himself constantly to prevailing social patterns or goes down trying.
(The gl'eatesl ability is adaptability, ;1s Samuel Crothers once said.)
l'urther, it is dca1· that there is re<)m in these charts for ;di the free-
<lom and accident of choice any reader would like lo have: implied.
Consider, (or instance, the wartime peaks in comn1odity prices. So
[g8]
TllF. g-YEAR RHYTHM
far as ou1 patt(·rn is nincen1cd, then; is nothing there that c;11ls for
a wanimc peak at any time, or anywhere. !'\or is there anything
that Lelis us if and when war will br:-ak 0111. Nor, when it breaks,
have we any way of measu ring in advance the int.ensil}' of the lever-
age which this force will exert 0 11 the price level, once the force has
been inttoduccd into the pau tnl.
Davis, by measurinll the in tensity of war's effect o n commodity
prices, has worked out an inLcresting equation for com pari11g such
a .>kirn\i)l1 as Ollt- Spa1l.ish-A1ncrican \\:;ir witl l our erron i 11 \ \lorld
\Var I. By comparison wi Lh 1917- 191 8, the ;1ffair of 1898 was no
war at aJI. • But these are mauers which ma}' be seen on ly after the
event, by noting the dcviaLions of actual commodity prices from
levels callc:d for by long-tenn patterns. As previo usly noted, we
lta\•e nc> '\Va}· (>f kr10,..·ing ill 3d,.-ancc ,,·l1t:11 '''Ctr is going to break.
And we can only reaso n by analogy in j u dging what fin a l effect a
war might have.
There is nothing in the first two .?4·year rhythms in our nation's
histoi;'. for instance, to suggest that t11e third one would see two
wallime peaks in commodity price '.evels. There does n ot seem ttJ
ha,•e beC'n a11y patter11 <le111antlit1g e' 'CI\ one ,..·ar iriAation . [f 011r
peak of 1!)19 did ba,·c lWO e-~rlier precedents in the peaks Of 181,f
and 186,), occuni.ng aL nearly corrcsponcling poin ts on the cyck,
there appears nothing lo suggcist that a wartime peak was boun d co
come th en. All one could have done in 1914, when world war cli1l
hreak, 1>as to have n·asoned by a:1a logy co the conclusion that
wurlrl w:ir at that poinL 0£ the cycle, where two major 1hyt111ns were
moving up together, would resu lt in aver)' large leverage on prices.
It did.
Quite sim ilarly, and again only by analogy, it was possible to
reason in 1939 chat at this point of time - when I.be !i1·year rhythm
was moving downward - the ;1dvent of war would not have the
pric:c lev<·rage of years Stich as '!)14-• !Jl9• when war and two up-
si>eeping cydes accompanied each other.
• AnalyJis of Ecor1omie 'rime Serie.1, p. 554.
[99]
CYCLES
1l would seem that such an interpretation, viewed from the van-

tage point of 1946, had justified i tself. Tt must of course be staterl
Ratly that analogy can always be dangerous. ll must always be used
warily, with full knowledge of the dangers.
Before we leave this subject of wartime commodity prices for
a while, ano ther warning, by way of qualification of the analysis, is
wonh registering. Plc-.isc note, on Fig. 9. r.he last 9-year rhythm
which starts rising on Llie declining 51-ycar rhythm in 1941 or 1942.
This shorter rloytl1111, due to reach a peak in rnid-1916. represenL<.-d
a strong rhythm ic force. Though it was not reinforced by a rising
54-year cycle, to act in conjunction with it, it was given incalculabl e
war leverage as a partner in its i942-1 94fi rise. T he combined
result. - as our 9-year rhythm reached for its peak - wa.~ a very
heady inAation.
By analogy we could have reasoned in advance that the 1916
inflation would not reach 1919 heights, because of its position on
the 54-year wave. But we had no ready way of estimating accurately
in advance whether the inflationary Jc,,cr•gc of \\lorld \ \-'ar II
would he greater than in \\Torld \Var I, or by how much.
Jn the forties, of course, there were vastly greater quantities of
money in circulaliOtl, stc1r11ning fron1 '\'arti111e gt>"'cmment defic its
that. far exceeded those connected with \\1orld \\l:tr I. nut statistical
comparisons of this kind threw small advance light o n comparable
leverage, in view of such factors as the b>Tcater will-to-save on the
part of the people, Llle tendenc.y of many to hoard money and
uct.lits a.s a store of value, and the abolition of the Cree market
cffectccl under OPA and 01.her government conll'OI<. Such variables
made dillicult a.ny predict.ion based o n estimates of the comparable
war leverage factors during the two periods.•
13ut b'Tanting the hypothesis of much. greater war leverage behind
prices in the larger war of the fonies, one could still ask in 19.16
whether or no t it wou ld prove great enough to compensate for the
• For !urther discussion of ways ro meaJu.tt tJ1c distortions of war, sec Cha~
tcr XIV.
( 100 J
TllF 9 - VF.A R kHYTH~I
lack of any price leverage in the declining YJ·) car rhythm. If so,
then one might ex pen prices rising on the 9-year rhythm to reach
a peak matchinf.\" or cxcrrding that of 1919. Jf not, then the price
level cou ld still be expected to reach a substan tial t\146· '!H7 l'eak,
at the <.TCst of the 9-year w<ive pattern - though probablj• well
unilcr the peak of 1919 wl1cn the 9·year and 54-year rhythms had
both been rising in ninjunction .
Our patterns could 1,>ive us no definitive ad vance a nswer concern·
Ing the heil(htof chc p<:al..1v Loe 1c-d<.IJc,J.T ltct could onl y indicate chc
approximacc t iming. But it se<.·mcd safe, even from the early van·
cage point of '!Ht, to assume that a very suh;tantial 'l· to 5-year
rise in comn1odi ty prircs and stock market kvels lay ahead. The
course of the !)-year 1hythm ma<le that fnrecast ckarly possible.
Those wloo made their pl;uis accordingly, on the basis of the l)·year
rhythm's projection. profited well.
ln making jttdRments as to timing, one fact a lwa)S must be
unclersn>red. The 9·year 1hythm, and also the other economic
rhythms we are to consider, man ifest themselves f!Togn:ssively in
v~rious sons of husint:ss activity and prices. For example, note
fig. 7 in this Chapter ai:a in . And compare, in Chapter VIII, Figs.
3 <ind 5·
from 194fJ--1947 onward, the 9-year rhythm, as well as the 5'1'
year patten1, is due LO be rca(:hiug for a bottom around 19.~1-1952.
\Vhetlicr, in declining after their excess wartime inflation, whole-
sale prices ancl stuck mark<'t values may temporarily fo.11 below the
tren(J )ine is anolhC.'T t(Ue.)tiOtl the }»(t{'Tfl~ tin nnl ~n~\\'CT. R ut past
ohsen-ations tend co >how that wartime di.nortion in prices above
the trend sugge.•L~ some compensation below it that can be expected
ultimately- if precedent prevails.
( I OI j
VI I I
The 3 ~-Year Rh_,J tlznz 1
LEARJ, Y TNOICATED in the Feeler;,) Reserve Index of lnclustrial

C Production, in pig iron procl11ction, in common stock prices,
in scores of other product.ion and price series, runs a rhythm that
has bec:n variously estin1atcd at from !IC) or 40 lo 4.u 1nonths in
length."' It would seem tliat -to.8 months is t.h<~ kngth that provides
the closest fit for many series from 1900 10 dai:e, bnt. the true length
has as yet not been clcter mint:cl for certain. There may be two
rhythmic tendencies, for instance - one slightly over 42 months,
perhaps, and one slightly Jess tf1<1n 40 months- that. combine to
produce a Huctuation with a seeming lc:11g1.h of almost 41. This
41-month rhythrn h<•s con.tinucl' t.r ue from its discovery back in
1923.
I t is comparatively easy to find rhythms in anything- even ran-
.. PJ'l°tfe.'>SOr \\>~. T.. CrttTI,, if'I 1 9~3. sllO\.\'Cd a rhyth1n or ap1)f()Xi fn at~ J~· .jC)
rnor1tJ\s in rnontl•ly commercial paper rates i 11 NeY; )'()rk frorri 1$66 to 19:tt.
Joscpl1 Kltchi11 arottnd the sa111e time f<tulld cJ1c :)lime rhythm i11 batik clearings.
\<.:holts:tle ]>rices and inLerest tales, bc1th i1t tl1c Ur1itcd Scates arid i11 Great Brit·
ain. Chapin lioski11.s irt early 1935, as a rcsttlt of a tifnc-c;hart :(r1aly~is of pig
iron procJ11c.tion and rna1))' otl1er eOOJ)()tui<; seric;;s, <X>nclucled 1lie cycle ltad aa.
a1>1•r·()xlmatc 40..5·n:1onch J)Crioc.ticity. Later he came LO believe tile lc;ngtlt to l)c
41.c> months.
Pro{cssor Sch110lJ)t:Ler's a11a lysis of five ..\merican S}'Sle111~1ti c series gave a
roeati cJur:Ation of 42.05 mo11tl)S. i\rmsLf()t)f; c1( 1J)c n(:ll Telephone Company
(lc;ccrmined the lt:t1gtl1, in general 1Jusir1Css con(litions in the U11ited States sinc.;e
i885, to be 10.78 months.
[ 102]
THF. 3~-YE AR RHYTHM
dom numbers - although, o f course, s11d1 rhi• thms do not occur

with enough 1·cgularity or with enough repetition to justify bel ief
in their reality. NevcrtheleM, it is true that rhythms in almost any-
thing can be fo11rnl after the event. If, ho wever, the n11mhers are
rnndom and the rhyth ms arc merely r ead into thcrn, 1.hey will not
continue as the ~ries continues to unfold. For example, in dealing
c:irils face up, one mii:,..f1t get a black and a red. a blacl:. and a red, a
black and a red, ten times in a row. This is a rhythm of unusual
rcgl1 l;1rity, bt1l conc(;:i\':il>I)' tfle res11lt <>f c l1:"1.nc:(;' .lf, '''''"'P\l(:r. f1a\~i11g
rnadc this observation, we fiiul 1.hat the ncxr ten cards also come up
bla('k and red. we t·an begin to suspect that the cards have been
stad..cd, and that there is a re;il rhythn1, sou1chow incorporated in
tl1c arrangemcr11..
\Vid1 the rhythm that concerns us here, as wiLb m;my other
rhytJ1ms, thousands of hours o f labor m111t doubtless srill be spent
before \•te can maln«1i11 111c>rc ll1ar1 tentative co1lclt1sions rL-garding
periods. If now w~ call lhis rhythm one o( 11} years, it is primarily
for convenience of reference. Professor Sd1111npe1.er likes to call
it the Kilchin C)de, in recognition of one of the lint disro\'ercrs,
and obsen·es that " the .K itchins can he best oh,en•ed, Cor all coun-
tries, on the chart of rates of changes."
A chaning of the Fed cm 1 Reserve I ndex o( Industrial Produc-
tion. given in Fig. 1 with" 41-mon1h pa11e n1 added, shows how, o n
an artual f>Yoductiun chart, the rhythm concealed in tbe figure&
rtlrl)' elude.: tlte ~\·c.:ragc businessr11an. T r11c, tl1ere is mt1c;~1 c"·icle11ce
of rhythm here, but the: existence of the pattern is not obvious
until it is rev•·aled by tlie rate of diange tl'rlmique. 1'o sc:c 1.his, look
at Fig. 2, where these same prod uction figures arc graphed for the
rale of change. l·or an example of the same kind of an:ilysis applied
to pig iron production, see Fii;s. 3- 4.
A ra te-of-d1ange graph is based on a simple calculation. Here,
i11>teail of plo tting the actual tonnage fi!,'llres, the tonnage for each
montb is plotted in terms of its percentage of the figure for a
previous fixed interval. Fo1· inst<Ul(:e, on the basis of a o ne-year
[ 103)
CYCLF.S
fndex
2so ~~-~~~~~~~~~~~~~~~~~~~~~~~~~~~
22$ n
200
ISO INDUSTRIAL
12>
100
n
1, 1\ ,, /,
I '
I / ' I I '\ II ''· II ' \
'''
so
\
'/
I
\
v
I
I \
"' '
' '\ / I
'.
"• • •
•
• •
J902 1904 1906 1908 l9JO Ult 19!4 1916 1918 1m 1922 1924 l9Zb
Fit:. ), 'J Ill :,l~~- \'f-\Jt }{H\' 11 fM

111e ;t~~·yc;tr rlt)'tlln' in chc Federa l R.c ~r·n•r Index .,1 l11d11s1ria) Produc:ti01),
.J ou uary 1919 - DC(c·1nh,·r 1945. J\ rt·~ulnr 41 -1no1ltJ1 t:yl'lc hn' been acldctl ;illll
l'rnj"rtcd tlu·ou,,;f1 ••Jl)<f, co indic:.itc :tpproll.itnalt:ly 'o\·h:tt \'r'1ll happe11 ii the
th)·rl\llt (OOti11u1!'!. (A£ce1 Chapin H.o\kirl!->.) Rolin fra/('.
in1e1'\·al we would comp•re January wirh the previous January,

Fel>ruary with Lhc previous Febru:iry, etc. This uH:ihod eliminaLCS
t~la lt)' of Lhe .~)1e)1't~[crn1 \'ariatiot•S, 111ini1nizes lor 1~·lcrn1 S\iit1gs. ar1tl
t:11allles tis t<• ~c<· 111ore clearl)' the \."ilrlation <,f tc1t•cli11111 le11gtf1. l't!r·
~cntaRe relationships of this kind a1e familiar 1;nough. If we $3Y
rJlJ t l>11sir1ess tJ1i~ 111t1r1tb is 10 11cr t·<:r1t abo\ie bit ... l11<-ss a }'ear ago,
'i.'e a1·c: usi11g t11e sat11c S'JTt of co11i1.>ari~c>r1.
Jn Fig-. 4 che ag1ccr11<•11t of the :1r.1 11<d paucrn 1>ith rli e ide•I wave
is rather astonislting. with the majority of rh e highs and lows com·
""C..1
200~~~~~~~~~~~~~~~~~~~~~~~~~
J7!1
1.f>O
.,., ..., rn
·~
I ft ..-,,
INDUSTRIAL
RAlt. OF
125
100
'
'
.. ,. •
li902 1904 1906 1908 1910 1912 1914 1916 1!>18 1!120 1922 19'24 J!l'26
Frc:. 2. RA.'Ji; OF c:11A:-1"r.t

'J'hc· :{~-year rhytl1rn in the raL(' <•f cl1ange or lJ ..'i irulustrin.l pro<luc:tion,
J;1nuary 1920 -0c·cc..mbcr 1945. 1-l('rc. nrt dui te<l mu, Jug pcrcentago ol the
( 10.1 ]
THE 32-YEAR RHYTHM
lndeot
r-r-,.......,r-i-r-r-r-r--.-..---.--.-.-;;:n"""-r-r--r--.-.-.-.-..-r-i250
m tn 225
200
..
PRODUCTION J50
125
JOO
,,. .\\ />\ ,•\ /\\ - 75
',, ... //
/
\ ,_,..,,"
/
\, ,/ /
/
\ /' ' so
•" v v ..,
• "
.. • • •
1928 1930 'J932 1934 1936 1938 19;10 194t 1944 J946 l948 1950 1952 1954
IN 1NDUS1'RIAL PN.oDU(7r10N
·r·he rcgt1J~t r ·1 t-roonll1 cycle ";as c.Iete.rmi1\<!(l a11d i)rojecLed by J\.1r• .Hoskins
e.arl~· it\1938. ''l'he elates ((>r che three r.cxt cn~\1i11g f1lghs of tl1e projected
41·mor1th cycle v;oulcl IJc· :\1n·jJ 1947. Se plembcr 1950, anc1 Fcbro;iry 1954.. TJ1e
tW1) 11exc lo'"'-s of tl1e same cycli.: '''<>olc.I be~ June 194K a11d NO\'Ctnber 1951.
ing pretty closely within the fran1es al top and bottom that indicate
majority performance. And the point:; where complete agreement
is ];1c~kll1g <-lre as significa11t for 011r pt1rpc.)se as tli<:: c>r.f1ers. Notice,
fol' instance, the trough reached in 1908. It is delayed; it falls out-
side the pat.tern. Some force has entered the picture here which
our 3~-year pattern does not '1Ccount for, and the unknown force
is distorting the wave. But now no1ice the next trough, at the end
of JgJo. J-kre our figure~ have snapped right back into t.hc pat-
tern, just as if no previous distorlion had taken place. \·Vhatever
Per Cent
~~~~~~~~~~~~~~~~~~~~~~~~--...--...-200
175
"'\\'''
l"""T"I
PRODUCTION 150
,, ,, J25
,, \
\/
I
I
I
/
,,
I
I '
\
\
\i
I
I
I
,,
''\,
\
''
I
I
I
,, "
JOO
·J \1 \} ~1 \/ 75
1928
•
1930 1932 1934 1936 1938 1940 194% 1944 1946 1948 1950 1952 1954
w .. 50
IN J~DtJSTRIAL PRc)f.>CCTION
Federal Reserve Index of Production Figtircs slto,v11 in Fig. 1. A regular
41 ~month cyde I1as bee11 added. (After Chapin Hoski11s.) Rali(1 stale.
[ J05]
CYCLES
s.ooo
•.ooo ..... ..... ..... ..... .... .... m
•
.,..,
•
f
~
S,000
I
"•
I ' '·
I'•' '
•i
I
I
' I
''\ I
J 2.,000
I
I
I
'I
'
(
(
I
I
I
\ I
I
I ;
I
I
I
0
I
I \
' I
I
I
\,I
, ,
I
'\ I
1/
(
1 1.000
PIG IRON
500 • '-'-' ..,., '-'-' ..... u.. LU '-'-'
1902 1904 1906 1908 1910 1912 1914 1915 1918 1920 1922 1924 1926
Fie. 5. TH< 5\4-YE-'• R.l<YTH.'t

Pig-iron production in the United Staie:..-., Jnr,ual"y 1901-Fcbn1ary 1g.ii6.
A regular 41-month cycle ha.< beeu ad!led and proj<cted through 19:;4, to indi-
ate i1t ~11eral terms "'Tiat \Vill happen if the 1l•ytl1n1 ro111ir1t1cs.
~Jlt t 1egular 41-Jnc•t1th cycle was d<termi11cd :11t<I 1•mjcacd by }.fr. Hoskins
Per Cent
sso
soo
250 ,,, ....... ....... ,.,, .... ....... ~
200 -
' I '
150 I' fI ' 1\ '\ '•
I \
\ \
I ' I
I
' f
\ \
I I
'I I '
100 ' '' I I '' I
I
I
\
l
I
'
I
I ,'
I '\
so
I'
"'
\' y'
I
'
PIG IRON
2$
w ~ w .... w ~
RATE OF
1902 1904 1906 1908' 1910 1912 1914 1916 1918 1920 1922 1924 1926
Frc. 4. R."n oF CHANGE
~fovin~ perccnragei (rate of c::l1a11gc) of pig-iron produclio1l, Ja11uary igo1-

Fthruary 1945, sho~·ing tlle S!4·yca.r rltyth1n.
1'fte rJlart sl1ot\>"S chc percentage 1J>oi1 1l1c ~1c;111a1 prod1tclio11 f<>r (:;:id• mc>ntl1
ii' of 1l1c corrcspondi11g a)l.OnLh one year before. A reRular .1 •·mor1th cycle h3.:"
l)ccn addod and projectecl through 1954. to indicate i1l. s:;e11era l t.crms bo\\1 the
[ 106)
THE 3k-YE AR RH YTll:-1
t. I
'\ "1
~
/l
1,
f 3,000 !
,,
,,
1,
I I ' 'I '• , l 1\

,1
il
,,
1\
,.
0
~
I
I
'I I I
I
I '
I
I 1 \ ' 'I ,I ·- 2,000.:;a
' ' I I
I
I ' I
I
\ I
I
\ ,'
' I
\I
I
\,1 \1' 'i
I
·'' I
I I
I
' \
v I
\I
I
v '/ i
l,000 J
l tlt 1'30 1932 1"4 1936 1938 JMO 1.9U 194' 1946 1948 1950 19SZ 19$4
IN' PIG TRON PRODUCTION
ca.rl)' in 1938. 1'hc 4.latC':i <1( the tbrc.:(; next ct1)uiu~ highs of the regitlar ti• ·
111onlh idea l t!Ja"e are j1Jl)' 1947. Dcc.:c:rubel' 1!)!><>, au(l !\·lay ' !)5•1· .EJ1&t1ing
Jo"'' j1l the rcgul:ir 41°montf1 idea.I v.·:j"e are ScJ>tt·ruber 1948. :111d t 'cbn1ar)'
1951. (Af1er Ch•pin Host.ins.) /IJitio scok.
Per cent
350
• 300
..., ,,.. rn rn rn ,..,., .,, 250
'
•'
" '' '
,,
1\ '" ' I
\
,,
I
•'
\
,,'
'' \ \ '"
I
•
\
'\
200
ISO
'' I I '' ' \ ' \
'' ' \ ' \
' I\ ' I'

\
\
\
' '' '
\
' \
'' '' 100
'I '
I \ I
•I '' ' '

'
I ' I
I ' '' '
'' I'
\:'
' I I I
I ' ' I' ' \' \ I I I
• "
?
,1
( If " i' v 50
PRODUCTION
CHANGE
"' P1r. lRo-111 f'Rooucr10N

movin~ pcrc<:rll OAC ''rill bcha\'\: the pa:,t. ir Ll1e rJ1ytJ\n1 continues acting ~1 $ i11
ll tviJJ be no•c.·tl tJ1at Lhe l1jgJt pt•in.:s in rhc rt-guJar 11-1nonlh cyc;:Je, f'ittccl
10 the ra~f<hangc dar.1, ate rc;id1ed 14 rrMUllhs prior to Ille hi,gli i1l the
rcg11lar 4t·m1ln1l1 t)cle 6tted to tht.: actual d:ira. Tlte lot\"S lcatl b) two mon1Ju,.
(After Chapin HO$kins.) Ratio ualc.
[ 107]
C:VCl.F.S
the force lhat distoncd 1he pre' ious rhy1.hm. ii has been d issipa1cd.
and we lo:ffe additional evidence lhat our paue1·n is real.
There ar<: similar distortions in relation t<.> lhe ideal pattern I.hat
called for a continuous downturn [rom t\1'9 to 192 1, and a bo1tom
in 193 1. Actual perfonnance of 1he real curve was dil.u1ncd so tl1ac
there was an extra Wa\'c in i919- 1920-a rliswnion that followed
tl1e i \1·111istice. Tl1<:re \\'as a1sc> ;1 c!ela}· itl tl1c LiJ11i11gof 1l1c bc>tturr1 i11
1932. But thereafter, following <.ach distonion, tl1e anunl pedonn-
3.rlce ag.iin f:itls inLO li11c witll tJ1c i<leal ,,·ave.
In Fig. 4. t11en, we ha\•c a d~inonsi.ration of the way in which
rhythms act, which is even mon' important. than i{ Fig. 4 showed
us a p<:rfeccly pat1<'Tned pcrfonrance. }'or - as we shall sec- the
performance on this chart show'! us rhythms intemc1i11g, wi1h, of
cour·sc. c hance distortions as well. Study of such interartion is the
funcbun ental task of Llw<c today who are analyzing 1hc rhythmic
pauerns in our social ancl economic organhm. '!'he)' <an be studied
bo1h in the ratc-of-<:hange figure! and in the actual figur<"s of which
1he rate oC change is lhc coun1e·:part. I t is lO he hoped that some
of our grca1 basic indu,trics, like sied, which a ffen with rhciropcra·
1ions t11c life of our whole economy, and which have al s1akc such
f.,'TCat in,,csrr11<~nls 111 tl1is eco1l<)IJI)'· t\.·ill e"·er1tt1all}' lc·11<I t:11coL11ag('·
mcnt. ;ind support to research of this order.
S1cc1 is, o[ course, an in,·olvcd irulustry. To es1;iblish within any
degree of accuracy the periods llf 1.he many rh}t11111s which affect
it.s operations would n:quiTe <·xtcnsive rc,carch. Not. only is tlic
in<llJSLl'}' cvi1le11r.ing- t~•e 5·1-}'t>:lr, !}' }'4";1r. 4'ncl 3!~~·t·;11· 1'J1yc.l)tr•s '"c
have been cradna: " it also shows C\'idenc.- of the 18-•e~r rh\, chm we .
sl1alJ sl1<>I tf}t meet operating in otlic~r i1nporta11t i11flt1st1it.-s; ;111<1
there is funh<:r eviden~e that it involve~ rhythms of :1n)tmd 6, 7},
and 15 ycar.1, along wit.11 shorter ri1ythms o( 6, 9, 24. anll '.13 months,
Regar ii less of the amount of worl yet 10 be done on 1he p1ohk111,
1J1c cursory data i.(· already ha\'C in hanrl arc useful enough to
suggest that a search for mar~ w(111ld be well wananLcd. As l;ne as
rnidsunnncr of '91'1· (or in,.tauce, an iron and steel trade journal
[ 1o8]
THE 1J:\"-Y£AR RHYTHM
was debating whether the indu.<tr>'s capacity would be sufficient

to meet the postwar demand for steel. No one familiar with the
pootwar· implications of tltc various cycles, in com hi nation with the
industrial trend lin e• we have explorer(, coul<l be seriously con-
cerned with such a question until :tfter the hal(-century mark, at
the earliest.
Interestingly enough, the 3z-ycarrhythn1 which we sec in pig iron
produnion is also duplicated in scon:5 if not hundreds of other
economic phenomena, including stock prices, wht'Te peaks Lend to
come ahou t five months earlier, and valie)·• one month earlier,
than in p ig iron proclnnion.
The existence of thi; same rhythm in stock prices is provocative.
(See Figs. 5- G.) Tron is a tangible crnnmodity. a thi11i.;. ;i vi.;ihk real-
it)' that would cxisr. in some form or other if nu man had ever been
on earth, or if all men disappeared. Rut stc~ck.< and prices have a
different sort of reality. A stock ccrtilicate is a slip of paper which
depends for iL• being on states of mincl; whaLever the assets be-
hind it, only the mind can provide a linl: he1wccn them and the
paper. Furr.her, prices also have a hasic reality that is purely mental,
for - rega1dless of whar. the figures be that prices are quoted in -
the ligures can only be an index to a mental attitude focused as
desire. Stock prices rcllect, far more~ than some t:otnJHentators ever
pauoe Lo realize, the psyd1ological states prevailing in a cornmu-
11.ity.
The stock market which bas rhytluns is not located in \\Tall
Stn!el. It is only an 1•xp1 ession of ideas, and ide:is arc expressions
o( people. Tf there is rhythm in the action of tlH? market, it obvi-
ously must spring frorn tin: fact that the rhythm is flowing in the
human organism. In studying th~ market we arc studying men, and
studying them in one of the fundamental aspects of the ir nature.
As a study for exploration, the quest.ions regarding the nature of
reality and its relation to the mind !lave hcen le(L by the ec;ono111i.;ts
la1-gcly to the philosophers, and perhaps to the psychologists. Ilut,
in an era where all the sciences are drawing closer together, the
[ 1og]
CYCLES
}nd1•.--~~
. ~~.--~..-~~~~~.--.--~..--r~~~~.--.--~~~~~~~
...... m
300
200
no ,1'\\
?20
I
t\
\\;1
,1 ·
A\
\v \'I, \\ , , , \/ ,
/\
i'\
\
t\
,'\
'v'I
/,/\I'
', I
v \
1'', ,\ ;\
80
80
COMMON
•O ..... LI.J ..... "-' w u...
20
1906 19® 1910 1912. 1~14 1916 1918 1920 1922 1924 192&
FlG. 5. Ttr:r. 3~ ·'1'•.AR R H\'TI-1\f

Do\'"·J 011n indw11i•1l rommo11 ~It pric~. January 1~1-t.faTCh 19-16.
(Scale prior LO 1914 :tdju~tM-1 to confor1n 10 sttbsequc111 l):tsis.) :\ tegul;,r
41 .u a.>11Lh cycle has bee"u acf(fcd a.11d 1>r<•jt·c1c·d tlU"01.1glJ '!J;;1. co i ndicale "-J>-
p r(>,..jruatcly '''llal \viii h :tppc11 if tJic· r P1 y1l1r11 COJ)ti11uc$.
'This regular 41 -1110111!1 q•cle, like Lhc C)•cJc jtl l\ig Jror' i>rodL\CLi<)rt a11d tl1e
2$0
200 ......, ....., ..... ....., ,.,.., ..... ....., ,.,...,

140 I \
,, •\
:·
I \ I I
I
100
\
'
' ''
!
/l
,,
v
._.... .......
COMMON
RATE OF
20,__-!,,~-',~-.J,',...;'--J.,.,~.J.....l..
' --'-,..,J--'C...'-+,-,!-...L...J.--1--''--"--'--..1.,.,,.1.__,L~--'
19o:z 1904 1906 l908 1910 1912 i ttc t9t& t9l8 im 11n 19:2• 1m
Fir;. tt. 'R.ATY. OF CHAr\Glt

T Jtc 3~~)•ear l'l)ytltm ir1 the rate or l'.hiltlgc (If c:ommOJl f.lOC::k prices. i ll le11ll$
of llll)\'i11i; yu::rccntagcs or
Lile l)Ov•·Jon('$ lndt1strial :'\V<.TaAe~ l'ofUJ, \'Jl in Chart ,;.
l .tkc ('harts 2 ancl 1 1 1his chart shO\\'$ Ll1e ()'!rtc·nt.-ge t11ott the a c.cual prices
of each iooulll are or ct1c corresponding 1tu111tJ1 one year lx:.£or~
[ 110]
300
,\
200
,,'' '\ 160
I
80
60
STOCKS
....... •o
L-L.JL.J....l.-L--1.....J......L.L_..1-L.JL.J---L-'---'--'--'--'-'--'-''--'....I.-'---'--''°
1928 1930 1932 J9l4 1936 1038 1940 1942 1944 10"1G 1943 1950 1952 19$4
m (;o)J't'ON Sru<::K PRKXs

cy<'le in the-: Federal Rc~e1'\'C- lnd<..·x f()r 1ncl~1strit'll Prodtt<tion. ~·as <lttt:r1r1incd
a11d 1•rojccll·cl h}' l\·tr. lf<>Ski11s c::1rly j11 · ~~8.
· 1he <la1e1 projt"Ctrd £or the three ncxl ensuing J1igfil in d1c T~ular 41 ·UH>ttth
eye le are Frl1ruary '!J-17. July l!)~O. and nec.-embcr 1953. !be t"•t• ncx( IO\\"$ :.re
1\ugusc 1918 n11d J :i11u:u·y 195!1'. ( Afccr Chapin J.fcn;kins.) R.tlli11 St..·ale.
~~~~~~~~~~~~~~~~~~~~~~~~~>SO
I
I
..
\
\ I
.
I\
\
200
140
' \ I \\
1 I
·--'"-'-
\ I
--'=.--i----''..i 1'00
\ I
I I
1•
"
{
60
•o
'-'-'
STOCKS
CHANCE
, I t 20
1028 1930 1032 1934 1936 1935 1940 10"2 •••• 1946 1948 1950 1952 1954
JN Co1'otOS .~·roCJt t•k1CF....:;

.1\ reg\1Jar 4J·monch c;yrlc flnJ bc:1.:n ;.ddcd n11d projer.ltd throligta 1954 1u
in<llc;1le apy>1(•Xi1nn.tcJy hO\\' cite 1novi11g pcrcc111ages \vil l behave if the rllythm
OOfllinttC.'i.
(1\(ter CltJJ'in Host.ins.) llalioscol~.
[ 11 1 J
CYCLES
economist has mud1 10 gain b~· building on the a.-.-umulated lore

of other ficlcls of study he has often been too husy to investii:;ate
personally. \Ve ma)' rcco1,'11ize. at least, that what we see in fig. 5
is primaril)' - i11 at least one ~cnsc - a community psychological
chart.
Figure 6 shows t11e rnt.c of change in th e Dow-Jones imlc:x of
industrial common stock prices, together with the tracing of an
ideal 31-· year (i.e .. 11 -month) wave. Note, firs1, 1he distonions in
the • 932- 19'3 period, when both trough and <rest were dclay<:d.
Herc we find a delay <)llit.e parallel to the lag~ in the pig iron pro-
ducr.ion chart. Tile rircumscances also <tre parallel.
On the whole, the irnpression this chart kaves us with is
on e of amazing regularity, with th e majoritr or tro11ghs and cresrs
alike do.1cly ;1ssociated with the ideal pattern as .1hown by rJie dotted
lines. lVh cre the departures from the pattern do >how n p. we have
other data that go far lo explain the deviations in tenns of inter·
acting waves of other 1><·riodicities. And inwfar as rJ1esc other wa' <'S
are rhythmic, they are presumptively predicwhlc.
One point of historic interest may he mentioned in connection
with th e divergence from pattcm at the end of ' 026. l\ote that
around the hq; inning of •92'7 the a(·11rnl wave ha lts in mid-air, as
it were, while on itswaydown. Jt fai ls tocontin tl(: on 10 the !Tough ,
hut starts mounting - nearly a year earlier than the ickal pa ttern
called for- to the p<·:ik that we know wa.' 1cad1t•cl in 1929.
The other forces meeting al conflucnC'C in this period pro\'cd over-
powering in 19•7 so far as the short !!~·year wa\e w:i.s concernccl.
Thus in early 1927. instead of continuing downward to its year-end
trough, as would have h;ippcned in following it. icleal pattern, the
marker. sta11.ed n1oun1ing. And it kept climbing un1il it had almost
reached its par.t<:-rnccl c rest in 1929.
A rather similar divergence in market action from the pattern
called for by the 3tyear wave occurred in 1914. At th;1t time, in·
steacl of falling into a trough, as the projec:tccl pattern \vould have
forecast, stock market averages showed buoyancy. \ Ve may postu·
[ 11 2 ]
l'llt: 3l YEAR RRY T H ) I
late that in 191.1. as i11 1927, it '"·as an exrraorfli rl31')· inflationary

force thac k.ept chc mal'l:et steady when che pa!tcrn ra il ed for de-
cline. Jn '!l'15· as in 1928, when the rattern again called for a rise,
the inAationary forces still present. made that rise almost mc:teork.
This rise continued in the stock market, with l'<:w i1llcrruptions,
un1il the tum-down in t he iclcal 9-year wave came in n1i<l-1946.
The ensuing break in 1h<" market in August-Sept<'mbcr is history.
The 1946 break came ju~t aft"r d1e 9-year wave started re~eding.
although the shoner ~~·year wave was still mounting.
'Vlien tlu: time comes in early 1947, as then indicated by the
projcc;tcd pattern, for the !!~·year wave to turn dtn>n again, and
add its down-thrust to that of the 9-year rhythm. observl-rs will be
interested in wal.<:hing lo see whether post·l!JU!) history is approxi-
mately paralleled.
It has alrcacly been •talcd that in the early forties our d1arts en·
ahkd one l.o foresee a very substantial and heady rise in stocks and
commoditk-s into 1916. Those who look hack from the vantage
point of subsequent history may hav~ occasion to note dtat the in·
ffation o[ the lorries only added to the tolling of the hell when 1950
came. For- to repeat - with 1.hc !l'rear rhythm turning down in
194G, and the 31-y,;;ir rhythm schedu:cd to turn down in 194·7, the
markn will 1.hc:n he left with no dominant rhythm 1.0 suppol'l i l -
at least, none that we huvc I.raced.
\\·'hate\•Cr tlic war le\'~rage ClTI'tOlJJlts to 11p to 1917, \\'C 1na)' postu·
late tentatively that i1.s inRationary force can hardly act after that
year, when there will be no major market rhythm to act through.
Many volumes have, and will be, wriuen on the $1thj1'<'t of in-
terpreting stock market panerns. Here only the mo.st cursory illus..
tratic>11s c:ar• l>c.· giverl <>f lite \Va}' in ,,,-J1ich tl1c 1naj1>r economic
rhythms lllay be sec:n a1. work in the market's operation. Our pat-
tern~ show up clearly here, as d~cwhere; the !!~·)'Car wa"e is one
of the most important among them; and the incidence of such
C\'ents as ,.,·ar.s ancl , ... ar i11f1atio11, tl1ougll iL may c)i~turt tl1c ,,.a,,e~
does not fundamentally change the pattern.
[ 113]
CYCLES
Out of 1nore than ;;oo different kind.~ of economic series that hav<
been analyzed by various research workers, something more than
half show ftuc:tuations which are of che 3t-year interv;i l type. ~'vlany
o( these can serve as extrcrnelv. us<:ful baron1eters, not onlv' in t.he
stock market but in rn;iny otlier economic organizacions- particu-
larly when the shorter waves are viewed in connection with the
longer rhythms, and what we have learned about their working is
applied to the interpretation.
One <!anger should he emphasized. This, scrangely enough. is
the danger of traditional " thought." The average businessman
who uses rhythmic patl.erns to guide his planning is using ;ul in-
strument so foreign to his way! of thi111'ing about his work chat he
may kt fault.y judgnients interfere w.i th accunae reading. l\·fost of
us, taught to think in conventional tenns of cat1sc and elfect, and
trained in a system where the ecluc.at.ional wares are often cut to
the lowest corrnnon cknominator of mass intelligence, find it harcl
to remember tl1at the human mind is very finite. It can never hope
t<J gTa.s p i11 atl}' sitt1aLio11 all the ca•1s<:s tl101t 1rta}· \V<>rk tc>getl1er to
produce any given result. That is why judgments are so fu ulty when
we reason solelv on the onlinarv (:a11se-and-effect basis. Some if noL
' '
most of th<: actual causes will escape our knowledge, and tl1ose
we do take into cakulat.ion n1ay be pure assurnption. If despite t.his
we reach a correct conclusion, it is a triumph of: accident- or of
irltllitiOtl.
Awareness of economic patterns, and such knowledge as we can
gu..it1 of tl1ei1· ope-r·ation. li.elps ~1,•oid sotl).C of t.l'le pitfalls inherent
in the limitations of out ability to think accurately about ihe in-
finite world that sunounds us.
Readers willing to pursue further some o[ the problems inherent
in cause-and-effect thinking will find Chapter XII! devoted to the
subject.
IX
The 18-Year Rliythm
MF. 18-YEAI< rhythmic cyck hu been calculated t0 have an

T aver.sgc length of :uound 18~ years. IL is hardly l o be regarded
as a cloubling of the 9-)ear rhythm we have already seen in act.ion.
It shows up for the most part in an entire I)' different s.·rics of cace-
gorics. Figure 1 shows ir. in 1..he index for real escate acLiviLy in the
United States, after adj ustnH:nt for Lrencl, as compi led hy Roy
\•\'enzlick. Figure 2 shows ii. in building activity, according w the
\ ·Varren and Pearson com posite index. All building activity timing
mentioned hereafter in this Chapter refers to this particular index.
Choice of another index 111igh1..slighdy advance or retard the timing.
Ilowe,•er, no matter what cndex be 11~cd, chis 18-ycar rhythm seems
one of 1..lsc clearest, most regular patterns revealed in our economic
life.•
• Clarence D. Lor1g, Jr., \\ho bas made one o( tl-.e mosc atmsi"e scuclics or
this rhylhru, declares: ••Ar. fi.t'l cons.i<lr.r.atit>n, it may u:e1n that considerable
' 'arincion c,->Ci:;w eve1l in the lJn.ited Statt.'9 h1•11.rPPn thr <l11rati()O esti1natc'i of
vttrJc)US \\rritcrs. Act11ally. ll'lc crclcs the1nsCl\•cs v;;iry 0\'er time, :1nrl I.ht~ \'aria-
tion! (in cs1i111a tes) arc tluc tv di!Cercnccs i11 L11e period <)f 1i1ne studi01.l. The
clcJse~t approxima.ti(>t1 is J)l'Obably tlu: one--ht1J11lrL'tl·)ear study c>I' jt1hn Riggle-
man. '""hid1 places che avmge duration for the cntir-e till'. C)'CIC1 at se"·t-.11t.een
yt"":ars and (or the last tl1ree cycJcs at sli~h1ly more than eigt11t.-er1 )tlls. ..• 11
might be ~up(K:tSed that the i11dustry in c~1ch locality v.·<.c.ild hilve great ir1r!l,,·id-
ualit>'. , •. ...-\ ctuall)·, although sucl\ i11di\•i<lulllli1.y cxiscs. l<>r1g C)'Cle.s do appear
in all cl~ies <ind tl1c agrcc1ncnt of Lhe t11rrd11g 1•oints of thcM.: cycles is S:ttr-
fJrisin,gly ltigl1, t:~pecially '1.' i1 l1 rt-:t.tJect to tl1« crttll~ILs." (l~u.ilrl1ug (;yrles and
t/1,. Theory of ln•1n·t,nt!t1t, l'rir1<,;c:tc>11 Uni\.·~nity PreS&. PI>· •1;. •!i~·}
[ 115]
CYCLES
160 \
I
i· •
/I
I I I I
I I I I
I I I I I I I
I I I I I I
I I I I I
120 I
\ I I I I
\ I I I
I I I I I
100 I I
I I I I I
I l I I
I
80
'I\ I
I I
I
I I
I \
I I I I
' I
I
I I ! I
I
\
I I
I
I
I
I ,,
I ' \
60 I
17S$ 1800 1810 1830 1840 1850 1860 1870

' .
1880
Fie;. i. ~rtt£ 18%-' rl!AR "Rt1\'"t1r~r
Real est.ate acti\'it)' in the United St.ates, 1796-1946 (aflet' Roy \ ·\:enzlick i11
the lleal e.~tate 1lnal)1st). Data are for January o( eacl\ year..>\ regular 18lf3-year
C)'Clc h:is l>ccn added and projected to 1970, co shol'o· in a ge11eraJ ~·ay 'vhat \\'ill
The existence of a similar rhythm in the marriage rate (sec Fig. 3)

could readily lead to some false assumptions. It would be simple
c:nough to reason that people who get married st.art to think about
house hunting and home building, and that this results in building
projects. Or one could stare his reasoning in reverse, and end in
the >am<~ place, by saying that. building stimulates prosperity, and
, ..•lien peo1>1e feel prospc1·ous t.hcy get m:..rriecl, and wl1en married
tJ1cy huy the homes that con tractors have previously ercctccl to
attract. customers who feel prosperous because money is being spent
to erect building~ for people who will get married.
Such reason-why arguments chase their tails. Nor is this state·
n1ent of them unusual, except in btevitr. \Vhole books have been
written that incorponite sud1 "reasoning" into learned syllables.
\Ve: should be prepared to ignore reasoning of that kind, hut we
are justified in asking why our building statistics should formulate
[ 116]
11\ Rl'.AL E.sTATE _l\CJ1VM'Y
ha.ppcn if tl1i:s rhythm con1inuQ. Ratio :C<Jle. Ideal cops f:111 1888.6, 1907.0
(i .e. June 30, 1907) , 1915.3, 1(>13.6, a.nd 1g6t.o: Jowi 1897.8, 1916.t. 1934-5>
195.1.8, 19/1.!?.
r.hernsclves in such exceedingly regular rhythms. Before research

0£ the future supplies us with a definitive answer to this particular
'luery, we may haiai-cl " postulate that man's mating instinct and
man's building instinct may be aboriginalli• associated in his being.
just as they extend down the biological scale to the birds. lf both
instincts proved ro Row in tl1e race in a common rhythm, it should
not seem too surprising. Perhaps economists could well join forces
with hiologists in tl1c new kind of research such a problem suggests.
The 18-year rhythm is evident not only in real estate activity, new
!Juilding, and the marriage rate, but bas even shown up in acreage
planted to wheat (sec Fig. 4.)
As far b;ick as the eighteenth century, which is about as far as
available statistics on the subject go, we can sec r.races of this 18-year
rhytl1m at work. H. A. Shannon, for instance, published in 1934 the
results of research on the number of bricks produced in England
C111 I
C Y C L ES
150 BUILDING
,,,
l:t<l •
" !• A
I
' '
I ,, ''
I
\
I \
I
I '
, II '\ I
I
''
x
I Patl:: cf I
90
I '
I
'\
'\
........
laJ7
I
v ''\
lts7
I \ P11kof
' I
I
'
,
I
I
\ '
''
60 \ ''
\:/
\ I
''\ '
so
"'
Fu.;. t. 'J'1lK 181/s·'\'EAR R r1YT1rM

Rt1ifding activily j 1\ 1111: tt11it1:d Scates. 1850 '936· 'Tl\e: C:ornp(1site-
RiAAle1ntJ11, \ Vc111li<.k, and 120City-lndex of bu ilding :_1(;:1i\<'iC)'· { ..\ftcr ' '\rarrcn
ancl Pc-arson.) \ 'nlue of 001lst1ut,;Li<111 cc,n1.r<1cls :t \~: 11'l lc•1 l in rhirc)··seven states,
F. \\:.Dodge Corporn 1io1t. 1 92~- •9•15· ,'\ rcgtil-.r 181;3·yc:1r cycle J1as l>et:11 ;.iJcled,
ancl \ Vales from 1785 to 1849; since bricks wrl'c then taxed, records
were kepL \ Varrcn and Pearson, inteq><>laring population data,
drc'" the chart sho~-n .in Fig. 5 to illt.istratc: tl1c· J>r,Jducti<>n pattern.
A distinct 16- to 18-ycar rhythm is shown. Uuilding statistics for a
givc11 c;irc111nscribed area, suclt as a p•1rticular cil)', do 11ot alli.•ays
reveal the same rhy1hm. Glasgow, Scotlan<I, for instance, has la;id"
building cycle or around 30 years in length ~i ncc t he middle or Lhe
last century. London's cycle in rc<:.·nt times seem.1 to ha ve heen
around 2:; years. !n Hamburg, o n the 01.her hand, the rhyLhm for.
rnerly proximated the length of that in the United States (sec
Fig. 6) -
[ 118)
TUE 18 VF.AR RH YTH M
I '
8
CONSTRUCTION
A
'
11
fl
II
7
~1 'I
, 1
· I II
I\
I \
! 1
' ' \P.anicof
1 I'\
i\ t\
1,
I \ I\ 9%9 ' ;J \ I\
'
t \I I \ , ,.., \ It \
\
\ / \ 1 11 \ I
I / ,/1 \ ,'
I ' ,,
I
I
I I
I
\
\ I
/
I ' 1\ I It \
' I I
I ' I I \\ IC
f.'
\I
,,
\
\ /
f
\fI ai °'./
.,
I
J
I
rN llUILJ)lNC J\C"nvrn'
:trad projected 10 19;0 to <dtO\7.6, 1926.0, 1944·3· i962.6; low• haUw;1y between at 1843.5, i861.8. t8So.~.
1898.5. i916.B, 0935.a, 1953.5.
American cities vary as between themselves; in sornc t11e 18-year

rhythm is nor so evident as in others. Further, areas where the
rhythm is pronounced niay show a difference in timing of peak.~
arul valleys. In Florida, for instance, peaks have tended to precede
New York peaks in recent times. But on the whole, as Long pointed
out, there has been remarkahlc agreement. Here we shall he con·
cerncd only wi1.h national averages.
The length of 1.ime between peaks in over-all building activity
in the L'nitcd States bas varied frolll lO to 19 years; the 18-year span
is an average. The peak of t11c twenties was reached in 1925; thaf
of I.he forties seems to have come in late 19,12, in which wanim•
r119 J
CYCLES
42 MARRIAGES
,
I-
I
I I
I I
I
I
,
I
I
I
\ I
\ I
,;
1000 1940
l·1c. g. T 11P. 181h ~Yt-:A1 R.tty·r11111 JN ~fAJtR.IAGl!S
~{arriages per io,(')()('I ud1.1lc males in gTeater St. .L.cluis, 1886-1934. (r\ftcr
\'\fan-eil a1)1J Pear.son. l.)aca b)• Rf>y \\'t:nzljd ln the R~al 1!..ttate /1n1ilyst .) 1-\
reJo..'Ular 18 Y~rycar cycle l1a1 bc:t11 acl<lcC..
WHEAT ACREAGE
600
200
Ftc. 4· nir. 18'h·YJW< R11rm:>< IN \VHl!AT ACUM.•

Acreage pfanr<:<l to wheat in l\ew York, 1880-1930. (Airer 'Varren and
Pcanon and King.)
[ 120 l
TU£ 1S-Y£AI( RHYTIJM
year total new construction, public and private, amounted to some

13.5 billion dollars. Building costs and rentals, incidentally, t~nd
to sta)' up for periods well after building activity has started to
decline.
150
BRICKS
140
130
.120
i
~ 110
!t
..
ll
~ 100
80
70
60
Fra. :;. BRJCIC Pa.ooucr10N

Bricks produced per capk• in Englanrl an<I \Vales. 1785-18.fg (after \Vam:n
and rc:arson).
St.l'angely enough, in the face of a cyde t.haL is so regular, so clear

to the eye on any chart, and so often repeated that Lhere ~•n be no
reasonable doubt of its rea lity, few busines>men refer to it in their
operations, and neither does the home-buying public. ·rhus many
of the foreclosures of the deprc.\.Sion in the thirties were the after-
math of properties bought at top prices between 1925 and 1929. \Ve
shall perhaps see simil:u· foreclosures in the fifties on rea lty bought
at inflationary prices in the forties.
[ 12 : 1
CYC.L RS
The small Opt·ra rors are n oL Lhc o nly o nes read ily caughL in the
snare~ oft 1c n est-building urge wh~n the rhythm rises Lo its peak.
\Varren ar:d Pca1'lion have co1npilcd records to . how that, in lal'gc
cities like: Chi.:.1go and l'\ew York, most of 1.he skyscrapers have
been built ju;t at the top of building boo ms. :vrost of L11ern, in
consequence, had lo srnmble Lhroui;h a s uhscquent dt'prcssion car·
~2<00r
...be•.•~~~-r-~~~-r-~~~T'"~~~~~~~~~~~
::ONSTRUCTION IN HAMBURG
1800
I
I ' ,,
I ''\ II '
I ' '\ I
I
\
1200
I
I
\\
'I \ I
I \
\
I \
I \ I
\ I
\
I
I ' \
I '\ I
I
I \ I \ I
GOO I '\ I I
I I I
I
,
'I I
0
1880 18'0 1900 1910 1~20 1930 194-0
}.. 1u. li. 1'1:1r.111v<.: C<..1NsTRt:CIJo :-.1 ' "' Jof,\11.iRt:RG
'ft)tal t1t1n1lJGr o ( 1lC\\' builcling:; c1)nstrt1cted i11 tile city of 1-t~ull hurg. Gcr 4
:m;iny, 1885- 193,r;;.
rying vaca1:L o ffices LhaLwere ncv<'r occu pied unti l the next cycle
came along to produce ll1e n ecessary tenanLS. ~'fc:1nwhite many of
the b t.i ildj r.gs \'''''' ' 1 hrot1~h w llat :.pet:i;1liau t.all tl1c ·• wri nger."
The fi nding> a1" illustTaLed in Figs. 7 a nd 11.
Fron1 tlloc: t" r' > cJ1arts, one n1ig11t t:o 11c:l11dc tlklt i11 tlac.: tti,•cnties
\Vriglcy was o ne o f 111<: shrtwdesL operato rs in Chicago. In New
Yo1·k, one of the uesLjudges of the ma1·ket in Lhc C't rly 1.hirties seems
to hav<~ been Rocke fe ller, who huilt Rockcfo lle r CenLer almo;t at
1he bouom o f th~ building cycle, ;ind tli cn (a.1hed in as the rising
demand for sp:u;c took every square in( h availa))e in the Center
d uring the recover)' period in the late lhin ie• and the fortic.-s.
r 122 1
THE 18- YEAR RHYTHM
IAdP
''"""'-·
'#-.i.111 l,111!1<1 ,
1.••,,,. 11111,
,I"' ~tlh~711J•.
s.,.. .
0.l, N I!"
~ ..
lttliol*f Nold, ltt;lml,....b., •··•tr r1tu.,
210
ISO
s....n111,
41L6r:t
~
DV*nl'9 81J9,.
C.." Eo.el..ll)t
.....;....
..... ...
11.'I'. C.l>lld tl<:lt·
'"'''"·
-..
1'11r.ltott b:N-it.
)mt I I•
~wITT lh!oM
1~0 >WI HOIJH
11011'•.
$':1y•t$;)t.I
"'... , " font • 111,
..... ._
·-
W-'f ft~.ic:..._
uo
.......
•• ...
llllflfl
tnt>tt h •o,
111.....
60 ..,.,._
.... ....
711t1 lltllo~
CfJ '#Jlt.111,
LI,.':?,,.
....
~
MUI & (JQ•O•,
!),\.. &•.
11,11, ei..1,..
'· ••M ..."
•••
'~
0
1830 ll!SO 1870
1'011:0•(~
1800 1910 1930 ....

Flc. 7. N1(w \ 'oil&. SK\'SCR..Ji.rP.RS
Dates when cerL:tin large b\JiltlinK,s \Vere Q>Jl&tructc<1 iJl New York City aJ1d
the C.tl1npositc- Riggleman, \\1fn1Jid:., ancl 120.Cil)'·lndex of building ac-
tivity 183<>-19~6. (After \-Varren anti Pearsou.)
1nde•r~~r-~-.~~,-~--,r-~..-~~:'.: :~~;;~::'.::~~r-~:::::::::;~~~~~~
...........
... -....,....
,1111 OI llfll<
Jl!O
M1110JMotl
P><•.
.......
· - · 1"111tfll
•1 $<.inl< T1A'll1t,
rso ~ A.i.1..111 Uo1k,
,.t, Wllfl"\I' ...,.." '•
-
lit'~" , .... Sl'I~'- kl!
Clkai:I• Tiii•
",...... "'""'
120
90
60
F IG. 8. CHICAGO SK VS(:R.. \Pl!llS

Dates ~·lien
certain large buildings '~ere oortslrucrc.. '<1 j 11 Chit<'lgo aJtd Lhc
Compnisite - Riggleman. \Vtntlick. and 1tc>City-lndex of building ictiviry,
1830-1936 (after \V•rren and Pc:mon).
[ 123]
CYCLES
·rhere ha.< bel·n an obvious explanation offered for the building

of .so many of our sk yscrnpers at peak prices in 1.ime of peak activity.
•
Thal is the 1.ime when they can ordinarily be most easily financed.
fiut why invesLOl'S ancl bankers should bt! so t•ager to risk their
money at the very time the risk becomes unduly heightened is a
mystery our statistics do not explain.
·rhe patlen 1of our rccul'rent building booms has bcc:n so sim ilar
for many years t hat itS repetition scenis almost rou tine. T he cn-
tr.Ulc.e of government inro huilding activity ha.< not <liscemibly
<:hanged the paucn1.
The 18-year rhy1.l11n, falling from 1925, rearhc<I iu low around
1933, and left such a wake o[ foreclosures in the nation lhal, in th<:
depths of the depression, incipient rebdlion was ahroa<I in the land.
In pal'ts of th e !\fiddle \Vest, for inst•mce, former.<actually handed
together to use fon·c "gainst sherifb entering with wriLs. Auction
sales of foreclosed p1opcrties had 1heir purpose defeated by !he
refusal of altending ''isiwrs to bid. Refusals to pay taxes were com-
mon. Hcfore such rebell ion could hecomc a d 1allcnge even to
ft•leral authority, tl>Cr<' was organized a series of cred it aids to save
farmers an1l <lcfoulting homeowners, together with the hanks and
insurance compani<'s which had invested hc:ivil y in mortgages.
'\-Vhen, after 1933. a ccrr.ain dei,.,.ce of stabilization in the silua-
t.ion had been est.ahlistwd, the renewal of building activity at
the start of th e next cycle was laund• ed largely on governmen t
funds. 'f'he up turn in building came almost cxan ly in accord witl1
the rh )·tJ:aroic pattc;m. \>\'11cL11c:1 v1 ••l•l :s1J1,. l 1 a 1c a1e\\ a l o f l.Ju ilcling
1
enterprise would have begun without the airl of government is a

subject only for a<'adcrnic speculation. There was nu great upsuri;c
in the building of industrial plants, office buildings, and commcr·
cial structures. l.lut on the strength of large govcmmenl funds,
mad1~ available for lending to tl1e huilders of homes, 1.tu, nt!w cycle
got under way, and th e home-building activity conduCLed in it was
largely under government control.
T he govemn1ent continued control of the financing - even
[ 124]
Tile 18-YE AR RHYTHM
after the time when die banks would have been glad LO enter the
mongagc 111arket again - by offering tem1s more fa\'orable than
the hanks had tr<>dilionally extended. By controlling the financing,
it was simple for 1.hc government also to maintain control of. dlC
cost, the planning, and even the architectural design of the houses
built with \Vashington aid. Additional public funds invested
throughout the country, in pu blic works like new government.
buildings and gigantic enterprises such •• great darns and power
plants. carried the LOtal \•olume of building steadily forward from
its previous depression lows.
1' his was the situalion as building activi ty mounte<l toward new
peaks until approximately i g40, after d;c outbreak of the war in
Europe. At this time there was still an abundance of commercial
building space availa ble in most parts of the country.
The Amc1·ir.an government. then decided that vast expansion in
American manufacturing facilities iuust he rushl'<l. But industrial
leaden in general hesitated to invest large funds in expanding
plants for war purposes, only to be left in postwar <lays with an
investment on their hands that might then be won.hless. The gov-
ernment. on die theory that national defense needs were vital,
promptly devised measures that '''Ould permit industrial plant ex-
pansion with a niinimum of private investment. IE an efficient in-
dustrial o perator showed willingness to honow or invest funds
for constn1ction purposes, he was permiued to amoni1.e his invest-
ment out of profits al a rate far more rapid than previously per-
mil!ed under income tax laws. \Vhe1·e industries showed reluctance
Lo borrow even under these terms, die goven1ment itsc•lt bullt vast
in<lustrial plants which were leased to operators, with or without
option to huy.
As a result of such measnres, government runds in\•Csted in in-
dustrial plant facilities from 19·10 through i943 amounted Loa total
of some ~5 billion dollars. \\Then the United States actually became
involved in war al the end of i gp, buHding of homes and t:am1
structures had to cease almost entirely, under government decree
[ 115]
CYCL l!S
- the famous L-41 regulation of the \Var Production Board. The

chi ef exception to cessation of home-b uilding was the hastily built
housing erected for war workers in plant communities. But the
continuing boom in industrial consrruction carried total Ameri-
ca n building activity st<'adily forward into new highs.
The pc;ik of this activity seems 1.0 have been reached in the third
quart.er of J !)42 - aho11t two years ahead of the " ideal " peak in
the 18-)·car paucrn. Total new construction, public and private,
de<;lincd from 13.;, 1.oilliun dollars In 1942 to 7.7 billions in 1943,
3.9 billions in •911• and arounrl 4.7 billions in 1915.
As the nation was drawn into its economic organi1.ation for war,
the ncc<ls of government for land wc'l'e intensified also, and b'Teat
an·as were taken over for puri>05es connc~1ed with wanime pro-
duction and military training. The outcome was that in J uly of
1941 a P11blic Lands Subcommittee of the H ouse of R epresenta·
tives could announce that 21 per cent, or aln1ost one-fourth, of all
the land in continental United States was ownccl by the govern-
ment. 1' he figures ,,f course include national park areas, as well as
mountain and dcscn wasteland.
\ Ve may expect con1.inucd efforts, in postwar )'Cars ahead, to
return to private ownership some p3rt of these land holdings and
also a good part of the wartin1e indu<trial facilities wlokh the gov-
ernn1ent 3Cquired. J:lut whether this happens or not, it is worth
repeating tl1at government operations apparently do not altcr the
rhythms prevailing in a nation's economic life. Government under
t;ikinp;.~ .l\,f'l"'1n r.1fh('r to foll<>'"' the r11,•tl1ms. Tl1c fuct Ll1at go,·crn·
ment, by the sheer weight of its l:oldings, may now be said to control
a good part of the American rea I estate 1narkct and the activi tics
of the building industry does not necessarily indicate any change
in the 18-ycar real csr.ne cycle.
11iis has imporc.ant implications for those who have advanced
t.he theory that government planning, and timely government
action. may be used to overcoml cycl ic depressions. In view of such
a hope. the course of real estate activity in the Unilecl States during
[ 126)
THI:: 18-Yl'A R RHYTHM
:he imrncdiatc postwar years ber.<nnes a matter o( challenging in·

tercsL to all concerned with theory in naLiornil planning projects.
As alreadj' suited, sunislical downturn in building in late •!H2
c;:ull(: some two years prior to the Lum called for iu the established
pattern. A prior judgment based only on the projected pattern
would have declared that, from 1944 on into the liftics. build·
ing activity would clo exactly this, gradually declining toward a
bottom due around 1953·
On the other hand, by thr f'nrl of 'Vorld \Var II swollen mass
incoincs had created th e same enormous demand For new housing,
and for office ;ind loft space, tli~t haci been cxperknced fol lowing
\\'oriel \Var I. ny 1946 1l1e nation was conscious of an overwhd.m·
ing housing problem that seemed quite comparable to tht~ one it
had experienn·d in 191 9 and the early rwenties. ' rJ1ose who basecl
their 1946.judgmcnts on events following \Vorld \Var I, when a
building boom promptly got. under way and lasted Fo1· years, would
natu1ally assume t.hc inevitahilil)" of a si1nilal" great expansion in
American building activity in the years following, and a boom in
the huilding industry t.h at would last-once undc1· way - almost
indefinitely.
\\'i ll government ancmpL~ to use all possible 111easurcs in pro-
moting new housing suc~ceed in dupfi(:ating, after \'\lorlcl \Var II,
the building boom that followecl \Vorld Viar I? Those who look
back from the fifties will know the final answer. For the present,
we must content ourselves with noting that. the 18-year rl1ythm
was juu near Ll1e 1><><,,.;n1tl11g of iu q«:lic rise when \\'orld \Var I
cndecl. The rise in the pattern exu:nded through '!J'.5· "fhc buiJct.
ing boo111 which rode this rising rhythm with incn~asing momeut.um
lasted well through 1929, when the wtal of new construction, pub-
lic and private, amounted to some 10.7 billion dollars.
To loave told the averag·c man in early 1946 that no similar boom
was prohable following \Vorld \Var II would have seemed fanta.,tic
proph ecy. Tltc need of housiug seemed enormous all over the coun·
try. Economists r.alked about millions of purchasing power avail·
[ 127]
CYCl.£S
able. The fact remains that projections of the 18-year rhythm

seemed to say, long before the war, that the over-all real estate mar-
ket in t11e United State• after \Vorld \Var II would fincl itself in a
situation hardly comparable to that which it knew after 1918.
\Ve can of course expect a large amount of building in this coun-
try in the fate forties - this Collntry engages in vast building opera·
tions in "abnormal " as well as in " normal " timrs. • There is a
minor rhythm in r esidential building of 33 months (see Fig. 1,
ucxt <:hapcer) which suggests that home building will ri$C to •n-
ot11er peak around 19.18. Projectioru of uur 18·year rhythm merely
suggest the probability that O\'Cr-all building opera tions will not
reach the scale some observers have forecast; th~t they will in gen-
eral be declining, irregularly, from the '!112 peak; that they will
reach a postwar low in the early fifties.
The i8-year rhyt111n is found not only in our building and real
estace activiLy. It evidences itself in the activities of many iadus.
u·ies who.se operations are allied to those of the building industry.
The construction induscry, using the products of heavy industry,
follows a pattern related to all enterprise, and full employment in
production and distri bution in other industries usu:illy occurs only
when building activity is at high levels.
Not only is building activity reRected in the volume of car-
loadings, for instance; it is also interesting to !101.C tltat railroads
have doue most of their construction work at times when other
building was active.
F i311rf' 9. adopted frorn 'r\7anen and Pe2Tson, indicates a close
association between huilding activity and deviations from the nor-
mal trend in pig iron production. That t11is 18-ycar rhythm should
show up in pig iron, on wp of the g-ycar and 3~-year rhythms, c1n
l1ardly surprise us.
'rhe volume of lumber production and of furniture produced
moves with, and lags slightly after, the index for building activity
• .E''t°n in t11e \\•:tr yt::.rs. £or instance, private co11itru<.tion tntaJed rome i.8
billions in 194t, 1.6 biWon1 in 1945, 1.6 billions i.t1 1944, z.7 billions in l9f5-
[ 128 J
THE 1 8 - YEAR RHYTHM
·- an associaLion we should find natural. In the past, Lo ral loans and

discounts of American bank.s have also u~ndcd, with variations in
volume, to move with the 18 -ycar rhythm in building, with a lag
of around three years (see fig. 10).
Pet Cent
',,
110 145
120 ,, ,\ ,, Pia lion

I 12&
/ 1
1
I
I
I ~
I
' \
'' •2
II l l , ' \ . '
I
•
90
I
I
I I'
I
l
,
t
I
I 'I
t
\'i I
/
I
'1
I
I
I
''
I
I
'
10s .;
.."'
asJ
j!&o, ' ,
I
,,,
"o II
I
I '' I I ' l
CD \
~ I I
30
'" }~
Buildllla "y I ••
'
0 45
' o
18SO 1870 1890 1910 1930
Foe. g. PIG l•<>i<

Tb~ 1~1'->=• rhr lhm in p;g iron prcxh•ct;on, 1870-19!6. Data ap~
as dt\·iations Crom nonual trend. The Compo6itc - Riggleman. \\'tnzlick. arid
••O·Citr·Index ()( building activity. 1830-•986 has been added. (After \Varreo
antJ Pearson.)
Similady, prices of common stocks have tended in their major

swings lo move with the building rhyLhm (sec Figs. 11 - 12- 13).
The •ll!·ycar rhythm, as it slwws iL>clf in the •ale• 0£ a large in-
dustrial comp;ony. is charted in Figure 14.
The building cycle movement, as compared to that of industrial
stock averages, has been interpreted by \"/arrcn and Pearson as
fol lows:
J>1 ices of industrial stocks appear to rise in the early part of the build-
ing qcle, then fall and rise again. Thi> may be because recovery in
clothing, automobiles and other d3ings with a 5hort <-yclc combine in
[ 129}
CYCLES
>uch a way as Iv fonu a minor peak bcforo bui!Jing i< pro::e<.'tling ac·
li•'l~r. ~1nd the111·l'Jr/1 :11101he1· peak a/Jn11t tlzc.· ~·'''''' rin1e ~,,., buildi1Jg. •
.l'crthcr ITscarch in this fide! seems to i11di1·:ite that th e phe-
nomena referred co here by 'Varrcn a nd Pca1so11 :ipptar as the
9-year rhythm " ricks o n the back" of t he t!!·year wave.
lode: Per Cent
140
130
ISO
Loan$ & C>i$00tl•b
(\,,
120 ,,J /
Buildina / \
'
120
b /
·'
' I
I I
I
I I
l I
'., I
I
I
I
l
;!i 90 I
I I
~
I
I,. I ' I I
M
I I
'
I ' \I '
•
;_g GO ' I
' I I
I '
\
l
I
I
I l
m
~
30
' -' '
\ I
I l I
"'"
I '· .'
I
I
I
I •
I
I,)'
70
GO
61.il:lin~
'
-. .,
0
ltl:tO 1870 13~0 1910 19.30
183l 1853 JS]3 139J 19l3 lOll'IS & Di:s;oonts
FIG. ••>- THv 181.W\'EAR R11YT11\t 1:o1 Ltt4.NJ AND 1.>1scol'~rs
\ -'ariatiOJL' i11 the \'Olu111e t•f lo.-.ns and d.iscou111~. 18'3 1~36. t-Jt1..-o..~I as :t
ptT«."llfagc o[ oorn1:1I. _.ll<f ~·J li..'tl thtee )eotn lt> C'11fTNJ>Ot'ld ,,,i1f1 rhr 0MTipc:>1itc
- Rig&tcm:to . \Vc111J11 l:., :"Ind 1~o-City-lu<h·x of bt11ldiJ1S- .:tcrivjty. 18~0--1936
(after \·\ratte11 and Pt«1rsor1).
'.'.kst impon;1111, we find a tendency in majvr panics to cl.1te them-
selves by turn~ in the rhyll11ns of building '" tivity. l :sJally the
pt1nic.:s ha~.:e cor1lc" 'vitl1ir1 t\\'O LC> fo1tr }'c;tTSafter 1l1r. l,tJ i l<lin~ acLi\·it~·
• Reprintt•cr 1J) ('>(,·1·mis1ion lrorn l~,.orld PriceJ a111l l)ar R•lilclinf{ /11{l·uJlr)
(p. 14.1). G<:orge t'. \Varrtn and Fronk ,\ . P<arwn. PubH1hed h)' John \Viky Sc
Sc'Hl s, ln c;.
THE 18-YEAR RH YTHM'
~k. The 1836 peak was followed by panic in 1837. The 1853 peak
wa.s folloi.·ed b~ panic in 1857. The peak around 1 !171 was followed
by panic in 1873. The 1890 peak was followed hy panic in 1893.
The 1906 peak "'a' fc)llowed by pan ic in 1907, albeit a brief c redit
pan ic of 110 i,•n·at lcngr.h e>1· severi ty. The peak of 1925, 0 11 t he ocher
r ~
, •t \ ''
.· ~· ~,
\I r1"\• j '"
l I
'I
9ui5chng 1/
,..,'' ''' ,
~
',
' I
1810 1890 1910
Ft<;. 11. T11t: 18Jh·Yr.A1t RHYTH..( IN RA11.AOAD 51uc.ac. PRJCF.S

·1·he ClevC'1.an<1 TrtlJiol. C:o111f>:ln) index of price ,,f rnil1oad stocks and tltr
C:rt111posite - Rigglc•n1:1n, \'\:t:11Llit..k, and i 2o-Cit.y·J1\!rt:)( ot building activity,
183 1-1936 (afccr \'\';1rrt:11 nucl Pt:;~rw 1l) ,
hand, was fo llowed in 1929 by the most severe pa nic in our history,
and long depression afterward . It will be ilueresti ng to ohserve
wheth~r the precedent holds, an d whether the peak in bu ilding
acti,,ity reached in late 1912 will ultimately be followed hy com-
parable events. As 'Varrcn and Pearson have pointed out:
\\'hen it is apparent that a building peak ha3 been reached, it is a
warning of the danger of n seriou$ reaction. 1f there i~ abo an impend-
ing rise in tht: v~tflrc <>f gol<l, lite danger is acc:e11tt1atcci .•
And they contin u e, in an analysis whic.h m akes it cl ear that the

18-year rhyc11m is one of the mos: importa n t economic clocks we
loave:
• Ibid.., p. 150.
[ 131 ]
CYCLES
The building cycle is so long <hat Cew people experience two com-
plete cycles in their business li[e. Education, to he effecti\'e, must there-
fore be " book knowledge" rather than experience... .
For mariy in<lividun l~. a11 ltJlfa ..•orable firsc exJ)Crie11c:c 1r1cans a life-
time tragedy. The expedcnccs of the last fifty ycal'S have emphasized
Index
SOOrrrTTT1-rTT71TTIT1Tr!TTl'1'TT1n"n'MTrrnrrrr..TrTT1TinT:Tl'T,.,.,.,rnTTITnT1'n'T'"''"'"" ,,,.""m1
200
COMMON STOCK PRICES
100
Ill:!; Years
/
/ v....
Fae. 11...1~1-1e 18 ~..YuR RH't1"H~r m Co)JMOr<-' $TOCR PR.t<:v.s

Three-}·ear l'.UO\li r1g =-vcragc of the ."-xe-tloughLon in<ll"X of common stuck
prices in the United Staca 1854- 1945, together witJ, ~rt 13-yca.r mo,ing aver.ab~
crend ut.rapalatc;tl. The 18Y,,:-year lows a.re clearly in c'•idcnce.
the dependence o{ the individual and the nation on price movements

. .. price movements are important to the indil'idual regardless of his
monetary theories. T he next most impol'laut focr.or is the building
C)'de. These two forces urc che major factors in business.
The welfare of an individual i.' often determined by the time in
which he was bol'u. If he is old enough to swrr business at the low of
a building cycle, which is accompanied by a railing value of gold and
riting prices, his chances for success are very good. Conversely, if be is
[ 132 ]
TH£ 18-Y~.AR RHYTHM
Per Cenl
150n-rMM'l''TT1T1rn'TTrrrn"TTTT!'T1M'l''TTTTICTI'TTTT1rTT'TTrMn,~
, ...,.
, ,..,,,.,.,,.,.T
, T, .-nCTT
,T, TT'M
COMMON STOCK PRICES

125
I
I
I I
I I
I I
I
I I :
I \ I
\ \ I
I
\: I
Atlue.I Wave;, Srnoothed.
50....,,_...._u...o..LLI~~.LLI_..,,_._,,.......,-'-'-u...o..LLI"-'-LI..L......,_...._u.....u..+i-w..i..LU.J..LLL.l"-'-'u..LUJ
1870 1880 1890 11300 1910 1920 1930 1940
Fie;. t 3. T11E i8%·Y.E.AR RH Y'rH ~< 1N Co~tl'v1·0:-J S' l'OCK P R1t:Y.S

Pcrce11ta.ge devialions of che 3 ~ycar movi11g avc r~gc . Axc-Hooghton it).dex
18.;'1- 1939, frt,1u Ll1e l8-year mo'"·ing average trend (extrapolated) an<l '>mooch<.~d
by a ty,·o tS·year set:tio11 11toving a \'erage plotted to Lhe second section, together
~·ith a rcgtrlar ]\'f -sf1;'1ped 187-!ry<;.;;tT C)'cle.
'l 'he i1Tegularities of t.h c highs :.lW:C to be explair1ed by lhe concunc1\t pre:;c11cc
of a 9-year rhytlun. (See Chapter VII.)
10,000 -
5,000
AN INDUSTRIAL COMPANY
4,000
3,000
2.000
~ 1,000
18!1 Year&
5011 -
i
M
400
•oo
200
18:.S Yea!'$
100
Lu,;-~-:._;::;:;jt;:;c::::;~Iai,:i::v~e~aG"J:i:;:;::iLU..u..L.J..1.Ju.J.J..1..u....1..u..u.1.J..LU..u..!..!.1"'uh.u..u...J
1
SO I It
18ZO 1880 1890 1900 1910 1920 1930 1940
F rc. 14. T11e J8~·3-YEAR JU1y·ru.t.1 IN ·rHY. SAt.ns OF AN t!"ousnuAL <.:OMPANY
r •ss J
CYCLES
hon1 at such a tl•ir rhal he Starrs in husinC.. al the peak of a building

cycle, wltid1 i; ac~ompanicd by falling coinm<>clit) price>. his chan("
o( s11rr.~ss arc: s111all.
~rf\tch ,>r tht• Slt<.c.CSS or rail11rc of art ir1clivj<-J11 nl is <l11c to forces over
\\'J1ich J1c l:as r10 c-ontrol: l>l1l if he 1111,Jc.:rs1ar1lls these force~. lie n1ay
prc>ccrc l1iu.self fron'l 1l1r. 'v-orst rcs11lts of tinfavorah.c combinations ancJ
)')J"Oftt pcrso11aJJy frorn r~\'Orablc co1nbi11atinn\. •
Clarenct D . I .ong, Jr.. who-e study of the Ameri~.an building

cycle i< a k.ndmark i11 research, not<:s that" C)Cks in buildini; sinct
igoo '"·ere abOllt ~ tf1ircl 111ore se\·t·re tha11 J,t·for\!;· a11d dccl;1rN:
Fo11r grt:ic '"'uvc.s i.it the consLr11r.tio11 of IJ11ilrlings t1avc o<.:t urrccl
IJ{'\lf
si11ce the Cl\·jl \ .\lar :.\tld it is not c<)c' 11tl1tll to sa~· ttaat tl1cir Lurning
poitl.t!> m:-ir'<. the rr1ust exciting an<l r11et'11oral>I<: CJ>j~o<les in tf1c nation's
hisrory.t
Prof<"<sor Long. as pan of his resean h, rxa,11ined all prusiblt

explanations for rl1c consistent rlt)"thm and char.u:1erislic lenltlh of
the cvcle,
, r.rul ended wirh the conclusion:
'IN• are .cit, rh~1.-fore, wirh only the p~ychologir.al fccli11g on the
or
part of b11sinc~~ tllCll t.O OCerning 1J1c life :lSS<'IS in a" }')l'ObFfCssivt buSi·
ttt·ss." Tn a 11<.,.,. r.ou 11t1·y Sl1<.:l1 as the LI 11iLec.I Sta lc:'S . " r.>sycl1ologir.al life "
\\'Cluld 1Je ~hor1 cr a.11c..l 1c11laccmencs ''1"c>uJcl rcc.:over surJt1er, 1ha11 \vould
he true i11 olc..l a11d maLurc: cou11Lries. Thii may acwu11L fol' tl1e $hort~r
buil<Jj11g <')·r.IC!a i11 Ll1is co\1 nrr)' thara itl Ei11 01>t·!
\ Ve lta\ e seen th•t d1is 18~-)'ear rh)rlnn ;, imponant not ouly in

real r.s1;1w anrl building, here and :lbro.1<1. hut in marrial?C, pig
iron pr<>duction, '"'·hc;it ;J<Te-age, loa11s anrl cliS<'OJtlt~, railroort stock
prices, anc: in t.hc prices of inrl11srrial common srocks. Fi~. 1 f shows
it irt 1l1e sr.l<:s of a t»pical i11ant1fa<:tLLl'i11g com pan)' as ,.,·c;ll. Later
in the hook we shall mcc·t it again .
• T/,id .. pp. 17.1 177.

1 Bua'ldi"g Cytlrr and tire Tltt:Ot)· of lnvclt1nc111, Clarence D. l .ong, Jr.
(Prin~t()n l 1ni'''""''Y l'ress). J'Jge 150.
! Ibid., p. 165.
.X
Causes, Correlatio1is, C01ijecl:zlres
of 1hyth1ns in our economy we ltave

I
N THE FOR£GOINC "NAr.Ysrs
cm1fmerl our '"'" dption co the four which tnay be rqprded as
the nui>t important of the ma.ny that have: been isolated. These four
arc. tc> summ;a1i1.e:
,-\. l'l1c 5-i·)'·ear rlt}tJ1111 i11 '•1tole1ale price.s ttnd industrial innn,·atio11s.
R. ..-1 he 9-}'car rlt}tl•11i in \,:}tol<>~ale prices. security prit:es, J)ig iron pro·
ductio11, ancl i11dt1strial acti\"ity.
C. Th~ '!f2-yc'1r 1hythm in security prices and in husinC>S ~•ctivi ry,
wholesale and retail.
n . The 18.yc-.tr 1hy1hm in real estate activity, .ind in relat<'tl i11clustrial
c~ r1 lflr[)l'ise.
Various other C('(momic rhythms have been nbsen•cd. Figure t ,

for instance, shows a 33-rnonth rhvtltm in residential building
activity. A rhythm of this length is also imp<lrtant in many lines of
r!:tail trac1c.:, anfl i11 c:c>rlsur11ers• gc11>ds pro<luction . .>\. 2~·L1lc>nll1
rhythm in tcx1 ilc production is illustrated 111 Jiig. z_This length i.•
al•o fo und frequrrnly. Figure 3 shows a 3.1-<lay {i.e., mat krl. day)
rhythm in the rate of change of prices of r<1w sugar. }'igu1c 4 shows
Professor Pearson's index of Lhe purchasing power of bc:cf cattle
prices,• together with a 1ruc 11.8-year periodici ty added. Similar
• Oivjdi11g the p1Jcc of a gi\'c;n c:.nnolodjty, X. by :in jncfcx <.tl •)I ices of otl1cr
rommoclirir:..... eli111iri:'ltCI the effects of aJJ lbc f.-.cr.ors commor1 LO both X and
the t>thtr <Omln<>cli,ics.. lt gi,·t~ a q\10ticnt, the pt1rc:ha~ing: po\'\'Cr of X. \\'l1ich
provi<lc$ M>me ittdic:>.tlon of the £acLOrs peculiar LO that. particular COflllUOdity.
[ 1 35]
CYC LES
lndt:'X
• •
210
•'
MINOR RESIDENTIAL
:\ BUILDING ACTIVITY ,, /,
'' '·
170
\
'
.'
'I
;.' II'
<' '
I
I I
I' II
\
\ I
I "
/\
I
\ I
I \
f I
I
.'
.
\
130 \
I
i \ I
\/
' \
I
I I
\I f I
' ' \ I \I I
I
I' "
90 \; . I/' ,,
"• I \
~ \;
50 fforma • r· '
Actual
10
1919 1921 1923 1925 1927 1929 1931 1933 1935 1937 1939 1941 1943 1945 1947
FJG. 1. Rt.(\Jf)F.NTIAL BUlLDIKG
T l1e g3-mo11tJ1 rf1ytl1111 in residential l>tiiJ.-ling acti\•iLy. •9•9-•939· Rate of

change. A 1~·tJU>f1 th rn t)\'i11g p~rcentage of accual clata, together wit}1 a regular
33.month cycle (after Pcar.«>tl) . The regttJar cycle h<1s been <::xtended lO sl\OY.'
approximacely what will hapy.>c.11 if tlle rlt)•chm continues.
I
115 11
f\
'•
' . .''.
105 '' 'I "~ l
. I
I
I
\
I
I
I
!'
\ I
95 l
I I
.
'• ' '' ; . I
\ I
o
"' •'i'
I I
f \:
"~' I '''' " \
"
I
85 ~! •f
75 NonnaJ
TEXTILE
65 PRODUCTION
6
)919 192~ 1923 tsiS 1927 1929 1931 1933 1935 1937 1939 19.il 194J 1945 1S47
Fie, •· T£xm.Es
The t3·month rhyt.hm in textile produc::ti,Jrl, •9•9-•939· RaL<: of dtangc.
t\ 12·tnonth 1novl11g percentage of act\:a.l cJ11ta, tngt.th<.:r witl\ a t'egular 23.month
cycle (after Pcarso•') . Tl1e rt:gu Jar cycle ha.s been extended to illclicaLC a1>1>roxi·
rna<ely ,..1l\at \\·ill happen if tl1c rh)'tl1111 conLi11ue.s.
CAUSES, COR.R.F.l.ATIONS, CONJECTURES
llO
SUGAR,, "'"'
,, '\ ,,,'''
0
"·
'
100 ,
' ''/, ,, ' \•,'
90 ~ '
''
y v
)934
F tC. 3. StJC·"' J>R:rcl!.<
·t·he 34M rua1·kel-Oay rhythm in l"(l\\f sug.:ir r>ric;es, 1934. Rate of cli.i.nge. ~.\ 16·
<la}' Ol1)vi11g perce111age p( actual data, togct.l)tr '~itlt a regltlar 34·1narkct·<l:1y
.._y..,Ji;. 1·J, j,., 1.c11dcilt.)' p<:1t.i~t~ before and <\ft<:.r ·~J!S1• :'l.l> d<H.~~ Ille tcndcr1cy for
every thircl v.:av(! I<> l>e g1·ectter.
180
!GO BEEF CATTLE
HO
120 ' '\ ,/\
\ I ''
' ' '\ '\ '' I
,,
\
\
'' \ , \
\
'' \
80 - , '\ "' \
\I '
\' \I v
' lroe 14.8 Yea}
00 re1iodiettY
40L_--'-~J,.,.-'-~.,L.,--'~,.,l-~L-..J.,~L-.,.;.,~L-.,..,l~-'-.,..,l,,.,--L--,l,
t88o 1m 1900 1910 1921 l930 1940 19so 1960
FTC:. ;f. CATTLE PRICES

·1·he 14.S.year rhytlltrt in tlu: l'urch~irtg pol\•er of beef cattle prices. Pro·
fcssor Pearscn~·s jndex is cbarced here, ao<l a regt1J"'r 14.8-yeat cycle f)as been
ad<fc<.t. Th.c c..1·c:fc has bee1• projected to shOl\' more or le» ,..,}1a.t v..JJJ l1a1>pe11 jf
lhc. rhytl1m c;1J11Li11ues. Ro.tio scale:.
[ 137]
CYCLES
200 BLACK PEPPER
120 I
I
,\
II
'I
'\
I
\
', I
,,
,,
\
,,,
I
,A.
/ ' \ I \ I \
'
\
\\ I
I
I I
'\
' I
II ! ',,'\'''
\
I
I \
\ti
I \
\
80 \ , \
\ I I \ \
\f \ \_, \/
" \
40
1860 1870 18SO 1890 igoo 19)0 1920 1930 1S41J 1950 1960 1970
Ftc. !j. PEPPER PRlCf':S
'J~lle 14.S-year J'ltytl,rn in. t11e \\·l101esale J->l'ices <1f black f>l:pf>f:r. ;\ Tt~gular
1.1.8-ycar cy(le has l1ee11 adcled and projected, Lt) :tllow nt<>rc f>r Jess ..,.,.h~1t will
happen i( tl1e rhy1h1n continues. (.\ftcr ;in i<1c<t by Pearson, Ca.sset1a. and
Bennt:Lt.) Rat.lo scale.
rather regular rhythms appear in the prices for horses, wi1.h peaks
arol1nd 25 )'ears apart. ;1n<l ;1lso i11 tl1e prites for \ aric>us otlier li\•e- 1
stclck. F ig11re 5 sho\VS a cl1art, s11hrgc:ste(l b}' a st11<I}' I>)' l'(;arson.

C:asst<tta, ancl Rcnnett, illustrating a very regular rhythm in pepper
prices, with peaks at. approximately 15-year intervals. Figure 6
shows a 3-1nonth wave, also often found present in economic
phenomena.
These rhythms are introduced here merely to illustrate the
diversity of the forces with which we must deal; there are many
iuo
1.\0
100
60
40 MANUFACTURING
OEVIATIONS
20
1931 1932 W34 193$
FTC. 6. SAJ. F.<; OF A~
The thrce·monch rhythm in tlle sa.les of a large i11dusLtial c.:c>n11>a11y. Curve

[ 138 J
CAUSES, CO RRELATIONS, CO~JECTVRF,S
others. But to explore an)' of them exhaustive!)', after four have

been analyzed, could involve us - as Professor Sch um peter has re·
marked in a similar connection-in the law of diminishing returns.
In the rhythms of 54, !J• 3~. and 18t )'ears, we have mea~uring
scales wlticb provide, when the average businessman asl:.s what 1itnc
of life it is by the economic dock, as accurate a guide lo timing as
we 1nay hope 10 esiablish, for !he purpose of genera l economic
procedure. Jn Chapter X H we shall show how it is possible to go
fnnhPr for an individual business, and isol:nc t11e rhythms peculiar
to its O\\-Il operacio11.
These economic rhyth 1ns, as we have already noted, need lo be
n·acl - like the hands o[ a clock - in conjunction with each other.
And because Lhey are r~pctitivc they may he •:ombined, together
with a trend, t.o obtain a great deal of light on the near future of
the economy.
.But before reaching thal prohlem in synthesis. it seems well to
pause here briefly 10 answer the reader who is curious ahout causes.
\\' hat is the cause of these rhythms? Can we really accept anything
as true until we know tlle cause of it? If we judge by the progress
n1arlc i11 otl-1er .sr.icnces, \\'e can , . .·ell 11r1clcrsta11d that ultitlt.alc cat1se
i~ an extremely diflic11lt factor to discover. lt is more a problem for
philosophy tl1an for .s<:icnce. Nor hit. even aclruitted, in son1e fields
of thought, that, we shall ever discover ultimate ca use at a.ll - on
the theory that linite 111ind is by its very na ture unable to under·
•tand the ultimate working of an infinite universe. lint in the study
200
l «l
M;\-1 100
60
COMPANY 40
FROM TREND
20
1936 1937 1938 1930 19:W
ll\"Oll~TRJAL C'..O:\-Jl'A,'IY
,J1t>"-~ J.e\•iatio11 !ro1J1 5-monlh moving a,·erage. .-\ 3 mon1t1 q.-,-te i1 added.
[ 1!19 l
CYCL F.S
of various rhythms in ou r environment and in ourselves, research

workers have at kast 111ade great progress in isolating some sig·
nifkant data.
In the various categorie.~ of rhythms we know, some are as fomil·
iar to the man in the str eet as day and nii.:ht :tnd the tides; others
are relativel y u nknowfl 10 him. Analytical work on 1hese rhythms
has been developing in what might be called the iruerstiCl'S of the
sciences. '\.Ve have as yet only the beginnings of a science of rhythms
per se. But in each licld of science some group ol workers becomes
especially <;oncemed with the subject of rhythms in the special
field of phenomena he is observ ing; such a group eventually <:om·
pares no1cs with a C'orrcsponding gToup engaged in some other
scientific fiekl ; :111<1 slowly all concerned begin to suspet.t. r.hat they
arc dealing, however diverse the fields, wi th phenomena r.omchow
related.
'fhe obj ect of all tnu: sci entific inquiry is tO find the simplest
possible ex planation to co\·er the greatest possible number of
diverse phenomena. The further we go i n science, the more we he-
come convinced th:it rhe universe is a uni!)'. that i1s Jaws of action
arc relatively few and sim ple. and that !lie)' appl)' to all 1J1ings. \Ve
thus have a rii.:ht to suspect, wh<:n we find rh ythrn~ p1·esent in so
m:my kinds of ph enomena which our sciences study. tliat we may be
dealing with related parts of a whole.
\ Ve must guard, of course, against confusing relationships with
causes. Suppose, for iru.tance, that a meteorologist fount! a rhythm
of exactl)· 4 1 n1ontl\s in tl1c ,,·ea ther, and t..lrd\ ''-.: .tlic:a<l)· k t lt:\\' of
an e.~actly similar rhythm in the stock market. This would not
jusLify us i11 concluding that the weather caused rhyLluus in the
stock market. °P<'rhaps some other unknown Fiictor affected the
weather together wich the peop:e who buy :l.ll<l sell in the stock
market. Conceivably th ey could be totally unrelated - ale hough
th is would seem hard LO hclieve. Until we were sure, we could ouly
speak of r elationships, or of correlations - this last being a tech·
nical term for a particular me~urc of relatedness.
[ 140 l
CA USES, CORRF.l.A TIONS, COi< J ECTURES
The study of rhythm• has no t yet proc:cc<led be)ond problems

in correlation. F.ven LO this point our· progress h as been wav<:ring
and un~enain. The greatest single piece of progress to dac.e, in the
study of economic ph eno1nc11a, has undoubte<l ly been the recog-
nition 1liat we arc in facL often dealing with rhy1hms raLher than
mere oscillations. This recognition has recently clianged the whole
focus of econ omic inquiry. Until it was reached, millions of words
had been written in tl11• •1ues1 for reasons for oconomic changes.
P unu.i i11g :;Lach a method oE inquiry tl1c old cconon1ia h:.cl tll:adc
little pr~Tess toward ht~oruing a true science. Rather, it became a
larger ;i11d lal'ger debating soclet)', dc:,.roted to eXtl111ir1i1lg -rcast)ns
why a given event shoul<I 1;au.$c S(1mc particular result.
The mere abun<lancc of such reason-why material - .•o gTeat that
no •tudcnt. could ever master it, reconcile it, and put it to practical
use - was itself evidence 1h:tt the method was fruitless. True scien-
Lihc progress is mca,ur ed in the growing ability to reduce the
es'ICntial laws and rel•tion~hips in a given field of in<1uiry into small
compass, and 10 predict. In fact, it is possible to assert that the
abilit~ to p redict rr1easurcs tl1e status of a •Ciencc.
\Vii h 1.he discovery o( rhythm in economic phe nomena. some eco-
nomic in<juiry immccliately changed direc.Lion. The question was
tul l<J11ger 1i.1/1.)' lV<'. f1a\"C bl1sincss c.;y<:ll~S, and wliy ''lC 1t1eet. l'<:<~tltrent.
econo111ic slumps. Now rlw fundamen tal quest.ion was simply: Are
the crcles really rhythmic? And if >o, does the rhytl1111 spring from
witlun or without the econollly?
T o the first part or tl1e query ..ontinued research has been able
to give an affirmative answer almost beyond clouht. Some of Lhe
evidence is in the foregoing pages. 111 concentraring on the second
part o f the <jttestion, economic inquiry has p1oduced theories that
fall into two classes. H•>t.e lling has referred to them as theori<~S of
"fr('c and forced oscilla tions." Both terms are borrowed from
mechanics.
"forced oscillations" rncans rhythms caused by forces entering
from outside the economic environment. Jevons, for i11•tance, en-
[ 141 J
CYCLES
tertained a forced oscillation thCQry. H e beliC'' ed it possible lO
show that changes iu the sun affected agricultural production, and
ll1c resulting changes in agricultur.il production eventually caused
changes throughout ll1e econon1y lhat registered themse lves in fac.
tory production, prices, and even stock. mark.el values.
Mll!loncl>T
lht r •;.•'..,...'TTrn-rr=crr-rrmorrnm-rn,.,.,.,....,,.,.,..,..,..,.,...,..,..,,..,...,.,,..,....,.,..,..,
10111 ri, I
disc 1' ,' \ ,, 1',
1000 /~\ A I \ h• ... v I
Qvn1pot I \ I \ I \f, I \ I
$C0 A1ees / \ / '- I \'\ /.J I \
\ I \ I \ / \ I \ J ',
+•
01-'_,-.,,.'~=--''"'-"''----':..:-,,.._'___,
+soo
•oo ~ B
+iog ~· \
f11$l Dincn::.11ees ot ..\
/ "\
.
1-un11pot f ,
.,.., j \ l
; ,
l\
.,
'~-/
r·
~
/
,
,
\,
'~'
I
.-.., . Busi-
:~ I \ .i \ __,../ r;:-t-'\,-\-+-+----1
I...
lader
+10s
n~ss
-i03 - ,r ,..J \
+s:<
fie.final
c · 51'
w - 10%
I
1875 18HI) 1890 1900 l !JlO 1&20 l~JO
1~·1G . ; . SUNSl'OT:i A!\'.1) l)ROO\.C"l'l()N'

A. Sunspot ..\reas. n. 1:itSl Dlffcrc11ff:$ t) f Sllll$p0t :-\ rcais, :tncl c. ail index of
tr.S. · 1~o tal l'roductit>tl ExcJudi11g C 1·•) }JS. H nn<:l C 1·ycar nu•vi1\g a\•erag·es.
(After C•l'Cia.Mata and Sltaff11er.)
ll·h n)' ycai~~ later, in 1934, two Harvard rescar<;h workers. Carlos
Garcia·lliata and Felix Shaffner, re-examined the J evons studies and
l.l,C'\.ki...-...l Llicm, tllC')' c-nde<l ,,,itl1 tl1c co11c1u:tio11 tl1a.t S\Jnspot phe.
nomcna showed no correlation wiLh agrirulLural production, but -
that :solar phenomena showed a remarkable c·orrelation with in·
dll.\trial production (see fig. 7), business activity (Fig. 8) . and wiu1
stOC'k market prices (see Figs. 9 and 10). Since this panicular out·
come of their studies 'lpparcntly lc£t them a little surpdsed and
aghast, the two students threw up their hanrls and passed the prob-
lem over to u1e biologists and the psychologists.
ll may be ohsei·ved that - even thoug h they had discovered an
[ 14~ l
CAl.'5£5, CORRELA T IONS, CONJECTURJ>S
approach to the problem dillttent from J evons's- lhey were still

dealing with ·• forced 0$Cillation "data. Some of the ir data has since
b~en confirmed by no Jess an astronomer than Or. H. T . Stetson,•
together with others, •md it. m;iy be considerc<l that Garcia-l\.fata
and Shaffner r11aclc a real contribution t.o modern r~sea rch . Fur-
~.--.--.--.--.-.....,.--.--.,--.--.,---.---.-- C.klries
PCJ SQ.
.••.. cm pet
mi1n1te
;" \ Solal rad'l.a~
:
' .
...\l ,.
.•"\
·-
- .01
A.:
8u.,l· f----7----'"-<:---~
"·
··..:•,,_;:,--,f•''---';,.•-....,----j Normal
··. ...·
ness • •' ··............ ..••
act.
+ 10"
......... :
•...'
•
--,,,,,,,...·-..
/"''
•'\
'.......... ,.-'
rto1ma1111----/'-'--''"<:-~'-----""-.L...---"<------1
.. ,
....... , ... + .DI
Ult1•
violet
taO.
- l O"
•\
-zo~1~..::_:::_
.,./_____ B/,
, Ov$1nen
.,,.,,., , '\
_.i;c~·~·----,.;~~:::::'.'.:::::::._,_:~,._-..:j
1~1JO
l.<O
1.30
-30:<~ , Ultr1Jfiokl
A~ena-e
/ 1.10
1.00
I
19:20 19?! 1924 1926 1928 1930 1932
1:1c. 8. Sor AR P 11F.NO).t£NA ,,~n BtJ.~TN F..~~ J\C'n\'JTY
A. S<>l;tr racliatioo tltac hi, v:.a rja1ic>n o f the solar cc•r1 ~ta 11t - Ju11e i920-
Ju11e l9!i· invr·r1.e.J. li. (.A'>l. L. l'. Ayres Jnde" o{ ..\n1criC(tn nusines.ci ..\Cli\•icy,
j uJlt: 1920--0ecembcr 195i. C:. lJltra-,riolet svlar radiacio11, J1111c 1924-Dece1n-
bcr 19~2 (after Garcia -~hca and Shaffner).
AJLhough the pcri<>d iJ too short for statistical dcductio11, u-e comparison is
ir\Lcresting.
ther, in recogniiing the psychologit:al implications of their work.

chey have done much to enhance its sc.ientific usefulness.
The otlier theory of economic rhythms - that rcforn:d to a.~
"free oscillation~" - h:is also been called the "random shock"
• Rc.scarcl1 associalt-:, Jvf:.i5Sacht1s('tl.s Tnstittice of Tccl1nology, ar1cl direcror,
t .aboratOf)' o! C01Smic :lJl<"I Tenestrial Research. SC'<: his Sunspots and Thei'r
£Deets (\'l'ltlctlesey House, 1957).
(143 ]
CYCLES
theory. If, for instance, you take a pendulum and subject it to one
or more random blows, it will start swinging in regular rhythm, as
it is the nature of the pendulum to do. Exhaustive rnathcmatical
studies have been made with the purpose of demonstrating t.hat our
economy acts in the same manner, in rCs[l-Onse to events which come
A
- 700
; \ ~
/
,," \
\
650 >
0
!
-.012 /
Hew York Stoclo: Pricei ,," \
\ 600 "'
,, ...... ... ___ / / \
;!
-.010
~ - .008
A .,.,.,#1
,,
B
;
--- / \
I
\
I 500
so
168
ig - ........-'\
c I 64
·- -----. .
.006 /' '
..
~ -.00.t
""'·-... -.......
...... /
/' \
'
I
I
I1___
450
- l60c
•
~
\ <OO ~
c
·~ - ' 0
156 _,
.002
..."'Normal
London Stock Prioci
\ 350
'
\' - lS2
+ .002
\ - 148
Jan. Feb. Mar. Apr. MiY June JuJ~ Aue. Sept, Oct. Nov. Dec.
1"1c. g. SOLAR RADJA'r 10N A.N·o STOCK PRtcf.:S
A. Nt"'-' , ..ork s 1ock prices (Barron's a\•erage). B. Solar Radiation. inverted.

arid <"~. London stock prices, all by mor)tl1s~ 19:!9 (after Catcia.f\·I ata and
Shaffnt:r).
OJrrcspon<.lcncc lu.~t.,\1t:<~11 curves O'\'!:J: a short period of li1ne may be provoca·
li\'t:, Uul l>y il.l)cl{ JJlU\'C~ 11v1..l1iut; cxv..:111.. tlwt l\·1r tllC period sl10,..,·1l tlle curve,) do
wrn(!\\'ltaL correspond.
as erratic shocks, but set up rhythmic.al responses. E. Slutsky has

exphiined this point of view cogently in an essay, first published in
t927 in Russian, called, The Summation of Random Causes as lhe
Source of Cyclic Processes, in which he said:
The summation of random causes generates a c:ydkal series which
tends to imitate for a number of cycles a harmonic series of a relatively
[ 144 J
CAV~FS, CORRELATIOJ'/S, CON J 1'CTt:RES
smaJI 11un1l>er of si11c cl1rve-s. 1\l'tcr a 1c1orc or less consi<lt.·r:1 l11e numhcr
of periods eve'1· regime becomes <.lisarrange<I, the tran;i1ion to anothtr
rcgi1nf: occurring wmetimcs rath<'T gradlially, somc.."limes more or I~>
:ibruptJ)', around ccrtain cri1ical points.
A
G ____ _...'" London Stock P1loe1 '
,1-..-.. -- . -· ~.
/
100
90
\• ,I
i 2
i • I
'·' ...v~/. $4.wlspots
I
I
I
80 ::
",;:
I 6
.:! 8 -
I
I
\
1' c I
I
I
70
~
~
3
I \ I
10 I
' II
I \
.... ..., I
I
I ,, ~
I
\
\
\
I H... YOfl Steel Prtou
\
' ,,.. ,I
6')
\ I
I so
' ' ~- "
I I I
hL--------ti-~-----
1932 1933
1: 1c;. 10. S1 :~s1>on M"l> Smc;K PRJC£<J
A. 1.c-.11<l<>1\ itock. prices (llank1·r':.. t.1ag-uinc T1u:Jex.). 6. St•nspc>U in the Stirt's
Centr.rl 7.one, i11\~rtcd. and C. Ne:"' Yori.: s.tock 11rice1 (Don··j('Hlt"' Industrial
a\'CTngc:-s), all from January• 193:t t.hrou.gh \ ·fay 1935 (after (;:irci:t-l\rfata ar1d
~h:iffnt:r).
I LCrc again, chc t llr.,·es art: u11u.lt l 11\, l\hort ''>v.•ur10.nt aoy <1t·ductio1i.s '\•li!kt-
SOl'\ t(•r, For an ex..'lmplc, of t\.,·r> se1 ie5 th;:tt are prc~u ntably rc)a1cd M:e the C\tn•es,
of sun.spl>l'li and 1er1e-suial magn<·1i-.iu s.110\'\o-n tJ1\ page 64.
The "random shock " is the theory that Professor Sd1umpetcr

appa rcnLly preft-rs in accoun ti ng for t he rh ythms in the capitalisc
,,, onomy which he ana lyres in his monumental Dusincss (;ydr·s. He
finds it easy 10 hdie'c that inrllwations, acting as surh shocks, ac-
coun t for the cyclic movemcnL
r ' ·IS]
CYCLJ!.S
1~his much deserves to be said: The resources of modern maLhe·

matics in the hands of our eminent mathematicians today are quite
ample to demonstrate I.he fJOssibility that random shock does ac-
count for our economic rhythms. But such a demonstraLion as to
possibility is still far fro1n final evidence tl1at such shock, responded
to with rhythm, is indeed tl1e undt:rlying "cause" of rhythmic
economic fluctuations. Thete is no reason why it should not be
accepted as a hj•potl1esis by those who find it helpful. Rut in em-
bracing it t11ey should realize t11at, unLil we know considerably
tllore tlaa11 \.VC clc) toda~·, t11e}' arc er1gagi11g ir1 ar1 act (lf faith.
Funher, there are two characteristics of our economic rhythms
that this theory apparently <loes riot explairi . First, it fails to tell us
why, after a distortion, the rhythn1s snap back into phase with the
old established pattern, as has been observed frequently. Second,
it fails to accoLlnt for the (:orrcspondence between so many of the
economic and the " natural" rhythms.
Kondraticff, for whose work Professor Schurnpeter expresses ad-
111iratiorl, is l1itllself utl\villing to at:;,rrec '"'it.h_ tli.e raJ1do1n sl1ock
theory: " In assl'rting the existence of Jong waves and in denying
that they arise out of random causes, we arc also oft.h e opinion that
the long waves arise out of causes which are inherent in the C-<5t: nce
of the C<lpitalist. economy." "Kondratielf refrains fro1n speculating
on the problem of what t11is es.<ence might he. Nor docs he speculate
on t11c obvious comparisons with biologic::al org-anisn1s, where
rhythm as an essence of the organism itself can actually determine
\ \•l\etl1e1· tl1e organism \\•ill J'espond to e-:<f.er11a.l st:.i111Ltli or l>c i111~
moved by them. The rhythm in the sex Ji(e of women is an ahnost
perfect parallel.
Konchaticff makes bold to assert, however, that causes usually
assumed for our rhythms "reverse the causal connections, and
take t11e consequence to be the cause, or see an accident where we
really have to deal with a law governing the events." Sometimes
• It.alibi s11p1>lic(I. This and the fo!I0\\ itlg quolaLioos are from the Reviefl)
1
of Econo11~ic SJatisJics, r\o\'e111bet, 1935. pp. 105-115.

[ 146]
(:Al.SLS, COR RELATIONS, C:ONJ EC.TURI:~
improvnucnts in t~• huique bavC" IJecn alkg~,1 as a cause. But, says

Kondralictf, it would be agn·at enor" ii one helieved that lhedirec-
tion and intensity of those discoveries a11d inventions were entirely
accidental. .. . The clcvelopment of technique iL<clf is pan of
r.he long l\J•es." Similarly, ic i$ a r:1i.>1.ake to reg-Jrd wars as causal:
"H-'ars and t'e<'o/ucio11$ c:;in also be filled inco c!u? rflr1lu11 of the
long waV('S a nd do not prove to lie the forces from wh id1 these
tr10\:cments origi1)ate, bt1t rar•1er one of their s~1n1)to11ls.''
As for rl1e Tcsla1ts of l'~J"'''•r:ttinn: " TI.le o~ni ng up o{ ne\\• t:tJUn·
tries docs nC)L provokt: che upsw<:Cp of a long wave. On the co1urary.
;111c""r ups,...·cc(> 111akes tflc exploitatior1 of tlc,,~ <:otL11tries , l1C\V 1nar..
kets, and 1ww sources of raw mar<'ri•ls necessary and possihle:· Nor
does gol<l production pla)' a cauJ.1 pan, as Kondratidf 'iews it:
"Gold prod union, en·n though iu in~Teasecan he acondirion for an
advance i11 commodity pl'iccs and for a general upswing in economic
activity, is }et s11hordina1e to the rhythm o( the Jong WJ\'tS and
co11seq 11cr1tl )'cannot f)<' rcbrarded ns a L"-tt1sal ftntl random f.1< tor.~·
:.Jany other researd1 workers wl10 are working on evidence of
"fon~ecl osci llation" have produced data o[ interest and si3·nifi·
c.ance. Dr. Huntington of Yale is a distinguished repre,cntative of
these.
lrt l1is 1llt">llt1n1e11r;1J \\'Ork tl1~1t sftO\\'S tl1c inllucru.:c.~ <>f cli111;1te Oll
history. :ll11inspri11g.r of Civilizalion, Dr. Iluntingro11 states his
belief that we art: dealing in our uudies not with one force. but
with three, which create: tl11·ee dilf<·rcnt kinds o( rhythmic rcspo11s<:s;
he has summarized his conc:lusions :ts follows:
Our mrur important finding is that human life an<l Ille progress o(
ci1;ilizaliru1 are ittfluer1ct.·c.l 0}' three cliStinct l>lll closely i11terlocki11g
trpcs of cycles io addition to the more obvious cycles of tbe clay aud
year.
The mou familiar of these is cydcs of ordin•ry wearhcr. No one
tXistencc of tl•ese, for C\.'Ct'}'One sec.::i the <:ffCcl n( UilltSlJat
d<>11bts tile
heat or cold. 1ain 01· drought, sunshine or fog. Y•riat.icms in the growth
o( plants, induding all kinds of crop•, are directly depcnclclll on such
[ 1 47]
CYCLES
cycles, and are inOucnccd hy 1.hem far more than any other factor.
\\feather also influe nces re production and growth of animals. both di-
rectly and chrougb plants and the food supply. The indirect effect of
the '''catl1er 0 11 1nan, t10,\ic\:er, is for this very reasotl g·re~tlcr tl1a11 upon
ani111aJs a,11cl Cat' greaier tllatl ltpor1 pl~111ts. <~yclcs of ,.,·cathcr ha\.'e heen
arl.cl still ai·e amo11g tl1e r11icf stirr11Jlar1ts r.o it1vention. tll ri(L a11d fore-
sight.
T hc: next !,'1'C<1t type of cycles is (•;pccially t~vi<lent in animals. Tt shows
itstlf prin1aril ·y in i>l1~1o; iol(•8i'·;if prr)rt·s1;,r->s, P ...:pPri~l ly rpprod\1r.tlon.
TJ1e onl}' knol\1 r1 factor \v-l1ich st1ovis the reqttircd periodicicy and seems
<:011apttc11t to procit1ce cJ1e obscrvetl rl'Sl1lts is atmospheric 07.011e . . . •
Thus the most tenable working hypothesis seems to be that, in addition
to g·c1es of t.e11lJ)Ct'atllf<~ art<.( moiscure, th<.:re arc cycles o ( OZOOe \1t•h.idl.
are especjall}' associate<{ \vith t11<~ reprod1Jctio11 of anit•1als. l\·fat1 also
appears w be influenr.ed, but not so much as :rnirnah.
The third r.ydic type is primarily h uman and psychological. At any
1·aw it is only in hum an beings that we have a.' yet. found clear signs
of ic. It displays itseH in business and prices in a most interesting way.
Sio1ila1· rl1)1 tll.1US arc f<>.un<l i11 atrnosr>l1eric clcctricit)'. Tttc c\ritl<:nce i~
sufficient to 'vana11t tltc '\'Orking h)·potl1esis tl1at at111ospl1cric electric:·
ity, due presumably to the sun, hut perhaps also to the whole solar
s~·stem, is ;t cyclic: f~tc:tor c l<>sel}· associated \e:itl1 ps}·cl1<,logical r<!.actio11s.
The thi·ee types of cydes, clue supposedly to heat, ultra-violet radia-
tion an<l elcctric~tf radiatiot1, ar)r>ea1· (0 be so i11ti111ately cor1nc<:te.rl that
it is often difficult to separate them•. , .•
These three t.ypcs of cycles, as Iluntington ohserves t.hern, "re

os~<.><:ia.tc.:(I ¥-'1th solar pl.. c1101'l\Cr1a. '\'arii1t.ior1s ir1 he~t recci\·cd fro1u
the sun are primarily responsible for weather variatio11s th:it act
directly <>tl pl~1ncs. '/ariatio11s itl u1tr<.t\: i<>lct ra<liati(>n crea Le , . aria·
tions in ozone cont.cut of the earth's atmosphere. with a direct ef-
fect on hiological processes in animals. Variations in rad iated dec-
tricily seem <lin:ctly to a ffect. the psychologic<il proct:sses of man-
or, as Huntington also suggests, "the nerves."
Edgar L1wrence Smith has suggested that such an electric;il radi-
• J'fainsfni·ngs of l'ivilit.ati<;n (Jolt.I\ \\.'iley & St) tlS, Joe.), pp. 52~527.
[ 148 J
CAUSES, CORRELATIONS, CONJ F.CTU Rf.S
ation may react direct!)' on Lhe blood, and affect iu acid·alkaline

balance.
Reproduci ng a plT c urve showing 1938 changes in 1he ratio of
acidity to alkalinity in the blood stream o[ numc:Tous persuns tested
h)' Dr. \Villiam f. Petersen. or the Vnivcrsity of Illinois, Smith
compared it with the 1938 course of the stock market variations,
found it closely similar. He then sought the doctor's opinion. Dr.
Petersen, who is author of The Patient a11d the H'eathcr, <Xin·
firmed Sr11itJ1•s 0~11 i111preMion:
Offhand. I shou l<I sny that the pU curve certainly coin<.i<l•<l with
your stor.k market variai ions for 1938. . . . ha genera I, the popu lalion
is apt 10 ~ depressed and blue when the pH is low, and buoyant and
energetic and optimi>tic when the pH curve i~ higher. Thrre ;u-e, of
00 1.Jrs<~. n1a1ly cxccptio11s, Uut i11 gener;1l I '\\f0\1)d s~ty th;ir this \¥Ol1l<l
l>c the effect. This quite coincides, as you will notice, with your stock
market.•
The worl of two Germans, Bernhard and Traute Diill, has pTO-
duced some impona11t data bearing un Hun tington's hypot11esis :as
to a relation between change.~ in atmospheric electricity on the
psychological states of men. First, the Dillis sho wed synchronization
between solar en1prions of inca11clescent gas and curves in the in·
strumcntal records of atmospheric ekl:tricity at Saint·Cyr in France
and at ·runis in l'\ord1 Africa. Their work seems t0 ha,•c made it
clear that variability in solar activit)', mon· t11an the area and
number of sunspot.,, is the solar factor we '11ould consider in seek·
lng to establisl1 solar· t(·r1t:strial 1tl.1ti..:.nsl-.ip3; and tf1a.t. varic.ilJility
in the aver•gc potential gradient on eanh, instead of its a,·cr•ge in·
tensity, is the factor that shows most promise of yiclcling a cor-
rc:lation.
llaving estalilished a relationship betwc<'ll solar activity and at·
mospheric <•lectricity, the Dillis then plOl!l~I for three months
the daily deaths in Copenhagen. frankfurt am 1'-fain, an1l Zurich.
• Ti<kt in tire AU•frt o/ ftfen (Th• Macmillan Company), p. 81.

[ 149 J
CYCLF.S
The curves for t11e days when most deaths occurred from suicide,
mental disorders, and diseases of 1.he nerves, sensory organs, and
11er\tOUS S}'Stcr1·1, r(tn al1nost 1>arallel l\'itl1 tl1c t~ttl'\'C f<>r r11ag·nctic: di,s..
1.urbance on these same days. A similar plotting of suicides in Den-
mark, Germany, and Switzerland showed an apparent dose rela-
tionship between t11e course of the curve on 735 days when suicides
frequency (\\'ave.s per Setof\4)
10-: 10-1 100~1 10 1 10 2 10 3 io• 101J io' io' to• to' 10 10 io11
I • I , 1I d 1
Electric Waves Radio Waves
~ 10 11 1~ ~ 10 1 101 ~ ro 5 ~ ro• 10• ro• · ,~

\\!ave leneth (In Ctnlirnct<:rs)
Ftc;. 11. ''\~AV£ $ 1'ECTK1:i 1
A Specu'ttm of Electro11.1 agt)flic: , ,\ra\'tS \Villl Ler•gth$ frorn 10'1 tf> 1o·t•
were especially num<:rous, and a similar rise in the cu1·,.,, for calcium
flon:uli on the sun's surface.
lncrea.ses in the number of sunspots or in solar erupt.ions are
ordi11a1·il}: accc)tllpa11iccl l>}: ir1<~ rea.ses in ~<))ar radiati<>n . S<>lar ra<li;1.
t.ion consists not onlv, of heat waves and ultnl';o)et ravs , -which af-
feet us as sunburn - but of many other forms of waves which do not
register on our senses at all. Our increa.sing knowledge gives us
ground to suspect th<1t rrmny of these unscnsed waves affect us as
vitally (perhaps even more vitally) as those we know about through
ottt se11se <)rg«t11s.
The earth circling through space is a target for wave lengths of
c11or1r'l.OL:t:; t«ltlgc. \\Ye ktl O \'>', fv:r itl.st.n 11<,:(:, tlla c cl<.:ctroul~lgnct ic 'V\'a\•cs
alone cover a range of at least 25 " octaves " - an octa,·c here mcan-
lr1g the interval from one wave length to another ten times a.s great.
Sonic waves have a length of 5,000 miles or more - we use
similar waves on high-tension power lines. Some have a length of
only a trillionth pan of a millimeter - these being- the waves that
help ionize the air at high levels, permiu.in~ it. to reflect radio
waves and so making radio communication possible.
Figure 11 is a chart which shows some of t11e characteristics of
[ 150]
CAIJSF.S, CORRl::l..A 'l TONS, C:O:-TJECTL"RF:S
this enormous electro-magnetic wave barn!, and sen•es also to iruli-

cate how few of Lhe radiation> rcgi>ter on our senses. \ \le have d is-
covered the. others only as we deYelope<l inst• uments which scrvcrl
- as a suppknwnt to our live senses - Lo detect th<:m. Visible
light, it will be noted, occ11pies only about a single octave of the
radiation.
10 1' 10 11 10 111 10 1$ 10» 10 2\ 10 12 10%1 10«:1 102i

I
1 R•ys Second11ry Coamlc Kays
( :\ r:Tt:R Co.-.1r 1uN A1'u c:..\.LD\\'ELL)

C4:ntiu1eters: lhrtt il\, \Villi 1:reque11cjcs rrc>Tfl 10 "' to 10::~ \\'aves per Sc<:flTUJ
Our social sci1·nc.e.~ have tcn<k<l :o Corgi'! , in die bravur" of mod-

ern times. one fa<:t which to many of the :indents was clc.. r: J\lan is
a d1iltl of tl1e earth, and by that same toke n a part o[ it. ancl for this
reason hto un1s1 a lso be a chi ld or the s11n. If ou r rnorlern rcseJrc.h
should finrl clear evidence that Ll'.e rhythms of rite sun dominate
in man'> life, as drcy dominacc in man)' of t11e other phenomena
0 11r •r.iences stud), only rnen Loo proud to rcmernher their paternity
sltoulc! be surprised.
But hefore sud1 a hypothesis can be rnore. lhan a mere hint tr>
guide the dir<.'('tion of our research we need a !,'f'Cat many more data
tlian we h<Ht: )el amassed. Datil we do po>sess would indicate we
are 0 11 a track well worth following.
\Ve alr(:;rdy have good evidence that one of the 1naior rhyr.hms
fell by our economic S)'Stem - the 3,t-ycar rhythm -has a parallel
in solar aclivicy. It will be r~membt.-rL~l 1.hat this rhythm runs from
10 r.v 4" montl". T. £.Sterne. o f rhc Ilarvarrl Universi ty Observa -
tf>t'}t, h;is detcr11 1i 11ed Ll\at tile solar <;t>11Sl11.r1t 1 as 111e:.\Sllf(.:<l by tl\e
Srnitl1>onia11 f1h1.itutio11, ll uct11atcs in a compound 1'hyr.lun, one
p;in o( which a\·eragei; a lmosc exactly 40.1! rnonths in length. The
wave is quite dear. II will he fun:ter remembered, in this connec-
[ 15 11
CYCLES
tion, that Huntington has also shown a 41-mont·h rhythm in the

variation of atmospheric elecuicity- the potential gradient·-
o,•er a. lorager period (>f tin1<~ tl1a11 Stemc·'s stucly c:<>"·ers.
As for the other three rhythms to which the major portion of our
study has been directed, tl1(: evidence as to closely corresponding
rhythms in the sun is wort11 analyzing. Not all students have found
satisfying parallels.
H ere is a table compiled hy H. H. Clayton, which shows t11e ca).
culations made by various workers of certain major sunspot
r}1 ytJ1ms, in intenrals of vears: *
. '
Pt:n'od,\' in Y r.ars
A. Schuster 4.8 8.4 11. • 3 • 3.5
K. Smmplf 5.6 7.3 8.8 10.0 u .13 12.9 14.3
A. E. Dougla" R.5 10.0 11.4 13.5 14.3
D.Alter 7.6 S.1 8.7 10.0 11.37 1.1.0
H. H. Cl•yton 5.6 S.1 8.9.1 9.9 11.17 14.9
Clayton's 8.94-year period shown here might seem suggestive of
our 9-year economic cycle, but no more. On t.he whole, there is
nothing here to tl1row light on our four m•jor economic rhythms.
Great efforts have been rnadc to show that there is a business cycle
('orrespondi11g to 1.he obvious sunspot cycle of around 11 years. Rut
evidence of a direct i·el~tionship h<:tween business and the 11-year
sunspot rhythm seems open to some criticism. !:'or, while a para!·
lelism ha.~ been shown, for shore spans, hetween the 11-rear cycle
and business activity (sect.he wo7k of H. T. Stetson, and of Garcia·
l\ofata and Shaffner), these spans arc not. long enough to give us
conclt1si\.·e evicl<~11ce. Ori tl1e co11trary, '"''e 11a,:e t)ettcr e \:iclcn.c e that
other rhythnlS a.re far more important in our economy. Cenainly
none of tl1c periods listed by Clayton corresponds to the rhythms
we have been noting in the economy.
C. N. Anderson, of the American Telephone and Telegraph
Cornpany, has originated a different technique in research on the
same subject, with far more interesting results. Fig'llre 12 shows his
• FrtJrn "The Sun Spot Period," Smitllsonia·n ll·1iscellaneou.s Colle<:tion,
Vol. 98, No.•·
[ 152)
CA tJSF.S, CORRELATIONS, CONJECTURES
60
10 -
15 ?.Q 3~ 4o so &o 10 so
Period$ in Yetrs
100 150 zoo 300 400
Rela.li\'e lmport::t.l\Ce oi \!arious c:o1r'•1>011eJ) lS of a Harmclnic r\r•:•lysis o(

Sunspots, ,,,ith altcn1att cycles rever.scd. Ft1ndan1e11cal 312 yea.t's. ( ..<\fter
c;. N. :\r1derson.)
r\ list <>f sc>fne of L11t. tttc•rC import~nt Jcngths fol Jo,\·s:
FundanlCOl:\l 31 \ yc-..:1rs 9111 fla1·u1ooic 34.6 )'e.ll'S r']lfl l·lannonic tl'!.35 yt;tr'
:!Ud llarmonir 1,r,6 " 1c11l1 " 31.2 •· 18th " 17·33
3r<l " 10.1 " 11th " :eS.40 " 19th
.. 16...1.t ..
4tf1 .. 78 12th
..
., 26.00 ••
..
20th 15.60 .•
,jtll .. 62.1 " l3th
.. 24.00 .. 21st
.. 14.Hs ·•
6Lll
;th
.•
''
5..::.0
44.G "
11th
•.:;th .. t:i . .tS
20.So ••
..
2211tl
2~rtl
.. 14.18
13.!j6 ..
..
Rth 39.0 " 16rh t9.5e> st4tll .. 13.00 ..
arwlysis of comr>Qnents of the sunspot cydc:. Using sunspot darn

t.hat goes back to 1749, he found a Jong-term C)•cle of 312 years, with
a rhythm of 22 . 28 years as the chief compcuicnt. In ~·ig. 12 • the
c:omponcnt rhyt.hms of the cycle are shown hy lines which indicate
- in their relative length - their relative importance <is con1-
p<mc11u. Their posit.ion on t.hc scale indicate.s the length of each
component rhytlnn in years.
* Reproduced {rot'l\ 1J1e Dell System Tecllnicol Jou111al1 .~pril, 1939, J). s98.
[ 1:rnJ
CYCLES
in the beginning A.nderson used an itlteresting discovery made

by Hale. In alternate 11-year cycles in solar pheno1nena. most sun-
spots reverse their pohirity. I t would appear that alternate cycles
may be considered negative, and that the true cycle is twice the ap-
proximate 1 1.13, amounting to something over 22 years. Adopting
this hypothesis, and subjecting to hannonic analysis a 300-ycar
l790 1810 1830 1850 1870 1890 1910 1930
Fie. 13. SK~u1:rc.>~1zv.1> llt:fkf.SJl:N'l'ATION OF TJIE 18¥s-YE..\R R11v1·uM IN

SuNSro·r N11ll.tuERS
J)iagran1 o{ Lite 17th 11arn1onic of C. N. 1\nders()n.'s Jlatrnonic :'\nalysis o(
Si.1nspot Nt1rul>er)J \\'ith !ilteroate cycles re\•ersed. The c.ii;1gra1n is correct for
am1,litude and <lat.c.;s, a11cl appr<lxi1J"1alely correct for average sha.pc. (Compare
l\'itJ1 tl\e 18Ya·ycar rbytl>m tliscu:;scd in Cl1a1)L(~r IX.)
series of sunspot numbers with alternate cycles reversed, Anderson

found rhythms that dosdy conespnnd - both in time span and
in t.i1ni1ls·- ,,vitl1 out· l~OO•lOnli<:: <lata. Ins1>cction Of this cl1art in
Fig. 12 shows immediately that the 22.28-year rhythm is of domi-
nant importance. T his is very close to rhyth1ns we find in many
u1etcorological phenomena.
Next in importance in Anderson's chart are the two adjacent
rhythms of 17.33 and 18.35 yca1~. The rhythm at 18.35 year:; is an
aln10st perfect match for our 18;\-year economic rhyLlim thaL we see
evidenced so clearly in real estate - and it reaches its crest at the
same time. (See Fig. • 3.)
[ 154 ]
CAt.i s~~. CORR F.l~1\ T l0'.'15, CON .1F'CTUR ES
The 17.3.year rhyrlun shows up in epidemics of d iS<.<1se, in conon

1>1 ices, and in a n1 1rrtf>t'r ,,f <)tl1er eco11o rraic: ~cries.
It wi ll uc uotc<I LhaL t.he cornponent ac 14.8 yt«irs also is very
strvng. Seemingly th e sarnc rl1yt.h m shows up in the purchasing
po wer of beef ca11k prices (see pagt: tJ/), of pepper prices (s<:e
page 138) , and in bo nd prices.
Clustered around tJ1t !J· to 10-year period th~ reader will note
rhytluns at 8.91 , 9.17, !J 15, 9.8, and 10.1 ye;in. This <liagram does
nnr mran - either at this poin t or elsewhere - th:11 rhythms with
exactly t11ese periodiciti1•s have been isolated. It docs indicate that
either at or between these points there are waves of at least tJ1e
impottance ind icated. Tl1 c fact that five such waves show up in the
8.91- to 10. 1·)'Ca1· period suggests that l'u rther research ma y <li;;closc
counLerpans of the !)·year i·hythms we ha,•c e ncountered in eco·
nomic phenomena. As a suggestion of possibil ities, see Fig. 14..
Anderson's dam should ;i lso be examin<'l to see if the wave in·
dicated around 52 y<-ars niay not - when isolated - correspond
to the 54-year <'conomk cycle already d iscuMe<l. Some observers,
like Gillette, have suggested that t11c 54 may be viewed as a multi·
pie of the 9 or the 18. '>Ve may well view 1.he use of such multiples
\Vitl1 rcse1'\:e, tintil \ Ve h<t\'C more e\•iden<.:e \"·arrar"1.i11g st1cl1 an ap-
proach to th<': use of tlie data. 'f hey arc interesting, but hardly
c:orlr:lusi\'C.
A number of other research wo rkers have mad e extensive com-
putations on the solar rltytluns, and have traced va rious o Lher cor·
rcspondences between solar periods and rhythms in ter restrial
phenomena. • Unfortunately. the research is still in its infan<.y. In
view of the great possibil ities glimpsed in the shon period of study
descr ibed here, one would presuppose that man y rcso ur<'Cli had
immediate ly been pu t be hind· the work. On the rontrary. To date
tlic work has been co nd ucted with ex tremely lirni1.ecl funds. Th is
is one of the gaps in o ur knowledg<: that needs filling promptl)'. In
• Sec H. \\'.Clough, M o !llhly W eather R eui•w, April, •9$.~. U.S. Depart·
mcnt o( 1\gric:ulturc, pJ>. 9g-1o8.
CYCLES
A 3.91 Year Period in S11ft1pot1
D Comllo$at' (A+R + C)
120 - E Wltoltialo Price Deviations, Smoothed
100
so.,
- - -
:I
0
~
' 0
~
e
- -
5
0
: ..- ...-
0
~
0
Ftc. 14. Onllll Sul'\srOT Cvct.1'.S At1r.'D Pa.ice.~

Skt1etor1izr.cl R eprCi-et\lation, i832- 1939, o f (1\) tf1c 8.91·yt:ar, (B) the
g.J7•}'ear and (C) the !»·1,!:i·yea.r <:<>Olpone1lt.S, as usccl b>' C. N ...\ndcrsort ir1 J1i'i
Harmonic 1\r1alysi& of Sunspot Nuntber... ;\ltc~r11:it.t: C:)•Cle1 Reversed.
Also inclutlctf ~ire (1)) 1.he Combin:ition or :\ , I\ :tucl (;, n11d (£) Lhe 9.year
\V11.,•e ir1 \•\?holesale Price,, 18~ 2- 1 !)39·
le \4/·ill be notctl th:tL 111mjng points in prices gc:tlCrJtll)' precede the corre·
tponding tllrnir)g points in the comlJined solar series. Prit·c::s are 21o-n100Ll1et.1 l>y a
throe: 9-year section tn:Ol 'ing avcrage. (For an explanation uf Uae: ~011 mc>ving
a,.enge. see Appc::ndix Ill.)
particular, as Huntington has sui,,-gested, we need to analy'c the

solar cycles not only in terms of the number and area of sunspots.
tlocculi, etc., but in terms of the rate of change, or variability.
\\Then this has been don(:. it is to be hopccl that both biologists and
psychologists will cxccnd c11e research to determine just bow far
solar rhythms arc related to human rhythms - including those in
economic life - ancl what the relationship may be.
r. 1;6
. l.
CAUS£S, CORRIU.ATIO'.'< S, CONJl!CT\IRF. S
Tchijewski, a Rus.•ia1i, has conduc1cd a series of studies which

shortly before \\/oriel \\lar II were gaining wiclc attention in Eu·
rop('. but have had small audience in F.nglish·spcaking co untries.
It is his ~ondusio11 that increased elec1ri cal activity in the sun,
especially around sunspot maxima, inc1·cascs ionization of the air.
lie believes i.his i11 turn stimulates mankind both physiologic.ally
and psychologicall)'. with effects that can be traced throughout all
social relationships.
Dr. R•)'moncl Holcler '\'heeler, of the Psychology Deparunent
at the University of K.:msas, has developed a Llleory of tire effect of
cl1a11g<:s i11 lteaL radiation on sociCt}' atld civiliz.c1tion, which as a
study is related to this same type of inquiry. His exhaustive ac-
n1n1ulation of hi.<1.orical data tends, just as does Tchijewski's, to
support the postu l:uc 1.ltat solar rhytl1ms of many lengths and am·
plitucies arc connected with human rhy1.h1ns LhaL show up regularly
through history.
One imp<>rtant finding results from Anderson·s st11dies. In a
number of the solar rhyt.hms, the peaks usually come al the time
of, or after, the CTCSL of the rhythm in economic affairs. This sug-
gests that the solar rhythm is itself not causal - that the sun is
itseH responding to some rhythm which <ilso affects men.
None of the dac.1 we have to elate regarding the ultimate origin
of these rhythms i.< sadsf)'ing proof of anything. .But they clo suggest
that we arc just at the gate of a new field of exciting knowledge.
And they suggc>t a possible po;tulatc:
\\'e know t11at our sp;ice is permeated by radiation from the sun
- there are light waves, for instance, ot a wide band ot diUenng
"-ave lengths. \Ve know that here on earth some objects, such as
a red pencil, n'sponcl to waves of the" red lengths ..; that ntlter oh-
jects, such as a blue pencil, respond to waves of the" blue lengths";
;md that a third oh,jcct, such as a woman's dress. responds to both
"red and blue waves" and reflects a combination of 1.hese which
we regard as purple. Light of all wave lengths, including yellow,
is falling on these tltn-e ohjectS; but they arc " blind " and unre-
[ 157]
CY C!,.F.S
sponsive to any waves except those of a length which " fit " their
respective natures.
\Ve also know t11at there are, in the universe, waves longer than
light waves, with a longer time interval from crest to crest:, which
are used for raclio transmission. Here, similarly, certain waves may
be affe<..:ting· <>11c radio,. \1o•l1ereas ar1otl1ct raclio~ i11 tl1e next roorr1,
may be totally unaffected by them, but is being affected by other
waves of different frequencies.
Now, let us suppose that there are still longer waves in the uni-
verse - "Y " waves we may call them. Jtnagine some of the waves
with peaks which come 3~ years apart, ot11ers with peaks 9 years
apart, 18§ years apart, 54 years apart, and perhaps much farther
spaced. It is not inconceivable t hat these longer waves could directly
or indirectly affect the sun, the weaLher, animals, and human he-
ings, and that just as a red pencil may responcl to light waves of
only one length, so a particular kind of organism might respond
only to Y waves 0£ one particular length.
Such a state of affairs is wholly ilnaginary. But Lhe irlea might
serve to explain how so many different pheno1nena could oscillate
with rhythms of identical length, and how one set of phenomena
could fluctuate in waves with lengths entirely diffel'ent from those
of rhythms in adjacent phenomena. But such conjectures com·
pletely outrun the facts in hand.
That we have discovered solar rhythms only in the past. few years,
and are only learning about human rhythms now, may he put down
LU tlJ.~ !Nllllt: SUl l ur .>lV\'\' l1Ut));11l de·vclr1pn1CUL "t'\•l1icl1 kept US froJTl
knowledge of such vital matters as electromai,,.,letic radiation until
the nineteenth cc:nturv of our era.
'
There is, of course, an excuse. Fol' these rhythms are extTemely
complex. How long, for instance, might it have taken men to dis-
cover the rhythm of the tides, if t11c earth - like Jupitel' - had
nine moons instead of one? Suppose, in addition, that our earth
were so surrounded hy fog we could not see the nloons. How
complex the tides would be, and how mysterious! \Ve might even
[158]
CAUSP.S, C:ORRELATIO NS, CONJJ::CTURES
yet ha',. fJilcd to discover that the tidal sw~'<:ps o( our oceans were
gra,·iiational rhytlnns u n ited in a complex synthe.~i~.
The• \\Olk ahead o( us strctdu:s far and long. Fortunately that
l..no\\ledg<· of ullimate c;1usc, though it may be an ultimate goal,
is not immediately \'ital to the conduct of our study, any more than
to chat of other scicnc<:s. As the grcnt Eddington has declared: '' Tn
1pi1c nf popular opinions to the 1;on trary, scientific investig-;11.ion
don not lead l.o L11r knowledge of che int rinsic nature of tlli11gs.
\\lhenever we st;tre the properties of a body in tcnns of physical
q11a111.iti<'S, \vc are imparting knowledge of I.he response of various
111eL1·jc;1l ir1clicat<>rs t<> it.s prese11cc, atifl 11r1ll1i11g r1io1·e. To go ft1r-
1licr is 1he task of philosophy."
\\lhatever working hypothesis l'onccrning the cause of human
rhy1hms 1hc reader may prefer 10 adopt for purposes of study-
whctl1er he looks more favorably on the theory of free or forced
oscillation" or prefer s some other one he can •till find theanaly·
•i> of 1he rhythms full of reward. For, as Dr. \\lillford I. King, pro-
r~s;or of er.onomics at New Yori:. University, ha.\ l<li<l so well:
Jf 1hc oscillations in business ;u-c dominAt<:d hy force> followillg some

fll:llf1,.rr1atir~ll f<>rn1t1Ja, ~"· ht:tlLC-r tflat forrn\tla be or1t: rcprc!ie11ti11g a
<:u1111>o~il<:of sine cur,·cs. or ochcr,visc, tl\c patler11 of bt1sir1css ntust to
a r.011sitle1·able exte11l be re1>etitl,,e in natul'e a11ll l1c11t:e 1>rediccahlt:.
~"11r any cur\•c. or co1 11bi1l~tlio11 of cu 1"\•es, rcprcsc11te<l Uy mat11en1atical
formulae, can be reproduced indcflniLcly into the future.•
• Tl1e Cau.\·ts a/ E co,10111ic Fl11t:fuali1J11~· (Tile ltonald Press), p. :.-44.
XI
o TAK F. a complex wave an<l analyze it into its simple compo-

T nenLS involves cenain t.e chnic;1l knowledge, and so docs the task
of synthesiT. ing sin1p l.e rhytl~ms into a Lotalit)'. IL is not necessary
that the reader of Lhis study undersLand the tedani<.Jl•Cs involved,
and anyone desiring to avoid even th« semblance of a menLion of
metlwd5 may prefer to skip this chapter. But the principles are
simple, arul he 111ay firHl this umlerstmading helpful before ap·
proaching - in c11e next chapter- a practical application of the
knowledge to a specific business.
Ir should be niade clear that in speaking of waves the word is
used to describe what rnay be seen on the charts, and not lo describe
cxact.Iy what happens in society. As Sir .James Jcatas has said:
., A rr1atl1crr1atic:al f<>r111u1a <:arl n.e\•er tell us '":hat a cl1ing is. but
('JJtl)' llo,,r it bcl1a"·es . . . . Tl1is poi11t of ·vie\'\' brir1.gs ll.S t(;lit::f f1uJu
m<my of th(: difficulties an1l apparent irn:o11sisu:nc:ies of present day
physics. \Ve need no longer discuss whecher light consists of p;irti-
de.$ or waves; we know all there is to be known about it if we have
found a machcrnatic;al forrr1t1la which accurately describes its be-
havior, ancl w<: c"n Lhink oI iL a~ either particle~ or waves accord·
ing to our mood and the convenience of the moment." ..
• Tl~el1fysteriou.s U1liVet-se (fhe i ,faanillan Company and Cambridge
t:niv~rSily Press) , pp. 951-2.
ANALYSIS ANI> SYNTHF.SIS
Jeans tcntati\ ely identifies the wave with pro/J1Jl1ility, in a beau-

tifully dear passage. I le says:
?>lost physicists would, l think, agree ll\at lhe sevcn·dimensional

space irt \v"l1icl1 (Ile 1A ave-rnecl1 anics Jl i<:tL1r<:s the 1ueeting of tv.~o elec·
1
trons is pttrely fic.:LiLiuus, i11 v.·hicl1 C\Se the '"'aves ,,1}1ich acco1upa11}' the
electrons mus1 also he r<'garded as licti1ious . ... Yet ... ll\c waves
of a single electron are t<:al enough to record 1henuclves on a photo·
graphic plate. . . .
Some ph}si.ci,tc. mef't this situation by rcgarcling the rlPrrrnn.\ ...·:lv~
as waves of probability. \ Vhen we speak of a tidal-wave"'" mean a ma-
terial ,..,a,·c of water \\•hich l\·ets <-"-.et")'t!1j11g in ils f)3 1h. \·'lhcn l<\·e speak
of a heat-wave we rne:tr\ something whkh. although not material, warms
up everything in ir.• path. llut when rhe evening papers speak o r: a
suicide-i,v~tv~. the}' ([n nut mean tflat eacl1 Jler~~n i11 clu.: p~1tl1 of llle
wave will commit suific.k; they merely mean rhat. rhc likelihood of his
doing so is increased. If a suicide wave pas.es O\'~r London, the dcath-
rdte from suicide goes up: if it passes o,·er Robinson Crusoe's island,
the prohability that tJ1c sole inhabitant will kill himself goes up. The
"'a\'CS ,.,.·hich represent :in elcctroc. in the wa,·c·111cthaniC$ may, it is
suggested, be probability·wa\'cS, whose incen•it)' at any point measures
the probabilil y of the electron bdng at that point.•
Just as our reference tu wave~ of light is a mathematica l reference,

110 La sensor}' <Jrte, Sc) is 011r rcfcrenC(! to '"~a,'<'.8 itl tl1e economy. Tf1c
w;ivcs we talk about arc" mathcrnatical record of energies which -
su "ar as people are concen 1ed - appear in their Jives as pay cuu; or
raises, or the courage to buy a new automobile, or the sclf-Ocnial
in•'olvcd in makin)l the old hat do, or the determination to see that
the fanlily moves into a nother neighborhoocl hcfore :>iellie beco1nes
of school age. Trillion~ vf such daily events. built up Crom people'~
decisions and impulses, are tlie energies that distribute t hemselves
through sodcty.
Now, distributions of energy are easily recorded as curves. Fig-
• I bid., p. 130.
r .s.1
CYCLES
. .. .. . .
. . . . .. . . . .
. . .
. . . . .
. .
. . . . . . . . • . • . . . . •I
. . . . . . . .. .. . . . . . . . . . ' . . . . . . ..
I . . . . . . . . . . . . . . .
. . . . . . . . .
. .
. .
'. . . . . . . .
. . . .
. . . . .
. .
Ftc. i. GALTON's QUINC'UNX

A devil'c for iJltistrating the physical IJ:iSls of tt1c norm.11 frequency lai;11 (alter
Worih;ng and GeJloer) .
ure 1 shows a device called Galton 's Quin,unx, which puts small
shot into ;he hopper at the top ;md leis them fall through a s1nall
hole, past interfering pegs, into bins at the bonorn. They fall in a
w~y that illustrates the physical basis of the normal frequency dis-
tribu1ion Jaw.
Figure 2 shows a graph of this same law of nonnal frequency dis·
triburion, in which you can recognize the wave form apparent in
[ 162 ]
A:-IALYSIS AKD SYKTHESIS
Vp=0.797110
1fa=0.12'8¥o
'Vq=0.601v0
I
V~~f Nf 0.368110
Ftr.. i . FR.t-':Qt.'.t;...vcy D1s·rRl6l} T10N

:\ graJ>h c1f tht.: J::i~· of normal frcqt.1ency <listribi.1tio11, sl10,\ing chc Prcci$iOn
l11<lir;c;c; (;i ft<'·r \\.'c;trt-hjn.g and Gefi11er) .
"'=>-
c A "'c A
&~ c ·-
:>
B
·-'°"
B
·-
:; ~
..,~·5 c
-
-·
o~•
••
A •-
E"
•
"'
o~
•
Q :i
A
E
z•
if- P.r if if+94 m
Diameter, d b1ass. "'
(A) (BJ
Fu;. 3. Bl$ARIN<; '.BAL r.s
Distributio1l. <>f (A) ti\(; <liamctc;:rs and (R) Ille masses of a grot1p of l1caring
balls. lf A is a~surnc:c.1 11(1rm:.ll ft is necessarily ske,o;e<l. (•.\ftcr \'\rorthing ancJ
Gd (ner.)
the Quincunx. the purpose of which is to show a wave for1n by

1nect1;1t1i(.:al m ea11s.
Figure 3 show$ how t11e mass distribution may be" skewed," and
hence nonnom1al, when a group of bearing halls of equal density
and sphcricity. but of vatying dian1eter, follow che law of normal
freqtteDC)' distril>11tio1l , \\.'er e tills a11 ordi11ar}r 'l\'a\:e, Olle '"ro llld say
it is distorted. But even the" distortion" follows law.
[ t68]
CYCL£S
These illustl'ations are reproduced here for just one: rc:ason: t.o
show t.he reader without rnathenrntkal I.raining how a phenomenon
fam ili;1r i11 tlt1·cc.~ di111e11sio11s tna~· be intcrprct.C<l ir• r1l;1tt1err1atical
fonns.
n ~I A=l [al
n=!O A=0.3 1=9

b
I=IO
Frc. 4 . $c)()NO SYNTkF.SIS - T\o,'O Col.fPONEN'TS

Cur..·es rt:pre&cntir1g t\\'O s imple rot1nds an<I their co111bin:-ttio1\ (after ~tiller} .
Something quite similar happens when pig iron production, say,

hcr.omcs a chart on a piece of paper.
TJ1c ' 'ibrntio113 caused b)• sottnd.i may si111iJarly be n:1n<lc i1)to a
chart.. Prohably most. readers ai·e familiar, from thc:ir school days,
with the diagram of a simple sound wave as shown at the top o[
Fig. 4. The middle figtire in this diagram .~hows a second sound
wave. The bottom figure shows 1.he 1.wo combined in what we call
a synthesis.
Synthesis is, of course, the reverse of analysis. Figure 5 shows, at
the top, the curve made by the tone of an organ pipe. The sine
curves shown below are the harmonic con1ponents of this curve,
[ 164 J
A '.\J A L \' S I S AN l) SY" 'l'll 1!.bIS
1
-----r---
b
I
I
I
I
I
I
..••. I
J
0
o· .•
-------------A-- ------.I
I
' ••
I
J
•
I
I
II
2 ----J
I
10
II ...... - ..- -__...._ ~
~
•
:
_,
12 I
•
.
z ti>
I
Fie. 5. Sot.~o Sv1',,.1:E.Su - T\1fl!t..vc Co;\JJ'O"Dl"J'$

An org-.r.n pipe curve (cop) ao<l its 6nt twel,·~ h2rn1onic compone11t.s (after
Miller),
[165)
CYCLES
which has twelve components in all. The curve a.~ shown on t.hi.~
chart is really tJ1c gTaphic interpretation of a mathematical equa-
tion, the wave kngth being equal to 400: •
y :.... A0 +9<l.5si11 ( 6+ 76°) +66.osin ( 20 +319°)

+ 36.5 sin ( 3 0 + 337°) + 19.2 sin ( 4 0 + 354")
+ 10.3 sin ( !) II -1- 330°) + 8.4 sin ( 6 0 + 3·17°)
+ 6.4 sin ( 7 II+ 351°) + 8.9 sin ( 8 0 + 290°)
+ 4.3 sin ( 9 0 + 252") + 2.3 sin +·
(10 0 252°)
+ 2.2sin (1111+ 230°) + 1.5sin (1211+211°)
"I'he priru:ipk behind the prediction of tides, fol' lonp; periods

ahead, is quite comparable to the an;1Jysis of the pipe organ tone.
Called tlte ha111lonic method, it was first devised by Sir 'Villiam
Thompson in 1867. The tidal curve is broken down into a numlwr
of constituent curves. cat:h of which may he attributed to a periodic
cause, mainly astronomical. These constituents are sine or cosine
curves. Sine curves cle1·ive their name from the fact that, if yon
take a circle ar1<f plot tl1t: sirle.f of tl1e Htlg·les ru1111ing from 0° co 360°
as e<1uall y spaced ordinates, having due regard for si~'Jls, r.hc result·
ing points form one complete wave of a sine curve. RcpeLitive waves
found in natural phenomena - such as light and sound waves -
arc usually represented as sine curves or combinations of them.
All sine curves are periodic, repeating themselves indefinitely. fittl
S(>rltC pcrio<li<~ c:11r,•es are not si11e cttn'es. For an int(~resting <lisctts-
sion of the Sllbject St!e the ch <tpter ~lJl c:}''f'-11.·s 111 TJ1.1' r.o.'tt.tl'..~ 11/ P.rfJ-
nornic Fluctu.atio11.1 hy J>r. \Villford T. King (Ronald Pre.~')·
\\.1hen ti<fal ClJr\:es are ortce anal,.:zecl, cJ1e <:o,1stl t11e11ts mav l>c
' '
rccomhinerl hy 'yn1.hesis to predict t.he future tide for any 1.ime.
Figure 6 shows some o[ the constituents in 1he tide at Los Angeles.
Captain Paul C. \Vhitney. of the U.S. Coast and Geodetic Survey,
says of the work of analysis of such constituenLs:
• Reprod1l<:t!<l l)y perl'nissj4)0 fr<>rn l\'fi11t:r, Tlie !-;cience of 1\·fusical Sot1nd.$
(The :\!acmillan Company), p. 1 21.
( 166]
Al\ALYS!S Al'D SYN"THESIS
The actual computations necessary in mak;ng the harmonic analysis

are quite laborious, although the work has been reduced by sysccmatk
111ttl1ods. E\:en so, to <l<.:r.<nr1ir1c, sa}', 2,1 c:c>nstitL1C11ts at a giv·cr1 place
from a year's tidal series means computations involving about a million
figures.•
l 'tG. 6. 'fJDE PR.Elll(..110~
TI1e l'redicred tide at Los A11gelcs, C;.,llfc>rr1j:J, fc.1r ,\uI;,'1lSt 23-30. 1936: the
ObSel'VCCl tide; a11d f(1Ur (1\·f::, S::' J( , and 0 ) Of the ~ 1 (.:C)U$li lllfntS tJtat v,:ent
1 1
iut<t tl1c 1>rc;:tli{:ciot1 (ar1.e;:r \+\!hitttcy).
Tl1c.; L\\'V 111ytl1 f11!) ;-;f 1V\\'1t atl"ig. 6 ill u:1tr•1tc: tltc
Ll1c botto111 of
tide as it was predicted front a syntlicsis of th<~ c.onstituent curves.

and the tide as it was actually observed. Very c:losc: agreement is
in<lic;it.ed. The illustration shows only 4 out of the 21 constituents
used for this part.it:ular pn:dir.tion.
The l1armonic method of analysis, hy wliir.h the observed tide
~ "The Prcdiclio11 -Of Ticlei>," U. S. ~1\raval l11stitu.te Pro<:ee<li11gs., l)ecem·

ber, 1010, PP· 1 7"~-4.
(:YCLES
ct1n·e is se1J;1ratf:cl into tl1c: <":<)Sit1c <:urvcs, ,vf1i<:l1 ;ire c;1Jlc':cl c.(>r1stit11-
ents, is now used by all leading rnaritime nations in pred icting
Lides. lt is f1aserl 011 t.11«: pri11<":ir>lc: t.l1a.t any f)eri111lit: 111c>tic>r'1 or os-
cillation <::an always he resolved into 1.he sum of a series o( simple:;
J1armor1ic.: r.r1c)ti<>r1s.
Here l\'(-! (:C)Tll('. LJJ>Clfl a fac:t r>c:rl1a1>s cliffic~tllt. l>llt i tllpOrt.aflt
to
ur1dersLa11d. -rl:ie 11t11nt>e-r of <.:011stitue11c.s i1.t ;1 t1:1r111or1ic ct1r·ve is
theoretically unlimited. Constituent• are mere building blocks, as
it were. They are something like the pk ces in a jigsa•v puzzle -1·ou
can keep on n1ultiplying them indefinitely. by cutting them up into
still more pieces. This docs not mean t.hat the curve is not " real."
The 1.heoretical infinit)' of constituents inercly reAcct.s the fact that
you can theorctkally divide at1ything into an infinite number of
par1s. ln practice, we don't.
Jn predicting the tides, for instance, the const.it.uents which are
of practical importance depend on the precision requir(:d in rhe
predictions. :Each constituen t is given a symbol or name, set by
international u.-;age. Thus tl1e principal lunar constituent has a
period of 22.42 hours, which is r.hc semi lunar day, and it is called
:VI,. The period o( the principal solar constituent is 12 hours, or
half the solar day, and is called S,. T he principal lunar daily con.
stituent is 0 1lkd K,, and the principal solar daily constituent is
call<:d 0,. The solar annual constituent is called Sa.
Jn Fig. 6, which we have already ohscrved, the consti tucnr.s M,
and S, - after being traced separately- are shown combined into
a c:ttrve. i~t1tl K 1 a11tl Ot- 1Ji\..t t11 ctl ~cl''1 1.1tc1 y - arc .1]S(> sl'tfJ''\'11.
con1bined. 'f he prcdktcd curve is a synthesis of these and i 7 ocher
<::onstituents, or 21 in all.
:Because the mathcm;itic;1l computations involved in using har-
monic coust.ants for tide predictions are so laborious, Thornpson
developed practical machh1<:s for rapidly analyzing and summing
the various constituents. The predictor that has long been used
by tl1e U.S. Coast and Geodetic Sun'<.~Y is b;ised on Thompson's
original design, now 1nud1 improved, and is constntcLed for sum-
( 168 J
Ar-:A T.VSIS AND SYNTHESIS
ming 37 constituents. In about seven hours, it will make daily pre-

<llctior1s for a year at a g·i\>·<.:11 J)<)i 11t.
Predic1.ion of such a curve thus involves two principles: analysis
of the known curve to obtain its constituents; and 1.hen synthesis
of the constituents t.O permit extending the curve into the future.
Synthesis also serves to verify or prove the correctness of the analysis,
by reconstructing the original curve - or a close approximation
thereof - from the constituents.
,..l 'he ma<:hjn~s t.1s~c.l for ana1ysis an<l S)•ntl\esis are employect in
many fields. Professor A. A. ~tichelson devised an inge11ious har-
monic ;malyier and synthcsi1.er for So cou1ponencs, which has been
used to study light waves. Such machines are also employed in elec-
trical engineering for the study of alternating current waves and
other periodic c urves; t11ey are useful in fields as various as elec-
trici1.y, acoustics, optics, astrophysics, naval architecture, and aero-
dynamics. The rnachines accomplish what the mathematicians can
perform equally well by rather laho,;ous 1nethods which call for
the resolving of involved equations.
Harmonic analysis is based on 1.he mathematical principle called
Fourier's 'J'hcorem, for which Ilaron J.B. J. Fourier first published
the statement and proof in i822. The Fourier F.quation is not a
tool for laymen untrained in mathematics; but it has been s1.aced as
follows, in language a layman can grasp, by Or. Dayton Clat·encc
i\<liller of the Case School of Applied Science:
If all}' Cltn.•e l)e gi\.·cn, l1a\'ing a \\'a.\.'C lengtll /, the san1c curve c<.-tn
al,vays l)e re1'>rodtJc:c<l :.tnd in 011c p:trticul:\1· "''ay Otlly, b·y corn1)0l1ftt;li11g
si1npJc l1ar111ori.i c ct1r·\:es of st1itable amplitudes a11cl pl1ases, ir1 gc:11cr..t.l
jnfinite ir1 11ur11bc-r, l1~1virtg tl1c S<-tme a:xis. a11<l ha'\;ng '\la\•e le11gchs of
l, Vzl, ~Jal, anci s11c:c:tssi,,e alic1t10t parts of l; tl1e gi,•en c11rve may ha.\'C
any a1·bitrary form whateve1·, incliiding an)' number of straight por-
ti01L<, provided that the ordinate of the curve is always finite and that
the projection on the axis of a point describing 1l1c c:urve moves always
in t}1e s~t1nc directio11. •
• TJie ::,·c£ence of 1\!usical Sott11dJ f l 11e ~facmilla11 Co111pan}') , p. 93.
[ 169]
CY Ci, ES
And Dr. :l>filler explains the application of the Fourier Fquation

as follows: " ~fany of the curves studied by 1.his method can be
exactly reproduced by compounding a limited nwnber of the sim-
ple curves. For sound waves the n11mbcr of compon<~tus required
is often more than ten, and rarely as many as thirty. In some arbi-
trary mathematical curves, a finite number of components gives
only a n1ore or Jess approximate representation, while an exact
reproduction n:quin~s rhe infinite series of components." w
In ti"~ study o[ economic rhythms, harmonic analysis is only one
of our tools. The metl1od does give an over-;·1ll picture useful for
preliminary and reference purposes, and also gives to any dcsirecl
accuracy a mathematical description of the behavior. Rtlt harmonic
<tnal)'Sis i.~ of ttse prim;1ri1}'· ita e:onneclio11 , ..,ill1 a ''°'a''e- st1c~1 a.~ a
sound wave - which is n1orc or less exactly repetitive.
Harmonic analysis fails of itself to give t.he true kngt.h of a
rhythm; it ignores individual variation and changes in amplitude
and shift of phase. It also fails to unscramhlc a wave frorr1 its own
alrr•ost-l1arnl<)tl.i<:s. i\·f ltltiple 11ar111011ic a11al~:sis gi\:cs ;;1 11tLL<:l1 clc)ser
approximation to the length oE the rhythm; but ocher than this, it
has all the lim itations of ordinary harmonic analysis.
Some twenty other rnethods of analysis are available. The peti·
o<lOf,TT<'trll is a 1netl1od of ap1)roacJ1 to att<tl)'Sis \Vf>ic~l1 \\'C <:(tn use
when tl1e problern is to determine dominating waves in a series of
data the periocl of which may not belong to the Fourier st,qncncc.
This method is also a usefn l preliminary in clet.ermining rhythms.
For the student ot economic Jluctuations, the Hoskins Time
Chart is a tool of great usefulncs~. It 111akes pos,;ible a rather com-
plete analysis o[ a series with celerity anil ease, ancl suggests and
reveals rhythms that do not appear from 1.he study of an ordinary
graph through Jn(:re inspection or the use of a gniduated rule. 'fhe
method does have its limitations; it fails, for instance, to give either
the shape 01· the amplitude of the wave, or to give the length as
accurately as some of the other rnctl1ods do. Rut these shortcomings
• Ibid, p. 93·
ANALYSIS AND SYNTHESIS
can be rnet by suppkmcnwry use of the l'vioving Cycle Average, of

the Periodic Table, and by other methods.
Unfortunately. we as yet l1 ave no machines generally available
for conducrjng research work in economic rhythms on a scale match·
ing the importance of the subject. 'fhe so-called Cydograph exists
in the form of only r.wo machines - one is owned by its invento1
in Chicago, and one is the property of Andrew E. Douglass in
Tucson, Arizona'.
Electrical sl'':' ctrttm ;-tnal'ysis, if a m:o1ch.i11e .-nrJ tlle skill to •·11n it
were available, would save vast amounts of Lime that must now be
spent in multiple h;irmonic analysis. F.lectromecha.nical machines,
if they were refined and made generally available, would save much
time in periodogram construction, with the added advantage of
finding tltc: valnes for all positions for each period, as in the periodic
table.
Such matters are mentioned here in passing, and only briefly, for
the: purpose o( indicating to the general reader the scope of the
problems involved. The problems themselves can safely be left in
the hancls of t11ose with the technical training for their ultimate
mastery. Our own imtnediate in;.erest is confined, fron1 this point
on, 1.0 answcriug the questions with which our study began.
XI I
fCJ::N WE ASK '"·J1~t tirne it is <>n tl1c.: clo('.k, \·V(: are reall}'
W asking a spatial question. Its fundamental meaning is:
" \\/here is iny spo1.of earth in reference to 1.he eanh's course around
its axis ancl around the sun?" One <:ould plot the answer with a
graph . A clock is merely a kind of graph in mechanical form.
\\!'hen we ask whe[her a business should expand, or wht:ther its
stock is a g(>o<l irl\:est1r1c:r1r,, or '"·he(J1er it earl i><~tter raise fL1n<ls 1>}'
selling stocks or bonds, we are asking a similar sort o[ question.
F'or 1.hc essence of our qucstio11 is mere!)': " \\/here does this busi-
ness stand in reference to its coutsc of life?" Every business has its
rhythms. ·rhcse rhythms, wr;ippccl around its trend, combine to
make a path that can be graphed like an orbit. \\!e can discover
these rhythms by analysis, much as we discover the separate rhythms
in the tides.
A l>l1siJ1ess is an orga11- rnajt)f or minor - ir1 •J1e \11.1 hole ec.:ott·
01ny. \\!'c should therefore expect to find in it at least one of the
rr1ajor rl1}'(.l1tr1s r.hat "''<~ tlJ<!Ct i1a tit<~ c::<:c)notn}: as a \\,. llo lc. '·\"c t)rdi-
narily do. In addition, we often meet one or more rhythms seem-
ingly unique to that particular busintss. ~Vhy this should be so,
\\'C clc> Of>t )'·ct k 11c>w.
An illus1.ration of t.he sales chart for one business is shown in
Fig. 1. 'l' he business represented is an important public utility
company of the United States. The sales from 1883-1939 in physi-
l 172 J
Tll\'llNC A BUSINESS
cal units arc plot.t.ecl on logaritlnnic paper to show rate o[ growth,

in the fashion to which the reader was earlier introduced. It is
worth noting, at this point, that this entire plotting and analysis
was made in 1940, before the United States bccarnc involved in war.
Index
1000. rTTTrnrnrrrTTTTTTTT"rr1rrl"T"T"TTTrT"rTT1rT"'l"T"TTTT, .,-
, ...,..
, ...., .-,..-..-,......,..,..,..,-~
. ,.,
100.
10.
1.0
.001 U.LLLLLI..JU-1..U..U..U..LLLLLI..J..U..U..U..U..LI..JLI..J..U..J..l..U...U..1..LLI..Jl.J...l.J. lJ..!

1883 1890 1900 1910 1920 1930 1940
Fie. 1. A SAtns Ct·JART

S;tlcs of :t Public U1ilicy Company 1883- •939. Ratio scale.
Figure 1 shows us a company t11at has had a strong and healthy

growth - one whose growth is continuing in terms of annual in-
creases in dollars of sales. But the fact that the slope of the li.nc
0
is getting flatter shows that the 1·ate of growth is now on the down-
b"·aclc. The company is rapidly becoming mature. A comparison
with some of the charts showing the trends for 1najor American
industries, as indicated in the early pages of this book, will show
[ 1 73]
CYCLES
a marked similarity in the trend line. \Ve are, therefore, dealing

with a more or less typical major indusc.ry.
Nothing that mecL' the eye of the reader in Fig. 1 sugge~ts any
parcicular rhythm - there arc merely some apparently irregular
Per Cunt
200
180
!GO
A
HO
120
1900 1905 1910 1915 :920 1925 1930 19l5 1940
Fie;. :. A R.\'ff;: OF GROV.''l'H CHART

Rate o( gt'O"''lll., i903- 193~» of tl\c Po1.>lic: Utility C•>r.npa•1y, tl1c sales f>f \V})icl1
ate gi ...eo, in J7jg._ 1. ·1·11e curve represents a 3·ycar moving pcrccnca.gc:, an(l shr)v•s
(he perce11tage b)• t.;hicl1 snles (or c;,icl1 month. ~tr<.: ;JbQ\'t; or lu:l«)w 1J1ose t)f tlie
c::C)rrespoJ1dj11g; Ult)ll.th three years before.
deviations from the trend. Rnt with our knowledge of rhythmic

behaviors we know chat rhythms in figures are not always readily
apparent. If they were, everyone would know all about them. \Ve
must look beneath the surface of our d:tta. "Ve learned earlier that
calculating the rl<le of change will often reveal a suspected rhythm
with great clarity. \Ve therefore calculate the monthly rate of
cl1a11g·c i11 tl1is bt1si11ess, ttsing· in tl1is instance (t 3·}'e<1r inte1·,:aJ. '*"
That is, we use a 3-year moving percentage.
• TJ\e pos.\ible treat1nents of figurts to reveal rhythm are exceedingly nu..
mero\1s. They include, bC$\clc vario11s lll(lvif1g 1-,ercentages: (l) dt!viatiort <>f the
daLa fl'Ot\t various si1n1)le moving a\•erage trends; {2) dc''ia<ion of chc (Jaca
from "·arioos '\·~ighcc:d m.tJ\·ing avM-age t1·c11ds: (3) deviation of tl\e daLa from
various n\aLhen1aLir.al trends; (4) deviation of vari(u1s "'oviog av<..·ragc.-s <>f the
data ff(.>m various movir1g average 1.re11ds; {5} deviation of \'al'ious moving
a·verages o( tl1e data from '':.ltious nu'lthcm:.ltietl ll't:nds; (6) tlC\'iat.i1)fl of one
1l1atlle1•1atical u·et)d from another; (i) ''a.riot1s u1oving .sct•lon avc.:r;i.ges. 1\J.
[ 174]
Tl'.l.HNG A BUSINESS
Figure 2 shows us the result. If we have not. yet. revealed a rhythm

dear to the eye by simpk in.~pection, at least we have before us
a chart that shows wide swings in the rate of change.
Pet Cent
+20
JI
0
-20
+2<1 -
c
0
- 20
1900 1!>05 1910 1925 1930 1935 1940
FIG. 3. 'T\\'0 Ril:GULAR CYCLtS

Cycle B h:is a petiod (ler1gt.11£1·00111igh to high or lotv- to l<>tv) of ~}O Tl'.lf)tltll.~
(7~'2
yrs.) . Cydc Chas a periC>d of 220 months (•SY., yrs.).
Our prohlcm now becomes one of analysis. The out.come is t.he

two sets of waves shown in Fig. 3. \Ve have come up witb two per-
fectly regular rhythms - one of 18t years, and the other of 7} years.
They were concealed from us in Fig. 2 because crests and troughs
\\~ere so C(>ml)ined, at ,.·a1·iot1s times, t~aat crests so111eti111es aug-
mented crests and so1nctimes tended to wipe out troughs, while
troughs sometimes combined to deepen troughs and at other times
obscured crests.
But we have not yet finished our task. As one test of our work,
we must move from analysis back through synthesis . .Fignre ·1 shows
the two rhythms, which were isolated in Fig. 3, recombined into a
complex. Figure 4, when compared with Fig. 2 , shows a sug-gestive
similarity. Out work, however, is Still not finished. \Ve are dealing
in this business not only with rhythms, but with a trend. Figure 5
though tJu~ JneL11ods a1·e varioltS, chcy all J1a\'e 011e basic pur1>ose, r1arr1cly, to
subordin<l.tc or minimize the irregularitia (irJcJucling rl1ytl'1nis of other lengtl1s)
~o tllat a rh}<thm; i! present, c.1n be observed and studied.
[ 175 l
C¥CT.F.S
Per cen1
D
01------,1--~~-'\:=='"--;:~~
-20 .(B+C)
Frc:. '1 · S\':-.'TIIESJS
·1·:\e 7~-yc:1r :lllfl tl1c 1873.yertl' C)'Cles combined {Cllf\'e n ~ Curve n plus
Curve C).
180 I - .
)COi- E
140 - .
120~ .
-
. . . .
190) 1905 1910 19 i ~ 1920 1925 1930 1935 1940
Ftc. 5. 1·11E 'l'RENO Lr:-;v.

T :-t·nf 1J1c t•ublic U1ilil1
Cotnpan~·. the ac1t1a l s;1l1:s. nr \\'l'lic:lt are cl1artcd in Fig. 1, The (A>rnpa••}' is grov.·-
iug, hut at a dect'easing rat1·. Fc..>t a !J:ifopler exanl:plc, rcfc-r to tf1c hypotllctia.1
c:omi:·:lr)}' e,_fiS(;ll $.Se<l ift c:J1aptel' l.
~hows this trericl - the bask underl)'ing trend in the niteof growth .
Herc the graph is in terms of per cent. Tt i~ decreasing, and our trend
line slopes steadily downward to the right.
Now, in Fig. 6, the trend line delineated in Fig..5 is combined
witl1 the complex wave, created by synthesis, shown in fig. 1· The
total synthesi.~ gives us something dosely approaching the wave we
s1:arced out with.
For convenience, Fig. fl is repeated under the new synthesis in
Fig. 6, so that we roay make a ready comparison. \\.'hen the two
[176]
Tl~UNG A BUSINESS
curves are compared in detail, it is clear that th<:re an: certain

ftw:t11;itio11s in I.he true curve o f Fig. 2 for which there are rio
counterparts in Fig. u. On the whole, however, the comparison is
excellent, and conformity is sCCTJ to have been especially dose in
Per Cenl
Per Cent
2"0
180 -
160
G
140 -
120 -
100
1900 1905 1910 1915 )920 1925 1930 1955 1940
Frc. 6. ' l~REt\D L1N1: ANT> Cvcr.rs

Tr1:nd of tl1c Rat.e of Cro\'ith oi che Pul>lic Utilicy Corop;iny a:; :-;110 ..\'tl i11
Fig. 5. plus the coml>inatinn c1f lf1e 7~4· ~·ear an(l the i8~·~·Yc;-ir cycles, as s l1 0\\'T)
in Fig. ·1· Curve F = Curve: B pJu;o; Cut'\'e C.: plus E. for p11ry1<1s~s of cc1n1parisc1r1,
curve ,·\ i.s rcprc,Hlucc::<l agai11 a1)d labeUc<l G.
the past twenty years, or since the business h<is reached maturity.
' ·Ve may therefore: assume that we have isolated the major rhytl11ns
which move along this rrend line.
The conformity between our s)'ntltesi& and the actual curve tnay
be seen hcue1· if, for th<: past twenty years, we draw the derived
curve, as shown in Fi~. 6, in clot.tcd lines, and phice it on the a~tual
cun'e first shown in Fig. 2. This operation is performed in Fig. 7.
[ 1 77:
CYCLES
The synthesis of the two waves (7 ~- and 18} years) is projected here
for the next twenty years.
Figure 7 shows us, in other words, the application lo economic
prediction of the same principle that is used in predicting future
tides- i.e., the synthesizing oE the consLiLuents after they have
been isolated in analysis, so that the synthesis may be projected into
the future.
Per Cent
/-'
/ ' ''
120 .. -~_/ \
1001-- -- - ----'......,.,t..-------'~-----+--1
-----'-..__/
Fie. 7. PRo]Ecno~ AN'n Co"' PAJttsor-;

Tl\c 3-Year ?\:loving Percentage, 1920-1939. of the Sales of the P111>1lc l Jt.ility
Companr. Lhe actual sales of \V"hich :ire given ln Fig. 1 (s<'1fid li1l.t:), togecher
' '1icr1 the Synlhesis of (i) tll.e regular 7Vz~Ycar Cycle. {ii) the regular iS~·Ycar
Cycle, and (iii) the T1·end, as shown in fig. 6 (broken liue). The synthesis
and trc11~l arc projected to 1959, to givt~ a1\ idea of what lvill l1a1>1>e1l. if the
rhythms ~tr1d tlu; treud (;( HI tir111e as i1'l the past.
For this particular company, the projection has been cxu·emely

dose to the actual curve, as sub1equent experience has drawn that
curve from 1939 through 1945. The two rhythms on which the pro-
jection is l>ased g(> 1>ac:k •LS far as Ll1t s~1ic:s- [)l )'~a1.s - a1'lt.l prcLLY
well describe the series' behavior, at lea.~l since the company has
reached its 1najority. The probability of their continuing seems
l1igh, and the unwisdom of ignoring this probability would seem
real.
Jn Figure 8 the rate-of-d1ange projections have been convened
back inlo an actual physical projection. Figure 8, t11at is, is the final
part of Figure 1, enlarged. 1' his particular chart. showed that in
1938 and 1939 the plant facilities could be expanded boldly-in-
( 178]
'f!MINC A BUSINESS
cidentally, at a time of reasonable c.:osts. It showed that between

19~9 and 1946 refinancing shou ld be possible on a most favorable
basis, presumably bcc::;1use of satisfactory earning statements of a
kind that might not be pos.~ible in the following clecadc when earn·
ings would probably be drifting below 1939 averages. The marked
downf,rr'adc indicated in the chart after 1946 suggested the wisdorn
of considering whether earnings should be generously distributed
...
10001.-..-..-..-..-..--..-..-..-..-..-..-..-..-..-..-..-~~
eoo
. ~. ~~~~~~~
,. ......---............... --... ____ _

$00 ~!:----- ------------....-:::::
400:~----::-=-~;,;;.-:o-:::."'-;-.:-"'-"_"'_-;. ::~-..,,,
300
200
------ ---
l0011-1=L.JL.J-l-l,,.!,--L...L-L..LJ...!--'--'-i,...[-JL.J-L-l,_,.L...L..L..L.J
1930 1935 1940 19·1S 1950
FIG. 8. AN E!\"LA.ltG!:'~ff.N'I'
SaJi?s or 1l1e Ptibltc Utility. 193<~1939. (Tl1e last ltt) years of lhe CUt'\IC in
Fig. 1 are here enl:trgctl.) R11tio scale. There have been added the trend, and
:l. projection i11t<, Lilt~ £ult1re. based <1tl a 0011versio11 of the rate c)f cltange liglu·es
clJ:-tr1c·1J i1t Fig. 7. tt) give ail idea 1)£ \\·hac v.·ill happen if the rJ,ythtns and u·end
contint1c a" ir• 1.hc 1)asL. l;or a sinlpler example of tl1e .sarne sort, 1·e(er to the
hypothetic~tl co1npany i,1 Clta1)tcr 1.
from 1939 to 1946, or whether surplus should be built up to hdp

maintain dividends after the decline sds in.
Clearly, on the hasis oft.be patLern, 1944-l!J45 were years when
no new vice-presidencies should have been creatccl; 1945- 1946
were years when new employees shoulcl have been taken on slowly,
if at all; and 1947-l!J<!S show as years in which ernploycc.1 may leave
to be weeded out, Certainly our .chart tells us that they should not
be retained merely in the hope o~ a tevival in sales two or three
years lacer.
After 19.16 the company may begin 1.0 find itself with some
excess plant fad Iities; undoubted!)'. after 1916 has co1ne and gone.
and the downturn has definitely established itself, the company
[179 ]
C Y C L F.S
should consiclcr s<'rapping surplus plant anrl equipment that will

not justify ca:-ryi11g dw1·gcs du1·ing Lhc eighteen years or so that they
wiJI not be nee<led. For our chart shows dcli11i1.dy t.haL, il the
unde rlying growth t rend Cor this particular business continue.• to
flatten at the rate of t li c past twenty years, and if tlae cydes con-
tin ue, more than t•igl atcen years will pass hcfurc t.he company will
again reach tie levels of •!)16.
I nformation of this cha meter can he put LO work not only in con-
nection wit!. quest.ions affecting production facilities, financial
management. and employment, but in every department of the
business: inventory coratrol, purchas ing polk it•s, t •c Liming o( ad-
ve1·t.i.,ing and sales promotion, and the clin;ction of public relations
pol ides.
Because t his particular utility company reflects the o perations
of a whole cornmunjty, its own r hythmic anivity is closely asso-
ciated with the rommunity's life. H ence the projL'Olion we have
drawn for it will apply with not too great a divergence to the eco-
nomic acti vity of the 1-ruire. cir.y iu which it operates.
In 1946 Lhi s city \>'as enjoying a b"T<:at real estate boom. Resi·
dential r ents W('fC under government control, hut ren t~ for com·
r11c:rcial propertic.·s ,,:1.;re;: so;1rir1g. L 111d)<)rcls '''CI'e demancling r<~t\ta.1
increases of from 100 to 300 per cent front sornt: business tenants,
together wi Lh new leases r u11n i rag from five years up. T he b'Teat. busi-
ness activity reflected in Lite public u tili ty's saks 1·nabled many Jand-
'lords to get hea\'y rcnraIs on the long-term leases they demanded,
(u• tt:11i1r1t.) ''("c1c dc$pcr~'l tc. A s..J.aortat,"t'.:: of lofl bui~dings ru-1<l other
co1nmercial space left no place to move.
But if our chart correctly portrays the [uLure of Lhis community,
there will bea great letdown well before t he for1ies have ended, and
those with long-1.e rm leases at high ren tals may have regrets.
Interestingly enough, the 18t·year rhyth m which has been iso-
lated in the life of this util ity company is the one 1hat shows up so
dearly in real estate activity - the rhythm that ' ·l 'arren and Pe•r·
son have declared 1hc next tnost important in our economy, alter
[ 1So]
TBtl "(; A 8 OST NF.SS
price movements. For it is closely associated with the activity o[

many basic industries. \Vli et her the 7 ~-year rbythm, which we find
opcrJting beside it in this utility company, exists in ot.hc1· activities
of the metropolis where the utility is locarcd is a question whid1
only loug- study and exploration could answer.
This chan incl"ccl i:iises more questions than the prc~ent state of
our knowledge perini11 u• Lo dea2 with. For instance, when tbe
projection in Fig. 8 was clrJwn, it was not known that 19.p-1945
\'t·ould be '' Y::tr ~·c:it~." Ilul notice the JargC in crnse in sal""' r'rO.
j~•·tcd for those ,-ery ye:irs, 011 rJ1e basis of the rwo rh)'tltms isolated
in this company's operat io ns. There is much in tl1c analysis of
r~1ytl1r11 t.c.1 i11dicate tl1at t11an)' rises in prircs. sales, a11d otller itl-
dices of ecor101r1 ic: ac:1iviLv co111e 0 11 ~<~l1eclttle, \\'ar ot 11'J \.var. i-\11d
'
dci:liues come sirnil:irly. A' slated befon::, war· may, and usually
does, distort the 11.rnJ>lillldC of a w;ive - just as a st1"ttng wind may
sweep a high Lide higllC'r - hut it often sccnb LO interfere with the
timi11g of tbe wave to a v~11· small degree, if at all.
'!"he proj~'t:tio11 shown in Fig. 8 warrJnts a word o{ warning as
rl'g;1rd' the war quntion. Some observers, around 191•. put their
linger on cite r 94G peal.. and said, in effect, " Ah, that's when th e
\.\'(1 r is gf>i11g to c11cll '' TJ1ere is notl1i11g· i11 oi1r ktlU\\•ledg·e of
th)'t.l uns tliac could warrant this kind of interpf(::Wt.ion. It can be
ha1ar<l<JUt\ to read Sl1Cll '' rc.;asc>tlS " into the.tic rl1ytl1m_
ic pattcr11s.
The piccun::s M:en in the graphs al'e incvit:rhly very generalized
ponrayals of Lhe lant!S<apc t11at lies ahead 11£ us. In t his method we
ha,·e a tool for our journey toward the future, but it is not a key
which unlocks all !tl)'•tCt ies. "J'he further we go in c•ery science.
the more profouncl thc chasrru in our knowk1lgc ran appear. One
such g;1p faces us wht·11 we note the projection for the utility com-
pany (as show11 orl page 179) for the years ;rfccr 1916- The 1011g
dedine fon:c.a•t rhct·c agTecs very well with the decline in Arn<:l'i-
ca's 11aLional econorrlir a«tivit)' ,.,·l1i<:l1 tl1e fo111· c.·f·,ic:f econoJ11ic
rhythms are now forecastini; after 1946--1917. nut o nly one of the
four national rhythms-tile 18tyear-is a wn>titucnt of the
[ 181 J
CYCLES
utility company's wave. \Vhy, on the basis of different constituents,

the waves for the nation and the one company should be so similar
in character is not for answer here.
A similar analysis and synthesis is shown Cor an industrial com-
pany in Figs. 9, 10, and 1 i. Fig-ure 9 shows adjusted net shipmenu,
1871- 1940; a !·year moving a''erage has been charted here on ratio
ruling, to show change in rate of groWth. Figure 10 shows the
lndea . . . - - - .- - . . - - - . , .- -,
2,000
1,000
soo
200
JOO
50
20
10
t,,~o--1sso.,...,._~1a~9-o-~1900,..,_,._~19~1~0-~11~20.,....-1930.,...,._~11·,~o-~19~s~o-~1960
~-~1e~ro
Fro. 9· SAt.J:S OP AN l~"DUSl'JtlAL (..o).tPA_NV, 1871-1940

Three-)'eU mo,rir1gavcrage. Ratio JUle.
analysi$. 1A'e see a 9}-year wave (A): an 18f-ycar wave (B); and an
underlying growth trend (D). The effect of \Vorld \Var I on this
company shows in a lift of the growth trend from one level to
anotllar. The pat.tern and the rate of growtll were not affected at
all. \Vhen the company was expanded at this point by the war
forces, tbe trend line merely concinued !cs esta!Jfuhtal cour•e at a
somewhat higher level. Figure 11 shows the s~·n11lesis of A plus n
plus D, projected to 1970, and matches it against the actual data
(until 1940) t11at was first plot.ted in Fig. 9. It will be noted that
the syntllesis shows very close correspondence with 1he actual data,
as far as these data go. After 1940 war forces lifted accual shipment
figures above the synthesis line, much as in \\lorld \\lar I.
The data for another corporation are shown in Figs. 12, 13, 14,
15, 16 and 17. Permission to use the data here bas been granted
[181)
TIMING A BUSINESS
2,00<l
1.000
---- -- --
500
200
100 D Ttrrtati.,. Uhdef'tYln& Orowttl trend

-- ~-
S4
20
10
Fie. to . .-\KAtvs1s
(A) A regular 9',3·Ye01· Cj'de: (S) a regular 18%-Year Cycle; (C) the Two
C'.(>rr•l>it•c<1; (0) the underlying Gro\!/lll ·1·rt-nd of tl\e 111dustrial Company. c.he
<iu.:tual salc::s of lvl1icl1 arc cl1arccd in Fig. 9. Ratio scale.
lnd&x,_---,-- -.--.---..,---.---...----,--- -,.---,----,

2,000
1,000
~00
200 cutated Veluo

100
so
20
10
1~'-,o-__:1_8'-80--,.18,\~~.-~,•.Lo-o - 1- ...,-.--1-'920c,-__1.,..•a-o--1•._•_0__1_m
..__ _1_,•.,_&0~-1J•10
Fie;. 11. SYNTIIESIS

Sales of an inc!usttia.1 compan)r, i871-1940 (3-year 1novjng averages), as
shf>V.'t\ir1 fig. ~). togctl:i.cr •1•ith a ~yntl1esjs o( the regular 9%·}'C:;tr cycle. the
J8~·$-ye<ir (:yclc:, <incr 1l1(· nnc.fcrlyjng gro\vth trc11d, as slio~·n in Fig. 10. Ratio
scale. This projection gives an idea ol what will happen il the rhythms a.nd the
cre11d continue as in tltc past.
CYCLES
lod<•
400
3$0
300
250
200
150
0
1928 1930 1932 1934 1936 193tl 1940 1942
Fie. i2. SAt.f.:\, 19:17- 1941, Ol{ •.\NC.) 1 ·H1~1t INOU1"1'1t1AL Co~<rANY
A 9"ID<>nth nu1v i11f.~ alfc:1·a.ge o{ tltc cJat:i, <i1lju'ltccl J'~>r !o.Ct.1SCJ11al r·l1ythl'n.
Index
300
zso
D
200
1$0
100
Pu Cent
+10
0,/\, _/\ , /. ,' '
-10 E .\/' ,, ' '
1928 1980 1934 1938 1940 1942
F1G. 13 . •.\NAJ.\'$lS
(A) A regular 41.month ci·cle; (B) a regular 2.1-momh ci-ele; (C) a regular
33.monch cycle: (D) a 10-yea.r wa•-e; (E) reiidual showing 15-monch cycle.
[ 18.t ]
TIMING A BUSl "ESS
,..... .......~~--.~-.----,
350~~~~~~~~~~~~~--.~-r~.--
.,,,...,,., I
,,
\
' "
I \
/ \I I ~ I \
I ~
I
J ',11,
,I r\ I
I\( '
\.. \
250 J u 1 \ I \
/ I /1 I 1 I I
1" \ I \ I \ I l
\J 1 I \ / '
I
1, ( ,., \
,-, I \. \
I
-
I \..,,,"" \,,
I I •
150
f•'u;. 14. Svxrne1s

Abo,re, syntbc~is ,.,f tJ,e Jegular cycles •1JJ tl"le 10-yea.r '\\';t\'(', slu)v•tt in t""ig. l'.'j
(dotted line) ; b<:I''"'·· ac1u;,J >a.lie& o( this romp."i11y as ~l\O~n origir1ally in Jtig. tJ
(><>lid line} .
100
so
0'--1~..J....~'---'-~-'-~'----1.~..J---"C...,--'-~.J-,,,,.......,__.,~..._~.,.,..,.,..
1928 l!JSO 1932 1934 1936 193.8 1940 194Z
l"rG. 1.'i· ('.OMP.4.RJSON

TJu: QJr\·es sl10\.\-TI in Tig. •4 supcrim1Jo06ed.
CYCLES
on the r.ondition that the c.orporaLion not be identified even by

re[erence to its fidd o[ industrial a~tivity. It may ouly be said that
it is at'l i1nportant orgar1izatio11..
Per Cent
+2:0,---,-,...,--~r--.~-,,.-~r---r~,., ....r--.-~,-"7"1,---,-~,.---,
+10
Ok--/---'>,--F"'<--/----'\-,F-".._...~---'\-,F-".._...~__,1-;r-:>'<--,t.I
-10 -
-20 -
+JO
o i-.1---\~-+-~-\--l'~-+-~1---\~-+-~\---1-~+-~~----~+-<
-10
+10
-1~F~=~="'-:':7.c:..::::,,,c::7"'=~=,...c~"C:7=~
lndlJJI
300
250
200 This line Is Pu1e Con;cmue. tc tie Modiliod es Time Unfulds
lSO
100
Fie. 16. .PROJF.CTTON

1 'f1e 1·egl1lar cyclessl1ot\11\ 111 Fig. 13 projcctt'\\'
tile dt:Ht:t.:lio1l froni t1·er1d a.11d f1·om ruajor '~a,:e-S th:tt \.\-ill t.akt: plac:e if tile
rl1ytf1rns t,;1111ti1auc as i1l the past. 'l'J1e to·year ~'ave is projec.:tcc1 111erely a.~ a
c..onjt:<.:ttl re.
I<igure 12 shows its total United States sales by months, 1927-

'!H 1, plotted a.~ a 9-month movir1g ~1vcragc ah.er adjustment for the
12-1nonth (or seasonal) rh)•thm. Figure 13 shows the analysis, whkh
produces a 4 1-month rhythm of a peculiar pattern; a 24-rnonth
rhythn1; a .~~-uionth 1·hythm; a 10-year rhytftm; and a resi<lu;iJ shown
:tt the hot.tom of the cltart as F. where we find strong suggestion
of a further rhythn1 of approximately 15 months.
[ 186)
Tl~lJNC A B USI N£SS
Figure 14 shows, at the top, the synthesis or these variou.s rhythms,

and at the botto rn - Cor e<rsy compar ison - die actual line that was
first plotted in Fig. 12. The correspondence is close. In Fig. 15 we
+20
+10
•l-:;'---\--l--\c----1-4--1-~--J~t.-~-1-~'--·f.-.:.~----I
- 10
,....
-20
350
300
250
200
50
Flo. 17. F1NAr... SYN'T1·n:.~1s
Synthesis of the q·clc& :.t-,<l tlle \\'a"c sl10\\'t1 j,, •~ig. 16. Top. the 15-month
cycle: middle. 1J1e S)~nt.hois Of the 41-m:>nth, 21-m0t'ltl\, :tod 35·monlh cycles:
botlOJ.U, S)Titha4 o! all tour rhythm5 with the oonjcc:rural io-year \\'a\'C.
can compare our synthesis and the actual line in another way, since
one is superimposed on the other. H erc the doued line represents
our synthesis, and the solid line the actual sales lig111'c.
'Ve ohsen·e that our synthesis checks very well with the" real ,.
line. Titus we are ready for a further step. Jn :Fig. 16 our various
rhythms are proj ected into 19:;6. The 10-ycar wave here is pro·
jectcd as a conjecture, inasmuch as this panicular corporation has
tooshortahistory (dating back only to 1927) topc:rmitofanyopin·
[ •871
CYCl.F.S
APPROXIMATE TIMING OF FOUR MAJOR

I I
1925
WHOLESALE PR.ICES
POSSIBLY PRODUCTION
AND INTEREST RATES
1906 1925
REAL~ATE
BUILDING ACTIVITY
COMMON STOCK LOWS
MANY INDUSTlllAL SERIES
1915 1934
A
WHOLESALE PRICES
COMMON STOCKS
v
MANY IN DUSTRIAL SERIES
v
P~CESl\F' !Nvftco,,,fco1T1~s
PRICES OF MANY SECURITIES
y v
RODUCTION Of. MANY BUSINESSES
y v v y
'
I I I I
1900 1910 1920 1940
Fr(:. 1 R. D1.\C':'RJ\:O.l"-«A'11t: Jtv ... Rr~~ENTA.TIOK or ·rtt~
It s.hot1Jd he pointed <nit aK'.ain Llu1L 1ltt:llC rhytlnn$ n1;1nif<·'\t tli('n1st:lves pro-
~ressively i11 \'.1riot1s ~>rts <•f l1l1si11t:~li uucJ fH i<.-tS nt\(t tl'lnt 1hc tirninK is there(orc
only a1>1)r(1xi1nn.te. The <.liab'T<tm ruaS.cs 1111 pr<:tcnse ol mo,\·ing rclati1t·e sLre11gll1
( 188 J
TIMING A ll USIN F5S
RHYTHMS IN THE AMERICAN ECONOMY
54·YEAR RHYTHM
1952
1943 1961
1952 1970
1940 1950 1970

' .
1980
51·Yt1AM, 18Y;s·YEAR., 9•Yf.AR AND 1 i·f\{orvTJr RllY'l'ttfl.t'
o( the ""'r ir111.. wav(;o;, i.••l1icJl also varies with Lhe seriC$. Th(~ 11n>jec"Lio11S Crom
194(; to 198 0 shO\I/ the stnJtturc that v.·ill prcvnil if tJ1ese rhythms t(tutinue ..;\
ruler laid u1> and do\'l'n the page '''ill slt(i\\' rhycl1ms ir1 a1•>1 givc11 year.
[189]
CYCLES
ions as to the reality of this wave. There are only two repetitions to
deal with.
Figure 17 now carries the synthesis to a conclusion which - how-
ever tentative it must be because of the short span of the data in
hand - is still wonh the making. In this figure, rhythm A is t11e
15-rnonth rhythm found in I.he residua] graphccl in Fig. 13. Line B
is a synthesis of A with three other rhythms - the 4 1-month, t11e
24-month, and th(~ 33-month . In C the 4·rhythm sy11thesis is com-
l>inecl \\:itt1 th~ C:<>r1jcttltrcd l()*}:ear '1,\°:t\'e, to Q})tai1\ a S)'tlthcsis Of
five rhythrns as a tentative projection for t11e corporation through
1 955·
In the pattern for this corporation an interesting fac.t is to be
noted. The corporation's United Stat.cs sales in 1930 reached a
peak higher than in 1929; they reached another - Jessel' - pe;Jk
in 1932. Thus the corporation's pattern ckarly varies marked ly
from that which prevails generally in the American econo1ny. The
tentative proJection carries this ohservation forward.
Figul'e 18 illustrates the timir:g of the four major rhythms which
we have observed at work in the national economy. For those who
refer to this di<1!,'Tam with an eye to t he management o[ their own
business affairs, Jet it be reiterated chat che rhythms man ifest them-
sel\•es progr<':ssi''<:l}' i11 ,,.arious ~orts of btrsiness org-aniz.ations and
pric.c series. The approximate timing in the cliagram will not neces-
sarily agree preds(•ly with the timing fol' any particL1lar series
hitherto discussed. Further, chi~ diagram, as in the instance of the
1St-yci1r 1·1'1ytl1i11, doe.$ 11ot t.ake juto accounL partial ye.ars.
11\!ith all such qualifications, it still seems important t.o note that
from 1947 00\\.•arc:l Lllese fotJl' rnaJc>r rl1ytf1rr1s are cl(:c.:lining, ancl
that all four reach a bottom circa 195~. with the subsequent pat-
tern calling for a concerted rise.
XI I I
Avoiding So1ne Eco1zornic Illusi()rlS
(and some economists) find it hard to

M Mv uusrKESSMl'.' l
use the trend and the rhyLluns in L11eir economic calcula-
tions, because o[ an a priori !)l'ejndicc. That is, Lhey find it hard lO
believe any kind of human experience can be " detern1ined " by
factors outside the control of man's own conscious life. And so
they resist recognition of the evidence. 'fhey accept the need of
•1<ljust1r1<~r1t to tfle '\'eath<:1~. kn<.l,\ i11g it is 01Jt of rr1;111's C:<lr1trol. llut
1
they resisL passionaLely any suggesLion that changes in tl1e social and
economic climate may be beyond the total rul e of man's con-
scious will.
A man accustomed to think that certain " c1uses '' must always
have given" resulLs "and that we control results by adv;1nce manip-
ulation of causes niay feel cspet.ially confused. If, for instance, he
believes that raising 1!1<: price of gold will raise the price of com-
modities, then he thinks he knows the "cause '' 0£ the Alnerican
price rise in 1933· How then, li e may ask, could rhythms have had
anything to do with it? Isn't it common sense to think that this
rhythm business -whatever the evidence - is esoteric nonsense?
The fact is that most of the "causes" we commonly assign to
events are the real nonsense. Science knows this. It was one of
Freud's great contrihutions, for instance, t.hat he showed how we
do what we feel subconsciously impelled to do, and then satisfi
[191]
CYGI.£$
our conscious minil hy de,·isin~ adequalc "rea,oru." ~lost of us

talk, think, and plan i n r.hc same kinds o( patccrns o ur a11cr.swrs
used in <lays before science was born. T he o urnnuc is illusion.
'Ve must fruc ourse lves of n1any of the old patterns if we are t.o use
LlH~ knowledge chat is 11ow ;ivailablc.
T o tl1at end leL us consider in t.enns of motlern psydiology and

physics a few facL< important for our approach to economic S<:icnce.
l'. U. OtL~pensky • once asked his readers to make an experiment.
In1agine. l1e saicl, rhar }'<JU l.i\.·e in t,\.O cli111c11si«)nS, inslead of tl1rc:c.
An easy way to do this io r.o imagine you arc a hcing lik.e a piece of
paper, infinitely thin, living upon a table. You can look neir.hcr
up nor down, (or u p ;11111 dow n are in a thi rd dimension. You can-
not even tltinh up and down, or conceive it. For yo11 h<ivc no thick·
11css, and hence ca.nnot even imaf;ine t11ickness.
l\ow in the ccntl·r of d1h tabletop where you live, there is cut
a slot. Ju rJ1is slot Lhere rcvolvt$ a wheel, so hung that half the wheel
is always below die table, and half of it above. 1.'his wheel is solid
and you <:an,.,.., only rJ1e edge of it. I .ct us imagine its edge is painted
in four colorccl segments - black, while, blue, and red. As the
wheel revolves, and you observe it en d-on, you o f course do not
k11ow that it is a wh ee l you sec. Fo r you are a two-tlimensional being,
and therefore see only a sinµ;le li·n.e of color il lou g the tabletop.
O•x:L~ionally, as the wh<·cl slowly re,•olv<'S, you do •ee tl1at Iine
change suddenly in C'tilor. Red will snd<lcnly change to black, and
.,Jack to white, white to hlue, and blue to red ag-•in.
l'\O\\', jf }'OU t•IA)Cf"\'C this phc:TIOlllCflOJl long (:Dough, you ,,jl]
linally decide lhal when the red com~'S np, it will e,·entually canse
black: and when Lhe black appears, it will cvrutually cause blue.
You w ill think you kwJw the causes of the ph enomena you obsene.
1f a two-dimensiona l scientist is observing the phenomena, he
will eventuall y discovt.:r a" law" in Ll1i.~ contin1iit)' of event. Csing
tliis law, he will he ahlc t.o predict changes of rolor accurately. The
scientist, by tli c use of mathematics, might also discover that a lhird
• Sec Tertitttn O rs:a11u11a. (AJ£rcc) A. Knopf. 19!.t) . J>· G5.
[ 19~ ]
AVOIDJN(; SOMF. ECONOMIC Jf.Lt!Sl01'($
dimension was necessary to account for t11e real pheno1nenon he saw

in two dimensions only. nut. neiLher of you <X>uld imagine tit is tltird
dimension as a sensorv realitv. .'.'Jor could vou know the real nature
J ' '
of the causes operating there. 'fhc scientist would admit this
frankly, saying !tis law merdy described what happened, wichou1
l:xplaining it. Bul you, untrained in such fine distinctions, would
speak boldly of a "cause" being followed by an "(:ffcct." And
each "effect " would in turn become a new cause (in your way of
t11inking) resulting in a further effect. w!tid1 followed. 1f )'OU per-
sisted in this belief, you might eventually resent being told diat you
knew nothing about. the real causality.
This lit.tie exercise in make-believe illustrates how our senses,
being limited to three-dimensional experience, limit tltc ki11d of
knowleclg·e \11:t1i<~l1 \o\ c call " '~(>tn1n<>tl sense.'' 1\s 11t1111an anim.als, \1,•e
1
are tltree-dimensional beings. But the universe li;1s more t.han three
dimensions; for convcnknc;c, and to avoid mathematical talk, let
us cal I it a tlu·ee-plus universe. In such a universe, we shall err seri·
ously if we think perception alone can ever tell us true causes.
\.Vhen a historical event does occur, we shall prohahly i:cco;,rr1i;e
a "cause." Or rather, we think we can. But before we get co lhat
event, we are never able to recognize in society a present cause A
that will inevitably produce a future soda! effect n. After the stock
market has gone up. for inst•n1ce, almost any commentator can fine!
;1 so1.1t1<l •• t'f:ason ., for its action. B11t l)Cf<lre it goes ttp. no one can
predict for sure what it will do on any particuh•r cby. Jn short, and
tO rc:peat: Our recognition of cause is always associated with past
e\:ents that '"'e vi<:~\\' itl retrospect.
This hring.1 us to a vital fact: '\Nhat we call our recognition of
'' cartse '' and ''effect ' is so11Je!io1u ass<>c:iatP-d u•ith. ti11ie, ancl 'vitl1
1
our perception in time. This is important r.o understand. for what

we call time is apparently only a 1:1ode of perception. Psychology
ha.1 long suspected as much ; physics has laid the essential ground·
work of evidence. Following Einstein's original pub.lic<1tion of his
findings, in 1905, iV!inkowsky sliowed that elcctrk.al phenomena
[193]
C YCLES
m ust be r egarded as occul'Ting not in space and time separately. bu t

in space a nd time welded together so closely that a join t could no t
he detected. 'Thus th e phenomena of elcc:t.romagnctism may he
thought of as occurring in a continuum of fou r di111ensio ns- r.hree
dimc:nsio ns o f space ancl o n<: of time. And it is impossible to separ(l,/e
the space from rhe time in any ~/>solute maimer. l\.fathema tics w-
day uses a factor representing time inccn:hangeably with factors
representing space.
Ouspensky, approac:lnng Lhe problem from a psychological back-
f,>TOWld, goes .so far as ro suggest that time is the way we experience
space in its higher dimensions. That is, the unknown dimensions
of space are revea led to us in time. A ro ug h comparison, to i.l lus-
trate this idea, can be fo und in the percepcio n o f a soldier who
si-1uds at c.hc foo t of a hi ll . He will nor know what lies on the other
side of the hil l unti l he spends the time ncce,sary to walk to t11e
crest. But an aviacor, in the sky directly above him, can already pcr-
cci,•e what the solrlicr wil l not know for another half hour. The
aviator sees the soldier's future experience as present, bccau'e he
occupies a different .t(latial relacion to the event awaiting the sol-
dier's perception.
1\-lodcrn physics tells us that two individ uals sep;i ratecl b y enough
space will not C\'Cll be able to know what simultaneity mea ns. The
even t chat is .11.ill fum re for o ne individual may he past fo r another
located in a d ifferent frame of reference. F.ddington has dn1ma·
tized this relativity of time with a happy ilhJ>tration:
Suppme tl1at you are in love with a lady on Neptune. and that she
retur1lS the sentiruerlt. It "•jJ1 be some consolatior1 for the melanf'holy
separation if you can say lo your5"1f at some - jJOSlibly prea rran~cd -
moment, "She is thinking of me now." l lnfonunately a difficulty has
ai·isen bc:c.1usc v.·c h:tl'C l11,1 d to abolisll ]\To'"" Tlicr<~ i:, 11<> <.tUsolutc Nol.,.,
b ut only che various rela tive Now>differing according to the reckoning
of different observers.•
• The Nacure of the Ph11i<ol IVorld (The ~faonill•n Company and Cam·
briclge Univc.-.i1y Press), p. 49-
AVOIDING SOME ECONOMIC !Ll.USIO:-JS
Obviously, if time is rdativ(:, then our perct:ption of cause and

effect is equally rdative. For cause and effect are ordered in
time - cause must precede its dicct, by the very definition of the
words. If Now has no absolute meaning, neither can Cttuse. If time
is only a mode of p<:T(:civing a tr,ree-plus universe, so is our idea
o f 0111se and effect.
\\:lien we put a pot of water on the hot stove, we tend to think
t.hat is what causes it Lo boil. But we shall do well t.o recognize the
fact tl1at the water hoi l< uncler wh"t moy he r.ollc<I a "law nf
averages." \Ve may assume it always will. But we know of no abso-
lute Jaw that it must, and the formulation of the <:iuantutu law in
physics has shown us how we arc limited in our knowledg·e that it
always docs. \'Ve have no grounds for predicting that any selected,
single molecule of the water will boil. \'l'e predict on the ba.~is
of the g1·tu1p. As E<ldini,>ton phrases it:
vVhen we ask what is che characteristic of the phenomena that have
been succcssfu1 ly pr<.'<iictecl, the a11s"¥-·er is that cl1e}' a.re e([er.t.s de[>eJl<I·
i11g 011 tl1c average <.:011figltratio11s <>£vast 11umbcrs of inclividual er1titi<~s.
. . . The Ia \"'s govt-r11i11g tl1e 111ic:roscopi<.: clc111c11ts of tt1c: plt}'Sir:a I
\\'orJtt - iric.ti,:i<ftta.1 aton,s. electrons, <tt1ant~1 - <.lo 11ot rnake <.lcfinite pre-
dictions as co what the individual will clo next. . . . Rue short odds on
the behavior 0£ ir1dividuals combine iuto very long odds on suitably
sdcccc<l >Catiscic' of a numher of in<lividu;ds. . . . All the successful
precLiction.r hitll erto atlrib1tted to c~1isality are traceable to tflis. It is
true tl1at the <jUar1u11n la\\'S for indi\.·i<luals arc 11ot i11c:on1patible v.•ich
c~-tusalit\';
' ' .
the\· mer<:lv ignore it.
HLttt1<J.11 life i:j fJf(Jvcrlriall}' 11rJu:.rt.ai11; ft,<:v.· Ll1i11g:s 4tt C J11u1c 1..1.:1L.ti1l.
than the solvency of a li[e insurance company.. . . Th" eclipse in 1999

is a5 safe as lhe l)ala1l.ce of a life i11Sl•l'at1cc coo1pan)'': the next quantt1m
j um1; 0£ an ato111 is as lt11c:crtair1 as )'Oltr life a11cl u1i1l.e. •
Having come thus far, it is hor>ed that the reader now uuder-
stands ,,:e are sailing no esoteric s(:as. l>ut a1·c cutti11g' a\\•a~· so1ne
underbru'h to reach firm, sound ground . .Herc is Lhat ground:
• l l1id., l'I'· 300-30::.
[ 195 j
CYC f.F.S
In C!<'Onomic.s, as io ph~sics. we ca nnot predict "hat any given inr!i-

viclual will do. llu t we do have useful methods of predicting what
groups will do. And we can succeed wi1.hi11 tolerable margins of
error. In using v11r kno1"1edgc of rhyLl1Jlls tv clv 1·xactly this, we
rely, just a$ in physics, on statistics for largl' numbers of individuals.
In using them, we ignore the U$ual concepts of causal ity. Arni we
ignore Lhe ran 1hat the future where the pn•clined action will occur
is not yet present to the senses. ;\11 equation th3t will be balanced
in the future is jus1 :i.s va lid a.\"• +
2 - 4 .. 1s now.
A ftt:r an event has happened, the h uman mind will always seek
to associate it \ Vitft arlot.ll er e\'ent '\•hich c,;ar• l>t: tlt<,ugltt <>f as a
ca1Jse. Tl1is is a cha.racterlstit: rn(:r1tal 11ecd of m a11. '"·110 (Ti<·s ro
impose his own sped al kind o[ order on rh c whole universe, in the
process of trying 1.0 understand that nniver.~e. But it is no t a 11 cecl
of science. Indeed, it merely handicaps scien1ific inq uiry in to the
real na u1r<· of the universe, and of the phcnosnen;i of which we are
a parL T11e new economics, h;ivingadopted this appro;ich, has made
1f>Te pro~ress in a fc,v dl't..ade.s tl1an \\'"as accomplisl1etl tllaring \\'hole
11_
gen eratio 11s o f econon1ic a rgLL111ent O\'e r '"'hich ta 11Lc.: first~ the 11en
or the egg.
Economists \\'Ile> lll\'C ,,.ie\\'ecl :s<><:iety as :t rYle re 111ultiplit~atic)n
,.>f i11cli\•idt1als, <irlcl S'> t.ricd to dc.:s(:ril>c: it irl 1.cr111s of ea(:h lr1cli-
vidual's reason-why, were bouncl 1.0 fail. P eople handed togetlwr
i n a n c~:c>11c,111ic com11 11111i t }' arc 111ore l ha1 1 :t ~i 111ple ~t1n1 c.l f j 11di-
"idual beings. A< " mcmbc1s one of another" they for111 a hiologir.al
organi,111 ,.,}\OSC lvholc· i:> ul..-.ays gr~tc·r rJ1a11 the $t1n1 oC it.s parts.
They are cells snhjcct to 1.he mathematical laws which control ag-
gregate>. The sci<-nrist who studies them with lhe lecbniq ues of
trlJC science is r1<.• tl)()fe co11cer11cd 'l\:itl1 cl1clr itl< li vi tl11(J J <.:ont«!pts
o[ reason·why t.11 an he is intcrcsLed in th<: possililt: thong hr. of the
y•·ast <:ells iv hose I ifc iii a Lest tube he <'hares o n a grnplc . His only
J>l'opcr question i> not why they ac.t as they do, bu t how and when.
' Vi th people en ma<.;e, as with yc;tst cells. hi~ graphs will tell.
The indi,·idual nct.-d not feel bis intellectual dignity bun by this
[196]
AVOIDINC. SOMF.. ECONOMIC ILLUSIONS
resen'ed approach to the affairs in which he plays a part. To the

man who quails at. t.11is approach we may extend the promise that
when prices go down he can always discover what looks like a cause;
when they go up. he can also discover a plausible cause; and when
.
thev remain stable he will always. be able to cite a reasonable ,.Ca·
son. '\Ve need not argue with him about the reality of the causes he
thinks he M:cs in operation. \\ie have a n1uc.h briefer way out. \Ve
merely show him Fig. 1 and ask him what he sees. Depending on
FIG. 1. llt.t:SJON
\\rJ1at t·c>u sec here depends on Lile focus of yOtJT atterition . A c:avc c.~ntrancc:;?
Glo~·eri11g J:ip ~·restlcrs? (After Rul,er~ aritl Hartm.111.)
1J1e focus of his eve, he may sec an uni, or two foces, or the entrance
• •
to a cave- or what he will. Jf his fc1<:us on a phenomenon in two
dimensions will pby hirn such tricks, what 1nay not happen when he
views events in thrcc dimensions? ArHl how can he e\•er hope to
know the true nnture of an event in a tl1ree·plus universe?
This is not nihilism - it is the essence of sound approach to
knowledge. For only by knowing thc limitations of our working
tools c•n we properly value the results. And only by recognizing
the limitation of mind can we think trulv• .
The outcome is ii.self sufficient justification of the method-an
[ 1 97)
CYCLES
outcome that permils prediction, which is the goal of every true

science. The meLhod has an0Lhe1· advantage - it draws economics
close to all the other sciences, just as Lhcy, too, converge. " T he
separate sciences - e pistemology, physic.s, c hemisny, 1na Lhematics,
astronomy - arc approaching- one anocher with acceleration, con·
verging toward a complete identity of r(:sulls." Those were the
words of Spengler; he could well ha,,e added hiology, psycholoi,ry,
economiQ, and M>Cioloi.,•y to his list.
It is not the h1J>iness of economics to explain how the human
mind (li\>ides a series of events up into Past, !'resent, and Future;
this is a prohkm for ocher scicnc.cs. Bue for progress in Lhe ability
to make useful prcclictions in Lhe economic field we do need all Lhe
knowledg-c 1.hat th e other sciences have to offer on chis suhjct:t.
And we need a workiui; l1 yp0Lhesis (we ma y accept it wholly ten-
tatively) whid1 permits us to understand how the future alr1'ady
exists in the present - not as the result of a cau.se now in operation,
but as the continuance of a pattern. Sir Jam~ Jeans offers us a
physicist·s hyp<>Lhesis:
" Ir may he· that lime, from ics beginning co tht· end o{ eterniiy,
is spread hefore us in 1.l1c picture, hut we are in contat:t with only
one instant, j11.<t as Lhe bicyc:le-wheel is in contact wiLh only one
point of 1.he road. 'fh<·n, as \\fey! pulS it, events do not happen; we
merely come al'ross Lhem." •
\\.'hatever h ypothesis we prefer. we must avoid one temptation
as we look at our char ts tbat project patterns into the future. \Ve
mttst al\Va}'S 1t:"ga1J \.dL&:\l:•and-<-ITcc:t tl1inling ns only /HWtial. ''\'e
can even call it rlirce-dimensional thinking applied to a three·plns
en\·ir<>t1111ent. \·\111c11 '"'e arc temptc<l LO use il, ir1 csri111atirtg Lhe
futnre, let us reca ll the two-dimensional being on 1.he table tOf).
Let us imagine tha t he liked I.he color red very much, a11cl that he
was one of a whole colony of si:d1 beings Jivi11g a1·t)t111d the wheel.
\ ·Vhen red came into view on the tahletop ho1 izon, he and l1 is [el-
• The &f1surious Univcne (The ~t acntlllan Cc:.n111)a1ly and Cambr.i<tge Uni·
vcrsity Press), p. "7·
AVOIDING SOMF. E CONOMIC ILLUSIONS
lows felt gay ;in<I happy all over. They joined hands- touching
two-dimensional finger tips. They spun round and round in danc-
ing-. So plc<i.l't11t w;is their sensation that they started an inquiry
w ascenain what tnea.~ures could be 1.aken to keep the red always
111 \'le\\r.
A h"·o11p of invest.igat.ot·s was ;1ppointcd, with the ohje(:t of tnak-
ing an orderly inquiry. Tt was found that the red always appeared
after a period of community inalaisc: during which the color had
been blt'te, fo r ,.,.llicl1 no one cared. Follo~...·ing a long s1>ell dt.1ring
which v<irious community readjus:.mc1Hs to life with 1.l1 c hlnc had
been made (and these were catalogued at length), the community
suddenly felt a new ac<;ess of well-hdng. Simultaneou.~ly the red
appearc<L Th<~ investigators could report that t11e red was there-
fore undonbtedly the result of a balanced state in the cc011omy,
r eached as a result of t.hc readjustments in the blue period. Un-
ff>rtur1aLely. as che <lancing acti\:it)· UlOlJratt~<l cluri11g· tl1e l1a r>J.>}' reel
period, excesses apparently developed, which the invcstigato.-s saw
evidenced in the fact that the tempo of the spinnini; gi·cw exceed-
ingly rapid. T he investigatOt'S measured these mounting excesses,
ancl compiled them into voluminous statistic;1l reports. 1t was theiT
unanimous conclilsion that the sudden passing o( 1.he red-which
had happenccl so many tin1es in their two-dimensior1al history-
liad always been the <lirc:ct result of such excesses. The invesdgators
ended t heir report with recomrnendations for<• further st11dy as to
kgislalion that would kt:ep the excesses under control.
F"ntasy though that may seem, it has as much sense in it. as 1nany
of 1.he prevailing economic doctrines which dispute learnedly over
the " ca nses " in the world we 1ive in.
It is not within the province. of our charts to tell us about the
nature of that world. But it is 1J1eir virtue that they enable us to
avoid, in tliinking abouc it, some of the illusions inherent in cause.
a11d-e£Ie<.:t ;1rgu1r1e11t.
XIV
War and I ts Dislocations
is some limitecl evidence that wars and periocl< <>i war

T trPKE
acliviry may have a rhythmic periodicity in !he affairs of ll)CJl.
Dr. T.in Yut.ang, in 1\1)' C<>untry Gnd /11)' People, • reproduces three
chans prepared by l)r. J. S. Lee for a study, "The Periodic Rc-
currc11c:c of Interneci ne \Vars in C..:hina,'' published in the China
]011rnal of Sciences and Arts in 193 t. The Chinese people have a
continuum of hu1ory far exceeding that of any 01ht:r people with
cx1an1 records for study. Dr. T.ec found two 8o<ryear periods in
Chinese history, and the beginning of a third, whid1 when com-
pared show striking parallelis111 in their p<'riods of p•·ace ancl cli>-
order. He concluded that the plrallclism "far exc·ccds the limits
of probability" and is" perhaps too exact to he expected from the
proceedings nf human affairs."
Each period began with a short-lived but ll)i!itarily strong c!y·
nasty, which unified rhc country after cenLuries of c11~organ1zauon
ancl civil strife. ·rhen would follow four or five hundred years of
peace. Then a change of dynasty would be follc)\"Cd by successive
wars. The capital would be moved from the :'>lonh to the South,
sc1·c.uion and rivalry between Nonh and South woulcl ~Tcatc in-
tensified conllicts. subjugation of the dissidents under foreign rule
would result, ancl the cycle woulrl end.
Spengler, who believed that all cultures have a life course in the
• Jolin Day Co. and Reyna!&: Ilitch<od ('935), pp. 3<>-31.
[~oo]
WAR A N D ITS DISLOCA TIONS
general order of 1,000 years. and saw comparable phenomena

emerging at parallel period$ in the life course of each culture,
would have found Dr. Lee's data of irresistibl e interest. Spengler,
for instance, saw evidence indicating that in the " springtime" of
a culture its life spirit alwa)'s blooms in great architecture, such as
the cathedrals of the Renaissance. Dr. Lee mentions the undertak-
ing of vast architectural projects in the early pan of each of the
800-year period$ he cites. In the first Boo-year period (221 5.C. to
A.D. 088) t11ere was the building of the Great \Vall under the Ch 'in
dynasty, and t11c colmsal palaces. In the second period (58g-
1367 A.D.) the Grand Canal was built under the Sui emperor, and
also palaces of great magnificence. In the beginning of the third
cycle (tbe ~(ing dynasty [ 1368 A.n.J to the present time, and hence
uncompleted) the Great \Vall was reconstructed, new canals and
dams were la id out, and Peking was built..
The charts which Dr. Lee created, on the basis of the records at
his disposal, to picture visually the parallelism of wars in these three
Chinese periods, are reproduced in Fig. 1. \ Vhilc we may regard
with due reservations the wave lines intended to show the fre-
quer1'y of internecine wars per quinquennium, in view- among
other reasons- of the great antiquity of the history involved, no
student of the suhjcct can fail to be impressed with the regularity
in the parallels, and wit11 the thought that the subject deserves far
greater study than it Joas )'Ct been accorded by scientific research.
Spengler, convinced of the ' 'alidity of such parallelisms, declared
ti oat periods of war and social upheavals show up at corresponding
points in tl1e life or every culture. He believed, for example, that
the Napoleonic period in our " \Vcsten1 " culture corresponded
closely to toe Alexandrian period in the clas.sical, and predicted that
our present "C'..aes.1rian" age would duplicate, with wars of ex-
tenninat.ion wiping out whole nations, a similar period in the his·
tory of classical Rome.
But scientists muse still regard these historical parallels as more
interesting than scientifically fruitful, in view of our failure to
[ '101 1
CYCLES
fiation11I expansion St«;slli<lf• ~nd 1ivalry

unlflcaUon, peace, prosperity tJctwcen north: and south
culturel development northern ln\·as.lon
8.C. 100 8.C. 100
I 1 •5 I I I
.SI St'ln 161 . I . I
: . 0 1 < - - - - - - - - Hon - - - - - - * " K ,...~sst Ch1n~SiJ1. Dyo.asties~
o1 uo 1 n1 1 1
"'
Fig. 1-/\. Fii:st Chinese T::poch (Soo years: 221 u.c.- A.o. 588.
iitoi
• • •• -··
<J!
N;,ti(ln:il c:•Jl:tnsion •
0'>1.£~
-; .:: s s Secession end rtvall)
- ·.8 :!.S- unific;ition, pe.ece, prosperitJ i ..- e.. i between north end south
~
u--
... .c:
E ~
•
II)
·-
u~~ cultural de\•tlopnle.nt
, . . , , ....(111.111
1!1k1 ~-tis northern invasion
..
.il u
~
~ ~
~
iE -
~·
E5
We;
700 800
.
900 1000 1100
.!\ '
1200 1300
•••c c~
!! ,•
~
.. 30
20 -
-
-
~
:; a
-c
'5., -
~"'
U •
c•
-~
0.
10
;
1l . I /\ fl/I i
.• JI
lf ;;
.:: 3' "'.,,..
"'I
T'ani:
I ·; I
I ~ 1 I
I
' !:i.J-- Horth Sung TSouth Sunz -k.-¥i.ian-:
I
;
Fig. i·B. Second Chinese Epoch (780 years: 589-A.D. 1367.

WAR AND I TS DISLOCATIONS
xo.,.
,_ •
. . g ..
.:.$:
tfetion•I upam.lon .. ~.q
- ti< =.==£ t,tnificatlon, p.eae1, Ot°""'rft)' l •I,
ll~ ~ t;~~
<.>.t! - ··-
cultur111 dfr\ltloomtnt .ii1•
•
-·
ce
"u
1l~
~ ~
a;
§<;
~- \ .
'
0
1400 00> 1&00 1700 IBOO 1900 2000 ZIOO 2l00

•l!i ~ 30... .
c~-
:;
! ~
.~
to ... .
=·e .
t ~ 101- .
c
•
~~
•
Q
a·~
- I
i<
I
·- .
••·- ·
f\UllC
"' • • ' I
::
~I - Chin«
\
·~
,,
'n
Q
I ! •.!!
Fig. ) c. Third ci.;,,,..., J::poch (A.O. 131\R LO present) .

.-\JI tJ1rtc ch;,r1-, aftrr J. ~. Lee and 1.i,, Vui.a_ng.
devise ways of measuri11g the phenomena. In the present state of

our knowledge, we as ) el have no satisfactory statistical approach
to wars, for instance, that wou ld guide the unrlenaking. Hotell ing's
elfon to w<:igh wars stacist il'ally by measuring their i111pacl on prices
sc:cws hardly adequate. Our Spanish-American \•Var, for instance,
was only a fraca.~, compared to some of our other conRicts. It had
littk immediate effect on tlte econom)" And yet, by consensus of
the histories, it rnatked Ameri,,i's emergence as a lirst-<:la.<s power.
Conversely. a hiscorian five hundred years hence might possibly
find our panidpation in \\'orld \Var I little more than another
border ;u\venture. The sol<licr deaths that occurred were fewer than
our annual civilian toll frorn aucomobile traflic a kw y<'ars lacer.
T he pri vations that w<:rc su[ ered by the American popu lation
were c.onsiderably kss than in the greaLpeace ckpression of t930
1932. StaListicians similarly could hardly feel justified in taking
price inRation as a guide to measuring the importance of \ \!or iel
\Var I LO American lire. For \<\'orld \Var JI, which demonstrably
[ to3]
CYCLES
has created an unprecedented disturbance and distortion in our

econorny, had certainly by 1946 created no more price inflation
(and probably less) tl1an its predecessor a quarter century previous.
although ta.xing our resources far more basically. It is worth noting,
in passing, that \Vorld "Var II heg<Ul for America - economically
- in 1939, and not 1941. Pearl Harbor did little inore than crystal-
lize a de facto situation.
Until our social sciences develop far greater resources than are
now at our disposal for defining and measuring wars, we shall
doubtless continue to lack data that will permit us to make sound
generalizations concerning the cycles in their incidence, or that
will serve any other useful purpose beyond that. of extremely ten-
tative hypotheses.
In such a problem we see, exceptionally magnified, the difficul-
ties always inherent in studying an organism of which the student
is himself a member. As has already been noted, it. is extremely
important, for scientific progress, to try to observe as if one were
outside the fratne of tl1e phenomena being studied. \Vhat (:ould
be learned about. man, for instance, by one inquiring cell in a man's
body- even granting to this hypothetic.al cell the highest. gift of
observation that you please - and giving it, through the blood
stream, 1nobility and power to visit every nook and cranny of the
man's physique?
Every person is to be comparably regarded, in one aspect, as a cell
in die social entity. In thinking about certain surrounding social
phenomena, he must try to detach himself imaginatively from the
organism. His measure of success in this is one mc;u;ure of his scien-
tific probrress. It happens that our society has seen little such scien-
tific thinking about war. "Ve cannot. yet-by means of a knowledge
of cycles- predict the timing of war's arri1•al. But we have data
to make estimates concerning certain associated reactions, in the
event that war has come.
First, by knowing the cycles in,•olvcd we can - ideally at least -
determine what would have happened had there been no war. Aft.er
[ 204 J
WAR. AND JTS DISJ.OCATIO:-IS
Index
ISOr-~.--~,--~,--~.--~,--~,--~,--~,--~,--~-,-~-,-~-,-~~
140
130
120
110
100
90
,,
60 \\
\ I
,,
50 ~'
40
30 MAJOR WAR DISTORTION
WHOLESALE PRICES
20
10
0
1830 1850 1870 1890 1910 1930 19!>0
Fit:. t. \>\'A •t -T1~1'£ DtST<YR.'110N

<'..hart s.l1owing distortio11 in the \'\il1olesale Pri.ce Iudex in time of V\iar. Tl1e
solid lir•c ~lt<>ws Lile ll)rte -year iooving a\'erage of tlle lndex of \\lholesale i•rices
in the U. S., 1fl31·~1~J·15· The hr<)ker1 line sl10\\·s tile S)' lltl1esis of t.11e regular
9-year and ~;.1-ycar cycles. The sl1a(lccl area::; $lt<'>lv tlic diffc>:renc.:c:~ l>etwee11 tJ1e
Ioclex and the reg'lJlar pattern for the periods of the Civil ' Var. \l'orld \Var I,
aud World \Var II.
The 3-year moving a\'erage has bee1l. extrapolated to l!)-15· TI1c shaded areas
begin o ne yc:-ir pri(>r t() tl1e ()\ttf)reak of ''1ars. since in a 3-yc~•r movi1Jg avcr.tgc
tJ1e eff(~CL of t11e first yc.a r of war is ex.tcock;d one year back,,oard.
It is jnter(':5ting t<> n<>lt.~ tlitJt in spice of the m<1gnltuc.le of tl1e tlist<)rtioo, the
titni11g of lhc p eaks h~lppcns '> coincide with the n<)mt<JI ti1ni1\g of Lhe 9-year
cydc. lt i~ t"l.$0 ir1tcr<:.)tiog to note }u)w nearly cqtHtl ure tlu~ <lii;t o r·til)r'1&.
any given war we at least know what did happen. The difference
can be viewed as tlic net distortion due to war. 'fhus, in l!ig. 2 the
dotted line indicates what was due to happen, the solid line what
did happen, and the shaded area tlie action and reaction of the war,
in regaTd to average wholesale prices in toe United States.
Second, by knowingouT cycles we can have useful advance knowl-
edge about the timing of peaks and valleys, even in wanime. Short
[ 2051
CYCLES
Index
132~------------------~
128
l.24
I
'
I
)20
I
I
f16
'
'
I
))2 I
'
I
I
103
,_ I'
II "
I
104 I
,, I'
II '
\ 1\ ,'\
1ooi..;,-+':....:'~-J':...:'.,-7 "<
\
\/I \
,,
I ',/
96'---------------- - ---'
Fie . .3· Pi:w.s1s'l·J::..Nc~ o•· \\'Av"~ lKFcu1:NcK
Diagra1n to sho'" how shore rhythms often 001:aci11t1c to jnffue11ce a series,
despite the in1pact c.1f a l<111ger \\lave or S'.'>1t\e accidenlal discortion such ;ts ~J \\'ar.
The solid Jit1e shov;s the effect of the lorJg(..T thy1l11n or the l\'ar. ~J'J1e dotted
line sho\'~ )1(1\v a r~gtLlar G-1l1011tl\ r11ythtu contint1cs to jnflucncc bel1avior..1\.11
,,.a,•es arc c:xatll )· 2 per cent i11 al'llt)litude, bt1t the distort.ion c:-;11,1:),(~< l l)y tJ1e
r.t11itl rise: of Lilt: "U'-'.111.l" Ji1,i; ..,tn,.t,u.1(~ Ll1is fact. The vertical sc.' llc ha,, tx:._:,·1
grea1Jy exaggerated.
cycles ride on the hac.k of the war disturbance, as well as on the

back of longer waves. The war seems to make little or no difference.
In Fig. 3 the black line shows long waves and/or war; the dotted
line shows, for simplicity, one short wave riding the long one.
Third, only by knowing the cycles involved, and by adjusting for
them, can one know correctly the underlring trend in effect at the
outbreak of a war. It is necessary to know this, if one is t.o project -
[206]
\\'AR AND ITS DISLOCATIO~S
through the war distortion - the kvcls to which affairs will nor-
111;,lly return after the war is over - unless t11e long prevailing trend
is LO be completely overturned.
These charts sen•e parenthetically lO illustrate a fact LhaL has
been stated before: The forces represented by our rhythms are
quite evident!)• not the only forces at wo1·k in our society. There
are doubtless many sporadic; ones in addition. Bul the rhythms we
have do show a regularity which persists, ancl usually dominates.
rPp,::lrrllPss of ,,.h~r. <)t.hcr fc>rc:cs are prese11t <)fl t11e scene.
Otl 111atl}' of tlte ect)flC>tnic cl1arts and '.\r(t\:e p•1tt.err1s , ... c k11f>\\'
ahoul, a period of war registers iL~elf as a sort of volcanic explosion
above the level of the pn:vailing trends. All the waves revolving
arottn<l the trer1cls ;ire t<)ssecl l1igl1 in tl1e air, as jt , ...<.~rt~. Y<~t tile
actual wave patterns - as patterns - an: often Jess disturb<:d by
war than one would expect.
\Ve have already seen, for instance, that it would be logical to
suppose the st.eel industry, required during moden1 war to work
at top capacity, would at least have a steady rat.e of production in
such a period. :-lonetheless, in \'l'orld vVar I at least, we fine! (:vi·
dence that the wave pauern persists; it m(;rely goes on al higlKT
levels. StTikingly eno ugh, the pattern of waves in the stock mark.et
l.n1dt~rgoes relatively small disturbance, and has alWO)'S re1;overed
its periodicity if that has been disturbed.
Still more: Regardless of the dc;,•TCC of disturbance, once the war
"explosion" has suhsidcd and the disturbances recede, the pre-
viously prevailing trends an<l lcveL\ have usually been re·cst.abli.;hed
in our economy. Previous wars did not, in themselves, alter basic
underlying trends, insofar as our statist.ical data show them.
But wars often do alter certain seemingly established relation·
ships within an economi'· It. is clear, for instance, that no such fan.
tastic, unprecedented spurt in the volume of United States manu-
factures, as traced in the pattern on page ~s. could occur without
vast straining of the social gears.
One of these strains, as we should know, is clearly revealed in
[ ~07]
CYCLES
what happened to wage levels in thr United Slales, as these manu-

faclures were pro<lur.ed. An unprecedented shift in the pattern
of lJ. S. income levels took place. for ins11u1cc, it has been esli-
mated that in l !J3!l there were~ million famili es with an income
of no more than .$ •00 a )'ear, 5 n1illion with an income o[ not more
than $500 a year, 1 million more with an income o[ not more than
$700 - making 8 million fon1ilies of \'ery low income, rclati\'ely
speaking. ,\nother 11 lllillion families lived in a slightly higher
carcgory. still another !\ million existed in good m1ddle-class com-
fort. and 2t mi ll iora t1tJ1e1·s ,.,·ere recei\~ing t{>p i11,~c1mes.
There had hccn much talk of die re<li.uribu1ion o( income started
in 1935. But the rca l rcdi,tribution waitc:d f<>r the war <~m of the
forties. J\·f any of the 19 mi ll ion f:nnilies at tl1 c bottom of the neap
moved swiftly, after 1941, into incomes that had previo11sly been
enjoyed only by lhc 111i<ldlc class. It has hec11 poi1ued out, in this
connection, that the greatest increases in currency circulation in
this counLry, where the increase had been pheno menal even during
the prewar yt>ars, <x:c uned in some of the " poorer" districts of the
nation. The increase in just four years, from 1939 to the end of
1943, was :,20 per c.cnl in the .Federal Rese1·ve Dislrict of .<\tlan1a;
450 per cent in Richn1<,n1<l, Virginia. In .Boston , 0 11 the other hand,
ii. was 192 per cc11t.
Such increases occurred in a period when IJie former micl<lle and
upper inco111c groups were hcing suipped of their previous mar-
~rins of income over expenditures, hy income and oth<.T taxes never
j>tt:vivus1y C'\• en imag,i ncd.
In the same four years, income taxes rose from under 1! billion
1lolla1·s to almost 20 billion. The vast niunhcrs who had previously
been on the i11r.0111e fringes, and \\•ere no"' tl1rt1st t1p irtto in<:ome
levels of new importan ce, cont.ribu1.ed to these tax payments. But
hcing without long-term cc:onomic commitments - life insurance,
long leases, elc. - that are accmnulate1! over 1he years by people
used to enjoying generous incomes, 1.hese newcomers to the ranks
of prospcrit)' had large sums le& them after taXes for use in the
[208]
WAR AND ITS DISLOCATI0:-1$
merchandise man. This fact was reflected in government reports

that consumer expc11tliturcs steadily climbed to all-time highs, with
department store 'ales in early 1946 outranking comparable fig·
ures for all prcviou,, yc:us. Nleanwhile, the Census nurcau could
report. that 1,r,,000.000 American civilians shifted their homes, in
tlu.: \\•ar·years, co ·· i11tprO\'C '' tlletnsel,,es.
American citircns as a whole have realized vaguely that income
rlistrihution is a subject of wide political interest; they have seen
it become one o f the subj«ts fought O\'er in major µolilical can•·
paigns following the depression of the 193o's. Few of them, how·
ever, seem to know that 011c school o[ economi~ts regard income
distribution as one ~cy to the outlook on a society's stabi li ty. The
subject i"u1s no light 10 throw on tl1c analysis o( trends and cycles;
hut it does have possi hie hearing on iilterprcting cenain data t hat
analysis of rhythms and Lrends puts in our hands.
")he Jaw of income distribution has been st.·ucd by Vilfredo
Pareto (1848-19~3) in a formula that has been rephrased by Davis
as follows:
In all places and at all times, the disu·ibution 0£ income in a stable
economy ... will be given approximately hy the empirical formula
y = ax-v
,.,..JlCl'C )' is tl1c 1)11 1nl-,<'T of J)COJJJC llaving the inco11Jc uf x or greater. a
is a co11stant, ar1c.l vis a111,roxin1ate!y t.5.
This equation has implications that are basically simple. It means

that in every slllb/~ community there will always be a gi,•en per·
centage ol people in the top income hrai:ket, In tbe bottom bracket,
and in the middle. The distribution is a fixed constant, whether the
community is rich and prosperous or poor.
Obviously such a law, if 1.rue, itrikcs at the Cundamcntal tenets
of many professional world-improvers. Thus, efforts to discover a
flaw have naturally not hccn wanting.
The "reason " for such a law has baffled many. Pareto himsel
said:
(209]
C\' CLES
These results are very remarkable. Tt is absolutely impossible to ad-

n1ic L)t~tt tl1cy are d\1e 0 111}' to chance. Tht~re is mc>st ('.Crtai11ly a cause,
which produces the tendency of ir.comes 10 arrnnge themselves a<cord-
ing to a certain curve. The form of this curve seems to depend only
1enttously ttpou different economic conditions of the countries consid-
ered, sir1r:e tl1e ctfcct.s are \'Cl]' n e:ul}' the sa1ne for tlle cou11u·ics \•:hose
economic conditions axe as dilfereot as those of England, of Ireland, of
Gcr1ua1ly, of tl1e ltaJiara c.:iti(·s, a.net e\'e11 of Peru.~
Carl !:inydcr, in 1936, advanc.cd an explanation of the law, in the

thesis that the obscn·ccl distribution of incomes is only one ex-
ample of a more general Jaw o( inequality which has been referred
to as the law of distribution of spet:ial abilities. (Schoolteachers are
familiar with one forn1 of this law, in the fact that most gTa<le., arc
·· average," while the nurnber of those in the top fourth of the class
will be about t:qnaled by those in the bottom fourth.)
However that may be, the Pareto formula for soc.ial stability has
possible implications for the fumre of any nation like the United
S1.ates.t And it offers interesting material for those who wish to
speculate about the possibility of new trends in our economy.
Either the lJnited States se~xns due for a great wage deflation, or
new trends are in the niaking.
• c;ot1rs d'~co1101r1ie politiqtte, Vol. 2, p. 3;l2.
t It sJ1<n1lcl be 11t)1.e(J t11at ll1e Pareto ,,·alue 0£ 1.5 is given as a11 approximate.
N. 0. jfJlmso11, making a11 elaborate i11vesti.gati-Or1 of the l:-11\•, on tl1c basis of
in<.'t)tt"Je Lax. data in Ll1e United Stace-.s from 19t4 throt1gh i933, estimated \~lues
rar1gi11g ftl>tn 1.34 i11 1916 to 1.76 i11 1932. It has been pointed ottt that a fairly
,;;ide \•atiation has at times existed '"itlu>ul undue ro<:ia1 unrest, anrl it seems
a.greed I.hat researcl1 has a$ )"et produced no v;a}' of cstimati11g "critical valltes."
H. T. Da\•i:> correctly dc~duced tile coming of serious social distt1rbance in
France, on tl1c l>asis of t.ltc 1935 Pareto jncJex for 1Jlat 1\alion of 1.85. Figi1re.s
£or GerJl\atiy, l\'lli(:ft l.vould offer a good test for tl1e perio<l of 1920 to 1935,
seem not co 11ave been con1pilc<.I. It is perhaf>S worLI\ retoe1r1beriog 1.fl_a t 1\1oeri·
cai1 conditi01\S it\ 1932; reflected in the figure of i.76, \\'ere S\tch that some <>f
tl1c "best•• people spcculiitcd on tJ1e cnrr1i1lg of•• lhe man on horseback." .J\nd,
in fact, rcvOJ,1tio11a.ry political 1 economic and social cll,(tngcs <lid spring from
Lhat year, to become more or less penna11ently incorporated i11 the American
$Cene.
r2101
WAR A1'D ITS DISLOCATIONS
'Vorld "!ar II w:u in one respect like the 'Var o( the Revolution.
lt was fougl1t during a period o( decline of the gical 54-year price
cydc, 01· groundswell. that has underlain prices in this country as
far back as there arc figures.
'fhe three wars t.hat. crealed lhe three !,'TCat peaks fan1iliar to all
observers of wholesale prke charts covering the past 150 years were:
the 'Var of 1812, whid1 occurred when the 54-)'ear rhytlunic pattern
was approaching its peak in 1817; the \Var Between the States,
which occurred when this ... me pattern was approac.hing its peak in
1871; and \\'orld \Var 1, which occurred when the long price wave
was climbing wits peak in 1925.
Undou htcdl y, 1.l 1esc wars added great arnpli tu de to current swings
of the price rhythm. \·Ve have no way - a.~ previously stated - of
forecasting the <·xtCtlt of Such a distortion in MIVOllCC O[ its arri\•al.
llut it is significant that during all three wars tbe dis1.ortion excr-
dsed on wholesale prkcs i.eetns more or Jess comparable, as we note
it in retrospect on the price chans. This is the more interesting be-
cause \Vorld 'V~r T exe1 tc<l in some respects less or a strain on our
national resources than the two war periods preceding it.
l.ly sharp contrast, \Vorl<i \Vat II was fought wholly in a period
when our 54-ycar price rhylhm was downwar<l in its pattern. Thus
any upward distortion t:h is war period e~ercised on pril:es 1nay be
assumed to have been effected against the influence of the 54-year
rhythmic tcJHkncy, and consequently to have liccn less than it
would otl1er\\.ise have been.
Ia ''iew of the wo1 Id •iluation and tl1c enormous volume of money
in circulation, po55ibility of any real price <kdinc in tl1e late forties
•eemed incrc<lihlc 10 mo.<t 1946 observers. Bue skeptics could de--
dare that price levels are related not only co rhc volume of money
in circulation, hut also to rhc volume of lllOncy 1.urnover. Anrl they
could also ask pointedly whether world famine aitd rlccay could
really make for booming markets.
\Ve know, of course, from experience that business can be good
and the nation can enjoy prosperity in a period of declining whole-
[~ 11]
CYCLES
sale prices. Price levels are certainly not the only determinants of
business prosperity. It is nonct!tcless apparent that any extended
period of declining prices following \Vorld \Var H can n~sult in
serious proble1ns for a nation with an unprecedented debt,• and
with a dehtd1arge burckrlsome even at high price level~ - a nation
whose populace has been geared, by admonitions over a long period,
to expect vast inflation as a "result" of tl1is indebtedne5.1. \\'here
the debt is so large (see .Fig. 4), and so many are out on traditional
limb3 expc<:ting inflation to go on gr.owing up to the sky. national
re;1ctic)n to a dc.~Aatio11 1nigl1t for·c.~c c\:(:rl rncJt·c r;1<li<:a1 c.:llanges i11
our economic system than would have been effected by c.he inflation
millions have feared.
This is not a predicLion that it will. '.Ye have no scientific data to
warrant such a prediction. Our data here dc<tl with price levels
alone. Rut in the face of the data, one would be short,ightcd not to
consider possible social implkati.ons. Great changes in our economic
and social organization may be the outcome. ' ·\'alter Lippmann,
for instance, has dedared:
1f we fix our minds upon the fact that the capacity to produr.e is the
nation's 've;-tltl1 a11d up<>t1 tl1c disl<>r.atior1 of tltat capar.it)' as the su-
pren1e. evil lo be a,. .oidc:d, lVC sl1all, J belic,:e, ha\1e hol<l of tl1e sa,..i11g
trut11. "l'l1is is not. the economir.s "''e v.•ere ca11ght in school. Bltt it is tl1c
economics we are going to ha1'e Lo lean1 in order LO live in this cc11tury.t
• 1~1 early 1!).16 the Federal lleservt Banko! Ne,,· York eslin1;itNI chat tocal
indebtedness i11 tlte U11ited Sta.Les, t)(>th pub1ic au(I priv:itC, ~imot111ted to at
le;tst 35<) billic)r1 clc.•llars, or <lotible tlle prc,\•ar l'cak reached in 193<1. Trl tJ)e
iJtt(:n•(:tling period ntOSC Of the debt f\ad bt:t:ll Sbi{ led t'UltCI th~ ~hOtlfdtrS Of tl1C
federal go,·c1nn1e1)t. At the e1\cJ c>f '~J.f,1') the fct.leral net debt amout1ted to 65
per cenL <>f tJ)e nati(>n;tl indebtedness, oom1)ated to Ot>1y ;?2 f.>~r u·r•t at the end
of '!)<1.0. ]\.fcanv;hile. private iuclcl>tcdncss fell iron1 68 per cent c)f th<: total in
1940 tc) aruuncf 30 per cent by che end of 1915. Debt of the states a11d local
go·vcrn1"J1cnts fell in Lhe sarnc pcrio(.f fro1n to co under 4 l)t~r c";(:t•t. 1\veragc
interesL rate~ fell abot1t one-ha}{ between 19~0 anc1 '!l-f5· so that the interest
}Jurden 01) me eOOllOrny irt '915 V.":lS approximately tllaC in 1930, Or <t little
more, accor<ling to the cstin1ate.
t New York Herald T'ib1111c, Jan. 18, '911·
[ 212]
WAR ANO TTS DISLOC.>\TJOI'\S
Our three wars previous to that of the forties happened 10 fall

at about the same plar.e in the r.yde; the postwar pauerns were
more or less similar not, presumably, because of the wars, but as
soo,.....-,.---,,---,.---,.---,----.----.-.---,..--,..--,..--,-~
100 INTEREST-BEARING DEBT OF U.S.
10
.1 ':"
FIG. 4.. JNTE.REST-BEARJNC Du·r OF ·ral!'.. U NJ1"f.o STA'J'1t$

18g~194r;. Dar.a decennial 1B30-1850: annual th crt~aftcr. 1\ tren<i is sJ).o~·n.
projcct<:<.I tcntalivcly to 1960. Ratios.cal~.
It will be noted tl\at tl1e tre11d is co11cavc uf);vard.~ oourlter Lo all laws of
gro,\rth. One may Qn1c:lude tf)at tJ1is situation cannot ootttinuc pennanently.
the outcom.e of the phase of the cycle. 'l'his provides us today with
opportunity LO .1ee whether or not it is the war or the cycle that
governs post.w;ir behavior. If war governs, we should expcct-
following "\>Vorld \Var l l - a slowdown of about a year for indus-
try to readjust, and then an upturn to a normal peak corresponding
(215]
CYCLES
to that of 1920; then a brief collapse, corresponding ro the depres-

sion of 1921; 1hen eight years or so of advance; and 1.hen a major
collapse, to be enrirl \:cl '' The Second Postwa r Depression."
If, howe,•er, rhyth1ns govern, we should nv1. r.xpect the events
that followed \Vorl(l '·Var I to repeat th~mselves. Rather we should
expect the so-called First Postwar Depression to bt· the 1najor one;
we shoulrl find that the 8-or 1 0-ye~r period of crest which came after
\Var I will not be so conveniently available thh time to Sta\'e off
prompt retributions for our trespasses.
xv
Postwar T1·ends
more specifically the major rhythms we

B EFORF. 1:-lTE.RPRF.TTNG
have available for examining postw<ir probabilities of the near
future, we 1nay well J'eview some of the facts we now know about
the basic trends around which the rhythms move - trends estab·
lishcd in America's economy which in all probability will con-
tin11e:
1. l.\s tl1<.~ trends i11 S(> many ir1$t.an0'.'.:S sll<)"\\'_, lite Unite<l Staces 11as nov.•
acI1it\'e<l \\•}tat son1e comn1cntar.ors ha\'C called" 111aturit)'," and, t<> a
greater o r les.s cr degree, r<!1ar.i\.·e stabilit}· in fur1<lamental or under..
Iying aspects of life where once chel'e was rapid growth.
2. For matt}' years no\\' the nation's Jarid area J1as been constant.. The
area in farms ttuctuatc~s. h11t basic.all)' is 110 fc>nge r ir1crcasing. Our
population is still increasing but at a decreasing rate, and the peak
is apparently just ahead of us. Ou1· foreign tl'ade seems to have passed
its peak and - except as we give things away-would seem due for
the futu!'e to maintain no more than tbe levels reached twenty years
ago.
3. Some indices, like railroad mileage, have ar.rually started down.
4. Nia11ufacturir1g activity is stjll increasing, bltt at a decreasing rate.
And some basic industries, such as steel, seem to be reaching the
Sllllllllit.
5. I>e1>ressjo11s \\•l1e11 the national tre11<l s110,vs •• maturit}' •• - C\'en de-
pressions tha t in earlier times would have seemed unimportant -
will be felt deeply.
[ 215 ~
CYCLES
This is a brief statistical picture of the society we live in, of the

milieu in which we as individuals must create our life. One could
possibly wish something else, '" did Oscar \\iildc:, mourning that
he was a Greek bon1 out of his time. On the 01.her hand, it is natural,
sound and inevitable that at some point growth of all things slack-
ens an<l tl1er1 slOJ)S. E':er}' orgru1isr11 - ~t1cl scu~iCt}' is an (>rganisr11
in at least one sense, ancl so is <~&ch of its institutions - carries in
its very form of life the limitations of ils growth.
'Vltat we know about the supplies of some of our major mineral
resources n:inforccs what our trend chans tell us. '\iorld \Var JI.
for in.11.ancc, reduced our known natural oil supplic.1 to unprece-
dentedly low levels. Our resources of iron and copper were simi lady
reduced t.o the point where we shall soon he <:ompeting with other
nations for foreign supplies. In many respects, the long-term post-
war problem will he not a que•tion of how to mainta in wartime
levels of income - but rather how to maintain even pre\\'ar levels.
As Bernard tvf. Baruch said before the wal' ended:
In our domestic readjustments alone we face the gl'eatest task in his-
tory. Nothing cowpar<tblc lo it bas ever been faced before. Few seem
to rea lize ho,1t drastic::all)' our \Var is drair1i1lg our rav..· r11atcriaJ re·
sourc.:cs - especial)}' Americ..an oil arid Americ;tn metals- or \VJ1at tl1e
impact will be from our high deb' sn11r.turc.
Afr.er tl1e "'ar the great ra\•..- macerial reser,,.oit·s, to say noll1i1lg of the
great lnan1)0\.;er ctnters, simply \\'ill 11ot Uc lVitl1i11 tile U11itc<l States.
The principal natural resources of the world will be in other natiot•<.
"J'lt.e lteci1-0ck, fact is tll(lt afte.1· tlii.s war the; Unitcct State:s u.·ill fac~ a strttg-
gle to rer1ia;ri tlie fi·rst pori•er i11 tlze tt1orld. lt u.•ill 11ot be ea.f)' for America
to hold that J>laa.•
Thus, for the United States die upper limits of growlh seem
alrea<ly t.o press down on us from many and various directions.
l:ntil this lendency of our !,'TOWlh t.o proceed at. a decreasing rate is
• l<alics supplied. In a c'>PYTigf1t.ccl a1·titJe by Hetlry J. Ta.y)()r {n)rU the

Ne\\T Yol'k J•J..orld ~Telegram, F<.~l:>. g~. '91·1·
[216]
POSTWAR T RENOS
demonstrably changed, it is sound procedure to expect ic.s con-

t.i11l1ar1ce.
The patent commissioner who wrote in the 188o's that everything
worth while had been invented, and tl1at the Patent Ollice might
as well be dosed, 1nade just one mi.~Lake. He rensonecl ou t what
ought to happen. He ignored the trends. The reasons he sckcted
as a basis for his conclusions - in a time when the trends
were of course the reverse of today's-had a scope exceedingly
finite. llcasoning will al'\'f'a)'S be ::sil'llilarl y Jjr11itC(l i11. its rcsource.s,
p>lrt.icularly when applied to phenomena of which we our.1<:lves are
a part. To look dispassionately - that is to say, st;u.istical ly - at
what has been happening, what is happening, and wlwt will happen
if present. trends ~ontinuc, helps to avoid at least one common pit-
fall of logic. And to know how fundamental a rcvolntion must be,
to permit the establishment of new trends in an environment,
should obviate another pitfall in wish ful t.ltinking- something to
\\'llicli tl1c 1\111ericar1 tc1up<;rarr1ent is prc.>r1t:.
It is perhaps typical of our lack of needed progress in the social
sciences that even the rate of growth of a nation can still be a mauer
of opinion subject to political argument and commercial ball yhoo.
In the advanced year of 194.1 there appeared a laudably cheerful
advertisement of a great agency, for instan~e. telling first how .lvlr.
Clarence Birdseye made himself millions after various shortsighted
investors had turned him down, and then went sweeping into a
patriotic peroration, criticizing those who see in our fixed land
frontiers a limit to our naLional gTOwth. ' rl1c advertisement
branded such thinkjng as unworthy of the A1nerican tradition, <md
placed upon Americ;in business the burden of disproving this
reasoning.
Doubtk.1.> this is all noble sentiment, and ideal f<.>r <tfter dinner
speeches. But those who pr<l(:C<:cl to plan their business future on
t11c probability that our industries generally will continue growing
here;iftcr at the rate pursued in the past, may be due - with their
creditors - for a sac! awakening. tl·f r. Birdseye- or others like
~ 2 17]
CYCLES
hitu-will undoubtedly go on thinking up inventions (though

there will probably be relatively (ewer in ventions, as our text
shows); and he and 5omc others will undou btl'Clly tllake some new
million.5, and manage to keep a small part uf them after t.11(: gov-
ernn1ent gets through taking 1J1e lar~cst portion in taxes to meet
tlie interest on the national debc. Dut one sw~llow, here or there,
makes no summer.
Actually. even our rate of growth in creative capacit}' has de·
dined. J>r. Pitirim A. Sorokin, chairman of the Department of
Sociology at H arvard, has compi:ed an effective analysis to demon-
strate the decline. H e says, wich italics tliaL are hi> own:
\Ve observe, lir.t, rhnt wirh 1he end of: rhc: nincLeenth and the he·
ginning of r.hc r.wcnLicth century the rate of incrtflSC of discoveries and
invetitio11s de{init£•1)• slO'l.VC<l dnwn; sccorad, tllat l)Cl\'iccn the outbreak
of the "1-Vorlcl \\tar of 1914 and the year 1920 ~ven tltc abso/1tte number
of discot1eries and ir1ocntions decliud; third, that tlte climax (in num-
ber mid importance of discout'Ties) in most of tlie exact sciences was
Tt!aclied 1iot i11 llit: twr11littl1 ce11t:i.'}'J b ttt eitltl''t' in lier nineteent/, or
(for mathematics) tlte t.iglttee.nth century. The [ollowing figures illus-
tl..ttc th.is mo\;cmcr•t:
T1>t<1l Jo..'u,,,tl1t:,. Tot4l 1\:u1tibl':r To1al J\:t1rriber
of Sci•ntific of T eclrnical of Cc11gra/1J.ic Gra.rt1l
Period nistflVf:Tir..t l11uenfious Visrove.ries T oto/
li91-1800 149 113 7 269
1801-1810 tt8 128 6 36•
1811-1820 •SI> 1 57 13 456
1821-1830 388 2'!-7 16 631
1831-1840 141 313 9 763
18-11-1850 534 306 9 899
1851-1860 584 ·1•3 13 1020
1861 - 1870 !1~3 414 I•

.? 992
1871-1880 63; 490 17 J 1'12
1881-1890 663 477 13 1153

1891- 1900 62:. 182 2 1109
1901-1908 552 309 862
[ 218 )
POSTWAR TRENDS
Figure 1 shows how Or. Sorokin"s analysis is bonic out b y the

trencl in r.he number of C. S. paten Ls issued .
Dr. Sorokin adds. in l1is conu:ienc on the prr.scnt st.ate of the
science that thus i~ falledng:
1000~
_~--,~--.~-.-~-.-~-.-~-.~-.~-.~-.~-..~-..~--,~--=>
PATENTS
100
1
~ 10
J
1
Frc. 1. L1. S. l"ATr.~-rs Jssm:o, 1837-1g.14

:\(rend is 5l•()'h'll, projected tc11t;•tivf:I>' to 1960. Ratio scale.
'Vho. save almighty Cod, coulrl comprehcn<I ii• infinite chaos of

.. [acts,'' es1lcc:inll}r 11i11c:c: \\'C <lo1l't k110,.,.· v:hir.h arc relc\!attl and '\ll1ich
il'Tdevantr 111 the face or chis difficulty we eloborow ~ndl~.ss rncchanical
indic<-s and bibliographic•. digests and abstracts. indi= of iudkes, bib·
liogr.1pbies or blblingraplajc!l. lligQl.~ tJr Jigcst~. and a~tracl$ of ab-
~lract.S. llumttrt liCt• i~ too short 10 master st1d1 ;t11 over,.;hcJming a11d
indi\criminare agglnmcrati<>11 of faces. . . .
The more economists h:l\'C tampered with economic conditions, the
'"·orsc Lile): l1avc l.Jcc:ortte: tl)e 1r1<>re IJolitic11l ~r.ic11LiSLS J1a\:e rcfol'tncd
go\•ernme11ts, the more tire go,:er11r11C1)t!> i11 ncerl f>f re(o1·1n .. . . Il e·
SJ>ile a11 chc natural and $oc.;ial sricr1c:t-s al ot1r cii~posal, \Ve are. ur1ablc
either to control the sc>rit>-c:ulttira1 processes or to a\'oid tl1e historiral
catastrophes. . . . ~either happiness, nor safety and S<;Curity, nor e•·en
[ tt9]
CYCl, l'\S
material romfort has been realized. ln few periods of human history

have so many millions of persons been so unhappy. so insecure, so tntn-
g·r)' anlt1tior1s1
crime, mental disease, and other evidences of deep-seated social mala-
dies llourish apace, some of them on a s~-;1le hitherto unknown. \Ve
are '"itnessi11g a verital>le ··blackout" <>f hurrian culture.•
Sorokin is quoted here at ks; length than he des(Tves, in view
of t.hc fact that he has boldly lifted a warning voice in a nation where
cheap optimism is so readily sold by the yard .
\Veil. says an objector, let's grant it's all true, as of the past. But
aren't these facts evidence that we arc probably on the verge of
laundiing an entire new trend on the st.age of history? \Vhat about
the theory that. \Vorld ' ·Var II was the final explosi on of capitalism?
.~faybe the explosion had to occur to destroy the old soc.i<1 l orf.\·a n-
is1n that was through with growing? :Nfaybe we are now abotlt w
de\'elop a11 er1tir<; rtcv.• tr<;11cl i11 At1lcri<:a11 &11·cl\'itl1? \ "f;1yl)c.: t~lC facl
that America is an island in a world of ruins is a sign of it?
It would undouhteclly he possible to conjecture that the trend
Jines we see revealed in b;isic American inclustdcs stand for the life
pattern of what we call capitalism, and that the explosion of capi-
talisn1 i11 otrr time means that nt~\\' trer1<ls arc.~ to l><~ 1-,c>tr) . 1\ sc)<~i;1list
or communist, seizing on such an idea, could then propose the
adoptjon of his own particular eoonorni<: ism as a method of initiat·
ing a new growth pattern, and a new era of industrial expan.~ion.
But the idea is of doubtful v<ilicli ty. True, Professor Schumpcter
seerr15 to cntert.ain., on occasio11, tlle 11ypotllesis tl1at tl1e rl1ytl1ms '\\l'e
ca11 trace ii1 tl1c econorr1i(: orga1·1ism are <.:l1ar.1.<:tcristic~ of <.:apitalis1n
in operation. Kondralieff suggests a similar postnlat.:. The fact is,
however, that we can trace the major rhythm - the 54-ycar wave -
far beyond the birth o[ capitalism, back to the age of feudalism.
The life of capitalism, however you define it, has been certainly
not n\orc than 150 or 200 years; 1.he persistence of the rhythm has
• From C?isis of tJur .1ge by P. A. Sorokin p1.1blis.hc:d anc.1 <:fJ['l}'Tit~ht<..•cl b~· E.
l'. D1.1tto11 and Co .. 11\c., ~1 t,\' \ 'ork. C...i>J'~·rigltL 1941. (pp. 1~6-131)
[ 220)
POST'\V 1\R TR F.f\'DS
been Jong by co01parison. Thus, the wavc;s we see in opcTat.ion cat~

hanlly be merely a characteristic of capitalistic pht:uomeua, how-
ever you define them.
Quite similarly - and t.his may comfort some conservatives- it
seems equally doubtful that t11e trcrul lines we sec in our society
are an (:sscncc of the: capitalistic organization of this society. If they
were, 1.hen something radical should have happ<~ncd to those tTend
Jines around 1938, when capitalis1n as we !rad known ic in the
United Si.ates was transformed into something rathCT cliffen'nt
(m L1<.:h ;1s it t1r1cler\.\'Ctlt t.ra11sft)1·n1aLio11 in France, Ger1natl)', and
England).
The conclusion, convet·sely, can hardly provide fodder for pro·
fessional reformers. \Vhacever may he the inAuencc that has slowed
rlrJ\.\'r1 t.l1c rate (.>f g;rf)\.\'tll i11 c>tlt' soclel}~, tl1ere is no i·eason to beli<..:\'e
that. 1.he imf.~osition of anv otht:r (:conomic svsLem would of itsdf
' '
suJlice to establish a new g'towch trend there.
Capitalism in its fnlkst st:nse has hcc:n only a convenient ccc:i-
nornic word used t0 describe a wt.ole agglomeration of habits, re-
lationsh ips, thinking patterns, and met.hods of sodal procc:clure.
1 .. his <-tgglorr1<:ratic>1l \.\'a.salt •>rga11ic gro,,:tl1 -it '"'ras never pla1111ed.
never imposed by any supcrminds. No one thought it up, no one sat
dOl.\'n lO int.·ertt it; it S})r(1J'1g fton1 tl1c COtllffiltfliLy Sf>OJlt.atlCt)llSI~~. as
paTI. ol a necessary mode o[ expnes.;ion at a given point of time.
If a new growth tr«:nd is <~ventually lo make its appearance on
top of the old one in our society, it may well be accornpanit:J by a
new kincl of economic: expression which- when historians survey
it in retrospect- they will clescriLe as arr econornk ··system." l311t
this economic system, so called, will hardly be die cause o[ the new
trend. Rat.her, it will be an expression or outcon1e of it. ·ro undeT·
stand this, one need only understand man foT what. he truly is - no
mere creature of econornics, hut one for whom economic activity
is only one fonn of expression among niany.
H, with all these reservations, "e still cling hopefully co the con·
viction t.haL a new growth pattern is going to get started in our
[ 22 I .I
CYCLES
nation, we arc wholly within our rights. Germany seemed lo have

launched herself on a new trend in the nineteenth century. Russia
began one in 19 '7·
.But -we must be warned. \Ve now know enough about trend
lines to realize that old ones merge into the takc-0ff of new ones
only when fundamental, even revolutionary, changes have occurred
in the environment and its organization, and perpetuate themselves.
Even though our sdence caonot tell us when a new growth pattern
w1JI begin - or it the pattern ever will-it does tell us this inuch.
Ft:\\' reallj· \.va11t .')tl<:lt a re\:oliition.
Remember Pearl's Drosophila bot.lie. Imagine for a moment that
we Americans are t.hc Dro!K>phila. \·Ve are reaching the upper
asymptote of our curve. The invention of a few new gadgets by in-
dustry- indeed, the invention of a whole new industtv icself- is
. '
;1 force ir1 rl<> \\'a): a<l<:c.1uate t<.> <:l1ange tl1<! rclati<)r1sl1lps i•l our l>ottle.
\Ne must have :1 whole new bottle. And if we are an1ong those who
call for lwving the postw;ir national income 1naintained at sorne
level around 150 billion dollars - or 1.he cquivaknr. thereof in 1939
values-then we art: nc(:cssarilydem;mdingsomesortof completely
new social bottle into which the American nation is going to move .
Now, we can predict that this will happen, if we wish, because
science has no evidence with which LO gainsay us. Ttc.ao even offer us
a liHle helpful data, such as the fact that in recent wanirne y<:ars
the birth ,.ate has leaped upward in the United States, as in many
ocher parts of the world. This deviation from one long-established
t.r·<:t~d lir~e })a..S heetl at.~>Url) c:<l }))' rno3t ol>3Crv<':rs to be 111crcly a tcma
porary phenomenon, a warti1ne distortion. But it coulcl possibly be
sun1eLl1i11g 11111cl1 ttlc)rc signifi<:<lnt f(>r tl1e ftLtUT(~. \\'e are e11titled
to our faith and hope. \\!e can believe in che binh of a new world,
if we wish.
Only we must he prepared to face the consequences, whether we
arc right or wrong. If we are right., we st.ill face enormous upheav-
als (for the ending of the " old " capitalism has as yet not changed
our lives much), and some of us are perhaps going to he badly
[ 222 J
POSTWAR TRE:-IDS
jolted. If we are wrong, we may court serious personal disaster in

betting on our judgment in the market place.
If we do not expect such a i-cvolution in Ameriain rehttionships
of every cc.onomic, social, and polit.ical variety, 1.hen we must assiunc
1.haL the trend lines for our economy will stay Aatt1:11cd out. In 1..his
event, we must similarly assu1nc- on tl1 c ba~is of past perform·
ancc-that four major :rhythms clcdining 1.o gctbcr can possibly
make a very deep economic depression inclec'<I, stanjng with a crest
of .. good times" in ic116 or '47, and working downward to the bot-
tom of the rrougli in perhaps 1951 or 1952.
If we make this assumption, p<.:rhaps we should go further, and
ask ourselves whether o ur complex >ociety could withstand another
such depression Hild meanwhile maintain orderly processes in
operation. If our answer is a tentative No, then we may expect to
get our social upheav;ils ""l'way. Jn which event the old trend lines
as we have known then1 might eventually be superseded by t11c be·
ginning of new ones, regardless.
'fhese speculations are entered here primarily to demooslrate
that our econotnic charu-howt.·\'(:r helpful - will by 110 means
sufli<'e to answer all the questions on which judgments o( the future
must be based. \Ve can synthesi'e our rhythms to portray the future
or a given corporation, without. too great a margin of t"T·t·or. But in
cstim"ting the future of a nation we are ccrt:tinly dealing wit.h
biologic.al aud psychological forces on a scale too g,.eal. for com·
pletdy accurate measurement wit11 our still primitive instTt11nents.
In rca<ling the instruments we have available, much allowance muse
be made for error.
Those instn1men1s can neither affirm nor deny- as we ha\'e
seen - that the trends rn:uuring in the American econom}' 1nay be
in process o( being supersedccl by new ones. Apparently no con·
scious planning will in itself sc1-ve to change whaLever trends arc
now eswhlishr.d. Apparently indivi<luals can do liule enough to
create or hasten developments tha: must spring from the heart of
the whole social organism. \Ve can figure out where we are - mak·
[ 223]
CYCLES
ing due allowance for a margin of error, but that gives us no recipe
for getting clscwl1erc.
So1ne economists who do not quite re.1lize this are responsihle for
much of the literature of recent times which rq>ea1.s e ndless ex-
hort1lt.ions to be up and moving to some place where we aren't. In
the words o( John Chamberlain, eminent book editor of the 1Vew
York Times:
There a.re two kind• of economic piety at larg•· in the world today.
The old school piety consists of varfatious on the refrain, "\\Te must
have free en1e11'ris1:," or" \Ve must preser\'e 1bc 1\mcrican way of life."
The newer piety rnnsists of saying. "\Ve must s1.ihili1e che economy at
a high level of r.onsumpcion and employment." In each case chc appeal
is purely hortatory; it docs not tdl you how to get Crom here lO there.
·1·11e m<>re I read, tl\c 111<>re I man'el al the rcputatic)r1s tl1at ha\.·e been
made by repe;iting one or tile other of the pious exhortations.•
A-fr_ Ch;unbcrlain is, of course, quite right. But it is poMiblc to go

further, ancl say that the econrnnists of ooth the honatory schools
which amaze him err not in fai lihg to do the impossible - for
they cannot gi ve a practical recipe to acl1icve the ends they recom-
mend. They en· rather in failing to recognize the rc11l limitations
of their approach lO ll1 c problen1, an<! in foi ling to recognize this
problcru's real essence. For it is apparen1Jy a prohlern hasic"11ly
biologic.al in its cbaracter ; and if there is a sound solution to be
found in atl)' form or traditional human loi,.;c, no expert has as yet
produced the key. Churchmen. con vinced that che ult.irnatJ> key
will be found in a swccpiug religious •l·vival that will spiritually
reform the worsh ipers of today's paganistic materialism, ha"e al
least as much logic :in their side as tl1osc who propose n1aterialistic
solutions.
If t.hc intc.rp•·ecatio ns here a re correct, th en ordinary polilical
an•I cconon1ic tinkcrings will not get the f>l'<Jhlcm resolved - as
Sorokin says clearly. T he community as a whole will ultimately
• New YOTI< Timu, ~hrch 11. '9H-
[ 224)
POST\VAK l"R£NT)S
give us the answer, whether in one way or in another. ll-feanwbile

no individual, while he wait.s to see what that answer will be, is
going to effect <my " plan" that wi:J really " save " the nation. But
he can certainly, with some foreknowledge of what the problem is
,.i I about, be in a h<:tl cr position to assay the ' 'alue of the various
plans chat will he offered in the future. He will be able better to
understand the fallacies inhLTent in the multitudi11ous nostruins.
He will douhdcss be able to help his neighbors more sympalhet-
ically. as they find themsdvcs saddled with heavy problems not
entirely of their own making, in these dmcs of world-wide misery.
And, e<rually important, he should certainly be ahlc to help him-
self through understanding some of the limits of the possibilities
~hat surround him.
X VI
Postwa1· Rhythms
K ANAl.YZING and projecting the trends, both for the nation

I as a whole and for certain basic induscries, this study has sought to
nmke clear that a dcdining r&te of gTowth elevates t.he "cycle" to
greater importance for business and industry than ever before in
our history. L imiicd as our knowledge of such rhythms still is,
there are basic data to help us face che problem of the post.war era
with more than 111ere guesswork. \Ve have a tool whereby, for any
given series of figures, we can get a fund of additional knowledge
for guidance in facing the problems of any specific organization.
Our pat.t.crn did not serve to show us when \Vorlcl \Var JI would
finally end. As previously noted, we do not even know how to define
"w<ir." The ending of actual bostil it.ics in 1945 may conceivably
prove not. even to have been t.he real end of the war - much as
the 1918--1939 period was one in which war merely shifted fro1n
the rnilit<lry to tlae dif>lolnatic level.
But here, in summary, arc some findings of which we may be
relatively sure:
t. The underlying 5.f.-year rhytlun in wholesale prices is on the de-
cline. It ttnned in 192,'): the pattern is due to reach bottom in 1952.
2. The shorter 9-year pattern in wholesale prices - a rhythm chat ap-
plies also to iron, steel, and s1ockmarkct prir.es -which had its last
}1igh in 19371 reacheci [of a11otl1cr l1igh itt 1946. The r~\ttl~rn \V<t~"i
<luc t11e11 to tl1rn clov.•11 t1r1til 1951.
[ 226}
l'OSTWA It ltffYTH!.fS
3 . .Just as the pauem of the g.y<-ar r:iythm was <luc to TC"dcb a p<-al:. in
1945, or shortly :ihcr (it <lcpcud! Oil the business in question), so
the 31;2-}•car paLt<'rn - aJr11ost uulversal i11 Uu!>illeis - - ,~:as fore·
casting a peal:. in 19,17. Other peaks in che 3V2-ycur rhythm arc <lue
iu 1950 and 1951. Lows al'e due in '48 and '51,
4. The i8Vs-year paHe1 u in building activity apparently reached its
high ahout tare •9·1•· tl1e exact date depending 011 10c1hods of re-
porting and compiling <tati>tics. In general. huil<ling and real es1.1te
paucrns are due to ckcline to a low around 19:;3.
5. The i 5-ycar pa.•tcrn in .the i:ndelC 0£ the p\•rch:.sing f>O'ver 0£ bee£
caute prices. an txcPeding ly int(Testing index !or the farmt-r, was
scheduled for a hi&h around 1944. thereaCter declining to a low due
i11 i95i. OPA regulations and wartime lllack markets veiled the
meaning of free ma1·ket statistics. Bue <lcclincs after •!HG would not
be surprising.
It c.mnot b e repeated too often that these findings a.re not offered
as unqualified forccasu.. The data do not b"' back far eno11gh to
permit unqualified assur.incc that :he observed rhythms are beyond
the result of chance. Further, if all the rhythms are real and con·
tinue, they 1night still be so buffeted about by other rhythms and
sporadic forces that the actual highs and lows would be delayed
or acceleratecl.
Still, with .all these quaJifkations, it is wonh no1.ing that the ex-
p ecwndes facing us do not suggest. any lcngcliy postwilr boom. On
the <ontrary, the number of important rhythms that come to a low
together around 195~ suggest die passibilit)' of a b'rowing postwar
crhi!.. The reader now familiar with this method of analysis will
not expect th e crisis really to be :n·oided or halted by any pre-
ventive measures whkh the b'Overnment might decide to take. Jn
chc present limitc<l state of htirnan knowledge as to the ultimate
nature of these rhyth111s and their correlations, we shall suspect
tl1at any action adopted by guvemmenl woulcl be ameliorative in
character without h eing curative. Probably the most we can hope
for would be some kind of palliative aciion if the blows fall.
[ 227]
CYCLES
The individual, on the other hand, can doubtless help himself

to some degree in advance - keeping in mind the good example
of Joseph in preparing for Jean years when fat was still upon the
land. To give specific advice is not the purpose of these pages, but
a few generalities may be mentioned parenthetically in passing, to
indicate recipes that cinnot be relied on.
Obviously no individual, in times like ours, can rely wholly on
money saved to carry him th.rough any future periods of social or
economic crisis. In other times, if faced ~)' evidence that deflation
lay ahead, he could unerringly have put his holdings into money
and obligations repayable in money. But since 1929, governments
all over the world have learned how to demonetize money. Our
own government in 1933, for instance, called in all old dollars -
Just as some European countries have recently done; it can readily
call in the current issue when it pleases, and/or make it unspencla-
bl e. It has learned through experience how t.o resuict the with-
drawals from savings and other accounts. It can delay pay1nents
on bonds due. It can, and now does, contl'ol the expol't and import
of dollars.
In oc.her words, various monetary controls devised by DI'. Schacht
for the mark have already been quietly applied in various ways to
the dollar, and no American man planning for tlie future can safely
assume tliat he will ever again be able to do wit11 dollars exactly
what he will. In actual practice (if not in theory!) sovereign states
today hold the view that r.u1·1·enr.y anl<tlC', Ctll<l t.llll:S CtlC' Lile::' Slate's p1i1.>pc:1ty, LU lJ(! 11cltl at1U ll:>Ct..l
hy intlividuals only so long as that use serves the state'.\ purposes-

or, at any rate, does not come in conllict witb them .
Nor, for that matter, do present tendencies and trends in the
economy sugg<:st future opportunities for freedom that once pre·
vailed in tl1e uninhibited enjoyment of any kind of real property.
l\fetropolitan properties, for instance, will doubtless continue to
he milked by high real estate taxes for the support of unemployed
city masses. Grim realty taxes, comhined with rent controls, could
[228]
POSTWAR RHYTHMS
ultimately prove 11lrnost confis<:atory. Thus, income-producing real

estate is no longer a perfect hedge for inAation, and in a time of
deflation may have questionable merits indeed.
Nor are bonds a perfect refuge for savings any long~'r. ·rhe ulti-
mate value: of govcnnocnt bonds is something for conjecture. Of
some corporate bonds it is possible ro say t.hat in a period of national
deAation they have: been d emonstnitcd to be as questionable an
asset as stocks. And stocks now mean to Lhe investor only the right
to participate in wh;1t is left after management takes its salaries and
expenses and government seizes what it will.
Some of the tax laws that weighed down the American economic
outlook during t.he war years have been changecl - but it is p.roba-
bly too 1nuch lo hope that they will undergo, in anr foreseeable fu.
turc, the wholly drastic change that will restore past relationships.
Thus 1.hc man seeking to protect his ow11 future <ig:iinst the turns
of the economic wheel 1nust nowadays face diflicuhies of a rather
unexampled kind. He will doubtless wish at least to reduce his
dollar indebteclness to the rninimum, on the theory that freedom
from debt at least means freedom of action in time of emergency.
In an era when freedom of choice has been steadily whittled down
to the vanishing point fo,. almost every individual in terms of u1Li-
ma1.cly ,ignifir.ant. values, freedom from debt is one that can still
he chosen. In the era ahead of us it may be valuable to anyone who
esteems his freedom dearly.
The foresighted individual will doubt.less know that long-term
leases, contrar.ted for wlten properties are at their pe<ik prke, are
n111ch the sarne as long-term debt; mortgages inevitably are debt,
regardless of terms of payment; and even life it1surnnce policies
- unless fully paid up - are a form of debt which , unc.il paid up,
can often cost the insured rnuch of his equity in the policy if he
fails to continue premium payments and must let the in,urancc
lapse. But tl1Cs(~ are all details that interested individuals wiJJ find
lllorc: adequately discussed in numerous other quarters. Here it is
suflici~nt to note that hedges to protect principal in periods like the
(229]
CYCl.J:'.S
present will continue hard to find. while the retun1s on capital in

general bid fair to continue at around low levels.
In an e<:onomy wh ere the rate of l,'TOwth is rapidly approaching
zero. and major underlying rhythms in the expression of energy are
also on the decline, what would one expect t.o meet as a major syrnp·
Per Cent
I I I I I I
7 I- -
6 ~
-
5 Averaze Profit
lat half of period 4.951'
-
4 - -
3 - Average Profit
2nd holf of period 3.03%
2 - -
I - -
' ' I I I I
1910 1920 1930 1940
Fie. i. I>r<c.>1•11~ )N" 'J'HJ::.

S·r.t.t:1.. 1Noi.::;·1·Kv, 1901-1945
(From daca fur11ished by 1.1"1e .'-\u1erica11 lroJt & Steel l11s1itt1te.)
1:or pi.1rposes of coroparability Lhc c:hart Ml()\VS averages for L\\'(> 18--yt.ar
cyr.les, C':a('J1 con t.aining a y.;ar. Tilt ~it.iatiou i11 steel i.<; re1>re.1.;c.~1\La(ive f<.1r many
in<lt1.strics.
tom of such lowered vitality? In a profiL economy, it could well be

.:. cl t"rl in i 11g l":l f e n t: pr·,lf-1 t . A r'lrl t I•JI t ls r~ -x :~rt l}' \1.• } );l f wf> },;-. \'f;' rr,et. '--'"'f'T
a long period of )'cars, in the United States of Arn.crica. See Fig. 1.
T~1<:~ fc>rf:g·c>i 11g is not n1eanc to st1g;gesc tl1aLot1 1~ rll}'tllnls i1lcli<":atc
collapse of our ccononl)'. \'\le uiay only infer, if they conLinue, chat
around 1947 a fall-off will start, and our economy will experience
rather protracted declines in prices, production, employment, and
economic activiLy generally, reaching a bottom that would pre-
sumabl v be dated somei.imc in the carlv fifties.
' -
As has been noted, facLOrs like war-though not (in the past)
f 230}
ultirnately changing either trends or rhythm patterns- can for a
time emphasiie, n1agnify, or distort a rhythrn. There are doubtless,
yet unknown, oLher factors hc.1idcs those appearing dearly in war-
time whid1 nmv effect the same end. It niav be that such a factor
' '
will show up in conjunction with the postwar rhythms, to modify
their operation. But the d:ua here do nol suggest t.he likelihood
of the possibility.
It rr1a}' alsc> l>e !\tTf".~<i.Prl ngai11 th71t 0 11r knolvl eifgP. nf rhythms i!<\
sLill in its infancy. Concerning t.heir n~a.lity, some high authoi·ities
have f.(:.\tified t.o conviction; as to the exat:tncss of the timi11g of the
rhytluns, some of Lhe greatest. authorities still disagree. Nor, 0£
course, can anv . research assert chaL a rlwthn1 , - however real in the
past - will undoubtedly continue. Again, it n1ust be noLed that
individual industries and co1npanies and co1nmodities have their
own pauerns, which must be determined one by one. \Vith all Lhese
<1 t1aJ i fi(:ati<">11s, tl1c u1eri t <>f kn<)\vi 11g \vl1a t pr<>l >al>i Ii t. ies t~lC- rl'~}·th rr1s
indicate is obvious.
In studying these rhyt.hms for purposes of postwar guidance, it
seems imponant to note again that for Lhe first time immediately
following a major wat' in over a century, the projected rhythm of
building activities will be on the downgrade. As we have already
seen, building is closely related to the activity of many basic indus-
tries. H building activity does fir.d itseir inhibited in the coming
years, we cottld be justified in expecting heightened seriousness in
any decline Lhat appeared in otl1er economic activity.
One could have" pr(lved " in advance of the war's ending, with
almost irrefutable logic, that after \Vorld \Var 11 enormous pent-up
clemand for h(lusi ng would launch an almost unprec:cdcnr.ecl build-
ing boom, not to say a period of extensive land speculation. But
the building cycle is one of the most obvious, evident, and
dearly marked. As we have seen, it. cvid.enccs a simple, not
complex rhythm. Its regularity is apparent to the most untrained
eye, on almost any charted pat.tern. lf 1.hc pattern fails to repeat
itself in relative decline following \Vorld \Var II, the failure
[231]
CYCLES
would be the first L<> appear sinc.c national statistical records have
hcen compilccl. H ence we seem to be warranted in taking tht: post-
war implications of a clcdining rhythm in building activity with
so111e scrio11sr1css.
NoL the indicat.ion of decline in price levels, but the indication
of restricted building act.ivit.y for some year.s along with a decline
in price le\ Cls f.>f <":0111rr1oclities an<l sect1rities, is a slgtlf)OSI. u11~
1
precedented for any 01.her immediate postwar em in the past cen-

tury. Declining price levels would mean, of course, that tlw dollar
would rise in purchasing power. It would indicate, in effect, that
the dollar has rcceAtly been undervalued in a world where it has
survived, under war's various desLntctive forces, as the strongest,
most desirable currency on the g!ohe. From this point of view, price
declines could well he viewed a, growing recognition of the essen-
tial inner strength of the United States of A1nerica.
In seeking to time the beginning of the corrective economic
downturn following \Vorld \Va:· II, we may remember, from our
various data, that the 1929 stoc;k market panic waited for a turn-
down in the 9-year cycle in wholesale and stock prices, and followed
tl1c I.urn in the building activity cyde by some 3 years. A 2-10 4-year
lag betwcer1 the turn in the building rhyLhm and a panicky fall
in stock prices has not been uncommon. as previously noted. Such
data from the past .~ngg<:st.cd to informed observers that !lie real
post-war break in the bu ll market of the forties could come any
time aft.er 1.11<: middle of 1946 - after the t.urn-down of the 9-year
p<tlt~111 i11 jt.J 11~ of tl1<ll y..-:a1·. Tl1<:: IJ1~•1k. a11;vt'.tl 011 !it:l1~c.[t1 1c.:. ,:\fccr
the turn-down in the decisive 3ty<:ar, or 11-rnonth, pattern scl1ed-
uled for early 19.1.7, the decline coulcl he cxp("ted to increase in
intensity.
If 1950--1953 looks like a subsequent deep valley on the charts,
this much should be noted: 1t is also the beginning of a long trend
upward. Further, it is not the peaks and valleys of the rhythmic
patterns, as a rule, that people note so memorably in their experi-
ence. The periods prevailing between are often the ones that most
[ 232]
!'OSTWAR RHYTHMS
affet~tlives. Thus, the downturn in the 5,J·year price rhythrn was

hardly even not(:d for its significance by most commcnt11tors in
192:,. Real estate bo01ned in areas like New York long after the
passing of the national activity peak in 1926, and the (:ollapse of the
Florida speculation of the period. Not even the stock market panic
of l!J29 was of pararnount significarKC 1.0 rnon~ than perhaps a few
thousand people who were unhappily too extenclccl in tlteir com-
mit.m<:11t.s to meet it. The millions proceeded pleasantly on their
way, and many reached 193 1 and 1932 before they t:nnduded that
tl1i11g·s \vet'e 11c>t 11cJr111aJ.
Titus an ohscrver may hai ard the guess t.hat the turns in various
projected rhythmic patterns, due in 1946 and 1947 if the pauerns
continue, may at the time hardly be not.iced by most o[ the people.
\-Ve could know, as of 1946, that any coming declines would start
fro1n such high levels of national incorne and production that in
their (:arly stages of fall most <:conomic indices would still he above
all prewar norrns. In April of 1946, for instance, production was
proceeding at the rate of 150 billion dollan annually- the highest
rnte of peacetime production in American history. At the same
Lime, the Federal Reserve Board's prodnction index crossed 170,
compared to 109 in 1929. The national income, after an insif,.-nifi-
cant "reconversion" slump, was still aln1ost at its wartime peak.
And the T reasury could announce that sales of savings bonds r.o
the people in the first. quarter of 1946 amounted 1.0 over 2 billion
dollars, or three-quaners as much as in the comp;irablc quart.er of
1945 when the war was on. '\Ne may expect that, by aJI prevailing
defirution.s. l!J47 will be a good btL~iness year. Probably many
will feel themselves equally as prosperous in 1948 and lat.er.
If, following depres_~ion lows, a definite turn upward finally comes
around 1!)52, it shou lrl be an important one in rnany ways, notable
particularly because a rise in the rhythm of real estate and huilding
activity will coincide with a rise in the long 5·1-)'ear price cycle.
\Ve rlo not need to ai.,rrce entirely with Professor Schumpetcr's
theories regarding the 54-year cycle to regard his research as of para-
[ 233]
CVCl, ES
mount importance, and his ideas as highly significant.. It will he

remen1bered that he retraced historical dat.a in the periods covered
hy 1.hrce such cycks - following a technique used for rather dif-
ferent purposes by Spengler. Somewhat like Spengler, who identi-
fied each one of his cycles with the progress of a particular culture,
Professor Schum peter belic,·cs that each ;,4-year cycle may be iden-
tified by its association with some particular economic influence.
Similarly, we need not wholly accept Professor SchumpeLer's hc-
lid that a 11cw i11dus trial "(:a use·· lies behind each 54-year cycl~
to agn:e wholly with his observation that - during the last three
cycles, at least - 1.hc course of each one has procccc.led simulta-
neously with some panicular economic force of a new kind.
It will he recalled that Schumpeter calls the cycle beginning in
the late 17oo's 1.hat of the industrial. revolution. The one that fol-
lowed saw the development of the steam engine and the railroad,
to a point where this cycle changed the h<ibits of men everywhere.
The latest one, which began around 1898, Schurnpeter has asso·
ciated with electricity and the auto1nobile. This is a striking oh-
servation, and one confinl1ed in the •~xpcricn<:c of millions who
have lived through the period when electricity and personal mo·
bility have changed almost all the living patterns in our nation.
IL is at least interesting, >)nd not wholly without possible useful-
ness, to wonder whether the new 54-year cycle (which, if this
hitherto fundamental pattern continues, will begin about 1952) is
ag:1in to be a.~sociated with some great d1a.11ge in our living pat-
ten1s, stemming fro1n industrial or social de":elopme11ts 110,..,r l1ardly
visualized.
No such indulgence in pure conjecture is required to note that,
for a period of years after 1952, the upward sweep of various major
periodicities suggests a time of important constructive accomplish-
ments and prosperity in the general econou1ic life of the nation,
regardless of the leveling out of basic u·ends of growth. Because this
is as far ahead as it can serve most readers' purpose to look at the
present Lime, the discussion of the 1011gcr tcnn may be dropped
[ 234 J
POSTWAR ~HYTHMS
at dii.~ point. The reader wishing to continue the exploration into

later years ma)' reaclily fol'rnnlat.c some conclusions of his own h)'
studying and projecting the rhythms grouped fo1· cxmvcni<:ncc on
pages 18$-189 ancl hy clet<·rmining and projecting the rhythms that
may h<~ prc:sc:nt. in t.he series in which he is particula1·ly interested.
l\Icanwhile, it is apparent th<1t our n;ir.ion will have various
bridges to cross chat clo not look whoily easy of negotiation. Some
connnent«tors, li.ke che sober Sumner '\Nel les, h;ivc i11<kcd hinted
at" revolution." '.\;(r. \Nelles clecl;1rccl, during the war years:
:Every grtat c:on\'ulsion rtc:c>rc1cd in histo11· or: \\•hi<;h t11e migl1ticst
1
will be this global war, bas been followed by political ~nd social reper·
cussions which have wholly modified the. course of human events.
lt is i11cvi.tab!c tlt<tt after the present \\.'<tr tile strt1c.:ttire of Ot1r 1nod-
ern civiJizati<>n v.•ill he profountJly Lra11sfc>rrne<l.•
'rhe data we have for rhythms, with so many declining in concert

afler i947, do undeniably suggest a difficult period of adjustment
for the short-term future. 1' here is no reason, as we lt;i\'C seen, to
ass111n(: tJ1at. the rhythms will act differently just because political
go,;ern1nent, instead of the ~>t•.sir1es.s mart, 110'"' clon1 i11at.(:s tl1<~ J\r11cri~
c;in economy. There is equally no reason to assume that. gov-
cnunent planning by government administrators will prevent
rl1ythmic deprcssio11s, any more than planning by bankers and
businessmen can.f
Our knowledge, however limited, is at least. adequate t.o sugg<~st
that American business should not indulge in optimism on this
score, and shoulcl prep;ire for distinctly leaner economic pickings
in the postwar clcco1dc. Fven husinesses that see every "reason " to
expect a boom should be sure of sound grounds for tlteir optimism.
• ~1 e'\-\' Yl>rk Herald 1·ribu11e1 July 12, 1944.
t It is n<1t:'ll)Je, f<)T iost<111<:e, 1J)at e'•e11 in a Jna11agecl ec:onon1y l ike Russia
tlu,:rc \\'<lJ'i. £arni11c jn the t'''(:ritiC's, to;hc:n capft;tlist natif>llS <·:xperi1:n(c<l t.l1cir
0\\'11 first post\,•<ir depression. and (amine again in tl1c (.-;trly tl1inics, v.·hcn the
seoond post.,..·ar def>ressior1 arriver!. \·\'l1ile 1>r·ice Slatislics !or llussia <ll1 11ot lc1uJ
Lhe1t1sel\•es to tracing rlt)'tbn1s, events of this kir11l ~11~~rt:st rome kiod of cor·
related rl1ythm that evC'n a stacist econon1y meet$ jn one form if not a.t1other.
[ 235)
CYCLES
Countless numbers of American businesses Jo have rhyLhmsof their

own which seem more or less independent of the dominanL cl:o-
nomic rhythms of t11e nation . Th(:re are many busine.1.<cs on record,
for inst;ince, 1.haunadc their highs in 1930, and some even in 1931.
Similarly, the presidenLs of laq.;e and successfu l corporations have
been heard 1.0 s;iy privately 1.hat depressions were periods in which
their companies bad made the greatest relative gains. ·rhe depres-
sion problem for business management is often only a problem of
hci ng prepared.
No general work of this nature c;an hope 1.0 do more than issue a
few words of general warning to those who arc ope1·ating their
private and puhli~ lives on 1.he supposition that tl1c ride or general
postwar " prosperiLy ·· will he so great i11 a world sLarvcd for our
products that even weak ships will triumph anLly ride the waves.
The judgments here, and throughout the text, rest on three as-
sumptions that are basic in l11is treatise, and arc worch emphasiz-
ing again:
1. That a rhythm can be considered tu be significant only if it
has repealed itself so many times and so regularly that it cannot

reasonably be Lhe result of chance - and
2. That the rhythms whkh have evidencc1l chemsdvcs dearly
in the past., i11 our economy, cont.inue to operalc 1·cpcli1ively- and
3. That the u·end~ which have been tril(Cd in our i,'<eat ba.\ic
industries continue on their establishccl courses.
The evidence hruught 10 bear on the problem srrongly suggests
the probability th:.-.t 3l IC':Jst some of tl1c rl1ytl1ms arc: :sio.,,i.fic.anL.
and will continue their patterns as projected. The e\'idcncc
regarding the trend lilies, on the other hand, cannot rule out the
pos.~ibility that new logistic curves ;ffe even now ~upcrimposing
chen1sclves on American trends that have hcc11 long familiar.
The probabiliti es, of course, are of a different nacure. They sug-
gest that every businessman and every responsible head of a family
should be looking ahead economically with great caution as the
lace forties approach. They suggest the rise of problems in govern-
[ 236]
1nen1, as well as in business 1nanagemenc, which will require che
height o[ wisdom 1.0 solve.
Tl1e same rhythms that suggest caution in planning fnr tlic late
forties also foreshadow various new opportunities op<:ning up in
the fifties, of a kind such as our nation's youth may hopefully an-
ticipacc.
However difficult the interim, it will not be the economic end o[
ot1r \\-'orlcl, as thl>se <~alJgt1t irl CCf>r1ou1ic panic are son·•cti111es in#
dined to think. It will merely be some wi11t.<:r weather. \Ve cannot
choose the (:oursc of the: wt:at.lter, but we can at least he t.ltouglttful
in preparation when we note that the barometer is beginning to
fall. Tl the rhyt11ms are uhiniately conlinned by events, in their
indications that th<~ immediate postwar era will hring our ccono1ny
in to some economic hard sledding, we can take renewed hope it\
the very rhythm of dl<~ winter's coming. lt will help confirm the
knowledge we have gained that. 0 1u economic rhytluns do have pat-
terns that can be projected-patterns that assure us that we shall
also see, not too Jong hcH(:c, a new eco110111ic spring.
[ 237]
Appendices
1\PPENDIX I
The Ratio Scale
llE souo LINE in Fig. r, on page 3, shows the sales of a hypo·

T theli<:al business that grew 9<• per cent in the live-year period
from 1905 to 191 o, 80 per cent in the next fi ve-ycar period, 70 per
(:ent in the next, and so on. The rate of growth, that is, declines
10 per cent in each five-year period, ancl falls to a level of 20 per cent
in the five-year period from 1940 to 1945.
The curve in Fig. 1 gives little or no visual warning of the drastic
fall-01f in sales that lies irnmediatcly ahead (as shown by the broken
line in this chart) if the growth pattern continues in the future
as in the past. If this pattern does continue. the sales in 1950 will
be only 10 per cent more than the sales in 194.i;, and the sales in
1955 will show no growth whatever when compared to 19.50.
In order to make this situation visually dear, many business1ncn
and students of trend.~ prefer to plot all rlata pertaining to growth
on what is known as a ratio (or semi-logarithmic) scale. On such a
scale equ;Jl vertical distanc.es n~present equal fuirce111age change.
LeL us illustrate by charting the growth of another business using
both the arithmetic and ratio s<::iks.
The husi11 ess starts at a sales volume of $2,000 a year; the second
year it doubles its sales to $,1,000; the ne.xt year it doubles again to
$8,ooo.
On the arithmetic scale these fact.~ would be represe11ted in a
chart laid out as follows:
C Y CLl\S
Crowch or S4000
or 100" of
Second Ye•r
Gro•..-th or $.2.000
or 100" ol
£.irst YN T
2 -------~
o.........,._,,__
l•I Year
__,,2Ad. . ,.Y11r
!,---..,..-,.,_-----
Jul Ytv
---------'
F u::. 1. E QUAi. RATE 0 11 GRO\!fTtt- 1\ lllTHMF.TIC Sc:At.t-:
Sales or ll l1ypoth1·cic·:1l lJ1.1&lness OTKl&llization $111)'.'' ing cons1:1Tll rate of gro wtll.
Arithm .r tit: .icale.
On ;1 r:it.io s<:;11c, lhe sa me facts would be represented on a chart

laid out in this fashion:
JO
a
5 ti llr0'9th of $•000
g 01 lOOJI', ot
$et(ln\1 Year
'O •
~ Gtcwtb ef $2000
Cf ll:I01' of
j l
ftr•l Year
lit Year 1nd Yetr 3fd Y••r
Frc. 2. EQU-''· R ...nt u 1· CR0\\'11[ - RATIO SCAl.£

Salcio o r a hypolht t.ic:al busir1ess org1tniz.a:tiorl slto\11ing conscanl r ate (){ gro~· Lh.
Ro.tio u o/4.
APPENDIX i
Notice that on a ratio scale the numbers on the vertical scale

represent equidistant percent;1ge valu~s, and come closer and closer
together, in a way LhaL interposes the same verr.i<:al distance l>C·
tween 1and8 as between 2 and 4 (because 8 !>cars the same relation
to4 that4 bears to z).
A straight line plotted on a ratio scale means that the rate of
f,>TOwt'1 is constant. (Such a state oi affairs, of course, never exists for
long in any real situation.)
\Vhen the rate of growth declines as tlle business or the organism
gets older, this decline is represented on a ratio scale by a bending.
over in the line.
Thus, if the business cited here grew from $2,ooo in its first year
lo $4,000 in its second year, and then grew an equal amou.nt - from
$4,000 to $6,ooo - from its second to third year, these facts would
be shown on ratio scale as follows:
8,-----,,----..---.------------------,
s. Growth of $2<100
or 50• ol }
Seeood Year Sp11Ce$ iteoresent
Growth of $2000 ~ G1owth
Of 100% of
fll$l Year
llll Y1;0"1r 2nd Y0:ar Srd Year
f)(;. 3· EQUAi . .-'\;l.,t)Uf\"T OY GRO\\'T1-I - RA.110 SCALE

Sales of a hypotl1ct.ia11 Jn1sincss org:1r1izaLio1\ sl\ol'>·ing co11stant amount of
gro\\lth . .Ro.ti<> scale.
On the other hand, using the arithmetic scale, tlle line would
continue from the second tot.he third year at t'1e same slope as front
the first year co the second, giving no visual sugg~stion that the rate
of growth was declining. ·rhe .same figures plotted on t.he arithmetic
scale would look like thi.~:
[ 243)
CYCLES
_______ ] so"°' $2000

G....th
or
Sm>nd"Yt~t
of
Spacts [qu.11 in
Pt()(IOr11on to
Absolylq GrOwlh
}
Crowth 1>t $2000
orJOO• t(
2
------- _________ fiutYur
2nd Year
FtG. '1· EQ.uAr. A\co1:1'-rs Ol" GRO\li.'T11 - A i.i.rn1:-.rrr1c ScA.r..x
S<tlt:s nf a ll)'f.Kl1.l1e1icnl busin t~'i t>1ganizatior1 .sho,o,1ing constant t1n101int of
grol\'t.h. Arill1metic icale.
800.---,---.---.---,.----.----.-----.--~-~
400
TREND
------
"
!!200
8
50
25
190S 1915 192S 19:15 1945 1955
Fie. 5. T'RJ)~o OF A H YJ"()TllETICAL H -.;-SINF_,s OP.GA~rlATtON
Data 1905- •!115· '"iLh n pr•1jt:c tion to J95,;. ·r'J\e projccti<;r1 is l>aM!d on the
assump1in11 of a conliuuation in 1hc COt)StilDt de;tJioe of the rntc of gr,>v.·1Jl, u
discussccl i1• Chapeer 1. Ratio sea~.
APl'F.l'<DIX I
Further tc> c.ornpare the difference between arithmetic and ratio

scales, the sales o{ Lhe hypoLhetic.al business discussed in Chapter 1
arc plot.tcd on a ratio sc:;ile iu .l'ig. 5 on the preceding page.
eoo~-~-~- - 1 --·-,- --·.-----,.--,...,---.--..,,,
A /
SOD
r/-';:."'::,-;7
TREND & CYCLE.
300 __/ I
200
_, I
Flc;. 6. 'T"t-tE TREND AND THT. Cv<:r:e

Treft<"l of a hypo1hctical busjness orga11izati<"10, a:> shr>'"-'TI in Fig. 5, 'Aiith a
rt:gul:ir cycle of. 20 per cent a1n1>litude su1,erinlp()\it(f. Ratio scale. Noce that
\\•lt<:n one t1ses ratio scale, c.J1e v.·avc li"s c<1t•al absolute n1agnilude tll:rougl1out.
One advantage of such a plotting is that it shows directly the
falling off in the rate of growt11, and permits making a visual or
freehand projection into the future that is much more likely to be
fulfilled than any similar projecting of a chart. on the arithmetic
scale.
Ratio scales are usually labelei:I as such. Jf not, they can be recog-
nized from Lhc fact that the r1umhers on the scale come closer and
closer togetlier as they grow larger, as in Fig. 2 and Fig. 3 above;
or else, for equal intervals, the numbers get proportionately larger
and larger, as in Fig. 5.
[245]
CYCLES
Index
150,-- -- . - - - -.-------,.--- ----,
100 INDUSTRIAL
PRODUCTION
50
15
10
5'--- - -''----.....J"---.....J"---.....J
1880 1890 1900 J9tt 1920 1930 1940
1920
1•1c;. 7. GRO\\'Ttr or f '1DUs'r1t1A1. PR.Ooucnos 1>-1 11-tll UN1Ta> STA1'l'S
Data •SSf- 1937. together w;lll trend. (Aftor Davis.) 'll•c chllrt bas bttn
split into two 1>art' ro t!mphasizc tht' cha11ging: charucrcr o[ the trend. Rtltio
scale.
On ratio scale nil boouis and (1cprC.'isicuts are- sllO\vl\ in their trvc relative
proportions.
Compare this cl1:1rt ~·itli Chapter l, Fig. 3.
The interval fro m one number- let us say 1 - on the vertical

scale of a ratio chart to the number ten 1imcs higher, in this in-
stance 10, is called a "cycle." The same dis1ance is used to repre-
sent the next 1enfolll growth, Crom 10 to 100, and so on. In reading
a curve plotted on ra1io scale. one should alwa)'S look at the vcrti·
cal scale to see how many cycles are involved.
So that the rl'rtder may make additional comparisons of data
charted on ratio and on arithmetic scales, th e <lata of the rest of the
charts in Chapte r 1 have been plotted on raLio scales. and are given
in Figs. 6, 7. and 8.
All the char(j in Chapters Ill and IV and many o f the charts in
the rest of the book arc plotted on ratio scale.
[ 246 ]
A P P£~ DIX I
Index
250
SPANISH TRADE
ISO
""'
8. Tfl.'D£x ov SPAl\'ISH TRADr;

}"'1e.
Data 1530-1650., (After Davis and f-iamilt.on), together \fitt• tren.t l Ratio
uole.
(;()mparc this ch.rt "itb Ohapter 1, Fig. 4.
APPENDIX II
Mo·oing Ai1erages
I
arithmetic 1noving average is a flexible mathemati-
HF. s r:-rPLF.
T cal methocl of srnooth ing data. The data so sn1oothed are often
used a.s a trend.
)Moving averages may be of any number of tenns desired. The
proc<:ss of computing a simple aritl1mccic moving average is illus-
trated in cl1e following table:
C.OI\.1PU'f ..\TI0N Ol;· A 3 "YF..1\R :\IOVING ..'\\.'ERi\CE 01,. AN l'.\lDEX OF
WHOLESALE PRICES, 1931-1939

(A) (B) (C) (TJ)
Yef!r Index 3-year m.omng ; -yeflr moving
total of index, average of in-
centered dex, t·e1ltered.
(The data in
Col.C+J)
1931 '13
1932 65 201 68
1933 66 .2(l6 69
1934 75 221 74
1935 80 236 79
1936 81 247 82
1937 86 246 82
1938 79 2(12 81
1939 77
[248]
73+65 + 66
The first average in Column D is = 68. 'fhe sec-
3
'lnd averao·c is
. 65 + tl6 + 75 69. And so on.
"' 3
The rnoving avenige figure is posted and plotted against the niid·
die of the i11·011p of items being avl'.raged "for the same reason that
figures referring to a whole period are custon1ari ly plotted on a
(:hart in the middle of the appropriate spaces," as Croxton and
Cowden explain.
A 5-year moving average would be computed the same way, using
. .
live terms instead of three. Thus:
73 + 65 + 66 75 + 80 + 72.
5
65 + 66 + 75 -1- So·+ 81
And = 73·
0
'I n using an cvt:n number of terms it is impossible to gc:t an exact
centering. It is therefore usual to compute, in addition, a 2-year
11l<>vir1g; av(:t·agc of t~1e r·esult.. Tl1i~ J>roc.:e<ll1re is sf1<l\vt1 in the ta l))e
on the following page.
CYCLES
CO.\tJ"l..'TATION OP A 4 ·Y.t:AR t.tOVINC A\"E'R.AC l~ OF AN INDEX OF
wuor.FSAl.E PRICES IN T!tE u. s. A. 1931- 1939:

(A) (B) (C) (D) (1') (P)
Year Index 4·yeor mov- 4·)•car mov- 2-year mov- 2wyear 1110'tJ•
ing total of ing aver«[{C iug total of i11g average
i11dex1 ccn· centered. f/1e 4·ycar of tht:
tered (The data moving 4·)'Cllf
in Col. C avnage. moving
+ 4) average.
(T he data
in C:ol. 1·:
+2)
66 71
1934 75 74
1$J35 80
So
162
3~6
1 9.~7 86 163 81
1939 77
This method of computation p ermir.s of c~xact centering.

Jn practice it is usual to take 2-year 1110\'ing tor.als of the 4-year
moving tot•ls, center 1.hem, and divide hy eight.
APPENDIX IJ
The followin)?; chart tells what happens to a regular cycle when

the curve of which it is a component is smoothed hy moving
ai.:e1-ag·es:
ld:esl Ntne·Ye.ar Rh·ythm .

.4
Throc-Ycor Mo11ingAvcro11Q:e of in !Cleal Nine·"car Rh)1hm.

[.ff~c;:t: L~~~r>o<I AMplitu\Sc;:, B,,.! No Cho tigo in Ungt11 of \\',.•·11 ,
f,f•te·Yesr 11.~ing Average of an Ideal Nine· Year Rhythm.

Effe<:t: LC$Stnc0 Amplitude, But No Chanae In Lenitll of Wov.i.
c
Seven-Year l!o-.·lng A\!e1oge of en Ideal Nln~'tear Rhythm.
Ettect: lessened AmplitOOo. But No Chsngie In Leneth of Wave..
D
fline-Yc~ !Aoving Avcr:i,go of on Ideal Hin~Yo•f Rhythm.

Effect: Pllne·Year Rtiythm (Umlnsted.
E
ft(:, l . .EFJ1JiC'r OF 1'fOVJKG A \'J:RACE, or, DlFFER.E.N"l' LE.NG"11iS t:l'ON AN ID~A L

g~Yt;A..R RHY'l"' fil(
It is C\:i<lcr1t that ;i 1110\: ir1g ;1\•cragc J1as rlo cffcf.~t c>n tl1c f1eriod Ol'
lt:ngch of the rhythm in the seri.es being avcragccl. Tlut it. doe~ have
an effect upon the umplitttde of the waves.
()rle 11la~· gc11crali1.c 1->y ~ayir)g tl1at aJl)'. tn(>\ i.r1g a\'Crage \\'itll. a 1
leni:th less than the period of a rhythm diminishes the amplitude

of the rl1ythm . 'fhe more nearly the length of the moving average
approached 1.he period of the rhythm, the more nearly it removes
it. \Vhen the length of the moving average equah the period o[ I.he
rhythm it completely removes it.
On the other hand, when one computes the deviations of a c:urn:
evidencing a rhythm frotn a moving average of that curve, the
length of the rhytlun is unaffcucd, regardless of tl1e length of the
1noving average.
CYCLES
This facL is illustl'ated in Fig. 2.
A
Dtvl•flon.a of an Ideal Ntn•Ytar Rtiycllm from its Frte--Year Morini Averq~.
[lf"t Sm•ll Amplitude, but No Ct11nae in leflzttl 01 Wave,
B
Ocvlollono ol an Ideal Nln•·Y•tf Rh}l11!m frt1111 ilt Sf\11en·Ye1r t.IO'<IRIC. A11er1s:e.
Efftct: Lcisse.ned Ampli111c1e, bul f'Co Cl1111nge in lenztfl ot Wl\'t,
c
OtYbliom of an hf~ NN-Ylllf Rhytlrle hoin h:a Hin~Ya-at Mo¥llla AV'tf•I-•·
[ffld: Ille ()rieiaal Millc-YsH ltft,uu:n R~rs.
Fie. 2'. n1~\llA'l'JONS OF .\N IDJ!.AL. 9·\"Jt,\lt KH,'TH'.\t YR0\1 t.-10\'lNC A'\tl:RAC.l!S Ott
DlrJ'UU::HT' l.T..KC11LS
Howe\'l,;r, a' can easily lie >een. 1he a111pli1ude is ,·ery much af·
fectc<l hy 1he lengtlt o[ the moving average used.
APPENDIX I I I
The Sect£on frloving' Average
HE SEcno" moving average, like the simple moving average

T dt:scribed in App<~ndix TT, is a matbernariral device for remov.
ir1g irregti!aritie~ i11 ti1rtc serie.s.
Unlike we woving average, however. it doe~ not average consecu.
th·c terms or figures of a series. \·\ 'hat it does do is to average two
01 ffi()fC COTI5t-"Cllti\·e secticJllS OT C)'Cles.
For example. imagi11c a series of figures represenr.ing the monthly
sales for ten years of some husiness with a sca;ona l par.tern (such as
the ice-cre<im h11•iness). In 1.hc case selected. 1.lic cycle, being sea-
sonal, is twelve monLh' in length. The behavior for each year is
inRucnced by iL> seawnal pauern, plus accidcnwl distortions, cydcs
of other thah 12 months in lenb'th, and trend.
Suppase we wish Lo c·akulate a three 12-month section moving
avcr.ige of these fi!,'llres. \\'c take the first, second. and third Janu-
arys and average t11tm; tht' first, second, a nil third Februarys and
a\•erage the111; •t•>cl .'>O on. 1nontl1 b;• r1•<>nth, \111til '\·\IC 11:1\•e a\'Craged
all th<: months of tl1c nr\t three years. This avl'rage o( the fil'st. tfiree
12-mon th sections is 1he first sectio11 of che thrc.: 1 ~-month section
moving average. IL is plouecl in che second Y""'·
l\cxc we average the second, third, and fourth J:u1uarys, che sec-
ond, third, and Courth Fehruarys, and so on, until we have avcrai:ed
rhc values for all months in che second, t11ird, and fourth years. This
r2.sa ]
CYCLES
gives us our second section of the three 12-month section moving

average. \Ve plot these values in the rnidclle of the span being aver-
aged, namely, the third year.
\Ve proceed in this way unt.i: each possible consecutive group of
three years has been averaged.
The effect of such a series of calculations is to minimize any
oE the month-t.o-month irregularities in mud1 the same way that
a 3-monlh moving average would minimize t.hem. That is, hom the
scaridpoirtt of ;:1(:ci<lc11tal. \:ariacior1s frorr1 tl1e true C}'clc·, it is t.inim-
ponant whether or not we smooth the curve by averaging three
C(>11:sect1tive. r1tl1r1l1<~rs or 1->y ;l\'c1·agi11g tl1ree r1l11r1l;crs t\vt;l,,.e figltres
apart. In either case. minus distortions will tend to offset plus dis-
tortions, and accidental variations of one sort or another will tend
to be spread over 1.hree months ratber than be concentrated in the
month in which they occur.
This technique is frequently ul;{:d to rev(:al and determine a sea-
sonal variation that changes over a course of Lime. Suppose, for ex·
ample, that t11e seasonal variations have dec1·eased percentage-wise
with the increased size oE t1H: business. Then any average: seasonal
variation determinc<l for the entire life of the business will fail to
express Lhe variation that LOok place in the earlier yc;ars, ancl will
exceecl tl1e \'t1riatio11 no'"' prc:sc11t irt tl1e lJt1sir1<:ss figt1res. r\ sectio11
rll<>,,.i11g· a\'Crag;e \viJl en;1ble <ltle lt:> co1n_pute it <">Tl a \ ar·yir1g itistcacl
1
of a fixed basis.
This 1.ed1ni<1uc serve.\ al-10 as a powerful tool in co1utc:(:tion with
r.:}•cle a11al~:sis.
First it. is useful as a smoothing technique.
Second. it is useful because - even if the length chosen fo1· the
sect.ion moving average is different, from the length of. the cycle -
this fact has no influence upon the length of the cycle disclosed in
the resultant. Thus, in the example chosen above, if we had chopped
all .series of figures into se(:tions thineL·n rnonths long. averaging
tlie first. January with the: sec.one! February and the third !\-larch,
and so on, the cyde disdosed by this three 13-inonth section
[ ~54]
APPENDIX Ill
moving average would still show a 12-month period. Likewise, if

we had chopped the curves into 11-month sections, the resultant
pattern would not have hcen eleven months in length, but would
have been twelve monchs as it was in the original series. There
would, however, in these iustanccs have been a cercain dampening
of the amplitude of t he cycle.
\-Ve may ge11er;ili1.e by saying that the length of the section moving
average has no effect upon the length of any real cycle present in
the original data, but that, as the length of the section n1oving aver·
age departs from the true length of the cycle, the tLmplitudc of the
cycle in the resultant average becomes Jes.~. until eventuall)' it dis-
appears entirely.
The third useful ch<irac;teristic of the section moving average
is the fact that when a series is simultaneously influenced by two
rhythmic forces, and is long enough, a section moving average will
<:omplctely eliminate cit.her rhythm, no 1natter how close the length
of rhythms i1n•olvcd, kavi11g the other rhythm undisturbed.
For exampk, suppose a series simultaneously influenced hy two
regular cyclic forces, one of eleven months in length and the other
of twelve months. By t;iking 1111 11-montlt section moving average
of enough terms, the 12-mont.h wave will vanish, leaving the 11-
month wave undisturbed. Simil;irly, " 12-monr.h section n10ving
average o[ enough terms will eliminate the 11-month wave.
[ 255]

Cycles The Science of Prediction

Uploaded by

Copyright:

Available Formats

Cycles The Science of Prediction

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cycles The Science of Prediction

Uploaded by

Copyright:

Available Formats

This volume belongs to set nun1ber _C/dt.

f>.y EO\VARO R. DF.,VEY

o.11tl ED\VIN F. J) AKIN

I l,Vhy Trenrl.< A re Important 1

II P1ttterr1s i11 Growth of Orgariisms I()

111 The Growth Tr~rtd iu Our Basic Industries

IX The z 8· Year Rhyr.hm

XI Anal)•sis a11d Syntltesis 160

XII 1"iming a Br~si11ess 172

XIV fVar and I ts Dislocations 200

C llA .. ' 1 t : R Ill

1. Population of Co>1ti11e11tal U. ~'., 1$;0-1940

1.A /Jundonu of Atlantic Salmnn 55

C11APTEI. ' ' I

6. B11iilrli11g Con.1tr·1i.ctiori i11 J{a111f>r1.r g J :.?:!

1. Rrsi1ltnlial Building 136

ct-rA .. 'l'l'.R. XII

9. Sale:,· of a11 I 1idtcsr.1·itTl Com/><1rf'y. 1871- r9,10 1S•

i. J¥ar~1·i111e Distor1io11 205

1. U.S. Pate111> lssued,z8;7-z941

·r 15 1hc business of science co predict.. An exact science like

The debt of the authors to those whose names, cqu:it.ions and

probahilities is a vital pan of all our rnodcrn scien tific progress. As

ca n henceforth predict the course of life for humanity would be

Why Trends Are Important

ut> FACI1i of growt.h are common knowledge to most mothers.

Suppose we draw sue.It a trend for a hypothet.ical business organ-

Offh<ind this looks like a business that has been expanding

In other words, we see that by 1955 the momentum will cease

flr.. 1. T• ..~.._D OF A HYwrrttETtCAL Rusna:" Oac.,~17-"-TlOf'I.

nation. If a similar tTend ~hows up simultaneously in a number of

1915 1925 1935 1945 1955

1:1c. 2. TtlE TRF.'.\1> :\'.'<I> 7HT: Cvr.tl!

it \vas better. nusitlt~.ss, (:\'er\ 111 <lcpressions, '"'as

1920 1930 1940

Frc.!· G'RO\'/T11 nr -t~ousnuAL t•ioouCT•OK t~ Tl~ l1si-rw STATF.S

frc. 4. I~vli:X. OF SPA.NlSH TR.-\t1r.

Patter1is in G1·owth <!f Organisn?S

CROWTll TRl!.ND amounts to a pattern; and the pattern fa similar

l:.now ...,-jth fair accuracy where facilities will be in greatest demand

stages; but periodically repeated weighingi; give a rough yet suf·

FIG. 2. GRO\\>JII r~ 'i\'t:.1c11·1· OP A Po~tPKIN (Afttt-r Pearl)

FIG. 4· GRO\\'TH OF A POPUlATIO:i-1 OY }"t.lllT FLIES

the study of hun1an populations, he found the same law of growth

F1c. 5. C¥O\\.TJI OF n1E PoPL'"L1\TJOI\' OP S\\•t.Oi<N (After Pearl)

The similar·ities Pearl n1cntions show up graphically in his charts

FlG. 6. GRO\\'l'H: O.F TH& POl'UL1\.TION OF Tf·(l!.

probabilities on their side. Thus, when we have enough census

The period of time required by a given nation to go UU'ough the

Fie. 7. GROV."'Tf-f O.F TJIE P0.1!'0J.A7TOK· OF FltANCF. (.-1/ter Pearl)

prising the period of open French cont.rol following conquest,

1856 1866 I876 1386 1$96 l!l-06 1916 1926

Fie. 8. GRov.·n1 OF 1'11£ J:-.:01c.i::.-,.;ous NA'fl\'J::. Po1•t•u·110N

apparently effect, through a reorganization of its resources, a ch<ingc

Frc. 9. CRO\\'TII oF i.1;11; Port:LATION oF GERJ1.fANV

surge was dearly in evidence l>y 1870, and suhscciuent census

It is possible to clraw another conclusion from which Pearl ab·

implications. 'Ve are going to discover that patterns of growth in

nomic and social upheavals being experkncecl hy our own genera-

10.__ _.__ _,__ _,... _,__ _,............. _._.._....._,.,.,,,..-.....