American Economic Association

Human Development: A New Paradigm or Reinvention of the Wheel?

Author(s): T. N. Srinivasan
Source: The American Economic Review, Vol. 84, No. 2, Papers and Proceedings of the Hundred
and Sixth Annual Meeting of the American Economic Association (May, 1994), pp. 238-243
Published by: American Economic Association
Stable URL: http://www.jstor.org/stable/2117836
Accessed: 28-04-2015 11:21 UTC

REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/2117836?seq=1&cid=pdf-reference#references_tab_contents

You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

American Economic Association is collaborating with JSTOR to digitize, preserve and extend access to The American Economic
Review.

http://www.jstor.org

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions
 Human Development:
A New Paradigmor Reinventionof the Wheel?

By T. N. SRINIVASAN*

The first Human Development Report eral institutions.Others relatingto military

(HDR) defines human development as de- expenditures, appropriatelyenough for a
noting "both the process of widening peo- publication of the United Nations, echo
ple's choices and the level of their achieved Prophet Isaiah's dictum: "They shall beat
well being" (United Nations Development their swords into ploughshares..." (Isaiah
Programme,1990, p. 9 [emphasisin origi- 2:4). Alas, the United Nations Development
nal]) and disarminglyadmitsthat its notions Programme(henceforth,UNDP) is very un-
-"the primaryobjectiveof developmentis likely to succeed where Isaiah obviouslydid
to benefit people... [and] income is not the not!
sum total of human life"-are not really
new. It claims that it "differsfrom conven- I. Development:Objectivesand
tional approachesto economic growth,hu- PerformanceIndicators
man capital formation, human welfare, or
basic human needs" (p. 10) in viewing The first HDR claims that "economic
growthin income as being necessarybut not growth became the main focus after the
sufficientfor human development,and hu- Second WorldWar-and the growthrate of
man beings as ends rather than as means, per capita GDP became the sole measureof
and as participantsin the developmentpro- development -as GNP became the goal of
cess rather than as beneficiaries.Finally, it developmentin 1950s and 1960s, the ques-
focuses on human choices rather than on tion of promotingindividualwell-being re-
the provision of goods and services that ceded-in time, distribution was altogether
deprivedgroupsneed. forgotten, and the argument of "trickle
The report also introduced the human- down" was made to defend such neglect"
development index (HDI) which it claims (UNDP, 1990 p. 104 [emphasisadded]).
"capturesthe three essential componentsof In fact, income was never even the pri-
humanlife: longevity,knowledge,and basic mary, let alone the sole, measure of devel-
incomefor decent livingstandard.Longevity opment,not only in the mindsof economists
and knowledge refer to the formation of but, more importantly,amongpolicymakers.
human capabilities,and income is a proxy N. S. Buchananand H. S. Ellis (1955), au-
measure for the choices people have in thors of one of the early postwar mono-
puttingtheir capabilitiesto use." HDI is the graphs on economic development, divided
equally weighted sum of deprivationof a statisticalindexes of developmentinto two
countrywith respect to each of three com- groups, the first encompassingthe quality
ponents: life expectancy at birth, literacy, and texture of life as "end product," and
and real income per head. the second portraying economic perfor-
Many of the policy recommendationsin mance and explaining(or at least correlat-
the HDR's overlap those of other multilat- ing with) life as end product. Life ex-
pectancy at birth, infant mortality,general
health indicators, food energy intake
(calories per day per person), and the like
were in the first. A country experiencing
*
Department of Economics, Yale University, Eco- sustainedimprovementin indexes of life as
nomic Growth Center, 27 Hillhouse Avenue, New end productwas viewed as achievingdevel-
Haven, Connecticut, CT 06520-8269. opment.
238

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions
VOL. 84 NO. 2 HUMAN DEVELOPMENT 239

Arthur Lewis (1955 pp. 420-21), valued For monitoringprogressthe committee set
growthprimarilyfor its instrumentalrole in targets in terms of food energy intake
promoting human development:"The ad- (2400-2800 Kcal per day per adult worker),
vantage of economic growth is not that clothing, shelter, and eight other indexes
wealth increases but it increases the range of which included life expectancyat birth and
human choice-the case for economic literacy.
growth is that it gives man greater control AmartyaSen's (1992) distinctionbetween
over his environment,and thereby increases "capabilities"and "functionings"presum-
his freedom-economic growthalso gives us ably is the rationalebehind HDR's distinc-
freedom to choose greaterleisure"[empha- tion between the formationof humancapa-
sis added]. Lewis was emphatic that from bilities and the use people make of their
the process of income growth: acquired capabilities. Functionings are
"states of being and doings"such as "being
... woman gains freedom from healthy,"and "avoidingprematuremortal-
drudgery, is emancipated from the ity." The capabilityset is the set of vectors
seclusion of the household, and gains of functioningsavailable to a person. Sen
at last the chance to be a full human persuasivelyargues that capabilityto func-
being, exercising her mind and her tion is closest to the notion of standardof
talents in the same way as men.... It is
open to men to debate whether eco- living.
nomicprogressis good for men or not, The importof Sen's appealingarguments
but for women to debate the desirabil- for comparativedevelopmentanalysisis far
ity of economic growth is to debate from clear. Indeed, as Robert Sugden(1993
whether women should have the p. 1953) asks:
chance to cease to be beasts of bur-
den, and to join the humanrace. Given the rich array of functionings
[p. 422] that Sen takes to be relevant, given
the extent of disagreementamongrea-
Policymakerswere also clear about the sonable people about the nature of
instrumental role of income growth and the good life, and given the unresolved
aware of the importanceof its appropriate problem of how to value sets, it is
distribution.The approachadopted in 1938 natural to ask how far Sen's frame-
by the IndianNational PlanningCommittee work is operational. Is it a realistic
alternative to the methods on which
under the chairmanshipof future Prime economists typically rely-measure-
Minister Nehru is one among many exam- ments of real income, and the kind of
ples of such awareness: practicalcost-benefit analysiswhich is
grounded in Marshallian consumer
Obviouslywe could not consider any theory?
problem,muchless plan,withoutsome
definite aim and social objective.That
aim was declared to be to insure an Sen's argumentthat varyingimportanceof
adequate standard of living for the different capabilities in the capability
masses. In other words, to get rid of the
appalling poverty of the people.... frameworkis analogousto the varyingvalue
There was lack of food, of clothing,of of differentcommoditiesin the real-income
housing, and of every other essential frameworkis not an adequate response.As
requirementof human existence. To Sugdencorrectlypoints out:
remove this lack and insure an irre-
ducible minimumstandardfor every- The real-income frameworkincludes
body, the national income had to be an operational metric for weighing
greatly increased, and in addition to commodities-the metric of exchange
this increased production there had value. Similarly,Marshallianconsumer
to be a more equitable distribution of theory, combined with the Kaldor-
wealth. Hicks compensationtest, provides an
[J. Nehru, 1946p. 399] operational metric for cost-benefit

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions
240 AEA PAPERS AND PROCEEDINGS MAY 1994

analysis.It remainsto be seen whether international human-sufferingindex, compris-

analogous metrics can be developed ing 10 measures of human welfare, includ-
for the capabilityapproach. ing income, infant mortality,nutrition,adult
[Sugden,1993p. 154] literacy and personal freedom, was already
available (Sharon Camp and J. Joseph
The only conceptually appropriatemet- Speidel, 1987).
rics for valuingfunctioningsand capabilities The interesting issue is not conceptual
have to be personalized prices or values, but empirical:do other indicators besides
namely, sets of values that are specific to income provide additional information in
the situation, location, time, and state of assessing development performance?They
nature. These will vary acrossindividualsin indeed do if they add significantlyto the
different circumstancesbut will remain the explanatorypower of income in accounting
same for all individualsin the same circum- for the variance across countries in some
stance, so that they are not subjectiveand independentempiricalmeasureof develop-
individual-preference-based. ment performance.Since there is no univer-
Even grantingthat "humandevelopment sally agreed upon empirical measure, this
goes beyond the choices that the HDI cap- approachis not operational.
tures" (UNDP, 1993, p. 104), whether the Alternatively,differentindicatorscouldbe
HDI is an internationally comparable "mea- measuring different, possibly related, as-
sure of people's abilityto live a healthylife, pects of developmentperformance.If they
to communicateand participatein the life are mutually orthogonal, then each mea-
of the community and to have sufficient sures an aspect of developmentunrelatedto
resourcesto obtain a decent living"(UNDP, that measured by any other. At the other
1993p. 104) is arguable.The componentsof extreme, if they are perfectly correlated with
HDI, namely, life expectancy and educa- each other, then all indicatorsmeasure the
tional attainment,are "functionings"in the same aspect. In principal-componentanaly-
Sen sense but their relativevalues need not sis, one transformsa given set of indicators
be the same across individuals,countries, into an equal numberof mutuallyorthogo-
and socioeconomicgroups.Besides, the "in- nal linear combinationsand computes the
trinsic" value of a single "functioning," proportion of the (generalized) variance
namely, ability to live a healthy life, is not among the indicators explained by each
captured by its linear deprivationmeasure combination.If combinations,far fewer in
in HDI, since a unit decrease in the depri- number than in the set, explain a large
vation in life expectancy at an initial life proportionof the variance,then the infor-
expectancyof, say, 40 years is not commen- mation contained in the set is largely cap-
surate with the same unit decrease at 60 tured by these few combinations.However,
years. interpretingeach such linear combination
as some conceptuallyplausiblefacet of de-
II. HumanDevelopmentIndexand Its velopmentis not simple.
Data Base UNDP (1993 p. 109) reports correlations
between 0.729 and 0.871 among the three
The conceptual foundations of the HDI components of HDI. Further, an almost
thus cannot be found in Sen's (1992) capa- equallyweighted linear combinationof the
bility approach to standard of living and three accountsfor 88 percent of the gener-
poverty.However,inclusionof other indica- alized variance among them. But this find-
tors besides real income growth for moni- ing, in and of itself, says nothingaboutwhat
toring and comparingdevelopmentperfor- aspects of developmentare being portrayed
mance could be rationalized on other by the combination.
grounds.In fact, the multifacetednature of The weaknessof the data on each of the
the development process was well under- componentsof HDI can hardlybe overem-
stood by economists and policymakers,and phasized. GNP data of many developing
a single index of "living conditions," the countries suffer from problems of incom-

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions
VOL. 84 NO. 2 HUMAN DEVELOPMENT 241

plete coverage,measurementerrors,and bi- surely such differencesreflect differencesin

ases. Furthermore,the severityand quanti- individual choices and the socioeconomic-
tative significance of these problems are political processes that result in some indi-
likelyto varyover time within a countryand viduals or groupsin a society being poor or
are not comparable across countries suffering from ill health while others are
(Srinivasan,1994). HDR's have converted rich and healthy.Have the HDR's shed any
domestic-pricesGNP of 173 countries into valuable light on these? Also, have coun-
internationaldollarsusingpurchasing-power tries with low HDI values instituted policy
parity (PPP) exchange rates that were put changes with respect to nutrition, health,
together by Robert Summers and Alan and educationsectors?
Heston (1991), who use problematicproce- Takingthe second question first, there is
dures of extrapolationfrom data for a few no evidence that HDR's have led countries
years and countriesto many more. to rethink their policies, nor is there any
Relativelyreliable and recent (i.e., a ref- convincingreason to expect it to happen. It
erence period of 1980 or later) data for was widelyknown,long before the firstHDR
estimatinglife expectancyat birth are not in 1990, that in spite of her low per capita
availablefor as many as 87 out of 117 less real income Sri Lanka'sachievementsin life
developedcountries(Joseph Chamie,1994). expectancyand literacywere outstanding,in
As the United Nations Children's Fund comparison not only with neighbors, but
(1993 p. 8) candidly admits, many of the also with countries(developedand develop-
statisticsused for under-fivedeaths are esti- ing) with substantiallyhigher per capita in-
mates for 1991based on mathematicalmod- comes. This knowledgedid not demonstra-
els, rather than on recent measurementsat bly lead other countries to learn from Sri
the nationallevel. Lanka's experience. An even more telling
Different countries define literacy differ- exampleis that of the Indianstate of Kerala
ently. Besides, Jere Behrman and Mark with its substantiallylower rates of infant
Rosenzweig(1994)point out that for 19 of a and child mortalityand higherrates of liter-
total of 145 (includingdeveloped)countries, acy in comparisonto other states, including
there are no data on adult literacy since Punjabwith more than twice Kerala'sreal
1970, and for 41 more the latest data relate domestic product per capita. Yet such dis-
to a year in the decade 1970-1979! Also, parities in performance within the same
school enrollmentdata are not internation- country have not led to significantpolicy
ally comparable, since quality of schools, changesin the laggingregions.Surelysocio-
drop-outrates, length of school year, and so economic-political processes, rather than
forth vary substantiallybetween and within low levels of income and lack of knowledge
countries. about the feasibilityof achievingsubstantial
In sum, the HDI is conceptually weak improvements,precludedthe policychanges
and empiricallyunsound, involvingserious needed to bring about improvements.
problemsof noncomparability over time and Unfortunately,none of the four HDR's
space, measurement errors, and biases. address in depth the political economy and
Meaningfulinferences about the process of sociologyof the constraintsthat have so far
development and performance as well as prevented most of the developing world
policy implicationscould hardly be drawn from replicating experiences of the few
from variationsin HDI. countrieswhere successfulhuman develop-
ment has occurred. A short chapter in
III. HDR's and Development Policy UNDP (1991) reports a few interestingex-
amples of limited scope and generalizability
Hardly anyone would choose to be poor, of successful countervailingof forces ar-
and certainly none would choose to be un- rayed against reforms favoring human de-
healthyor to die prematurely!Nonetheless, velopment. It does not offer a deep institu-
if countriesdiffer as they do in their indica- tional analysisbut repeatswell-worncliches
tors of health, mortality, and morbidity, (e.g., "For participation,the remedylies in

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions
242 AEA PAPERS AND PROCEEDINGS AL4Y1994

enpoweringthe poor"; and "Managingpo- immigrationrestrictions,the calculationsare

litical and economic transition is an art just arithmeticand no more. In any case, in
ratherthan a science"[UNDP, 1991 pp. 71, the same global economic environment,
75]). some countrieshave done much better than
While high-soundingand catchy slogans, others, presumablybecause of their supe-
such as "A Global Compact for Human rior policies. These policy differences are
Development" and "Human Adjustment more importantfrom a pragmaticperspec-
Assistance," strewn about in the HDR's, tive than global market distortions which
could attractattentionto a good cause-and the developing countries can hardly influ-
it would indeed be churlishto carp at ideal- ence.
ism and vision-ultimately it is the strength It is said that a "peace dividendof around
and soundness of the analysis and realism $1.5 trillion [can be realized] by the year
of the policy proposalsthat would advance 2000" (UNDP, 1992 p. 9). When many de-
the cause of human development. The veloping countries are actively expanding
HDR's lack in both dimensions. their militaryexpenditureswhether or not
UNDP (1992 p. 5) asserts that "global they have a realistic military threat and,
markets do not operate freely. This, to- what is worse, when civil wars and ethnic
gether with unequal partnership,costs the and religious conflicts are increasing,esti-
developingcountries $50 billion a year-10 mating the size of a peace dividend and
times what they receive in foreign assis- viewingit as "openinga windowof opportu-
tance." Of course, no market, at any level nityfor both richand poor nations"(UNDP,
can be shown to be purely competitive in 1992 p. 85) are luxuries only the affluent
the textbook sense. The term "unequal"in bureaucratsof the UN can afford.
characterizingany voluntary exchange or
partnershipis meaningless.
It is claimedthat "real interestrates have REFERENCES
been four times higher for poor nations
than rich ones. Developing countries effec- Behrman,Jere and Rosenzweig,Mark. "Cross-
tivelypaid 17%a year on their foreign debt Country Data on Education and the La-
during the 1980's, while rich nations paid bor Force." Journal of Development Eco-
only 4%" (UNDP, 1992 p. 48). This differ- nomics, 1994 (forthcoming).
ence multipliedby a debt of $1 trillionyields Buchanan,N. S. and Ellis, H. S. Approaches to
a loss to developingcountriesof $120 billion economic development. New York: Twen-
per year! Another calculationassumes that tieth Century Fund, 1955.
2% of the labor force in the developing Camp, Sharon and Speidel, J. Joseph. The in-
world would emigrate in the absence of ternational human suffering index. Wash-
immigrationcontrols. Multiplyingthe num- ington, DC: Population Crisis Committee,
ber of emigrants by their average wage 1987.
earnings (net of remittancesto their coun- Chamie, Joseph. "Population Databases in
tries of origin)leads to a loss of $250 billion Development Analysis."Joumal of Devel-
per year due to immigrationcontrols. opment Economics, 1994 (forthcoming).
Such calculations cannot be taken seri- Lewis,Arthur.The theory of economic growth.
ously. After all, interest-rate differentials Homewood, IL: Irwin, 1955.
between the poor and rich are ubiquitous Nehru,J. The discovery of India. New York:
withindevelopingcountries,and restrictions John Day, 1946.
on movementof workerswithin developing Sen, Amartya. Inequality reexamined. New
countries are not uncommoneither. There York: Russell Sage Foundation; Cam-
are well-knowneconomicreasonswhy credit bridge: Harvard University Press, 1992.
marketsdo not equilibratethrough adjust- Srinivasan, T. N. "Data Base for Develop-
ments in interest rates. Without a convinc- ment Analysis: An Overview." Journal of
ing economic analysis of interest-rate dif- Development Economics, 1994 (forthcom-
ferentials and socio-political analysis of ing).

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions
VOL. 84 NO. 2 HUMAN DEVELOPMENT 243

Sugden, Robert. "Welfare, Resources, and 1988." Quarterly Journal of Economics,

Capabilities:A Review of InequalityRe- May 1991, 106(2), pp. 327-68.
examined by Amartya Sen." Journal of United Nations Children'sFund. The progress
Economic Literature, December 1993, of nations. New York: UNICEF, 1993.
31(4), pp. 1947-62. United Nations DevelopmentProgramme.Hu-
Summers,Robertand Heston,Alan. "The Penn man development report. New York: Ox-
WorldTable (Mark5): An ExpandedSet ford University Press, various years.
of International Comparisons, 1950-

This content downloaded from 128.192.114.19 on Tue, 28 Apr 2015 11:21:10 UTC
All use subject to JSTOR Terms and Conditions