The 1990 Annual Report of Berkshire Hathaway included the following letter(*) to Wesco

stockholders from the Chairman of the Company.

WESCO FINANCIAL CORPORATION

LETTER TO SHAREHOLDERS

To Our Shareholders:

Wesco has three major subsidiaries, Mutual Savings, in Pasadena, Wesco-Financial Insurance
Company, headquartered in Omaha and currently engaged in the reinsurance business, and
Precision Steel, headquartered in Chicago and engaged in the steel warehousing and specialty
metal products businesses. Consolidated net income for the two years just ended breaks down
as follows (in 000s except for per-share amounts)(1):

Year Ended

December 31, 1990 December 31, 1989

Per Per
Wesco Wesco
Amount Share Amount Share

"Normal" net operating

income of:
Mutual Savings $ 4,099 $ .58 $ 4,191 $ .59
Wesco-Financial Insurance
business 14,924 2.10 14,276 2.00
Precision Steel's
businesses 1,985 .28 2,769 .39
All other "normal" net
operating income(2) 4,030 .56 3,178 .45

25,038 3.52 24,414 3.43

Net gains on sales of
marketable securities 391 .05 5,920 .83

Wesco consolidated net

income $25,429 $3.57 $30,334 $4.26

(1) All figures are net of income taxes.

(2) After deduction of interest and other corporate expenses. Income was from ownership of
the Mutual Savings headquarters office building, primarily leased to outside, interest and
dividend income from cash equivalents and marketable securities owned outside the savings
and loan and insurance subsidiaries, and the electrical equipment manufacturing business,
80%-owned by Wesco since yearend 1988.

This supplementary breakdown of earnings differs somewhat from that used in audited
financial statements which follow standard accounting convention. The supplementary
breakdown is furnished because it is considered useful to shareholders.

(*) A few small sections of Mr. Munger's letter have been excluded: When Berkshire's report exceeds
72 pages, we have problems in binding it. Because of this limitation, either Charlie's letter or mine
had to be cut and I decided a coin flip was appropriate. In fact - as things turned out - I finally decided
nine flips were appropriate. -- W.E.B.
Mutual Savings

Mutual Savings' "normal" net operating income of $4,099,000 in 1990 was almost equal to
the $4,191,000 figure the previous year.

As usual, these "normal-income" figures come from an abnormal savings and loan
association.

Separate balance sheets of Mutual Savings at yearend 1989 and 1990 are set forth at the end
of this annual report. They show (1) total savings accounts declining to $286 million from
$293 million the year before, (2) a very high ratio of shareholders' equity to savings account
liabilities (near the highest for any mature U.S. savings and loan association), (3) a
substantial portion of savings account liabilities offset by cash equivalents and marketable
securities, and (4) a loan portfolio (mostly real estate mortgages) of about $131 million at the
end of 1990, down moderately from $154 million at the end of 1989.

As pointed out in Note 9 to the accompanying financial statements, the book value of Wesco's
equity in Mutual Savings overstates the amount realizable, after taxes, from sale or liquidation
at book value. Wesco would get only about $30.8 million, after paying income taxes, from
the liquidation at book value of the $47 million portion of Mutual Savings' shareholders'
equity which is considered bad debt reserves for income tax purposes. The $4.1 million
Mutual Savings earned in 1990 is an inadequate return (8.7%) on the $47 million amount at
which we try to maintain shareholders' equity, but this same $4.1 million is a respectable
return (13.3%) on the $30.8 million which would be the after-tax proceeds of liquidation at
book value.

The loan portfolio at the end of 1990, although containing almost no risk of loss from
defaults, bore an average interest rate of only 9.20%, probably near the lowest among U.S.
savings and loan associations and roughly the same as the 9.23% rate at the end of 1989.
Because the loan portfolio is almost entirely made up of instruments of short maturity or
bearing interest rates that adjust automatically with the market, there is now much less
unrealized depreciation in the loan portfolio than the net unrealized appreciation in Mutual
Savings' interest-bearing securities and public utility preferred stocks. That appreciation at
December 31, 1990 was about $11 million.

While the "spread" between Mutual Savings' average interest rates paid on savings and
received on loans remains too low provide respectable rofits, this "spread" improved again
last year. The "spread" improved because interest rates paid on savings declined. Moreover,
the disadvantage from inadequate "spread" has been reduced in each recent year by the effect
of various forms of tax-advantaged investment, primarily preferred stock and municipal
bonds. The negative side of this tax-advantage antidote to inadequate interest rate margin on
loans is the risk that preferred stock and municipal bonds, with their fixed yield and long life,
will decline in value, and not provide enough income to cover Mutual Savings' interest and
other costs, if the general level of interest rates should sharply rise. In view of this risk,
Mutual Savings' total commitment has been kept conservative, relative to the amount of its
net worth.

New federal legislation enacted in 1989, widely under the acronym "FIRREA," is now
causing Mutual Savings, step by step, to dispose of the preferred stock portion ($54.4 million,
at cost, at December 31, 1990) of its tax advantaged assets. Ownership of preferred stock has
heretofore helped preserved earning power because tax-equivalent yield is so high (about 15%
at December 31, 1990). Adding to our forced-disposition-of-desirable-assets problem, recent
changes in income tax law now make impracticable the replacement, as they mature, of
Mutual Savings' direct holdings of municipal bonds ($16.9 million, at cost, at December 31,
1990). The municipal bonds also have a high tax-equivalent yield (about 17.5% at December
31, 1990). By mid-1994, and possibly much sooner, we expect virtually all benefit from
tax-advantaged investment to vanish from Mutual Savings.

Mutual Savings remains a "qualified thrift lender" under the old federal regulatory standard
(which ends June 30, 1991) requiring 60% of the assets to be in various housing-related
categories. It will shortly change its asset mix as necessary to comply with a new standard,
imposed by FIRREA, which requires that 70% of assets be maintained in a more restricted list
of housing-related assets.

Until U.S. laws governing financial institutions are further revised, Mutual Savings expects to
keep its required 70% in housing-related assets within the following five categories:

(1) mortgages issued in the course of sale of individual parcels, as Mutual Savings disposes
of foreclosed seaside property in Santa Barbara, California;
(2) directly made, fixed-rate house mortgages with short expected lives;
(3) indirectly made fixed-rate house mortgages with short expected lives, purchase in the
open market in the form of mortgage-backed securities;
(4) a modest amount of directly made, long-term house mortgages with variable interest rates
that fluctuate with the market up to 25% per annum;
(5) a substantial number of directly made, long-term, fixed rate house mortgages given only
to persons of low-to-moderate income, many in minority groups, who have good credit, reside
within seven miles of Mutual Savings' office, and support Mutual Savings' loans with house
equities amounting to at least 20% of house value, with the maximum size of mortgage
permitted being about $191,000.

We will work hard to expand assets in category (5), covering small, long-term, fixed-rate
house mortgages for local people of low-to-moderate income. Indeed this category is
expected to cover a majority in number of all new directly made mortgages. We expect to
impose no loans fees and to charge slightly below-market interest rates. Therefore, each new
loan will cause an immediate economic loss, which will hit our earnings statement even
before we sell the loans, as we plan to do. The loans will be resold, not because they are
inferior credit instruments, but because we do not wish to endure the asset-versus-liability
maturity mismatch imposed by any long-term, fixed-rate mortgage.

FIRREA has increased pressure on both banks and associations to expand lending of the sort
covered by category (5). As a result, in our area there can be no lack of availability in this
category of market-rate loans, meeting legislative objectives, for persons with good credit.
Instead, all lenders face a shortage of qualified applicants. Given this shortage, as we now
complete with bigger, better loan departments of larger institutions, the most efficient way to
get our share of qualifying loans is to quote below-market interest rates and loan charges.

We do not resent making these loss-causing loans. We intend, with pleasure, to make more
than our share, which we can well afford to do. We regret that we waited so long to complete
vigorously for these loans and that we required regulatory prompting before we found a
satisfactory solution of such simplicity. We were formerly brain-blocked, because (1) we
didn't want to hold any long-term, fixed-rate loans, (2) we didn't want to impose on
moderate-income borrowers the risks implicit in the only kind of variable-rate loan we
were willing to make, (3) we had never routinely resold loans or deliberately loaned at a loss,
and (4) we were preoccupied with avoiding calamitous results which came to many other
savings and loan operators. Regulators, of course, have not demanded that we now lend at
a loss. That aspect of our program is the results of our initiative alone.
We have had trouble attracting a significant volume of loans, with satisfactory characteristics,
in category (4), covering our variable-rate loans which can escalate to bear interest rates of
25%. These loans have been in short supply despite our use of a very low interest rate spread
(about 2 percentage points over the one-year U.S. Treasury rate). Moreover, while we have
realized no losses on our variable-rate loans, we have encountered several collection delays,
partly attributable to an incompetent policy decision of the Chairman. These two factors cause
us to expect this category to shrink to minor significance.

Category (3), the short-term, fixed-rate, mortgage-backed security category, is a "last-resort"

category for us. But it could eventually amount to a substantial percentage of assets,
depending on what is available elsewhere.

As we select mortgage-backed securities, we will probably not be buying any complex

instruments. Despite our love of comedy, we are going to avoid the newest form of "Jump Z
tranches in REMICS." This refers to a particular contractual fraction -- the "Z Form" -- of a
pool of mortgages, now subdivided by obliging issuers, advised by obliging investment
bankers, into two new contractual fractions: (1) the "Sticky Jump Z." At this rate,
subdivision will soon get down to quarks.

We are deterred from buying such securities partly by our hatred of complexity. We also
dread the prospect of state and federal examiners, none of whom has a Ph.D. in physics,
reviewing, one after the other, our choices for soundness and billing us on cost-plus basis to
reflect value thus added. Some of the wonders of modern finance go on without us as
we yearn for a lost age when most reasonable people could, with effort, understood what was
going on.

In total, during the next years, our policies will very likely our housing-related assets
(exclusively of the one-time effect of development of our foreclosed seaside property) to
continue to produce close to the lowest average gross return in the savings and loan
industry. Incremental returns may not quite cover incremental interest and operating costs as
we invest each new dollar of savings. It is quite conceivable that Mutual Savings will decline
in size because it should decline in size.

Even so, we expect that Mutual Savings will muddle through in a manner satisfactory to
Wesco shareholders with moderate expectations. Our optimism comes mainly (1) from an
expected minor profit boost from disposition of our foreclosed seaside property and (2) from
an expected major profit boost caused by ownership of our large holding of Freddie Mac
stock. Both of these grounds for optimism are discussed below.

Mutual Savings has a buried value in a piece of foreclosed property: 22 seaside acres in Santa
Barbara, acquired in 1966. By the time Mutual Savings started development (into 20 houses
and 12 lots) in order to facilitate sale, the value of this property had appreciated by at least
$12 million. The built-in appreciation will now be captured through development, assuming
no large reverses caused by collapse of housing prices or unanticipated new regulatory
troubles.

The first house is nearly finished, and about 15 houses are under construction. We expect to
close sale of about half the parcels during the next year. There will be little or no profit added
to built-in appreciation by the development process. Seaside land development, under present
regulatory and market conditions in California, tends to be no-profit activity--if you are lucky.
It is full of queer happenings and closely resembles a Chevy Chase movie of extreme
duration.
In 1988 Mutual Savings made a large and unusual purchase. It increased its holdings of
Federal Home Loan Mortgage Corporation (widely known as "Freddie Mac") to 2,400,000
shares, 4% of total shares outstanding. Mutual Savings' average cost is $29.89 per share,
compared to a price of $48.75 per share in trading on the New York Stock Exchange at the
end of 1990. Thus, based on 1990 yearend trading prices, Mutual Savings had an unrealized
pre-tax profit in Freddie Mac shares of about $45.3 million. At current tax rates the potential
after-tax profit is about $26.7 million, or $3.75 per Wesco share outstanding.

Freddie Mac, created and long run by a federal agency (the Federal Home Loan Bank Board),
is now owned privately, largely by institutional investors. It is now led by a very smart CEO,
Leland Brendsel, and governed by an outstanding independent board of directors, including
John B. McCoy of Banc One and Henry Kaufman, former chief economist of Salomon
Brothers. Freddie Mac supports housing primarily by purchasing housing mortgage loans for
immediate transmutation into mortgage-backed securities that it guarantees and promptly
sells. In the process Freddie Mac earns fees and "spreads" while avoiding most interest-rate-
change risk. This is a much better business than that carried on by most (or indeed most of
the top 10% of) savings and loan associations, as demonstrated by Freddie Mac's high
percentage returns earned on equity capital in recent years. One ironic cause of the high
returns is that this creation of federal regulators pays no deposit-insurance premiums as it
replaces much of the former function of the savings and loan industry. Freddie Mac's high
returns on equity are caused by a strong competitive position that is likely to last a long
time. In its activities it faces only one other competitor of similar size, efficiency and
reputation: Federal National Mortgage Association (widely known as "Fannie Mae"), a
similar private corporation with governmental overtones.

At Freddie Mac's 1990 dividend rate ($1.60 per annum per share), Mutual Savings' pre-tax
yield was only 5.35% on its $29.89 average cost per share. Post-tax, the dividend yield was
only 4.4%, but this amounted to about 75% of the current after-tax yield from very high grade
mortgages. Moreover, Freddie Mac has a creditable history of avoiding really hurtful loan
losses and increasing its earnings and dividend rate, virtues that contribute to increases in the
market price of its stock. Following are figures for 1985-1990:

Year-End
Earnings Dividends Market Price
Year Ended 12/31: per Share per Share per Share

1985 $2.98 $ .53 $ 9.19

1986 3.72 1.13 15.17
1987 4.53 1.10 12.12
1988 5.73 1.25 50.50
1989 7.28 (1) 1.60 67.12
1990 6.90 1.60 (2) 48.75

(TABLE CONTINUED)

Freddie Mac's
Return Earned
on All
Average Equity

1985 30.0%
1986 28.5
1987 28.2
1988 27.5
1989 25.0
1990 20.4

(1) restated
(2) raised to annualized rate of $2.00 per share on March 8, 1991

Despite Freddie Mac's strong competitive position, its stock declined in market value by 27%
in 1990 (from $67.12 per share to $48.75 per share, in trading on the New York Stock
Exchange). One reason for the decline was unanticipated losses from apartment house loans,
particularly in New York and Atlanta. As a result, Freddie Mac wisely discontinued the most
obviously dangerous part of its apartment house loan buying program. But it remains the
guarantor or owner of some old loans (fortunately a small portion of total apartment house
loans and a really tiny portion of total loans) that will create misery for years. It was probably
ill-advised for Freddie Mac, given its position and financial leverage and the nation's needs,
(1) ever to finance anything except owner-occupied, single-family, non-vacation houses, for
which substantial down payments had been made by credit-worthy people, and (2) ever to
deal with anyone other than mortgage originators and services of obvious integrity and
competence. Just as it is unwise for an individual to risk losing what he has and needs in an
effort to gain what he doesn't have and doesn't need, it seems unwise for Freddie Mac to
stretch its leveraged resources beyond purchase from obviously responsible people of
carefully selected first mortgages on individual houses. Each lender, including the one
writing this letter, seems destined to learn through painful, personal experience two obvious
lessons from the past:

(1) The first chance you have to avoid a loss from a foolish loan is by refusing to make it;
there is no second chance.

(2) As you occupy some high-profit niche in a competitive order, you must know how much
of your present prosperity is caused by talents and momentum assuring success in new
activities, and how much merely reflects the good fortune of being in your present niche.

In common experience, including ours, lesson (1) is eventually learned, but lessons (2)
resists learning, despite high pain inflicted by multiple reverses.

As nearly as we can foretell, Freddie Mac's troubles with apartment house loans are endurable
in scale and will no more significantly impair its long-term prospects than the salad oil
swindle of 1963 impaired the long-term prospects of American Express. Moreover, the
present managers and directors of Freddie Mac all seem to have absorbed a catechism
appropriate for Freddie Mac and to be willing to endure political friction burns as necessary to
keep operations sound. We like our large position.

Strangely, Mutual Savings' holdings of Freddie Mac, while lawful to own under FIRREA, (1)
so far do not count as "housing-related assets" in the new 70%-of-assets test, and (2) must be
written down, in stages, to a value of zero for regulatory accounting purposes. As these
provisions start to bind, Mutual Savings will dispose of part of its Freddie Mac stock. One
option is the transfer of stock to another Wesco subsidiary in return for cash.

What future in the savings and loan business do we expect? We don't know anything more
than that we are satisfied at the moment with our temporizing strategy. We expect further
changes, possibly radical, in the bank/savings-and-loan-association field, to which we will
adapt as they unfold.
The present situation, with its many insolvent and almost-insolvent institutions, is such a mess
that further legislation seems inevitable. We can predict neither the changes, nor whether the
changes will make matters better or worse. But we do have some opinions. These opinions
are almost totally out of step with current thinking in academia, among government officials,
among banking executives and, most of all, among banking lobbyists. Despite this
unconventionality, our opinions are now given to Wesco shareholders because they may
provide some insight into our institutional nature and likely future action. We also hope, but
only slightly, that the opinions, set forth below, will have a wider, civic utility.

First, let us turn to banking, after which we will consider the savings and loan business.

The sum of all deposit-insurance losses in banking will probably be much lower than the $200
billion or so recently caused by savings and loan associations. But there are a lot of very sick
banks, and deposit-insurance losses are sure to be large. Moreover, even if there had been no
such losses, there would be much to regret in the nature of our modern banks as they have
increasingly emphasized lending for consumption (even lending at 20% for vacations in
Tahiti) and lending to financial promoters and real estate developers. We have come a long
way from an ideal emphasizing the banker's provision, to both big and small businesses, of
what Pierre DuPont provided to General Motors. Plainly, we have a two-forked banking
problem, with a questionable shift in priorities accompanying rising insolvencies.

Let us attempt to diagnose the causes of our problem. By and large, our problem did not
come because banks couldn't branch across state lines, sell insurance, or underwrite corporate
securities. Instead, it came because banks "reached" for higher yields on assets as they faced
higher interest costs that came from (1) decontrol of interest rates paid by insured institutions
plus (2) pressure from new competitors, including money-market funds possessing a large
competitive edge.

Exactly how great is the money-market funds' competitive edge? To see, compare the
average heavily regulated bank, paying high deposit-insurance premiums, with what has been
created in an extreme form of uninsured money-market fund. In the fiscal year ended June
30, 1990 one such $4 billion fund (The Common Fund for Short Term Investments) did all of
the following:

(1) kept its assets in liquid short-term obligations of the U.S. government and other credit-
worthy entities;

(2) furnished efficient checkwriting privileges and wire transfer service to its depositors;

(3) kept its total operating costs under two-tenths of 1% of deposits per annum as it avoided
costs of maintaining branch offices, deposit insurance, etc.;

(4) furnished no capital of its own as a cushion supporting promises to depositors; and

(5) paid very competitive rates on its interest-bearing accounts, as a result of which it grew
27% in size.

This example demonstrates the raw competitive power of keeping things simple. Indeed, in
this example all costs combined have been controlled so as to be roughly equal to what the
average local bank pays for federal deposit insurance alone. We are not dealing with some
minor competitive advantage. The new competition is a juggernaut.

How important has the new competitor become? Naturally, the competitor has taken a huge
bite out of the market formerly served by banks (and savings and loan associations) burdened
by much higher costs. How could it be otherwise? Here is a dramatic graph reprinted from
what is surely among the best magazines in the world, England's The Economist:

[GRAPH OMITTED: "From nowhere, Amounts outstanding, December, Source: Federal

Reserve"]

The money-market funds are, in substance, "non-bank" banks, furnishing interest-bearing

savings and checking accounts. And, by an odd stroke of good fortune, their light regulation
by an overburdened SEC has turned out to be more advantageous than no regulation at all.
The rules of the SEC force investment largely confined to reasonably safe and liquid
categories. This has spawned simple operations with very low costs.

The simple, low-cost(*), cream-the-market approach thus taken (or stumbled into) often
works well in business. For instance, look at (1) GEICO, a hugely successful auto insurer
almost 50% owned by Wesco's parent corporation or (2) various membership warehouse
clubs, in the form invented by Sol Price, which are now clobbering retailing competitors as
they get total "markup" under 10%. And this approach, as would be expected, is working like
gangbusters for the money-market funds, as you see in the graph from The Economist.

What were the effects on banks as these new and successful, low-cost competitors took more
and more of the market while, at the same time, each bank's banking competitors could bid as
they wished for funds, using the government's credit? Well, naturally, almost every bank,
being inherently saddled with much higher costs, and not wanting to go out of business, tried
to get higher contractual interest rates on its loans. And this caused greater emphasis on loans
for consumption and loans to financial promoters and real estate developers. Indeed, many of
our most decisive bankers, quite logically, stopped trying to make loans to their most credit-
worthy customers, accepting the disappearance of any important linkage between our best
banks and our best businesses. The banks had been forced into an entirely different market
niche (which already had some occupants): high-interest-rate lending.

And what can be expected when virtually all banks become specialists in high-interest-rate
lending? It is hard to know for sure, because, throughout the past, high-interest-rate lending
was hard to fund since it came from skeptical sources, instead of from government-insured
deposits. Really large-scale, high-interest-rate lending is a comparatively recent
phenomenon, made possible by governmental support in the form of deposit insurance used
by banks with altered natures. But such experience as exists gives a likely answer: many
bank insolvencies will come. Just as the simple, low-cost, cream-the-market strategy is a
common business winner, the opposite strategy, involving high costs and high prices, is a
common loser. High interest rate lending as a field has usually provided (1) some winners
and (2) many casualties, often coming in bunches after periods of "follow-the-leader" asset-
quality debasement. (Remember the widespread disasters in R.E.I.T. lending.) And the past
bad experience should naturally worsen as the high-interest-rate lending field both expands
and becomes overcrowded, driven by governmental support.

We are not alone in our diagnosis. Here is an excerpt from a recent Wall Street Journal
editorial: "When more efficient, uninsured and less regulated financial institutions creamed
off profitable lines of business, the [Bank of New England] was left concentrated in
commercial real estate. This artificially diverted money into Boston's building boom, which
inevitably became a bust."

(*) Total costs are low, even though they include fees containing a substantial profit element
that are paid by the "non-bank" banks to the "non-independent" independent managing
companies employed in conformity with mutual fund practice. While Lewis Carroll might
have liked the consistency of the nomenclature just used, it is not clear that it befits a banking
system. "Pretending" under misleading labels is not a good idea in banks. All "pretending"
habits tend to spread.

Granting the presence of perverse incentives, what are the operating mechanics that cause
widespread bad loans (where the higher interest rates do not adequately cover increased risk
of loss) under our present system? After all, the bad lending, while it has a surface
plausibility to bankers under cost pressure, is, by definition, not rational, at least for the
lending banks and the wider civilization. How then does bad lending occur so often?

It occurs (partly) because there are predictable irrationalities among people as social animals.
It is now pretty clear (in experimental social psychology) that people on the horns of a
dilemma, which is where our system has placed our bankers, are extra likely to react unwisely
to the example of other peoples' conduct, now widely called "social proof". So, once some
banker has apparently (but not really) solved his cost-pressure problem by unwise lending, a
considerable amount of imitative "crowd folly", relying on the "social proof", is the natural
consequence. Additional massive irrational lending is caused by "reinforcement" of foolish
behavior, caused by unwise accounting convention in a manner discussed later in this letter.
It is hard to be wise when the messages which drive you are wrong messages provided by a
mal-designed system.

In chemistry, if you mix items that explode in combination, you always get in trouble until
you learn not to allow the mixture. So also, in the American banking system. To us, a lot of
foolish, unproductive lending and many bank insolvencies are the natural consequences,
given existing American banking culture, of the combination of the following two elements
alone:

(1) virtually unlimited deposit insurance; and

(2) uncontrolled interest rates on insured deposits.

These two elements combine to create a Gresham's law effect, in which "bad lending tends to
drive out good." Then, if factor (3) below is added to an already unsound combination, we
think deposit-insurance troubles are sure to be further expanded -- and not by a small amount:

(3) relatively unregulated, non-insured, low-cost "non-bank" banks.

Moreover, when the government starts suffering big deposit-insurance losses, if it

continuously responds (in a natural, unthinking reaction) by raising deposit-insurance prices,
we think it creates a "runaway-feedback" mode and makes its problems worse. This happens
because the government, by adding even more cost pressure on banks, increases the cause of
the troubles it is trying to cure. The price-raising "cure" is the equivalent of trying to
extinguish a fire with kerosene.

Many eminent "experts" would not agree with our notions about systemic irresponsibility
from combining (1) "free-market" pricing of interest rates with (2) government guarantees of
payment. If many eminent "experts" are wrong, how could this happen? Our explanation is
that the "experts" are over-charmed with an admirable, powerful, predictive model, coming
down from Adam Smith. Those discretionary interest rates on deposits have a "free-market"
image, making it easy to conclude, automatically, that the discretionary rates, like other free-
market processes, must be good. Indeed, they are appraised as remaining good even when
combined with governmental deposit insurance, a radical non-free-market element.

Such illogical thinking, displays the standard folly bedeviling the "expert" role in any soft
science: one tends to use only models from one's own segment of a discipline, ignoring or
underweighing others. Furthermore, the more powerful and useful is any model, the more
error it tends to produce through overconfident misuse.

This brings to mind Ben Graham's paradoxical observation that good ideas cause more
investment mischief than bad ideas. He had it right. It is so easy for us all to push a really
good idea to wretched excess, as in the case of the Florida land bubble or the "nifty fifty"
corporate stocks. Then mix in a little "social proof" (from other experts), and brains
(including ours) often turn to mush. It would be nice if great old models never tricked us, but,
alas, "some dreams are not to be." Even Einstein got tricked in his later years.

We may be right or wrong. But, if we are right, if there are deep, structural faults in the
American banking system, it follows that merely giving banks the right to branch across state
lines, to sell insurance, or to enter investment banking (or all of the above) is not going to end
our troubles.

Instead, a good long-term fix can come only after the government considers more extreme
modifications in the system, each of which has powerful, vocal opponents. What are the more
extreme modifications to consider? We think the list includes:

(1) greatly reducing deposit insurance;

(2) eliminating money-market funds;

(3) bringing back some form of controls on interest paid on insured deposits;

(4) intensifying regulatory control of bank lending in an attempt to reduce loan losses;

(5) forcing more conservative accounting covering bank lending;

(6) forcing weak banks into other hands before the weak banks become insolvent; and

(7) forcing insolvent banks into competing local banks, or entirely out of business, instead of
into strong, out-of-state banks.

Let us next attempt a brief discussion of the merits and/or political prospects of each of these
seven government options.

Option (1): greatly reducing deposit insurance:

To many people, remembering former banking panics, this option, adopted fully, seems like
trying to solve the overcrowding problem by bringing back cholera. Accordingly, proponents
of this option typically would limit its effects by (1) bringing back bank "runs" only for small
banks (big banks, regardless of law, are "too big to fail" in all advanced countries) and (2)
bringing back deposit losses only to some rich depositors. Because voters don't like bank
"runs" of any size, and small banks don't like discrimination, it seems unlikely that reductions
in deposit insurance are going to be made on a scale that solves the structural defect problem.
Conceivably, "brokered" deposits could be removed from insurance coverage, in a move
driven by legislative remembrance of many abuses involving stockbroker-assisted financing
of despicable insured institutions. (Many stockbrokers could easily see that the insured
certificates of deposit they were paid to sell were issued by institutions managed by knaves
and fools, presiding over piles of junk loans and junk securities. The stockbrokers thus knew,
or should have known, that their government was being robbed. To sell certificates under
such conditions was a lot like finding currency in a post office bag and deciding it was ethical
to keep it.)
Option (2): eliminating the money-market funds:

This option is almost never discussed. This seems peculiar. The money-market funds came
into being without public policy input when some clever person combined (1) mutual fund
status under the S.E.C. with (2) purchase, under subcontract, of services from a bank. What
was created was, in essence, a virtually unregulated, uninsured bank furnishing interest-
bearing savings and checking accounts. The creation of such entities would probably not
have been authorized if new legislation had been necessary. Where else do we have virtually
identical regulated and unregulated entities operating on the same scale, side by side? If new
legislation had been needed, the following questions might have been raised:

(1) What do money-market funds do for "community" lending, lifetime services to the
elderly, etc.?

(2) Are they fair to existing institutions?

(3) Won't the new "non-bank" banks make it harder for the Federal Reserve System to render
constructive economic service?

(4) Since the public is already on the hook as guarantor of solvency of existing institutions, is
it wise for the guarantor to risk losses from allowing uninsured, cream-the-market, more
efficient operators to add to the competition? (This question would not be hard to answer in a
private setting. If you were guarantor of all obligations of your brother-in-law's hamburger
joint, you would consider it very foolish to allow McDonald's to commence operations by his
side when you possessed the ability to prevent it.)

(5) Considering all of the above (and more), are the money-market funds in the long-term
interest of the soundness and service of the total banking system?

These questions are still good questions. But possession is strength under law. The money-
market genie is now out of the bottle. And, considering his size, it would be hard to put him
back. The prospects of rebottling are plainly remote.

Option (3): bringing back some form of controls on interest paid on insured deposits:

This option, too, is now seldom discussed. Again, this seems peculiar. It is among the first
things you or I would consider if we had to guarantee all obligations of that hamburger joint
owned by a brother-in-law. We would no more guarantee an 11% obligation for him, when
we could easily borrow at 8%, than we would burn currency in the fireplace. In fact, we
would suspect dishonorable "monkey business" if an 11% transaction occurred.

One reason for present lack of legislative interest in interest-rate controls lies in the
knowledge that a former version of such controls constricted housing credit when interest
rates rose to high levels. No one now seems interested in trying to develop new controls,
more flexible in form and practice, that would avoid former defects. Nor is anyone much
interested in the success the Japanese (or the United States) had during a long period of
control of interest rates paid by banks. The interest-rate-control option, at the moment, seems
dead.

Option (4): intensifying regulatory control of bank lending in an attempt to reduce loan
losses:
This option is already being exercised -- erratically -- with effects both good and bad. It
certainly has successful counterparts in non-banking businesses. For instance, take
McDonald's franchised restaurants. If you want to use the McDonald's authenticating name
and arches on your restaurant, you have to operate in a very limited, foolproof way.
Moreover, the McDonald's approach once worked in banking. When deposit insurance first
came in, and long thereafter, most insured banks operated in simple, sound fashion, often
through ill-paid employees. But, based on all recent precedents, the government won't now
act like McDonald's, or itself in a former era. (If it wished to do that, it might now give
deposit insurance to all the simple, sound money-market funds, lending to big business
through purchases of commercial paper, and take deposit insurance away from all the banks
and savings and loan associations) Government, instead, will probably take the more limited
approach of concurrently: (1) leaving banking over-stressed by competition, (2) leaving
banking very complicated, (3) trying to prevent problems by writing massive, hard-to-
understand regulations that create more work for lawyers, and then (4) monitoring bank
operations through overburdened civil servants. These limited remedies may be better than
nothing, but their prospects for causing a real banking fix seem poor. It is almost a general
rule of American life that, when incentives are all wrong, controls (even criminal-law
controls) can't fix our troubles. We can expect limited good effects from Option 4 and the
continuation of important, basic problems.

Option (5): forcing more conservative accounting covering bank lending:

Bank accounting is a hot current topic, but conservatism is not the goal. Everyone is
wondering how much to delay loan write-offs, when loans go sour, so as not to over-correct
weak banks. We are not going to enter the lists on that problem.

The almost-never-discussed problem that interests us is that presented by newly made loans,
bearing high interest rates, that under current bank accounting tend to be treated as "born
good." The result is that all interest accrued, and sometimes some up-front fees, are treated as
fully earned, even though the final outcome of the whole loan transaction is far from clear.
To us, this is counterproductive accounting, even though we use it ourselves when pushed by
convention.

We think current accounting for many high-interest-rate loans has terrible consequences in the
banking system. In essence, it "front ends" into reported income revenues that would have
been deferred until much later, after risky bets were more clearly won, if more conservative
accounting had been employed. This practice turns many a banker into a human version of
one of B.F. Skinner's pigeons, since he is "reinforced" into continuing and expanding bad
lending through the pleasure of seeing good figures in the short term. The good figures
substitute nicely in the mind for nonexistent underlying institutional good, partly through the
process, originally demonstrated by Pavlov, wherein we respond to a mere association
because it has usually portended a reality that would make the response correct.

Under prevailing accounting, banks now ordinarily report increases in both earnings and
equity capital during any transition they make toward less conservative lending. And then, if
more lending of that type is done, and is accompanied by growth in institutional size, good
reported figures will continue for an additional period. If an increase in institutional size is
deemed necessary, it is, of course, assured by the bank's access to the government's credit
through deposit insurance.

We think acculturated corporate nature, in American financial institutions, simply cannot, on

average, handle temptations implicit in this sort of accounting. Indeed, the succumbing to the
temptations, in a manner not consistent with long-term institutional interest, often occurs
through a subconscious process. The subconscious process includes bad effects from both (1)
"social proof", and (2) a "reality-denial" mode that creates bias in people stimulated, honored
and paid in proportion to institutional size. Under our present system a Columbia Savings,
and many less obscene versions of its model, are almost inevitable.

Of course, a large minority, even a majority, of bankers will remain sound, despite the
temptations. But this outcome is not sufficient to protect the deposit insurer from
unacceptable ultimate losses. In due course, given present conditions, the deposit insurer will
suffer from what some wag called the problem of there being so many more banks than
bankers.

What should now be considered are mandatory accounting changes, including changes in
accounting to shareholders, designed to force "back-ending" into reported income of revenue
from various types of gamy lending (and letters of credit), in lieu of allowing "front-ending"
to continue. The changes would cause American bank accounting, by fiat, to imitate what
some of the best European bankers have long done by choice. Eventually, credibility might be
returned to banks' audited financial statements, now often regarded as fairy tales.

Despite the obvious (to us) accounting defects that bedevil our system, we don't think any
wise and important accounting changes will be made. Typical bank reaction to such proposals
is, at best, that of the man who asked, well before his ultimate sainthhood: "God, give me
chastity, but not yet." Also, time periods for accomplishing even the simplest, "no-brainer"
changes in accounting convention tend to stretch into years.

Option (6): forcing weak banks into other hands before the weak banks become insolvent:

This option is also a hot topic. Usual governmental practice at the moment is to force merger
only when all shareholders' equity is gone and the deposit insurer has a large loss. This is
"bonkers," due process gone mad. It seems entirely logical now to commence the forced
merger or closure of many of the nation's 13,000 banks and to do it in many cases before a
weak bank is insolvent. Because the need is so obvious, laws and customs may possibly
change to cause more of this to happen. And interstate branching may be allowed in order to
enlarge the number of potential bank buyers.

While these steps seem helpful, they won't fix the problem of deep structural fault in the
system -- at least within any acceptable time period. Look at the present carnage in airlines.
Even when we are down to fewer than a dozen significant operators, messy airline failures
continue. If we wait for an airline-style solution in banking, we will have to endure years,
maybe decades, of suffering.

Option (7): forcing insolvent banks into competing local banks, or entirely out of business,
instead of into strong out-of-state banks:

According to Martin Mayer, writing recently in The Wall Street Journal, the FDIC now
typically deals with an insolvent bank by choosing between two options:

(1) forcing the insolvent bank into a competing local bank, or entirely out of business, thus
dampening local competition; or

(2) first, replacing all the insolvent bank's bad assets with good assets, and, second, selling it
to some skillfull out-of-state buyer, after which process the new bank can help clobber the
remaining also-weak-and-also-insured banks in the area.

Mayer believes it was "insane" for the FDIC to do as it did in many instances, which was to
select option (2). Accordingly to Mayer, the FDIC thus arranged that "overcapacity was
rigorously maintained." Mayer raises an interesting question. Coming back to the analogy
earlier used, if you or I were really unlucky and were guarantor for seven local brothers-in-
law, each with a troubled hamburger joint, what would we do when the first one went broke?
We would surely reject the idea, of, first, fixing up the defunct joint so that it was better than
the others, and, second, guaranteeing the obligations of a new and more skillfull out-of-state
operator who wanted to enter the market by taking over the improved facility.

Mayer is right insofar as he implies that there are too many banks and bank branches, just as
there were formerly too many filing stations, sometimes three or four at an intersection. The
departed filling stations "never will be missed," so perhaps the FDIC should "have a little
list," like the bloodthirsty figure in the Mikado.

Beyond that, we are not certain that Mayer's conclusions will always prove right. The basic
banking system is right out of Alice in Wonderland, so maybe it's like non-Euclidean
geometry and only Alice-in-Wonderland-type cures really fit in. After all, the scenario which
troubles Mayer has a perverse beauty, at least to a government. The bank failures cascade, on
and on, refreshed by new governmental acts, so that the FDIC can be saving a large part of the
banking system each year for a long time.

And we must admit that, if we were the FDIC and were thus forced to participate heavily in
our present banking system, like it or not, we would occasionally do what Mayer finds
objectionable, in those rare cases when we saw a chance for greatly improving banking
culture in some community. We would, for instance, occasionally sell a sick bank to John
McCoy (of Banc One), even when this brought a new bank to a state full of troubled banks, if
every in-state bank seemed too weak or foolish to be selected as an alternative buyer. We
would figure that (1) some subsequent insolvencies of other local banks were in our long-
term interest, (2) we were supporting a sound model, and (3) eventually, as the example
spread, our troubles as deposit-insurer of a silly system would be reduced. We would then
have a pleasant lull before the silly system caused new troubles to pop up, maybe even under
McCoy's successors at Banc One.

While Mayer's subject is interesting, we probably don't have to worry much about worldly
consequences. Outside science, it is amazing how little impact there can be from a powerful
idea, published in a prominent place (such as the Journal). Everyone's experience is that you
teach only what a reader almost knows, and that seldom.

If our foregoing comments about systemic irresponsibility and chances for a rational cure are
right, or substantially right, it is hard to be optimistic about coming legislative "reform" of
banking. Perhaps the best we can hope for is Menckenian reform where old error is replaced,
not by truth, but by new error. It is also possible that we will see exactly the same old
systemic error repeated, but bearing bells and whistles in the form of new bank powers. This
outcome is roughly what is recommended by the banking lobby, which has evidently learned
nothing from the history of the savings and loan laws.

Let us next turn to the savings and loan field. Here, faced with a more disastrous mess, the
legislators were so outraged that they attempted what they thought was extreme reform:
FIRREA. This legislation took a "back-to-basics" approach and has since been interpreted by
regulators who seem to believe, understandably, that they must act as though they were tough
"bouncers," given the job of bringing order to a drunken brawl (a description that understates
what the regulators faced).

This regulatory approach is now squeezing out (1) much folly, and (2) some non-folly needed
to keep institutions healthy. Most executives we know at other associations concentrate only
on the negative side and are outraged at instances of regulatory elimination of non-folly.
They tend to construe present FIRREA enforcement as the equivalent of Mark Twain's
prescription for preventing children's stuttering: "Remove the lower jaw."

Our view is different, even though we are much harmed by FIRREA. We think the system
needed new rules, interpreted by tough "bouncers," and that the "bouncing" process, done
with sufficient vigor, inevitably involves some lumps for the undeserving. There may even be
some deaths from "friendly fire." Nonetheless, the process must go on.

What concerns us is the most important question of all. Did our legislators, through FIRREA,
even with their "never again" mindset, fix the most important systemic error in the savings
and loan industry? We think not.

As the dust has cleared, the best savings and loan associations are clearly worse businesses
than the best banks (which themselves have plenty of troubles). This conclusion is supported
by both (1) stock market prices and (2) action of governmental liquidators in response to
market conditions. Stocks of the best associations now sell at much lower price/book-value
ratios than stocks of the best banks. And governmental liquidators are constantly selling
association branches to banks while almost never selling bank branches to associations.
FIRREA has not made associations, on average, as desirable for owners as banks. The two
institutional types remain different and unequal, while quite comparable in essential residual
function, now that Fannie Mae and Freddie Mac exist to perform a lion's share of the finance
function supporting housing.

The savings and loan system, in a modern era in which the government is always a large net
borrower, still tries to use short-term savings accounts to finance long-term housing lending.
This is, in essence, a very bad idea, violating the logic of an elementary prescription: "If a
thing isn't worth doing at all, it isn't worth doing well."

To be sure, some fix of systemic maturity-mismatch risk is now attempted, through

encouragement of variable-rate loans. But the variable-rate loans typically "cap" interest rate
escalation at a few percentage points, which must be done for moderate-income borrowers to
prevent both (1) unacceptable hardship and (2) sudden falls in non-housing spending. This
compromise is like having building codes in California protect only up to 5 points on the
Richter earthquake scale. The compromise is almost sure to bring back, probably at a remote
date, another horrible collapse of the savings and loan system.

As we say this, we are not critical of the best California associations, such as Home Savings,
Great Western Savings and World Savings. These people have logical operations bearing one
big systemic risk that cannot be avoided by permanent players. If we had to play forever
under current rules, we would try to imitate them. But we would have a big disadvantage:
"we don't know how to get there from here," because they have such momentum in systems,
particularly in loan origination. Fortunately, no one is sentencing us to play forever in a game
with a systemic risk we don't like and in which we are at a big disadvantage. Instead, we have
temporized with a different, acceptable "there" in a form combining (1) a big holding of
Freddie Mac, with (2) financial flexibility to adapt as we choose to new conditions.

So much for ridicule, pessimistic speculations, and excuses for our defects, always easy to
provide. As any responsible calamity-howler should, we will now risk playing the fool in
public by attempting to say what we would do with the bank/money-market fund/savings and
loan system if we were Congress:

(1) Because we have a help-housing bias, we would keep government-assisted housing

finance for low-to-moderate-income people. We would do this by forcing pension funds to
maintain a significant portion of their assets in housing-related assets in the form of Freddie
Mac and Fannie Mae mortgage-backed securities representing interests in fixed-rate
mortgages. This requirement strikes us as fair, given the tax exemption possessed by the
pension funds. And the pension funds are the logical suppliers of housing finance because
they by nature have (a) massive assets, and (b) liabilities with maturities matching
homeowners' needs for long-term, fixed-rate credit. Our reason for specifying Freddie Mac
and Fannie Mae securities as a conduit for housing assistance is our belief that these entities
would assure loan quality better and more cheaply than would any government bureaucracy.
In quantitative terms, we would leave housing finance more assisted than it is now,
particularly for first-time home buyers who have won their spurs.

(2) We would merge the banks, money-market funds and savings and loan associations into
one banking system, with insured deposits. The new banking system would be separate from
both (a) industry and (b) the part of investment banking likely to disappoint investors. It
would have the following characteristics:

(i) There would be one federal regulator that also served as deposit-insurer, in lieu of the
truly crazy, inefficient Balkanization of our present regulatory and insurance apparatus.
(Eliminating Balkanization would do more than reduce costs, delays, confusion and
competition in laxity. There is a system-design advantage in making the deposit-insurance
loss payer and the bank-controlling loss preventer one and the same. The system then
becomes more "responsible" in the Frankelian sense, requiring that systems be organized, to
the extent feasible, so that decision-makers, not others, bear consequences of decisions.)

(ii) There would be no bank-holding companies, but the new banks would have a monopoly
in offering check-writing privileges, debit cards and credit cards, except for credit cards
offered on behalf of a single vendor. (The new law would permit tax-free spinoffs of existing
banks, newly organized banks, and non-banks to help existing corporations come into
compliance. Spun-off non-banks could include specialists in high-interest-rate lending to
businesses.)

(iii) Flexible, government-regulatory-run controls would set a ceiling on interest that could be
paid on bank accounts. (If you are going to guarantee the credit of an entire industry, there is
a limit to the competition that is desirable. Besides, many banks will behave badly in their
important function when they are under the extreme cost pressure, not normal in business, that
occurs when one's competitors are all financed without limit by the government, through
deposit insurance.)

(iv) All capital satisfying regulatory requirements would have to be in the form of stock,
either common or preferred, except for "grandfathered" debt.

(v) Stockbrokers (and others) could buy for customers all the insured certificates of deposit
they wished, but they could not, in exchange, receive commissions or other advantages from
the banks issuing the certificates. ("Abuse it and lose it," is our motto.)

(vi) The federal regulator would have clear power, exercisable without an excess of "due
process" or "second guessing," to close out or force sale or merger of weak banks well before
they became insolvent. Banks could ordinarily avoid such calamities, after a first warning, by
raising new capital through "rights" issues, or in some other way. (There is nothing novel in
such a system. Close-out orders, issued well short of insolvency, have long been standard
practice under regulatory practice governing securities and currency traders.)

(vii) Bank accounting for all purposes would count much revenue as profit only after all
significant risk had been removed from the transactions generating the revenues. Bank
dividends, of course, could be paid only from the more conservatively reported profits.
Income tax would be deferred on the deferred revenues required by this new conservatism in
accounting. (It is a terrible mistake, a novice's mistake to try to control important behavior
with an all-stick-and-no-carrot approach. Therefore, the carrot-providing tax deferment
would be wise.)

(viii) There would be no 2,000-page mass of government regulations. But there would be
some rule for business and real estate loans such as: loan as you wish, but no new loans count
as bank assets unless supported by substantial equity, a stipulation that would create a large
margin of safety.

(ix) Deposit-insurance rates would promptly be lowered from present levels, but under a new
system so tough that risk of loss to the deposit insurer would be reduced, even after taking
into account the effects from lower rates.

(x) The whole system would be designed to have the best businesses small and large, again
become intimate with the best banks. The banks would again concentrate on being (1)
relatively low-interest-rate lenders to high-quality businesses, and (2) lenders to consumers
who are not "fiscaholics". High-interest-rate lending, to people with weak credit, would be
forced into non-banking systems retaining no common-management or common-premises
links with banking.

There is, no doubt, much wrong with our recommendations. But there is also much wrong
with our present system, which has helped cause a questionable shift in banking priorities and
a big mess, with every prospect for more of the same. In contrast, there is little in history to
suggest that our recommendations would be as bad. And even if the new system had serious
faults, it would probably be a better way station on the path to a banking system befitting a
great country.

In recent years the government has tried to maintain a useful, relatively trouble-free banking
system by making the banking business bear increased competitive burdens, and, when the
system has responded by working worse, the government has increased both the burdens and
the permitted scope of banks' activities. After such revisions the system
has again worked worse. Surely it is time to reverse our approach. We should act like the
artillery officer who, when he has put one shell over the target, next tries to put a shell clearly
short, expecting to get the desired result in due course.

Some people might worry that banking would get too profitable under the system we
recommend. To this worry there are three answers:

(1) The prospect of better profits, with less risk, would tend to (a) reduce governmental losses
as many billions of dollars worth of foreclosed thrift and bank assets are sold off by the FDIC,
and (b) enable the government, through tough capital standards, to cause eager private
augmentation of banking capital by shareholders, precisely what is needed.

(2) Based on past experience, the nation's bankers (including us) may, on average, be up to
the challenge of not earning excessive profits, even in an easier system.

(3) If excessive profits came, they could easily be reduced in due course by a new
governmental tax, charge or burden.

We now quitclaim legislative reform to those who make it their business. We also assure
Wesco shareholders that this reform-minded section of our letter to shareholders is an
unlikely-to-be-repeated aberration. It was caused, in part, by a combination of (1)
overwhelming disgust with the present scene, and (2) long association by the writer with an
eccentric fellow who may not share all the notions herein expressed but who encourages this
kind of writing.

This eccentric, who heads Berkshire Hathaway, Wesco's parent corporation, believes for some
reason that accumulated wealth should never be spent on oneself or one's family, but instead
should merely serve, before it is given to charity, as an example of a certain approach to life
and as a didactic platform. These uses, plus use in building the platform higher, are
considered the only honorable ones not only during life but also after death. Shareholders
who continue in such peculiar company are hereby warned by our example in writing this
section: some of the eccentricities of this fellow are contagious, at least if association is long
continued.

Consolidated Balance Sheet and Related Discussion

Wesco's consolidated balance sheet (1) retains a strength befitting a company whose
consolidated net worth supports large outstanding promises to others and (2) reflects a
continuing slow pace of acquisition of additional businesses because few are found available,
despite constant search, at prices deemed rational from the standpoint of Wesco shareholders.

As indicated in the accompanying financial statements, the aggregate market value of Wesco's
marketable equity securities was higher than their aggregate carrying value at December 31,
1990 by about $46 million, down significantly from about $98 million one year earlier. The
consolidated aggregate market value of all marketable securities, including bonds and other
fixed-income securities, exceeded aggregate carrying value by about $61.3 million. As earlier
emphasized, about $56.2 million of this unrealized appreciation lies within the savings and
loan subsidiary and includes $45.3 million of appreciation in stock of Freddie Mac.

The foregoing paragraph deals only with unrealized appreciation of securities above "carrying
value." Wesco also has some unrealized appreciation in securities that is already in "carrying
value." This has happened because Wesco's insurance subsidiary at December 31, 1990 had
about $40.9 million in appreciation in common stocks (mostly stock of The Coca-Cola
Company). Under a peculiar accounting convention applicable only to insurance companies,
this appreciation, minus the income taxes that would be due if the stocks were sold, is already
included in Wesco's audited net worth, even though the gain has never passed through any
audited report of income.

As indicated in Schedule I accompanying Wesco's financial statements, investments, both

those in the savings and loan and insurance subsidiaries and those held temporarily elsewhere
pending sale to fund business extension, tend to be concentrated in very few places. Through
this practice of concentration of investments, we seek to better understand the few decisions
we make.(*)

(*) It is interesting to compare Wesco's approach (deliberate non-diversification of

investments in an attempt to be more skillful per transaction) with an approach promoted for
years by Michael Milken to help sell junk bonds. The Milken approach, supported by
theories of many finance professors, argued that (1) market prices were efficient in a world
where investors get paid extra for enduring volatility (wide swings in outcomes); (2)
therefore, the prices at which new issues of junk bonds came to market were fair in a
probabilistic sense (meaning that the high promised interest rates covered increased statistical
expectancy of loss) and also provided some premium return to cover volatility exposure; and
(3) therefore, if a savings and loan association (or other institution) arranged diversification,
say, by buying, without much examination, a large part of each new Milken issue of junk
bonds, the association would work itself into the sure to-get-better-than-average-results
position of a gambling house proprietor with a "house" edge. This type of theorizing has now
wreaked havoc at institutions, governed by true-believers, which backed their conclusions by
buying Milken's "bonds." Contrary to the theorizing, widely diversified purchases of such
"bonds" have in most cases produced dismal results. We can all understand why Milken
behaved as he did and believed what he had to believe in order to maintain an endurable self-
image. But how can we explain why anyone else believed that Milken was pad 5%
commissions to put "bond" buyers in the position of the house in Las Vegas? We suggest this
cause: many of the foolish buyers, and their advisers, were trained by finance professors who
pushed beloved models (efficient market theory and modern portfolio theory) way too far,
while they ignored other models that would have warned of danger. This is a common type of
"expert" error, as we have earlier indicated.

Wesco's Pasadena real estate comprises a full block containing (1) about 125,000 first-class
net rentable square feet, including Mutual Savings' space, in a modern office building, plus
(2) an additional net rentable 34,000 square feet of economically marginal space in old
buildings, which it would probably be wiser to destroy than improve. This real estate has a
market value substantially in excess of carrying value. The existence of unrealized
appreciation is demonstrated by (1) mortgage debt ($4,524,000 at 9.25% fixed) against this
real estate exceeding its depreciated carrying value $3,163,000) in Wesco's balance sheet at
December 31, 1990, and (2) substantial current net cash flow (about $1 million per year) to
Wesco after debt service on the mortgage.

The modern office building is 99% rented, despite a glut of vacant office space in Pasadena.
We charge just-below-standard rents and run the building as a sort of first-class club for
tenants we admire. With these practices, a prime location and superior parking facilities, we
anticipate future increases in cash flow, but not in 1991 and 1992. The next two years are not
likely to be good years for most owners of commercial real estate.

Wesco remains in a prudent position when total debt is compared to total shareholders' equity
and total liquid assets. Wesco's practice has been to do a certain amount of long-term
borrowing in advance of specific need, in order to have maximum financial flexibility to face
both hazards and opportunities. It values its AA+ credit rating.

It is expected that the balance sheet strength of the consolidated enterprise will in due course
be used in one or more business extensions. The extension activity requires patience, at least
for people like us.

The ratio of Wesco's annual reported consolidated net income to reported consolidated
shareholders' equity, about 12% in 1988-90, was dependent to a significant extent on
securities gains, irregular by nature.

When Berkshire Hathaway bought into Wesco in 1973, the present stock (adjusted for a later
three-for-one split) traded at about $6. At yearend 1990, the stock traded at $47 7/8 and it has
paid modest dividends, increased every year, during Berkshire Hathaway's stewardship.

The financial results for Wesco shareholders have not been bad. But they are not outstanding,
considering the power of compound interest and the generally favorable business climate.
And now, after all these years, Wesco continues to have (1) a very strong balance sheet, and
(2) a shortage of direct ownership of businesses with enough commercial advantage in place
to assure permanent high future returns on capital employed. In contrast, the parent company,
Berkshire Hathaway, is better positioned. This outcome was explained in Wesco's annual
report last year, to which we refer Wesco shareholders, new and old.
On January 24, 1991, Wesco increased its regular quarterly dividend from 20 1/2 cents per
share to 21 1/2 cents per share, payable March 12, 1991, to shareholders of record as of the
close of business on February 28, 1991.

This annual report contains Form 10-K, a report filed with the Securities and Exchange
Commission, and includes detailed information about Wesco and its subsidiaries as well as
audited financial statements bearing extensive footnotes. As usual, your careful attention is
sought with respect to these items.

Charles T. Munger
Chairman of the Board
March 8, 1991