The International Dollar Standard and

Sustainability of the U.S. Current Account Deficit

by

Ronald McKinnon1
Stanford University

Brookings Panel on Economic Activity:

Symposium on the U.S. Current Account

March 29 and 30, 2001

Falk Auditorium

Revised April 2001

1
William D. Eberle Professor of International Economics, Department of Economics, Stanford University,
Stanford California 94305-6072. E-mail: <mckinnon@leland.stanford.edu>
 Revised April 13, 2001

The International Dollar Standard and

Sustainability of the U.S. Current Account Deficit

by

Ronald McKinnon2
Stanford University

For more than 20 years, the world’s richest, most mature, industrial economy has
drawn heavily on the world’s limited pool of saving to support high consumption—in the
1980s by the Federal Government, and in the 1990s by households. Over the past decade,
personal saving has fallen more than government saving (as manifested in budget
surpluses) has increased. The huge deficit in the current account in the U.S. balance of
payments, about 4.5 percent of GNP in 2000, reflects this saving gap. In order to support
a normal level of gross domestic investment, i.e., historically about 16 to 17 percent of
GNP, America has had to draw heavily on the saving of the rest of the world. On a flow
basis, the U.S. is now attracting more capital net than all the developing countries
combined.

Correspondingly, the international balance sheet of the United States has declined
from being a net creditor to the rest of the world at the beginning of the 1980s, to being
the largest net debtor—to the tune of an incredible $2,300 billion by 2000. The
cumulative effect of this private foreign borrowing over the last 10 years is now reflected
in the balance sheets of both American firms and households. For example, the
indebtedness of the personal sector is now a record 1.1 times disposable income, while
firms also show very high indebtedness relative to cash flow.

Should Americans worry about this anomalous situation? After all, the dollar
remains strong and the United States is unique in having a virtually unlimited line of
credit with the rest of the world, which is largely denominated in its own currency, i.e.,
dollars. Consequently, American banks and other financial institutions are relatively
immune to currency risk because both their assets, which are largely claims on the
domestic economy, and their (deposit) liabilities, of which a substantial fraction is owed
to foreigners, are dollar denominated.

In contrast, other debtor countries must learn to live with currency mismatches
where their banks’ and other corporate international liabilities are dollar denominated but

2
William E. Eberle Professor, Department of Economics, Stanford University, Stanford California 94305-
6072. E-mail: mckinnon@leland.stanford.edu. This paper has benefited greatly from the general
discussion following its presentation at the Brookings Institution on March 30, 2001.

1
their assets are denominated in the domestic currency. Indeed, this mismatch was the
genesis of the great Asian currency crisis of 1997-98. Because Thailand, Korea,
Indonesia, Philippines, and Malaysia had large outstanding (short-term) dollar liabilities,
they became extremely vulnerable to a currency attack, with the resulting devaluations
bankrupting domestic financial institutions. But however precarious and over-leveraged
the financing of American borrowers—including American banks, which intermediate
such borrowing internationally—might be, they are invulnerable to dollar devaluation.

Compared to other industrial countries, does this invulnerability to currency crises

simply reflect the greater strength of the American capital markets and wisdom of
American regulatory authorities? No. The fact that the United States is the preferred and
highly favored international borrower is pure serendipity. How did this accident of
history come about?

The International Dollar Standard

In the immediate aftermath of World War II, confidence in the currencies and
financial systems of all the other industrial countries had evaporated. To prevent capital
flight mainly to the United States, European countries as well as Japan imposed tight
exchange controls. The relatively stable-valued US dollar was the only major currency in
which international exchange could freely take place. In the late 1940s, under the Bretton
Woods monetary order, other nations declared official exchange rate parities against the
dollar, as the central numJraire for the system overall. Rather than creating asymmetry
among currencies, this official monetary order simply recognized it [McKinnon, 1996].
Thus was the dollar enthroned as “international money.”

When the system of official exchange rate parities broke down in 1971, the dollar
was not dethroned. To the present day, the dollar is still the vehicle currency in the inter-
bank spot and forward exchange markets, the currency of invoice for primary commodity
trade and for many industrial goods and services, and the main currency of denomination
for international capital flows—particularly at short term and interbank. Outside of
Europe, governments use the dollar as their prime intervention currency—often
unofficially pegging to the dollar, and U.S. Treasury bonds are widely held by foreign
central banks and treasuries as official exchange reserves.

This provision of international money, i.e., providing the central currency in the
world system, is a natural monopoly. Consider first a world of “N” national currencies
without official interventions or foreign exchange targeting by governments. In
organizing private interbank markets for foreign exchange, great savings in transactions
costs can be had if just one national currency, the Nth, is chosen as the vehicle currency.
Then all foreign exchange quotations—bids and offers—at all terms to maturity can take
place against this one vehicle currency. The number of active markets can be reduced
from N(N-1)/2 to just N-1. In a world of more than 150 national currencies, this is a
tremendous economy of markets for the large commercial banks that make the foreign
exchange market. The dollar’s interbank predominance (being on one side of almost 90

2
percent of interbank transactions) allows banks to cover both their forward exchange and
options exposures much more efficiently.

Trade in goods and services shows a similar pattern of using one national money
as the main currency of invoice. Exports of homogeneous primary products such as oil,
wheat, copper, and so on, all tend to be invoiced in dollars with worldwide price
formation in a centralized exchange. Spot trading, but particularly forward contracting, is
concentrated at these centralized exchanges—which are usually in American cities such
as Chicago and New York, although dollar-denominated commodity exchanges do exist
in London and elsewhere.

With the exception of large European countries, exports of heterogeneous services

and manufactured goods also tend to be invoiced in dollars—even exports from Japan. In
intra-regional trade in Asia and Latin America, the dollar is overwhelmingly used for
invoicing trade in both primary commodities and manufactures. And all countries trading
directly with the United States itself see both imports and exports invoiced in US dollars.

Once settled on for whatever historical reason, the dollar offers huge economies
of scale for its continued use as the central vehicle in international exchange. (The major
exception is the strong regional role played by the euro for countries on the fringe of the
EU.) Of the other 150 or so countries in the world system, the more countries A and B
use the dollar in international exchange, the more attractive (cost reducing) it is for C or
D to do so. In effect, the dollar could now only be deposed by some cataclysmic event—
such as massive inflation in the United States.

The Nominal Anchor

In periods of reasonable confidence in American monetary policy as over the past

decade—and in the 1950s and 1960s, these dollar prices of goods and services are
relatively invariant to fluctuations in the dollar's exchange rate. In contrast, if any other
country allows its exchange rate to fluctuate against the dollar, there is higher pass
through into its domestic goods prices as well as problems with the debt servicing of its
short-term dollar liabilities to foreigners. (Again, Europe is a partial exception.) The
upshot is that most countries are reluctant to let their exchange rates float freely against
the dollar—what Calvo and Reinhart (2000) call “fear of floating”.

What are the monetary implications? To an important degree, other countries

subordinate their domestic monetary policies to prevent, not always successfully, short-
and medium-term fluctuations in their dollar exchange rates. Of course, high inflation
countries must let their dollar exchange rates depreciate over the long term; but, in
noncrisis periods, even they strive to stabilize their exchange rates from one day (or one
week) to the next [McKinnon, 2001].

Consequently, as the Nth currency in an N currency world, only the United States
has the freedom to conduct its own monetary policy independently (except in great
crises) of exchange rate fluctuations—what the Robert Mundell [1968] called the

3
“redundancy problem”. Pretty well ignoring the dollar’s exchange rate against other
countries, Fed Chairman Alan Greenspan can focus just on stabilizing the American price
level and the purchasing of the dollar in terms of real goods and services—which he has
been quite successful in doing. Then the American price level becomes the (informal)
nominal anchor for international monetary system.

When the dollar-based nominal anchor seems secure as now, unlike the 1970s
when inflation was high and variable, this reinforces the willingness of other countries to
target their dollar exchange rates (again putting aside the euro zone as being a quasi
independent monetary regime). They become very reluctant to see their currencies
depreciate against the dollar because of the longer-term inflationary threat; and they may
be even more reluctant to see any substantial appreciation of their currency against the
dollar for fear of losing mercantile competitiveness in world markets in the short and
medium terms. In particular, they are most unlikely to jettison, or even stop
accumulating, their huge official exchange reserves—mainly dollar denominated and
often just U.S. Treasury bonds. Such a sell-off would provoke a sharp appreciation of
their currencies against the dollar. Willy Nilly, foreign governments cannot avoid being
important creditors of the United States.

America’s Soft Borrowing Constraint

Although this central monetary role for the dollar is all well and good for
promoting more efficient international exchange, an incidental consequence is that the
United States itself is given a much softer constraint on its own international borrowing.
As the rest of the world’s income grows, the demand by foreign enterprises and
governments to build up their stocks of international liquidity rises commensurately. So
America can provide these liquid dollar assets—whether liquid claims on American
banks, hand-to-hand currency, U.S. Treasury or government agency bonds, or various
kinds of private bonds or stocks (albeit somewhat less liquid)—which are claims on
American firms and households with no well-defined time frame for net repayment.

The closest analogy is to consider any central bank issuing fiat money within its
own national monetary domain. Although bank notes and coins may formally be the
liabilities of the central bank, in practice, they never have to be redeemed because of the
private sector’s ongoing demand for domestic money. Analogously on an international
scale, the collectivity which is the United States can issue liquid claims on itself to the
rest of the world that “never” have to be redeemed.

For the last 20 years, the United States has chosen to exploit this soft borrowing
constraint by absorbing capital on a net basis from the rest of the world. But an efficient
dollar standard need not depend on America’s running current-account deficits to provide
international liquidity. Even without such deficits, the rest of the world could still have
built up the dollar liquidity it so craves.

In fact, in the 1950s and 1960s, the U.S. ran large current-account surpluses.
However, long-term capital outflows—including illiquid direct investment abroad as well

4
as development aid—were greater than its current surpluses. This payments gap was then
covered by more liquid and generally shorter-term capital inflows: foreign firms built up
their liquid stocks of U.S. bank deposits and money market instruments, and foreign
governments built up stocks of U.S. Treasury bonds. Like a giant international financial
intermediary, the U.S. lent long to, and borrowed (less) short term from, the rest of the
world [Despres, Kindleberger and Salant, 1966]. By these gross capital flows, the U.S.
satisfied the world’s growing demands for dollar liquidity while remaining a net creditor.

If, in the new millennium, the U.S. could return to current-account balance or
even began to run surpluses, the rest of the world could still get the liquidity it needed
quite comfortably through greater long-term lending by the United States. But if we
accept the hypothesis put forth here that the American line of credit with the rest of the
world is indefinitely long, why not just keep borrowing to cover current account deficits?
Wouldn’t American consumers be better off if they continue to borrow indefinitely to
keep their expenditures above their incomes?

Financial Fragility

There are two big problems with continuing the status quo ante of running large
current-account deficits:

(1) excessive borrowing and declining creditworthiness of individual American

households and some firms;
(2) increasing protectionism as the large trade deficit continues to erode
America’s industrial base.

Under (1), the over-leveraging of American households is aggravated by banks

and consumer credit companies financing themselves too cheaply on international
markets. Either directly of indirectly, they can easily sell liquid dollar deposits and other
financial instruments to foreigners to finance the proliferation of domestic consumer
credit cards and mortgage lending. The resulting incredibly low net worth of American
households with moderate incomes makes the macro economy less stable. For example,
the large household debt overhang could well aggravate the cyclical downturn in 2001 by
inducing a sharp rise in household bankruptcies—and a sharp decline in consumer
spending more generally.

To a degree, the American corporate sector is less vulnerable to over-leveraging

from the economy’s soft borrowing constraint in international markets. Foreigners can
and do buy equity claims on American corporations, as well as industrial bonds and
commercial bills. Thus the debt-to-equity ratios in most American companies, while still
uncomfortably high, need not rise as result of foreign capital inflows. The problem of
over-leveraging by companies is more one of U.S. tax law and corporate governance.

However, nobody, including foreigners, can buy equity claims on American

households! Thus, insofar as the influx of foreign capital softens household budget

5
constraints, it takes the form of a greater buildup of household indebtedness. American
banks as (international) financial intermediaries are “special” in two important senses.
First, they lend to domestic economic units—American households and small firms—that
cannot finance themselves through the direct issue of stocks or bonds in primary
securities markets. Second, foreign claims on American banks are an important
component of international liquidity for which the rest of the world’s demand is rather
strong. Although heavily indebted American households don’t seem to be borrowing
from foreigners, they are doing so indirectly as intermediated by domestic banks and
finance companies.

The Dutch Disease and Protectionism

Under (2), there is also a political-economic restraint on American trade deficits:

the transfer problem. Foreign saving can only be transferred to the United States through
large American current account deficits, i.e., allowing American expenditures to rise
above income. For any given level of income, this means a reduction in American exports
(broadly defined) and an increase in imports. Because of the peculiarly heavy state
intervention and protectionism for agriculture and some services around the world, the
industrial sector typically bears the brunt of adjustment to swings in the trade balance.

To accommodate the trade deficit other things remaining equal, American

manufacturing industries must contract on both the export and import-competing sides.
Boeing will have much tougher time competing against Airbus Industries in aircraft,
Xerox against Ricoh in copiers, Ford against Toyota in autos, Caterpillar against
Komatsu in heavy equipment, and so on. Indeed, America has, or is, largely exiting
certain industries—such as photographic equipment including the latest digital
technologies—altogether. Where the U.S. has a technological lead in computers,
integrated circuits, and internet-related equipment—the rate at which American firms
farm out their production to overseas affiliates will be greater because of the need to
transfer capital net from the rest of the world.

A purist might say “If this is what the market dictates, then so be it.” But in some
sense “the market” is biased by international monetary considerations that give the U.S. a
uniquely soft long credit line with the rest of the world. If American consumers exploit—
or issuers of consumer credit cards cajol them into drawing on—this credit line, the
resulting capital inflows and strong dollar lead to a trade deficit. This international
monetary version of the “Dutch disease” has led (is leading) to an unusual shrinkage in
America’s industrial base.

More important, the political obstacles to preserving free trade are increased when
the trade deficit is large. First, a declining American export sector reduces the supply of
lobbyists in favor of keeping foreign markets open reciprocally with the domestic one.
The second is the perception, whether or not it is correct, that a large trade deficit reflects
“unfair” trading practices by foreigners—and that the government should do something
offsetting to protect American industry.

6
 During the “Goldilocks” period of the American economy from 1995 through
2000, these underlying protectionist pressures were dampened by the unusually low rate
of unemployment and the economy’s rapid rate of growth. However, once the economy
slips into an cyclical downturn with rising unemployment and widespread industrial
bankruptcies, then protectionist pressure will reappear with a vengeance—as with the
many episodes of Japan-bashing before 1995, and what could be China bashing in 2001
and beyond. In the longer run, the political economy of preserving free trade on a world
scale would be much easier to sustain if the center country’s trade accounts came into
better balance.

Tax Cuts the United States Can Afford

If today’s large budget surplus, i.e., the government’s contribution to national

saving over 2 percent of GNP in 2000, were to be reduced by massive tax cuts without
generating a substantial increase in U.S. personal saving, America’s huge current-account
deficit would increase.

Beyond credit cards, an important aspect of the low personal saving problem is
that Americans are putting aside far too little in their pension plans and then taking out
too much. Both are tax driven. Thus tax “cuts” should take the form of much higher
ceilings on personal tax deductions for pension saving, while allowing older people to
accumulate indefinitely within their pension plans without facing tax penalties.

Eliminating, or at least substantially re-crafting estate and inheritance taxes (E and

I taxes) could further increase the incentives for American households to save in order to
pass more wealth on to the next generation—both inside and outside their pension plans.
Upon retirement, people would no longer be so eager to convert the capital in their
pension funds into annuities and thus consume it all over their lifetimes. Instead, retirees
would be more content to leave some portion of their defined pension contributions as a
lump sum if they knew it would not be subject to E and I taxes. Indeed, putting all of
one’s retirement capital into a fixed annuity is risky: there is no margin for error should
one be hit with some unexpected expense.

These are but two examples of how tax cuts might increase the propensity to save
in American households. But a proper menu of tax and other institutional reforms would
go well beyond what can be covered in here. The bottom line is that, if one takes the
balance of international payments into account, tax reforms that demonstrably increase
private saving should be at the forefront of what the new Bush administration is
considering. But this is a lot to consider and perhaps too much to hope for.

Is the Dollar Standard’s Survival at Stake?

Like most writers on the subject, Catherine Mann in the extraordinarily

comprehensive statistical analysis in her book, Is the U.S. Trade Deficit Sustainable?

7
[1999], treated the United States as just an important large country albeit with some
special economic features—but nevertheless similar to other countries with high debts in
ultimately having to pay back what it has borrowed. Writing in 1999, she concluded that:

All told, this calculation for the investor constraint along side the borrower
constraint supports the notion that the U.S. current account is sustainable for at
least two or three more years, or even longer as judged by the investor’s
constraint.
[Mann, 1999. p.163]

From her analysis, if the American current account deficit remains high, capital
flight from the dollar, higher U.S. interest, and a weak dollar should now be with us—or
eventually will be. But a sustained flight from the dollar to force a correction in the
American current-account deficit would also undermine the international dollar standard.

My own view is that the only real threat to the dollar-based institutions of
international exchange could come from chronic inflation in the U.S. itself. Absent
monetary instability in the center country, the dollar standard is robust and could continue
without the U.S running up against significant borrowing constraints from the rest of the
world. Any incipient run on the dollar would be offset by foreign central banks
accumulating dollar reserves in order to prevent their currencies from appreciating—
with a consequential loss in their international mercantile competitiveness. However, for
reasons adumbrated above, the world and U.S. economies would be better off if the
American current-account deficit was smaller or non-existent.

Most people consider America’s ability to attract vast amounts of capital from the
rest of the world in the 1990s to hinge on the extraordinary boom in the America’s
“Goldilocks” economy that made the U.S. a great place in which to invest. Indeed,
Catherine Mann worries [page 174] that if growth in the United States slows down and
growth picks up in the rest of the world, a sharp reversal of net capital flows could
occur—possibly leading to substantial dollar depreciation.

However, from the perspective of the monetary economics of the world dollar
standard put forth here, faster growth in the rest of the world will increase the demand for
international liquidity. Foreign firms and financial institutions—including central
banks—will become more willing to accumulate dollar bank balances, U.S. Treasury
bonds, and so on. This liquidity effect would amount to a countervailing capital flow
back to the United States.

8
Appendix: Disappearing U.S. Treasury Bonds,
Should We Worry?

In opting last January 25 for cuts in tax revenue in the year 2001, Federal Reserve
Chairman Greenspan de-emphasized the usual Keynesian argument for a countercyclical
economic stimulus in 2001−02. Taking a longer-term perspective, Greenspan worried
that large fiscal surpluses—on projections which assume no tax cut—would eliminate the
stock of U.S. Treasury bonds held outside the U.S. Social Security System. The Office of
Management and Budget (OMB) estimated in June 2000 that debt held by the public, i.e.,
excluding that held in U.S. government accounts but including that held by the Federal
Reserve, would be fully redeemed by the year 2012. Signs abound already that the open
market for treasuries is becoming less complete inter-temporally, and thus less liquid,
with the discontinuance of new issues of one-year bills and possibly 30-year bonds.

If the fiscal surpluses continued after all Treasury bonds had been retired, the
American government including the Fed would have no choice but to start acquiring
claims on the private sector with its surplus tax revenues. In the domestic capital markets,
Greenspan sees the government granting credit to, or acquiring ownership claims on,
private agents to be far too intrusive.

However, if the tax cuts were geared mainly to inducing more private saving
through, say, individual 401K plans, the net worth of American households should
improve. Thus macroeconomic fragility from over-leveraging would be reduced. In
principle, households could demand more treasuries with their increased saving.
However, the demand for a safe and “neutral” liquid asset by important financial
institutions—the Fed itself, commercial banks, insurance companies, and their foreign
counterparts including other central banks—would almost surely dominate bidding for
the extant stock of Treasuries (much as it does now). With saving-inducing tax cuts,
therefore, the Greenspan “problem” of overly intrusive government in the private
financial markets would be solved without increasing the current-account deficit.

In contrast, if tax cuts go beyond incentives to increase private American saving,

then the supply of U.S. Treasuries to the world markets would certainly increase.
However, the cost would now be an increased U.S. current-account deficit without
reducing the precarious financial fragility of American households. And the share of
outstanding Treasuries held by foreigners would surely rise.

In the absence of any American tax cuts (or expenditure increases), can the world
dollar standard survive the elimination of U.S. Treasury bonds from international
markets? In the year 2000, foreigners held about 42 percent of U.S. Treasuries not held
by U.S. government trust funds or by the Fed, and about half of these foreign holdings

9
were by official institutions such as central banks or treasuries. Official foreign exchange
reserves can be huge: more than $350 billion by Japan, $150 billion by China, $100
billion by Hong Kong, $100 billion by Korea, $100 billion by Taiwan, and so on for
lesser amounts across almost every country in the world. Most of these exchange
reserves are in dollar assets, with a high proportion being U.S. Treasuries.

Free of default risk, U.S. Treasuries are seen as the “risk-free asset” in the world’s
capital markets. Because the American Federal Governments owns the dollar-creating
central bank (the Fed), it can always create the means of settlement on its own debt—
whether held domestically or by foreigners. Under the world dollar standard, no other
country can similarly create international money at will.

Undoubtedly, the presence of U.S. Treasuries as an attractive asset has

contributed to the very elastic line of credit with the rest of the world that the U.S. has
exploited for the past 20 years. But having such a safe reserve asset, with assured
international purchasing power, is also a great convenience to other countries. With the
dollar so commonly in use as a vehicle and invoice currency, finding an equally liquid
replacement as an international reserve asset would be difficult.

But not impossible. In the absence of U.S. Treasury bonds, foreign central banks
and finance ministries could experiment with holding dollar assets such as bonds or
stocks that are claims on the American private sector—or on foreign issuers of dollar
denominated debt, which could be inherently more risky. In any event, credit risk in
official reserve holding would be more of a problem. And now Greenspan’s dilemma
would arise in an additional guise. Foreign governments, as well as the American one,
would intrude on the financing of private American companies!

References

Calvo, Guillermo and Carmen Reinhart, “Fear of Floating”, University of Maryland

Working Paper, Sept. 2000

Despres, Emile, Charles Kindleberger, Walter Salant, “The Dollar and World Liquidity:
A Minority View”, The Economist, Feb 5, 1966. pp. 526-29.

Manne, Catherine L. Is the U.S Trade Deficit Sustainable?, Institute for International
Economics, Washington D.C. 1999.

McKinnon, Ronald I., The Rules of the Game: International Money and Exchange Rates,
MIT University Press, 1996.

_____________, “After the Crisis, the East Asian Dollar Standard Resurrected: An
Interpretation of High Frequency Exchange Rate Pegging”, in J. Stiglitz and S.
Yusuf Rethinking the East Asian Miracle, World Bank and Oxford Press, 2001.

Mundell, Robert, International Economics, MacMillan Company, New York, 1968.

10