CRS Report for Congress
The U.S. Trade Deficit: Trends, Theory, Policy,
and Sustainability
Updated June 17, 1999
Dick K. Nanto
Specialist in Industry and Trade
Foreign Affairs, Defense, and Trade Division

Congressional Research Service ˜ The Library of Congress

This report briefly surveys recent trends in the U.S. trade deficit and the economic theory and
policies surrounding it. The Federal Trade Deficit Review Commission was organized in June
1999 to develop trade policy recommendations. This report points out some problems with
and recent advances in traditional economic approaches to trade theory and policy,
particularly with respect to inter-industry trade, strategic trade, macroeconomic trade theory,
the twin deficits, the exchange rate, and intertemporal savings and investment. It also
examines the sustainability of the trade deficit and its relationship to employment. This report
will be updated as conditions warrant.

The U.S. Trade Deficit: Trends, Theory, Policy, and
This report briefly surveys recent trends in the U.S. trade deficit and the
economic theory and policies surrounding it. After dropping to $74 billion in 1991,
the U.S. merchandise trade deficit increased by $49 billion in 1998 to a record high
of $248 billion. Even though the reasons for the rising deficit seem apparent, it raises
questions about the theoretical analysis that underlies U.S. policies to deal with it and
its sustainability and effect on the U.S. economy.
The Federal Trade Deficit Review Commission was organized on June 10, 1999,
and is responsible for developing trade policy recommendations by examining the
economic, trade, tax, and investment policies and laws, and other incentives and
restrictions that are relevant to addressing the causes and consequences of the U.S.
merchandise and current account deficits.
Economic theory is evolving with respect to the international trade side of the
U.S. economy and the relevant policy implications. The standard macroeconomic
approach to explaining the trade deficit is encountering some obstacles. First, the
savings and investment equality is an ex post identity, a framework for analysis, and
not a behavioral equation. Second, the linkage between macroeconomic conditions
and a nation’s exchange rate has not been established; exchange rate changes do not
translate completely into price changes for imports and exports, and exchange rates
tend to overshoot the equilibrium rates. This brings other inefficiencies into an
economy. Government policy may also affect exchange rates. Recent weakness in
the Japanese yen, for example, has been exacerbated by Japanese government policy.
The United States does not seem to have a transparent method of determining if and
when it should intervene in currency markets.
A third issue is that a focus on the savings-investment relationship in the
economy can lead to dubious policy prescriptions — such as the assertion in the
1980s that eliminating the federal budget deficit would do the same for the trade
deficit. A fourth problem is that a static macroeconomic approach does not account
for why the United States may be borrowing from abroad and how those funds are
being used. There may be an optimal level for a nation’s current account deficit, and
borrowing may be justified if it is used to increase productivity. Recent U.S. trade
data indicate that a rising share of U.S. imports has been for capital goods that should
raise U.S. productivity and economic growth.
The policy implications with respect to trade still center on the belief that free
trade is optimal, but extensions of the theory now allow for a strategic trade policy
aimed at assisting certain industries. The conventional economic conclusion that all
intervention into trade flows has no effect on the trade deficit, however, is yet to be
demonstrated empirically. In terms of sustainability, if the recent large capital inflows
reaching nearly $200 billion per year continue into the next century, a U.S. foreign
debt would develop equivalent to a quarter of GDP and foreigners may end up
owning more than half the U.S. federal debt. If the capital inflows do not continue,
the U.S. trade and current account deficits of today will be unsustainable.

Recent Trends in the U.S. Trade Deficit ..............................2
Trade Theory and Policy..........................................5
Intra-industry Trade..........................................7
Strategic Trade Theory.......................................7
Macroeconomic Theory.......................................8
Problems with the Basic Macroeconomic Approach.................10
Exchange Rate Effects and Determination....................11
Twin Deficits..........................................17
Intertemporal Savings and Investment.......................18
Sustainability of the Trade Deficit..................................20
Trade and Jobs................................................25
Conclusion ................................................... 26
List of Figures
Figure 1. U.S. Merchandise Imports, Exports, and Balance of Trade (balance-of-
payments basis), 1980-98......................................2
Figure 2. U.S. Balance on Current Account, Merchandise Trade, Unilateral
Transfers, and Investment Income, 1992-1998......................3
Figure 3. U.S. Bilateral Merchandise Trade Balances with Selected Trading
Partners, 1998..............................................4
Figure 4. Monthly Changes in Japan’s Yen-Dollar Exchange Rate and Foreign
Exchange Reserves (Percent)..................................16
Figure 5. U.S. Imports of Merchandise by Major End-use Category, 1990-9719
Figure 6. Net Capital Flows (Government and Private) and Current Account
Balances for the United States, 1965-97..........................22
Figure 7. U.S. Foreign and Federal Debt and Foreign Assets in the United States,

1991-2002 (forecast)........................................24

The U.S. Trade Deficit: Trends, Theory, Policy,
and Sustainability
After dropping to $74 billion in 1991, the U.S. merchandise trade deficit
increased from $199 billion in 1997 to $248 billion (balance-of-payments basis) in
1998. This surge in the U.S. trade deficit can be explained primarily by the strong
U.S. economy (which is buying more of almost everything — including imports), the
Asian financial crisis (which has reduced U.S. exports to Asian countries and lowered
prices of Asian exports to the U.S.), and inflows of capital (which drive up the value
of the dollar and make imports into the United States cheaper and American exports
more expensive). Even though the reasons for the rising deficit are apparent, it raises
questions about the long-term implications of this imbalance, the economic analysis
that underlies U.S. policies to deal with it, and its sustainability.
In concert with the rising U.S. trade deficit, the exchange value of the dollar has
been rising, particularly with respect to the currencies of Asian countries (except for
China and Hong Kong). The strength of the dollar in the late 1990s bears a strong
resemblance to the strong dollar in the first half of the 1980s. This was the period
when the U.S. trade deficit grew by six times. In 1998, the dollar gained so much
strength with respect to the yen that on June 17, 1998, both the United States and
Japan intervened in currency markets to support the yen.
This report first provides an overview of trends in the U.S. trade deficit. It then
examines the economic theory for analyzing the trade deficit and formulating policy
implications. This section is somewhat technical and touches only on the major points
in the theoretical debate. It is intended to point out some of the pitfalls of the
standard interpretations of the trade deficit. It includes both a discussion of the
traditional comparative advantage and macroeconomic approaches to the deficit and
a brief review of exchange rates, the sustainability of the trade deficit, and the effect
of trade on jobs. This report will be updated as major events warrant. More recent
trade data can be found in CRS Issue Brief 96038, U.S. International Trade: Data
and Forecasts, by Dick K. Nanto and Vivian C. Jones.
As for legislative activity, in the 105 Congress, H.R. 3579 (Livingston,th
Supplemental Appropriations) included a provision to establish an Emergency Trade
Deficit Review Commission to study the causes and consequences of the U.S.
merchandise trade and current account deficits and develop trade policy
recommendations for the 21 century. Although this measure passed the Senate, thest
provision was removed in Conference. It then was inserted into the Legislative1

U.S. Congress. House of Representatives Report to accompany H.R. 3579, 1051thnd
Congress, 2 Session. (Emergency Supplemental Appropriations bill for the fiscal year

Branch FY99 Appropriations bill (S. 2137) and became the Trade Deficit Review
Commission Act of 1998 (19 USC 2213). The Federal Trade Deficit Review
commission is responsible for developing trade policy recommendations by examining
the economic, trade, tax, and investment policies and laws, and other incentives and
restrictions that are relevant to addressing the causes and consequences of the U.S.
merchandise and current account deficits. The Commission is composed of twelve
members appointed by the Senate and House and is to submit a final report to the
President and Congress not later than twelve months after the date of its initialth
meeting on June 10, 1999. In the 106 Congress, Section 312 of the Legislative
Branch Appropriations Act, 2000 (S.1206, Bennett, reported by the Senate
Appropriations Committee on June 10, 1999) would make some changes to the Trade
Deficit Review Commission. It would extend funding for the Commission and its
termination day to 90 days after the Commission submits its final report. It also
would exempt Commission members from certain pay authorities and provisions of
the Federal Advisory Committee Act. Other legislation related to the U.S. trade
deficit includes trade provisions aimed at specific countries, such as Japan and China,23
and legislation dealing with fast-track trade negotiating authority.
Recent Trends in the U.S. Trade Deficit

Figure 1. U.S. Merchandise Imports, Exports, and Balance of
Trade (balance-of-payments basis), 1980-98
1000$ Billions
800 $919
Imports $671
200 -$248
-200 Trade Deficit
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
Source: Trade data from U.S. Department of Commerce.
Ending September 30, 1998, and for other purposes.”
See: CRS Issue Brief 97015, U.S.-Japan Economic Ties: Status and Outlook, by2
William H. Cooper, and CRS Issue Brief 91121, China-U.S. Trade Issues, by Wayne M.
See: CRS Issue Brief 97016, Trade Agreements: Renewing the Negotiating and Fast-3
track Implementing Authority, by Vladimir N. Pregelj.

As shown in Figure 1, the U.S. trade deficit grew from $25 billion in 1980 to a
peak of $160 billion in 1987 as the burgeoning Federal budget deficit along with rising
U.S. interest rates drew in international capital, appreciated the dollar, and hampered
U.S. net exports. The deficit then declined to $74 billion during the U.S. recession
of 1991 but rose to $199 billion in 1997 and further to $248 billion in 1998 despite
significant shrinkage in the Federal budget deficit. U.S. export growth had been
strong until 1998 when exports declined. Imports have continued to increase.
Standard and Poor’s Data Resources, Inc., a major economic forecasting firm,
projects that for 1999, the U.S. merchandise trade deficit (balance-of- payments basis)
may reach $327 billion. Much of the deterioration is occurring because the Asian4
financial crisis has cut U.S. exports to countries in difficulty, such as South Korea,
Japan, Indonesia, and Thailand, and has driven capital from Asian markets to higher
quality investments in the United States. Prices of imports from Asia also are lower
because of their weakened currencies.
Figure 2. U.S. Balance on Current Account, Merchandise Trade,
Unilateral Transfers, and Investment Income, 1992-1998
50Investment Income
-100 Transfers
Current Account
-200BalanceMerchandise Trade
-250 -$233.4
70 72 74 76 78 80 82 84 86 88 90 92 94 96 98
71 73 75 77 79 81 83 85 87 89 91 93 95 97
Source: Data from U.S. Department of Commerce
In terms of the U.S. current account (which includes trade in merchandise and
services, investment income, and unilateral transfers), the picture is similar —
primarily because the deficit in merchandise trade dominates the other balances. (See
Figure 2) In services trade, however, the United States ran a surplus of $78.9 billion
in 1998 — down somewhat from 1997. Net unilateral transfers (transfers of money

Standard & Poor’s. Review of the U.S. Economy, May 1999.4

not in return for goods or services performed, e.g. remittances, private gifts, pension
payments, and government grants) have been fairly stable at about -$40 billion. In
1998 it was $-41.9 billion. A significant change in the U.S. current account has been
in the balance on investment income. This is the balance of income received on U.S.
investments abroad less income paid on foreign assets in the United States. This
balance dropped from a surplus of around $30 billion in the early 1980s, to $20 billion
in the early 1990s, to a deficit of -$5.3 billion in 1997, and -$22.5 billion in 1998. The
switch from surplus to deficit in this balance largely reflects the net inflow of
investment into the United States. This inward investment is the functional equivalent
of foreign borrowing and has pushed the net investment position of the United States
to a negative $1.545 trillion in 1998. The United States has become the world’s
largest debtor nation, and the servicing of that debt has increased the U.S. deficit on
current account.
Figure 3. U.S. Bilateral Merchandise Trade Balances
with Selected Trading Partners, 1998
Deficit Surplus
4.3United Kingdom
4.2Saudi Arabia
2.4Hong Kong
1.7United Arab Emirates
0 20-20-40-60-80
Source: Data from U.S. Department of Commerce
The rising overall U.S. merchandise trade deficit is reflected in bilateral trade
with specific countries. (See Figure 3) In 1997, U.S. trade deficits with Japan and
China accounted for nearly two-thirds of the total U.S. deficit. Trade with Canada
and Mexico, the two NAFTA (North American Free Trade Agreement) trading
partners, also remained in deficit. With Japan, after declining from a peak of $66
billion in 1994, the U.S. deficit fell to $48 billion in 1996 before rising to $64 billion
in 1998. The U.S. deficit with China has surged from $30 billion in 1994, to $39

billion in 1996, and to $57 billion in 1998. With NAFTA, the U.S. deficit has shrunk
somewhat as it eased from $39 billion in 1996 to $36 billion in 1998.5
Although there is no economic reason that bilateral trade with any particular
trading partner should be in balance, the chronic U.S. deficit with Japan and the rising
imbalance with China have attracted criticism, raised political tensions, and have
induced calls for changes in U.S. trade policy. Trade barriers and market- opening
measures, moreover, clearly affect trade with individual nations. The continual trade
deficit with Japan combined with that nation’s weak-yen policy and capital outflows
from Japanese financial deregulation along with trade barriers in China have arguably
contributed to the rising U.S. trade deficit.6
Trade Theory and Policy
International trade theory takes two basic tracks in analyzing aggregate trade
flows. The first harkens back to the theory of comparative advantage along with
neoclassical economics and the second focuses on the macroeconomic foundations
of trade flows. Although the theory of comparative advantage and neoclassical
economic theory do not focus on the trade deficit, per se, we discuss this briefly in
order to provide a picture of how various aspects of trade theory and policy are
The theory of comparative advantage and traditional trade policy analysis
conclude that under certain idealized conditions, a country maximizes its well-being
by specializing in producing those goods and services in which it has a comparative
advantage and importing those in which other nations have advantage. The policy
implication of the comparative advantage argument is that barriers to trade reduce
national welfare and free trade brings optimal benefits for the economy as a whole.
Import competition may hurt some domestic competitors and impose some
adjustment costs on certain actors in the economy, but, in the long run, free trade
tends to lower prices and provide benefits to consumers that should offset the
negative effects on industries that are hurt. For the economy as a whole, therefore,
national well-being is increased. It is up to the political system to determine whether
income is then transferred to the industries that are hurt. According to this analysis,
protectionism is self defeating. It benefits certain trade interests but at a cost to the
rest of society.
Criticism of this model for trade policy based on comparative advantage and
traditional trade theory comes both from some policymakers and theoretical
economists. Some decision-makers assert that the policy implication that free trade

See: CRS Report 97-183, The U.S. Trade Deficit in Manufactures: The Rise of China5
and NAFTA, by Dick K. Nanto.
In July 1998, a senior official of the U.S. Department of Commerce stated before the6
House Committee on International Relations, Subcommittee on International Economic Policy
and Trade, that the U.S. trade deficit is largely due to distortions in the trading relationship
with Japan and China. See: Trade Imbalance with China and Japan Priority in Reducing
Trade Deficit. Inside U.S. Trade, July 31, 1998. Internet edition available at

is always optimal is too broad-brushed and unidimensional and does not account for
the complexities of the real world. Adjustment costs from free trade may be too high,
or, in some cases, other countries may pursue industrial and trade policies that are
“unfair.” Governments may subsidize certain industries, maintain trade barriers, or
allow companies to engage in dumping or other unfair trade practices. Most
economists contend, however, that retaliation for trade barriers and protectionism
abroad usually ends up hurting U.S. consumers rather than solving the problem.
Economists acknowledge, however, that traditional trade models do not describe
or predict accurately trade flows in the real world, particularly in terms of
manufactured products. The models either ignore or provide unsatisfactory treatment
of five major deviations between the real world and the theoretical models. These
include: (1) increasing returns to scale (larger firms can produce goods at a lower per
unit cost than smaller firms), (2) learning-by-doing (comparative advantage can be
created as economies learn how to produce certain products and entry costs are high
for latecomers), (3) external effects (benefits exist from research and development and
technological change that cannot be captured solely by those directly involved, plus
a concentration of industrial processes — such as in Silicon Valley or Detroit —
produces benefits for other firms and society at large), (4) inter-firm strategic rivalries
(in contrast to the pure competition economic model in which all firms are small and
produce identical products, in the real world, the actions of one firm in an industry can
affect the operations of others), and (5) the existence of abnormal profits or losses by
firms for systematic reasons (i.e. for reasons beyond simple randomness or “luck”)
which government policies can influence greatly. The comparative advantage7
approach also does not account for trade barriers, exchange rate manipulation, or
business subsidies by individual countries that may alter the competitiveness of certain
A large body of economic literature has been developed to study these deviations
from the pure theory of international trade. This literature is summarized in the
Handbook of International Economics, edited by G. Grossman and K. Rogoff. An
important conclusion of this literature is that international trade theory is evolving and
some cherished axioms are being abandoned. The pure theory is not repudiated,
rather it is being extended. The critical question is whether or not the above problems
with the pure theory are significant enough to invalidate the analysis. So far, most
economists retain their support of the position that as a rule of thumb, free trade is
most beneficial, but acknowledge that in certain industries the deviations between the
model and the real world are significant enough to alter the standard policy
A complete review of the literature on how trade theory and policy are evolving
is beyond the purview of this paper. Here we examine two areas in which deviations
from the standard model bring new insights into real policymaking. They are intra-
industry trade and strategic trade.

Brander, James A. Strategic Trade Policy. In Handbook of International Economics,7
Vol. III, Edited by G. Grossman and K. Rogoff. New York, Elsevier Science, 1995. P. 1397-

98, 1447.

Intra-industry Trade
The theory of comparative advantage goes a long way in explaining why one
country specializes in producing one product which it then trades for another. Saudi
Arabia, for example, trades petroleum for aircraft. What the theory of comparative
advantage does not explain, however, is why Europe buys aircraft, automobiles, and
machine tools from the United States, while the United States buys the same products
from Europe. This is intra-industry trade or trade in products within the same
industry (e.g., American Jeeps for German BMWs) and contrasts with inter-industry
trade which is trade between countries for products of different industries (e.g.,
American power generators for Chinese-made shoes).
Intra-industry trade arises partly because of economies of scale in production.
The average cost of manufactured products tends to fall as the units produced
increase. If one country tries to produce the full variety of products that consumers
demand, it will be able to do so only at relatively high costs. Companies often need
extensive production runs to be competitive with respect to price. They produce
large quantities of certain product lines for sale in both domestic and export markets,
but they are not able to produce each type of the product (intellectual property rights
also come into play). The conclusion of traditional comparative advantage analysis
that, for example, Europe could produce all high-fashion goods, the United States all
automobiles, and Japan all consumer electronics makes way for a more complex
system of intra-industry trade in which all three economies produce and trade with
each other for all three products. This is the two-way trade of products within the
same industry, and it counters the idea that countries should abandon entire industries
that may appear uncompetitive at some time.
This intra-industry analysis may provide, for example, a framework to analyze
whether a government-backed loan to a company (i.e. the Chrysler corporation in the
1970s) is economically efficient. This extension of trade theory, however, does not
address the size of the U.S. trade deficit.
Strategic Trade Theory
Another response to the criticisms of the traditional approach to trade has been
to develop a theory of “strategic trade” in which governments play a role in fostering
the development of strategic industries and pursue trade policies that condition or
alter strategic relationships among firms. The primary strategic relationship in which
government intervention might be justified is interdependence: the profits of one firm
are directly affected by the individual strategy choices of other firms. In economic
models, one way to account for this interdependence is to introduce oligopolies
(industries dominated by a few firms producing differentiated products) or imperfect
competition into trade policy analysis. The appeal of this approach is that it allows
trade theory to address some of the practical concerns that dominate the political
debate over actual trade policy.
The policy implications of the strategic trade approach tend to justify
government intervention in favor of “strategic industries.” Intervention might include
subsidies for research and development, military contracts to develop dual-use

technology, or export support for commercial aircraft. The strategic approach also
helps to understand how apparently modest government intervention can have large
effects. If a comparatively small subsidy determines whether a foreign or domestic
firm enters an industry in which there are significant learning requirements and
economies of large-scale production, a sizable long-run impact can arise if the
domestic industry eventually becomes viable. The approach also helps explain how
trade and industrial policies have had a major role in influencing the current
international pattern of specialization and trade. It further provides a theoretical8
rationale for deviating from the policy of free trade.
Strategic trade policy prescriptions, however, pose problems of their own. Many
economists still are not convinced that the risks involved in implementing them justify
the deviation from free trade. Selecting a strategic industry is difficult in a political
system in which traditional industries (such as steel, automobiles, petroleum, and
textiles) have major influence on national economic policy. A true strategic industry
may be small and not represented well in political power centers. Strategic trade also
can be used as a rationale for protection of industries that may be strategic primarily
in terms of size or political clout but not with respect to the economic definition of the
Studies of strategic trade also indicate that even nationally successful strategic
trade policies have a “beggar-thy-neighbor” aspect; that is, one country’s strategic
industries may advance at the expense of those of another. This may induce other
countries to retaliate and attempt to neutralize the gain. Also, countries that compete
with each other with strategic trade policies tend to make agreements to ameliorate
or prevent such rivalries. The United States and other industrialized nations, for
example, have multilateral trade agreements curbing excessive use of subsidies and
other government assistance to industries.
Strategic trade policies, moreover, are not aimed directly at the U.S. trade
deficit. Whether a deficit exists or not does not change the rationale for a strategic
trade policy.
Macroeconomic Theory
Macroeconomics deals primarily with aggregate flows in an economy. It is
largely separate from considerations of comparative advantage or the actions of
particular microeconomic actors, such as business production or consumer choice.
It is in macroeconomics that the problem of the level of the trade deficit is addressed.
The macroeconomic interpretation of trade and current account deficits is that
they are caused by an imbalance in domestic saving and investment. Businesses,
consumers, or governments do not save enough to finance domestic investment, so
they use foreign capital to bridge that gap. They borrow from abroad to finance

Ibid., p. 1446-1447.8

investments or deficit spending. This implies that the relative level of trade flows is9
determined outside of specific trade policies that are aimed at imports or exports —
including strategic trade policy. Trade policies may change particular product flows
but are unlikely to change their aggregate level.
According to macroeconomic theory, for example, a policy to restrict imports
of steel may cause foreign steel inflows to decline, but the overall balance of trade
would not change. An import barrier causing a fall in U.S. imports of steel would
have two major effects: exchange rate appreciation and a multiplier effect on the
domestic economy. A decline in steel imports would first push up the value of the
dollar as fewer Japanese yen, Korean won, and other foreign currencies were required
to buy foreign steel. This dollar appreciation would make imports into the United
States relatively cheaper and U.S. exports more expensive for foreign buyers. The
second effect would be a domestic demand stimulus as steel consumers bought less
steel from abroad and more from local producers. This would be multiplied through
the domestic economy raising incomes and economic activity. Higher economic
activity would induce more imports. The combination of these two effects would
eventually bring international trade flows back to the previous level determined by the
balance of savings and investment.
The same logic applies to export promotion or even an increase in exports that
may occur for any of a variety of reasons. Paul Krugman, one of the foremost
international trade economists, for example, states the following:
...[C]onsider a national economy that finds one of its major exports
growing rapidly. ...[T]here will be strong negative feedbacks from the
growth of that export to employment and exports in other industries.
Indeed, those negative feedbacks will ordinarily be so strong that they will
more or less completely eliminate any improvements in overall employment
or the trade balance.10
The policy implications of a macroeconomic approach to the trade deficit,
therefore, is that trade policy measures have no effect on the overall balance of trade,
although they can influence specific flows of imports or exports or change the
distribution of the trade deficit among trading partners or among specific traded
goods. According to this approach, trade policy may do many things, but it is unlikely
to alter the aggregate size of a nation’s trade deficit.

The accounting identity that produces this result is derived as follows:9
Income(Y) =Consumption (C) + Business Investment (I) + Government Expenditures (G) +
Net Exports (Exports [X] - Imports [M]) or: Y = C + I + G + (X-M). By an alternative
definition, Income (Y) = Consumption (C) + Taxes (T) + Household savings (S) or: Y = C
+ T + S. Solving the above two equations yields: T + S = I + G + (X-M). Rearranging the
equation yields: (X - M) = (S - I) + (T -G), or the trade deficit equals the savings-
investment shortfall plus the budget deficit.
Krugman, Paul. A Country is Not a Company. Harvard Business Review, v. 74,10
January/February 1996. P. 40-51. Krugman assumes the economy is at full employment.

This macroeconomic approach to the trade deficit is reflected in policy
conclusions, such as the following from the Cato Institute:
Contrary to popular conception, the trade deficit is not caused by
unfair trade practices abroad or declining industrial competitiveness at
home. Trade deficits reflect the flow of capital across international borders,
flows that are determined by national rates of savings and investment. This
renders trade policy an ineffective tool for reducing a nation’s trade11
In recent years, this basic approach to the trade deficit and trade policy has been
extended. Economists have developed new models that attempt to deal with some
shortcomings of traditional analysis such as: (1) the inability of macroeconomic
models to explain movements in actual data, such as the exchange rate, (2) the
inability of macroeconomic models to indicate when exchange rates are over- or
undervalued, and (3) the static nature of macroeconomic models which do not
consider dynamic and intertemporal effects on variables such as productivity and
consumption at different time periods. Recently, the Asian financial crisis has raised
the issue of the sustainability of the trade deficit. The macroeconomic approach also
pays little attention to the time necessary for adjustment, the adjustment path, or12
whether the adjustment is partial or complete.
Problems with the Basic Macroeconomic Approach
The accounting identity underlying the macroeconomic approach to the trade
deficit relies heavily on the equality of savings and investment. The problem with this
is that while the savings-investment identity always holds ex post (with an adjustment
for errors and omissions of several billion dollars), it is less useful in determining
policy and in forecasting future changes.
The economic profession agrees that “savings = investment” is an accounting
identity and not a behavioral equation. The identity captures a final result but does
not show how to get there. It is a framework for analysis, a macroeconomic
relationship that overlays a microeconomic world. It provides a way to sort out the
many forces in an economy that affect trade and to highlight relationships that might
be difficult to grasp otherwise. It also is useful in categorizing the sources of a
savings-investment gap in a country. The identity, however, does not show the
direction of causality. Do levels of savings affect trade, or do levels of trade affect
savings? It also does not provide a mechanism to show exactly how a domestic
savings-investment gap determines international capital flows and the trade deficit.

Griswold, Daniel T. America’s Maligned and Misunderstood Trade Deficit. Cato11
Trade Policy Analysis No. 2, April 20, 1998. On Internet at <
For a simulation of some macroeconomic effects, see: CRS Report 96-301, Effects of12
Trade Policies on the U.S. Trade Deficit and Other Macroeconomic Variables. By Dick K.

A survey of the state of macroeconomic theory with respect to trade policy is
beyond the purview of this report. To catch a flavor of the debate and the policy13
implications of it, however, we will examine three issues that seem pertinent today.
These are foreign exchange rates, the so-called twin deficits, and the trade deficit over
Exchange Rate Effects and Determination. Exchange rates and the role they
play in equilibrating imbalances in trade flows among nations are a key to
macroeconomic trade analysis. Exchange rates are like international prices among
countries. They are a central part of the mechanism by which macroeconomic flows
of goods, services, and capital are brought into balance. In theory, by allowing
exchange rates to adjust, countries can bring their international accounts into balance.
Exchange rates also play a key role in linking capital flows to trade flows.
Consider, for instance, an inflow of capital which is usually considered to cause an
offsetting increase in a country’s trade deficit. How does the process work? The
standard interpretation is that the capital influx increases demand for a country’s
currency which causes the nation’s exchange value to appreciate. The higher value
for the currency makes imports cheaper and exports more expensive to foreign
buyers. The net effect is that imports rise, exports decline, and the trade deficit
increases. (The inflow of capital, if not offset by actions by monetary authorities, also
may lower interest rates which then stimulates demand for products — including
In the real world, however, exchange rates do not always determine the prices
of traded goods and services. The problem is that international markets tend to be
segmented: a product may have different domestic prices in different countries. The
“law of one price” does not hold in most cases. Prices for commodities, such as14
petroleum and coal, tend to be the most similar across countries, while prices for
sophisticated manufactures or for services tend to be the least similar. If different
prices for the same product can be charged in various countries, then an exchange rate
change will not necessarily induce the expected response in import (and export) flows.
On brand-name products, in particular, the pass-through of exchange rate changes is
usually incomplete.
For example, between January 5, 1994 and April 19, 1995, the Japanese yen
appreciated by 34% against the dollar as it rose from 113 to 80 yen per dollar. Prices
for exported products from Japan to the United States should have risen significantly,
but, for example, the U.S. sticker price of a Toyota Celica ST Coup rose by only 2%
(it went from $16,968 to $17,285), while the suggested retail price of a large-screen
Sony Trinitron television receiver actually fell by 15%. Between 1982 and 1985, the

For more detail, see: Grossman, G. and K. Rogoff, eds. Handbook of International13
Economics, Vol. III, New York, Elsevier Science, 1995.
The law of one price states that under perfect competition (and ignoring transportation14
costs) the price of an identical product must be the same across markets.

dollar appreciated by 21%, but U.S. import prices fell by only 4%. Between 1985 and

1988, the dollar depreciated by 44%, while U.S. import prices rose by only 10%.15

For shipments to the United States, economic studies have found that, on
average, an exchange rate change induces a price response equal to one-half the
amount, although it varies by industry. Measures of trade barriers, moreover, tend
to play no role in explaining the differences in prices across markets. An implication16
of this lack of response of domestic prices to exchange rate changes is that a currency17
depreciation will not necessarily eliminate a nation’s trade deficit, although empirical
studies indicate that for most countries over the long run, a real depreciation
(adjusting for domestic inflation) is likely to improve a nation’s current account
balance while a real appreciation is likely to worsen it. In the short-run, however, the18
opposite is likely to occur.
For practical purposes, the lack of response in flows of imports and exports to
an exchange rate change means that the exchange rate will often overshoot the
amount of change necessary to bring trade and capital flows back into balance. This
introduces more distortions and inefficiencies into the economy. An extreme case
would be the situation in Indonesia in which its trade gap was not closed despite a
drop in the value of its currency from 2,500 rupiah per dollar in July 1997 to about

14,000 per dollar in July 1998.

A further problem with exchange rates and the macroeconomy is that the link
between the two still has not been sufficiently quantified. If the savings-investment
relationship determines the trade deficit, and one of the mechanisms by which the
trade deficit changes in response to macroeconomic variables is through the exchange
rate, then economists should be able to establish linkages between exchange rate
movements and underlying macroeconomic conditions such as relative interest rates
and the money supply.
The economic literature on this point, however, is less than persuasive. Several
economic models have been estimated that seek to explain or forecast monthly or
quarterly exchange rates with traditional observable macroeconomic fundamentals but
have generated few encouraging results. In the words of a reviewer of this literature
for the Handbook of International Economics,

Krugman, Paul R. and Maurice Obstfeld. International Economics, Theory and15nd
Policy, 2 edition. New York, HarperCollins Publishers, 1991. P. 454.
Goldberg, Pinelopi Koujianou and Michael M. Knetter. Goods Prices and Exchange16
Rates: What Have We Learned? Journal of Economic Literature, Vol. 35, September 1997.
P. 1244, 1270. Note: Two national markets are segmented if buyers in those markets face
systematically different common currency prices for the same product. (P. 1245)
Ibid. P. 1248.17
In order for a real depreciation to improve the current account, exports and imports18
must be sufficiently elastic respect to the real exchange rate. This condition holds for most
industrialized countries for trade in manufactured goods in the long run but not in the short
run. Krugman and Obstfeld, International Economics, p. 450, 468.

“The dispiriting conclusion is that relatively little explanatory power is
found, and the models contain little forecasting ability compared to very
simple alternatives. Existing structural models have little in their favor
beyond theoretical coherence. Positive results, when they are found, are
often either fragile or unconvincing in that they rely on implausible
theoretical or empirical models. For these reasons, we, like much of the
profession, are doubtful of the value of further time-series modeling of
exchange rates at high or medium frequencies using macroeconomic19
In commenting on another study, the same reviewers concluded that
“macroeconomic variables cannot be very important determinants of exchange rate
volatility.... There appears to be a growing general consensus for this conclusion; it
is the rule rather than the exception that large movements in exchange rates occur in
the absence of plausible or detectable macroeconomic events. The theory, however,20
appears to do better with annual changes and in predicting long-term trends.
In view of volatility in exchange rates in the 1997-99 Asian financial crisis, more
work on exchange rate determination is being conducted. The current trend is to
view exchange rates in terms of microeconomic analysis — not unlike that used to21
analyze stock prices — but this approach is still primarily a research agenda. What
everyone does seem to agree upon is that exchange values (for floating rates) are
determined by supply and demand, and that capital flows rather than trade flows
comprise most of the foreign exchange transactions. Analysis of the exchange rate,
moreover, is complicated by the fact that the U.S. dollar is a key international
currency. Other nations use it as a unit of account, a store of value, and medium of
exchange. They often acquire dollars for purposes other than to purchase U.S.
exports or invest in the U.S. market.
There is little evidence that macroeconomic variables have consistent strong
effects on exchange rates (except during extraordinary circumstances such as
hyperinflations). Without this link between a nation’s macroeconomy and its
exchange rate, part of the adjustment mechanism by which a nation’s saving and
investment affects the trade deficit disappears. We are left with the demand effects
— an increase in exports stimulates the domestic economy which then draws in more

Frankel, Jeffrey A. and Andrew K. Rose. Empirical Research on Nominal Exchange19
Rates. In Handbook of International Economics, Vol. III, Edited by G. Grossman and K.
Rogoff. New York, Elsevier Science, 1995. P. 1705.
Ibid. P. 1707. This observation was based on: Flood, Robert P. and Andrew K. Rose.20
Fixing Exchange Rates: A Virtual Quest for Fundamentals. National Bureau of Economic
Research Working Paper No. 4503, October 1993.
Frankel, Jeffrey A. and Andrew K. Rose. Empirical Research on Nominal Exchange21
Rates. In Handbook of International Economics, Vol. III, Edited by G. Grossman and K.
Rogoff. New York, Elsevier Science, 1995. P. 1709-1710.

imports and offsets some of the original increase in exports. Demand effects alone,
however, cannot account for all movements in trade balances.22
The inability of macroeconomic theory to explain even simple exchange rate
movements also raises questions about the standard interpretation of international
trade policies. If exchange rates do not reflect either underlying trade flows or
macroeconomic variables (such as interest rates), then the standard macroeconomic
conclusion that all export promotion or import protection is negated completely by
market forces is subject to challenge. A country might be able to erect trade barriers
or promote exports without eliciting a major response from exchange markets. The
current state of macroeconomics allows one to state that reducing U.S. imports will
tend to raise the value of the dollar on exchange markets, but whether a combination
of the effects on the exchange rate and domestic demand are sufficient to offset the
reduced imports to keep the balance of trade the same does not seem to have been
established empirically.
In terms of actual trade policymaking, however, U.S. commitments under the
World Trade Organization and other trade agreements preclude unilateral
protectionist measures or excessive export promotion activity. Economic studies also
have shown that import protection does cause higher prices and a narrower choice of
products. This benefits domestic companies manufacturing those products but harms
consumers who might buy them.
The lack of an empirically established link between the exchange rate and
macroeconomic variables also implies that little can be said about whether an
exchange rate is at the “right” level or not. Although observers agree that as
exchange rates rise and fall they tend to overshoot, there is no agreed-upon formula
to determine what the appropriate level should be and when an exchange rate has
dropped too low or risen too high.
Some countries maintain a fixed exchange rate that they determine primarily in
response to forces of demand and supply. Hong Kong, China, and Chile, for example,
intervene aggressively in foreign exchange markets to maintain their fixed rates. They
have doggedly fended off speculators who have bet on a depreciation. They also may
be willing to sacrifice some amount of domestic economic growth to maintain what
they consider their “right” exchange rates. The United States, however, has adhered
to the policy that it intervenes in currency markets only in exceptional cases and will
not sacrifice the domestic economy to protect the exchange rate.23

In order for all increases in U.S. exports to be offset by greater imports caused by22
increased demand, the U.S. import share would have to exceed the inverse of the demand
multiplier. For example, if a $1 billion increase in exports raises a nation’s gross domestic
product (GDP) by $2.5 billion (multiplier of 2.5), then for imports to rise by $1 billion,
imports as a share of GDP must equal at least 40% or 1/2.5 and must respond
proportionately. In the United States, imports of goods and services account for about 13%
of GDP.
A major defense of a weakening exchange rate usually requires raising interest rates23
to attract capital inflows or stop capital outflows. Higher interest rates, however, cause

In June 1998, however, as the yen dropped to 146 yen per dollar, the United
States and Japan did intervene in currency markets to bolster the yen’s value and
weaken the dollar. The weakening yen not only was causing Japan’s trade surplus to
rise but it was imposing greater difficulty on exporters from the Asian economies in
financial crisis, particularly South Korea, whose exports compete with those from
Japan. The weak yen also was putting pressure on China to devalue its currency and24
was hurting U.S. exports to Japan.
In terms of U.S. policy, one may ask just what are the limits to exchange rate
changes that will trigger intervention? Are they established in some rational manner
or are they primarily determined by the “squeal factor” — allowing the dollar to
appreciate or depreciate until businesses that are hurt complain loudly enough to
compel intervention by the U.S. Federal Reserve or Treasury? Should the United
States have an exchange rate policy much like it has a monetary policy?
If other countries regularly intervene in exchange markets, moreover, then the
resulting exchange rates can be partly the result of government policy, although with
floating exchange rates most intervention appears to have only temporary and
psychological effects. One way that governments intervene in foreign exchange
markets is to accumulate foreign exchange reserves. Japan’s foreign exchange
reserves, for example, soared from $98.5 billion in 1993 to $183.3 billion in 1995, and
to $223.7 billion in May 1999. This still is small relative to world capital flows, but25
the increase from 1993 is more than double the size of Japan’s trade surplus with the
United States.
Countries also may regulate capital flows. Some countries restrict outflows of
capital or may cause a surge of foreign borrowing by local institutions (e.g. Thailand
in the mid-1990s) by liberalizing their nation’s capital flows or may induce a surge in
capital outflows by increasing the share of pension funds or other assets that can be
invested in foreign securities (e.g. Japan in 1998).
Figure 4, shows monthly percentage changes in Japan’s official foreign exchange
reserves (including holdings of gold) along with changes in Japan’s yen-dollar
exchange rate from January 1996 to June 1998. The accumulation or disposal of
foreign exchange reserves is functionally equivalent to intervention in foreign
exchange markets. When Japan’s central bank accumulates foreign exchange, it
refrains from converting dollars, marks, pounds, and other foreign currencies into yen.
This weakens the value of the yen and strengthens the value of foreign currencies,
particularly the dollar. (As the volume of Japan’s foreign trade increases, its need for

economic growth to slow.
U.S. Department of the Treasury. Statement by Treasury Secretary Robert E. Rubin24
on Japan. Press Release RR-2522, June 17, 1998. On Internet at <
press/releases/pr2522.htm>. Sanger, David. U.S. Intervenes in Currency Markets to Support
the Yen. The New York Times on the Web, June 18, 1998. On Internet at <http://www.>.
International Monetary Fund. International Financial Statistics. Japan. Ministry of25
Finance. On Internet at <>.

foreign currency reserves also may rise, but the level of Japan’s currency reserves
seems out of line with those of other major trading nations. Germany’s reserves
[including gold] are about $80 billion, the United Kingdom’s about $40 billion, and
the U.S. government’s about $70 billion.)
Figure 4. Monthly Changes in Japan’s Yen-Dollar Exchange Rate and Foreign
Exchange Reserves (Percent)
(X = Intervention to weaken a weakening yen,
O = Intervention to weaken a strengthening yen,
z = Intervention to strengthen a weakening yen.)
Change in Yen-dollar
Exchange Rate
5Weaker Yen
-5 O OO
Stronger Yen
-10Change in
Foreign ExchangeO
-15 Reserves
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3
1996 1997 1998 99
Source: Data from International Monetary Fund, Japan Ministry of Finance
If Japan’s policy is to stabilize its exchange rate, changes in its holdings of
foreign exchange and in the level of its exchange rate should move in opposite
directions. When the yen is weakening, Japan should be selling foreign exchange
(buying yen), and vice versa. If, however, Japan is accumulating foreign exchange at
the same time that its currency is weakening, it is pursuing a “cheap yen” policy. It
is intervening in the market to further weaken the yen. In Figure 4, such policy
episodes predominated during 1996 and into mid-1998. They are indicated by an X
along the line showing changes in exchange rates. When Japan is accumulating
foreign exchange at a time the yen is strengthening, it also is pursuing a policy of
weakening the yen. These episodes are indicated with an O and add to the episodes
in which Japan pursued a “cheap yen” policy. The times are indicated by a z when
Japan was selling dollars to strengthen its yen — which is the policy the United States
was encouraging Japan to take during much of this period,. These episodes occurred
primarily after the onset of the Asian financial crisis (from mid-1997) and primarily
after the value of the yen had dropped to levels that threatened to further destabilize
world markets. Even during the Asian crisis, however, Japan has, at times, reported
net increases in foreign exchange reserves. After intervening in foreign exchange

markets to strengthen the yen by selling $17.8 billion in foreign exchange in April
1998, the Bank of Japan added foreign exchange reserves in ten of the next twelve
months to bring its total holding of foreign exchange back to the level of $223 billion
in April 1999 — the amount Japan held before the April 1998 sales.26
From the standpoint of economic efficiency, excessive volatility in the exchange
rate adds risk, makes investment analysis more difficult, and may reduce economic
growth. For U.S. exporters to Japan, for example, the depreciation of the yen from
80 yen per dollar in 1995 to 146 yen per dollar in June 1998, was equivalent to
Japan’s imposing an 82% tariff on imports from the United States.
Twin Deficits. A focus on the savings and investment identity also may produce
policy prescriptions that can be incomplete. For example, in the 1970s and 1980s, the
United States experienced so-called twin deficits. A sizable volume of economic and
popular literature pointed out that the dissavings of the federal government in the
form of its budget deficit was generating a twin in the form of a rising U.S. trade
deficit. In order to reduce the trade deficit, the policy prescription was that the
United States should look beyond trade policy and put its resources instead into
increasing its savings rate. (This led to a 1989 policy proposal by Japan in trade
negotiations with the United States, for example, that in order to reduce the U.S.27
trade deficit, Americans should curtail their reliance on credit cards.)
How valid is the twin deficit connection? An econometric analysis indicates28
that for industrial countries, budget deficits indeed were positively correlated with
current-account deficits over the 1976-1985 period. Over the 1986-1990 period,
however, that correlation evaporated. In the mid-1990s, for the United States and29
Japan, the two deficits are negatively correlated and moving in opposite directions.
The U.S. budget deficit has shrunk and is turning into a surplus, while the U.S. trade
deficit has been rising. Japan’s budget deficit is rising, while the country runs a
ballooning trade surplus.
Although U.S. government budget deficits do reduce national savings and may
induce U.S. borrowing from abroad, a lower budget deficit does not imply a lower
trade deficit. In addition to federal government dissavings (deficits), private, business,
and local government savings also come into play. The conclusion now seems to be
that the twin deficits certainly are not Siamese, identical, or even fraternal and are
useful for policy purposes only as a generality.

Data on Japan’s Foreign Exchange Reserves are available from Japan’s Ministry of26
Finance on the Internet at <>.
Frankel, Jeffrey A. The SII Outcome: In Whose Best Interest? The International27
Economy, October/November 1990. P. 70.
For a discussion of this issue, see: CRS Report 97-985, Why the Budget Deficit and28
the Trade Deficit Haven't Been Moving Together, by Gail E. Makinen.
Obstfeld, Maurice and Kenneth Rogoff. The Intertemporal Approach to the Current29
Account. In Handbook of International Economics, Vol. III, Edited by G. Grossman and K.
Rogoff. New York, Elsevier Science, 1995. P. 1763.

Intertemporal Savings and Investment. Another area in which the traditional
macroeconomic approach to the trade deficit is being extended is to consider the
effects of international borrowing on variables such as productivity and consumption
at different time periods. If American citizens, for example, are borrowing from
abroad to finance current purchases, they must be doing so according to some
valuation of future versus current consumption.
According to a review of economic literature on the intertemporal approach to
the current account, economic models that fail to integrate investment, saving, and
growth make it virtually impossible to understand persistent current account
imbalances. Why, for example, are the current accounts of Canada and Australia
perennially in deficit while that of Japan is in surplus despite wide swings in their real
exchange rates? The standard macroeconomic approach to trade deficits, moreover,
offers no valid benchmark for evaluating the size of external imbalances. The
International Monetary Fund in its dealings with currency crises considers a current
account deficit that exceeds 5% of gross domestic product to be a warning signal that30
the deficit may be unsustainable. But that observation is based more on experience
than on theory. In practice, policymakers usually just strive to avoid a negative
current account. In considering trade across time periods, however, efficiency often
calls for an unbalanced current account (as explained below).31
One recent approach to this problem is to treat the international borrowing to
fund a current account deficit in terms of standard investment analysis. A current
account deficit represents the borrowing against future output in order to purchase
more goods and services today (which generates a trade deficit). This potentially
could burden future generations with lower consumption in order to repay the debts
being incurred today. In standard cost-benefit analysis, this borrowing would be
justified if it results in enough consumption being made available in the future to offset
borrowing costs (interest rates).
For a nation, rigorous cost-benefit analysis of all international borrowing is not
feasible. Capital flows occur because of both pull and push factors. Although inflows
of capital may be conceptually equivalent to international borrowing, foreign investors
may have their capital “pushed” into the U.S. market for a variety of reasons,
including considerations of political risk, which may be unrelated to differences in
interest rates or other borrowing criteria.
A stylized approach to assessing whether trade deficits are “good” or “bad,”
however, is to examine how a deficit in trade and corresponding imports of capital
(borrowing) are used. If the capital imports are employed to increase productivity
(e.g. used to buy capital equipment) that will bolster economic growth, the capital

Sugisaki, Shigemitsu. The Global Financial System: Status Report. Address at the30

11 Conference of the International Federation of Business Economists, Vancouver,

November 18, 1997.
Obstfeld, Maurice and Kenneth Rogoff. The Intertemporal Approach to the Current31
Account. In Handbook of International Economics, Vol. III, Edited by G. Grossman and K.
Rogoff. New York, Elsevier Science, 1995. P. 1793-94.

imports and corresponding trade deficit may be considered to be “good.” If the32
international borrowing is used to finance current consumption, it is considered to be
“bad” (in the sense that future consumption will be lower than it otherwise would be).
The logic of this analysis is similar to that for an individual. If the individual is
borrowing to finance investments in productive assets (e.g. an education, automobile
for commuting to work, or preventive medical care) that tend to increase productivity
and earned income, it may be considered to be “good.” However, if the individual is
borrowing merely to increase current consumption (e.g. vacation travel, clothes, or33
entertainment), it may be considered to be “bad.”
Figure 5. U.S. Imports of Merchandise by Major End-use Category,
Consumer Goods24%
600 16%
Automobiles & Parts
Capital Goods except Autos
Industrial Supplies & Materials22%
0Food, Feeds, Beverages4%
1990 1991 1992 1993 1994 1995 1996 1997 1998
As shown in Figure 5, U.S. imports of capital goods (excluding automobiles)
have been rising. In 1998, they accounted for 30% of total merchandise imports, up
from 23% in 1990. Imports of automobiles, some of which are capital goods, but
most of which are consumer durables, eased somewhat from 18% to 16%, while
consumer goods grew from 19% to 24%. In dollar amounts, imports of capital goods
(excluding automobiles) rose by $154 billion (from $116 billion in 1990 to $270
billion in 1998), while the U.S. current account deficit rose by $141 billion (from $92
billion in 1990 to $233 billion in 1998). In the 1990s, therefore, imports of capital

In a strict sense, the rate of return on the investment should exceed the interest paid32
on capital borrowed from abroad.
Lippens, Robert E. Are Trade Deficits That Bad? Business Economics, April 1977,33
vol. 32. P. 14-19. See also: CRS Report 98-508 E, U.S. Trade Balance: An Analysis of
Recent Trends and Policy Options, by Craig K. Elwell.

goods have increased by more than the current account deficit, and capital goods
imports have been growing both absolutely and as a percent of total merchandise
imports. In this sense, the capital borrowing from abroad seems to have been used
to finance productivity-enhancing investments. Consumption imports, however, still
are quite large.
The intertemporal approach to the current account deficit does provide one
policy implication that most developing nations have implemented. That is, if tariffs
are imposed they should fall on consumption and not capital goods.
This approach may indicate to policymakers whether a country is “squandering”
the capital being imported to purchase current consumption goods, but it does not
provide policy tools to determine the optimal size of a nation’s current account
deficit. The United States may be borrowing to enhance future economic growth, but
are we becoming too reliant on foreign capital? The intertemporal approach has not
yet developed to the point that it can indicate to policymakers whether or not a
chronic current account deficit can be sustained.
Sustainability of the Trade Deficit
Is the U.S. current account deficit sustainable? Is the United States likely to
encounter a situation in which investors lose confidence in dollar-denominated assets
and dump them thereby causing a rapid depreciation in the value of the dollar and a
currency crisis similar to that experienced by Mexico in 1995 and countries of Asia
in 1997 and 1998? This depends on whether the U.S. economy will be able to
generate the resources necessary to repay the funds borrowed from abroad.
First, it should be noted that even though the U.S. trade deficit has reached the
$200-billion level, the current account deficit is less at $155 billion primarily because
of the U.S. surplus in services trade. As a percent of total U.S. consumption,
moreover, the trade deficit at about 3% is down from the 5% in the mid-1980s. The
current account as a percent of gross domestic product also is comparably lower. The
U.S. current account deficit is not currently at the 5%-of-GDP level considered to be
a danger sign in international financial markets.
Second, it should also be noted that the foreign investments in the United States
have grown to the point that in 1997 foreign investors in the U.S. market earned $5.3
billion more than Americans earned on their investments abroad.
Concern over the sustainability of the U.S. current account deficit was recently
stated by Alan Greenspan, the Chairman of the Federal Reserve:
A more distant concern, but one that cannot be readily dismissed, is the
very condition that has enabled the surge in American household and
business demands to help sustain global stability: our rising trade and
current account deficits. There is a limit to how long and how far deficits
can be sustained, since current account deficits add to net foreign claims on
the United States.

It is very difficult to judge at what point debt service costs become unduly
burdensome and can no longer be sustained. There is no evidence at this
point that markets are disinclined to readily finance our foreign net
imbalance. But the arithmetic of foreign debt accumulation and
compounding interest costs does indicate somewhere in the future that,
unless reversed, our growing international imbalances are apt to create
significant problems for our economy. 34
In essence, whether the United States can sustain its current account deficit over
the foreseeable future depends on whether foreigners are willing to increase their
investments in U.S. assets. In short, the current account deficit puts the economic
fortunes of the country partially in the hands of foreign investors. Foreign investment
in the United States takes two primary forms: direct investment necessary to do
business in the U.S. market and portfolio investment to take advantage of higher rates
of return in the United States. The latter includes short-term speculative funds that
are seeking higher quality markets because of instability abroad and bank deposits.
As shown in Figure 6, net inflows of capital into the United States have more
than offset the recent U.S. current account deficit. Net capital flows and the current
account tend to mirror each other. (In 1997, net capital inflows were $255 billion
while the current account was only -$155 billion because of a -$100 billion statistical
discrepancy. This discrepancy has trended toward larger negative numbers. It went
from $25 billion in 1989 to $1 billion in 1994, to -$60 billion in 1996, and -$100
billion in 1997.)
Capital inflows include official reserve assets, other government assets, and
private assets. Private assets dominate the capital flows. In 1997, of the total $733.4
billion in capital inflows, $717.6 billion was in private assets (of which $93.4 billion
was in direct investments, $171.5 billion in Treasury securities and currency, $196.8
billion in U.S. private sector securities, $107.8 billion in nonbanking assets, and
$148.1 billion in banking assets). For capital outflows the story is similar. Of the
total $478.5 billion in capital outflows, private assets accounted for $477.7 billion (of
which $121.8 billion was in direct investments, $88.0 billion in foreign securities,
$120.4 billion in other nonbanking assets, and $147.4 billion in banking assets).35

Greenspan, Alan. The American economy in a World Context. Remarks at the 35th34
Annual Conference on Bank Structure and Competition of the Federal Reserve Bank of
Chicago, Chicago, Illinois. May 6, 1999. On the Internet at <
U.S. Bureau of Economic Analysis. Survey of Current Business, July 1998. P. 68-35


Figure 6. Net Capital Flows (Government and Private) and Current Account
Balances for the United States, 1965-97
300 $255
Net Capital Flows
Current Account
65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 97 99 01
66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02
Source: U.S. Bureau of Economic Analysis. Forecast by Standard & Poor's DRI
In terms of private assets, foreign direct investment inflows and outflows tend
to offset each other. In 1997, the inflow of foreign direct investment of $93.4 billion
was more than offset by U.S. direct investment abroad of $121.8 billion. In terms of
financing the U.S. trade deficit, direct investment is attractive because it is more stable
than portfolio investment. Foreign companies and actual plant and equipment in the
United States are here somewhat permanently Foreign owners cannot easily move
them out of the U.S. market.
Portfolio investment in securities and equities is much more volatile and depends
on relative rates of return, behavior of stock markets, relative interest rates, economic
stability, and other variables. In the mid-1990s, the booming U.S. stock market
combined with relatively high interest rates and currency instability in Asia to attract
considerable capital to the United States. (In mid-June 1998, the U.S. three-month
money market rate was 5.6% as compared with 0.4% in Japan, 1.9% in Switzerland,
3.6% in Germany, 3.8% in France, and 5.1% in Italy.) For now, U.S. capital markets
are appealing, but as Asian countries discovered during their financial crises, capital
can be fickle. It can flow out as fast as it flows in.
How much longer the United States can attract $150 to $250 billion per year in
net capital inflows in order to finance its trade deficit is an important question. Net
foreign purchases of U.S. Treasury securities have been running at an annual rate of
about $150 billion in 1996 and 1997. This almost equals the U.S. current account
deficit. According to one analyst, the continuation of the current account deficit and
strong dollar — and, hence, of the sequence of happy non-inflationary years with

domestic demand growing faster than the trend rate of GDP growth — depends on
foreigners’ willingness to purchase U.S. Treasuries at a rate equal to about 2 percent
of the American gross domestic product. The author concludes that recent trends in
the American balance of payments, and — more specifically — in the financing of the
current account deficit, are “clearly unsustainable.” He reasons that foreign and
international investors already own such as high percentage of the U.S. national debt
that they will not continue purchasing it at current levels. He warns that in a decade
foreign investors could end up owning all the U.S. national debt36
One factor working in the favor of the United States in terms of capital flows is
that the dollar has become a key world currency. It is used abroad for the same
reasons that it is used at home: as a unit of account, a medium of exchange, and a
store of value. Foreigners are willing to purchase U.S. paper currency (which pays
no interest) for use in international transactions. As of 1996, the U.S. Treasury
estimated that $209.6 billion in U.S. currency (primarily $100 notes) was circulating
abroad. This amount was increasing by about $15 billion per year and exceeded the
currency circulating in the United States. In other words, the United States has been
exporting paper currency in exchange for claims on real goods and services abroad.
The use of dollars as a key currency, moreover, means that American foreign
borrowing can be denominated in dollars rather than in a foreign currency. Even if
the value of the dollar falls on exchange markets, the dollar amount of U.S. borrowing
does not change. This is quite a different situation from that of Indonesia, South
Korea, or Mexico. As the value of their currencies fell, the amount of their foreign
debt, which was denominated in dollars and other foreign currencies, rose.
Foreign and international investors also invest their capital in U.S. equity and
property markets. Over 1996-97, U.S. equity markets attracted between $100 billion
and $200 billion per year from abroad. In the future, whether these markets can keep
pulling in such amounts, plus absorb more of the funds that have been flowing into
Treasury securities, is an open question. Much of the recent influx of foreign capital
into Treasury securities, U.S. equity markets, or even U.S. bank accounts has been
induced by the Asian financial crisis and threat of contagion in Eastern Europe and
Latin America. Finance capital has been leaving other countries because of instability
there. There has been a “flight to quality” despite countermeasures by countries, such
as South Korea, that have raised domestic interest rates to levels as high as 20 to
40%. Once the Asian financial crisis eases, however, some of this capital likely will
flow back to these Asian and other nations.

Congdon, Tim. Treasury Warning. The International Economy, July/August 1998.36
P. 34-35. (Actually, foreign purchases of U.S. Treasury securities plus currency amounted
to about 2% of U.S. gross domestic product in 1996 and 1997.)

Figure 7. U.S. Foreign and Federal Debt and Foreign Assets
in the United States, 1991-2002 (forecast)
8 40
Share of U.S. Federal Debt held
7by Foreigners (right scale)
6 30
Foreign Assets
420In the U.S. (left
U.S. Federal
210Debt Privately
Held (left scale)
U.S. Foreign Debt
00(Left scale)
91 92 93 94 95 96 97 98 99 00 01 02
The Standard and Poor’s DRI projection of the U.S. current account deficit and
net inflows of capital are both at about $230 billion from 1999 to the year 2002. As
shown in Figure 7, this implies an increase in the foreign indebtedness of the United
States of about a trillion dollars every four or five years. The foreign debt (net
international investment position of the United States with the sign reversed) has
already risen from $355 in 1991 to $1,328 billion in 1997. Following the Standard
and Poor’s DRI forecast, this position would deteriorate to $2,131 billion in the year
2000 and to $2,543 billion in the year 2002. This implies that U.S. foreign debt as a
percent of U.S. gross domestic product would have risen from 16.4% in 1997 to
about 25% in 2002. Foreign assets in the United States likewise have already surged
from $2,487 billion in 1991 to $4,993 billion in 1997. If current trends continue, they
are forecast to reach $6,745 billion by 2002 (or about 67% of the $10,076 billion
forecast for U.S. GDP).
One question is where this inflow of foreign investment would be directed. The
total U.S. privately held federal debt rose to $3,393 billion in 1997, but its level has
stabilized as the federal budget is beginning to generate a surplus. Purchases of
federal securities by foreigners, however, have continued. As a percent of total
federal debt privately held, the share held by foreigners rose from 19% in 1991 to
38% in 1997. If foreign holdings of U.S. federal debt continue to rise at $100 to
$200 billion per year, by the year 2002, between 47% and 67% could be held by
foreigners. The question is whether this level would be acceptable to the American
people or to foreign investors and lenders. For Americans, this is primarily a political
question. For foreign investors, this is primarily an economic question.

Looking forward into the next century, it is unlikely that net capital inflows into
the United States will continue at their unusually high rates of 1996-98. The Standard
and Poor’s DRI forecast merely extrapolates current trends. If the high rates of
capital inflows do not continue, the U.S. current account deficit also will have to
decline. Put another way, if the United States is able to reduce its trade deficit, its
borrowing needs also would decline.
Trade and Jobs
Considerable controversy exists as to the effect of imports and exports on
American jobs and wages. Most mainstream economists assert that international trade
results primarily in reallocation of jobs from less efficient to more efficient industries
rather than particular job gains or losses for the economy as a whole. This argument
was recently stated by Alan Greenspan as follows:
It is difficult to find credible evidence that trade has impacted the level
of total employment in this country over the long run. Indeed, we are
currently experiencing the widest trade deficit in history with a level of
unemployment close to record lows.
Certainly, the distribution of jobs by industry is affected by
international trade, but it is also affected by domestic trade. It is the
relative balance of supply and demand in a competitive market economy
that determines the mix of employment. When exports fall or imports rise,
domestic demand and relative prices have invariably adjusted in the long
run to leave total employment relatively unaffected. As economists like to
say, all imports are eventually paid for with exports. 37
Other studies, however, indicate that the trade deficit may have an impact on
U.S. potential employment. One study, for example, concluded that because of the
Asian financial crisis, many of the Asia countries affected are attempting to export
their way out of their problems. The resulting $100 billion increase in the U.S. trade
deficit could destroy 1.1 million U.S. jobs. The losses would occur in every state,
mostly in the manufacturing sector, where, even if interest rates are lowered enough
to stave off a jump in unemployment, 600,000 jobs would be shifted from the
high-wage manufacturing to the lower-paying service sector. 38
As for the North America Free-Trade Agreement, a 1998 study of its effect on
employment over the four-and-one-half years after it had been implemented concluded
that NAFTA had primarily served to accelerate trade, plant relocation, and sectoral
job gains and losses that were already ongoing. Approximately 191,000 primary jobs
had been certified by the U.S. Department of Labor in 1,638 plants as potentially

Greenspan, Alan. Trade and Technology. Remarks before the Alliance for the37
Commonwealth, Conference on International Business, Boston, Massachusetts. June 2, 1999.
On Internet at <>.
Scott, Robert E. and Jesse Rothstein. American Jobs and the Asian Crisis: The38
Employment Impact of the Coming Rise in the U.S. Trade Deficit. Washington, Economic
Policy Institute, 1998. 13 p.

threatened by increased imports or plant relocations to Mexico or Canada. Apparel
and electronics accounted for 40% of the NAFTA certifications. These potential job
losses, however, were balanced by an estimated 680,000 net job gains in the U.S.
economy from increased exports to Mexico and Canada since NAFTA took effect.
This study did not, however, attempt to estimate net job gains foregone arising from
increased imports from Canada and Mexico because there can be no net job losses as
long as U.S. output and employment continue to rise. Another study, however,39
estimated that over the first three years of NAFTA, the reduction in net exports to
Canada and Mexico had eliminated 167,172 American job opportunities, and in total,
NAFTA had resulted in a net loss of 394,835 jobs. This area of job losses and gains40
from trade agreements seems to be one in which data and methods of analysis are still
being developed.
This brief survey of the U.S. trade deficit and the theory and policies surrounding
it indicates that this field of economics is evolving. Recently, some basic economic
theory has been called into question as empirical studies fail to validate long accepted
theoretical linkages. The policy implications with respect to trade have developed
from the prescription that free trade is optimal to one that allows, for example, for
strategic trade policy aimed at assisting certain industries — particularly those in high-
technology. A strategic trade policy, however, comes with a new set of pitfalls.
The macroeconomic approach to explaining the trade deficit is encountering
some obstacles. Many seemingly straightforward theoretical linkages, for example,
have not been empirically established. In particular, studies have not been able to link
macroeconomic conditions to short- and medium-term changes in a nation’s exchange
rate. Also, studies have found that changes in exchange rates do not translate
completely into price changes for imports and exports. This implies that in order for
exchange depreciation to eliminate a trade deficit, it must overshoot the equilibrium
rate by quite a margin. This brings other inefficiencies into an economy.
Many nations regularly intervene in foreign exchange markets and influence the
values of their currencies. Recent weakness in the Japanese yen, for example, has
been exacerbated by the Japanese government’s intervention into exchange markets.
The United States, however, has no announced exchange rate policy nor does it have
a transparent method of determining when it should intervene in currency markets.
The inability of macroeconomic theory to explain even simple exchange rate
movements raises questions about the standard interpretation of international trade
policies. If exchange rates do not reflect either underlying trade flows or
macroeconomic variables (such as interest rates), then the standard macroeconomic
conclusion that export promotion or import protection is negated completely by

CRS Report 98-783 E. NAFTA: Estimates of Job Effects and Industry Trade Trends39
After 4 ½ Years, by Mary Jane Bolle.
Rothstein, Jesse and Robert Scott. NAFTA's Casualties, Employment Effects on40
Men, Women, and Minorities. Economic Policy Institute Issue Brief No. 120. September 19,


market forces is subject to challenge. A country might be able to erect trade barriers
or promote exports without eliciting a major response from exchange markets. U.S.
commitments under the World Trade Organization and other trade agreements,
however, preclude unilateral protectionist measures or excessive export promotion
The demand effects of import protection or export promotion, however, are well
established in empirical studies of the economy. Trade policies that reduce imports
or increase exports provide a stimulus to the domestic economy that usually causes
domestic production to rise. This, in turn, brings in more imports which offset, to
some extent, the reduction in imports or increase in exports resulting from the trade
policies. Also well established in economic studies are the benefits to competing U.S.
producers and harm done to the American consumer from higher prices and narrower
choice resulting from import protection policies.
The macroeconomic approach also is being extended to account for why the
United States may be borrowing from abroad and how those funds are being used.
Trade data indicate that a rising share of U.S. imports has been of capital goods that
should increase U.S. productivity and growth in the future. Treating foreign
borrowing with the same methods used to evaluate domestic borrowing might
determine an optimal level for the U.S. current account deficit. This has not yet been
done, however.
In terms of sustainability, it is difficult to see that the historically large capital
inflows of 1996-1998 will continue into the next century. As the Asian financial crisis
ebbs and world economies recover, some of the capital that has sought safety in U.S.
markets likely will return abroad. If so, the U.S. trade and current account deficits
of today will have to diminish. Put in opposite terms, if the U.S. trade deficit is
reduced, the need for foreign borrowing also will decline.
The ability of the United States to continue to incur trade and current account
deficits at the $200 billion level depends on the willingness of foreigners to lend and
invest funds in the American market. The deficit, currently at roughly 3% of gross
domestic product, is not at the danger-signal level of 5% of GDP used by the
International Monetary Fund. But whether foreigners will continue to move capital
to the American market, even if the foreign share of the U.S. federal debt rises to
exceed 50% or if the level of foreign assets in the U.S. market increases to the
equivalent of two-thirds of U.S. GDP, is both a political question for U.S.
policymakers and an economic question for foreign investors. A further issue is
whether net U.S. payments to foreign investors will continue to rise enough above
their $22.5 billion in 1998 and whether they will offset capital inflows or become a
burden to the U.S. economy.
Considerable controversy exists as to the effect of imports and exports on
American jobs and wages. Mainstream economists assert that international trade
results primarily in reallocation of jobs from less efficient to more efficient industries
rather than particular job gains or losses for the economy as a whole. Other studies,
however, document potential job losses in specific industries and point out that a
rising trade deficit induces a shift of jobs from relatively high-wage manufacturing
industries to the relatively lower-paying service sector