Statement of
Peter R. Orszag
Director

Foreign Holdings of U.S. Government
Securities and the U.S. Current
Account

before the
Committee on the Budget
U.S. House of Representatives

June 26, 2007

Chairman Spratt, Ranking Member Ryan, and Members of the Committee, thank you for inviting me to testify today. Foreign holdings of U.S. Treasury debt have grown rapidly in recent years and now are a significant percentage of such debt held by the public. A broader issue is the substantial deficit in the U.S. current account—which summarizes the country’s current transactions with the rest of the world, including trade in goods and services, net income from international investments and the compensation of employees, and net unilateral transfers (such as gifts, pension payments, and foreign aid). The mirror image of the nation’s large current-account deficit is foreign investors’ increased holdings of a variety of claims on the United States, including U.S. government debt as well as private-sector securities and assets.

My testimony today makes four main points:

• Foreign holdings of U.S. Treasury debt have risen rapidly. Between 2003 and 2006, for example, such holdings rose almost 50 percent. They now exceed $2 trillion and account for more than 40 percent of Treasury debt held by the public.
  
  
• Those increasing foreign holdings of U.S. government debt are part of a more fundamental issue: The nation is running a substantial current-account deficit, which is financed by increasing liabilities to and assets held by foreign investors. The current-account deficit measures the excess of the country’s spending over its income or, equivalently, of its domestic investment over its national saving. After the depreciation of physical capital is taken into account, the nation saved only 2 percent of its income last year, an unusually low level for the world’s leading economy. At the same time, the nation’s net domestic investment was 8 percent of its income. The difference, 6 percent of income, was financed by increases in net foreign claims on the United States and manifested itself in the current-account deficit.
  
  
• Economists generally agree that the nation’s current-account deficit cannot be sustained indefinitely at its current level relative to gross domestic product (GDP) because the nation’s indebtedness to the rest of the world will grow faster than its income. Moreover, foreign investors will not continue to be willing to purchase U.S. claims at current rates of return indefinitely as their portfolios become more and more concentrated in such assets. To be sure, views differ on whether a future adjustment in the current-account deficit will occur gradually or suddenly—but there is little disagreement that some sort of adjustment is inevitable.
  
  
• The necessary adjustment of the current-account deficit, which requires slower growth of consumption in the future, could take place slowly or rapidly. The more likely scenario appears to be a gradual adjustment without severe short-term economic consequences, but a sudden adjustment remains a risk—and possibly a growing risk as foreign net holdings of claims on the United States rise as a percentage of GDP. Policymakers can help facilitate the necessary reduction in the current-account deficit and reduce the risk of a severe economic disruption in foreign financing by taking actions to raise the rate of national saving.

Estimated Holdings of U.S. Government Securities by Foreign Investors

Foreign holdings of U.S. Treasury securities have grown rapidly in recent years. In 2003, for example, U.S. Treasury securities held by foreign investors amounted to $1.45 trillion, and by 2006, those holdings rose to $2.13 trillion—an increase of 47 percent.¹ As a percentage of total Treasury debt held by the public, foreign holdings rose from 37 percent to 44 percent over that span.² The increase in foreign holdings accounted for about 86 percent of total federal borrowing last year and about 72 percent from 2003 to 2006.

According to survey estimates, East Asian countries held a large share of foreign holdings of Treasury securities last year—about 63 percent.³ The two East Asian countries with the largest holdings were Japan, which held an estimated 31 percent of all foreign-held Treasury securities, and mainland China, with 19 percent. In comparison, the European Union held an estimated 15 percent, and oil-exporting countries in the Middle East, about 5 percent.

Foreign official institutions have played a significant role in the increase in foreign ownership of federal debt. Indeed, at the end of 2006, foreign central banks owned 66 percent of all federal debt held by foreign residents, up from 63 percent at the end of 2005.

The data on ownership by country and by type of foreign entity (official versus private) are imperfect.⁴ The surveys used to collect the data do not always capture the ultimate owner of the securities. If an owner entrusts securities with a custodian in a different country, for example, the ownership of the securities is attributed to the country of the custodian, not the owner. That “custodial bias” contributes to the large recorded foreign holdings of U.S. securities in major financial centers such as Belgium, the Caribbean banking centers, Luxembourg, Switzerland, and the United Kingdom.⁵ Similarly, some foreign official purchases may be misclassified as foreign private ones because they are conducted through private-sector traders.

Foreign investors also hold a growing share of securities of U.S. agencies and government-sponsored enterprises (GSEs), evidently reflecting a drive to increase the returns on their investments. At the end of 2006, those investors owned about $1.2 trillion of such securities, more than twice as much as in 2001. The countries with the largest holdings were China, with about 23 percent of all such foreign holdings, and Japan, with about 17 percent.

Examining only the securities of the Treasury Department and of agencies and the GSEs that are held by foreign investors, however, obscures the broader and more fundamental issue: the rising net foreign claims on the United States that result from the nation’s current-account deficit. The specific distribution of those foreign claims among different types of assets (U.S. government debt, equities, real estate, and so forth) may be important for considering some questions (for example, the potential for short-term disruptions in specific financial markets), but the overall level of those claims is more important in weighing other issues (for example, the vulnerability of the U.S. economy to adverse economic shocks). It is therefore important to emphasize that Treasury and other agency debt held by foreign investors represents only a portion of the total claims on the United States owned by the rest of the world.

According to the Bureau of Economic Analysis, the total amount of claims on the United States held by foreign investors in 2005 amounted to $13.6 trillion—9 percent more than in 2004 and 52 percent more than in 2000. A little more than 17 percent in 2005 was U.S. government securities, up from about 13 percent in 2000 (see Table 1). As noted, much of the rise in the share of U.S. government securities was associated with increased holdings by foreign governments, rather than by foreign private investors. The key point, though, is that however the claims are allocated among different asset types and foreign owners, the broader issue is the overall rise in net foreign claims on the United States assets; that rise is necessary to finance the nation’s current-account deficit.

Table 1.

The Fall in the U.S. Current-Account Balance

The current-account balance fell from -1.7 percent of GDP in 1997 to a record -6.1 percent last year (see Figure 1). At the same time, the outstanding amount of net international assets (holdings of claims on foreign entities by U.S. investors minus holdings of claims on the United States by foreign investors) fell from about -10 percent of GDP to about -20 percent in 2005 (see Figure 2).

Figure 1.

Figure 2.

To examine why the deficit in the current-account balance has increased in recent years, it is useful to examine trends in both net domestic investment and net national saving.⁶ Net domestic investment climbed steadily throughout the 1990s and then declined, on balance, in the early 2000s. On average, it has been 7 percent of national income since 1990 and in the past four years (see Figure 3).

Figure 3.

Net domestic investment can be financed either by net national saving or net foreign claims on U.S. assets. Since the late 1990s, it has been financed more and more by foreign claims, as the rate of net national saving has declined from an average of 4-½ percent in the 1990s to an average of 1 percent in the past four years. From that perspective, the low level of national saving has been responsible for the elevated level of the current-account deficit.

The decline in the rate of national saving in the 2000s largely reflects movements in both federal and private net saving rates (see Figure 4). The decline in the federal net saving rate from 2000 to 2003 accounts for much of the decline in the net national saving rate over that period. After 2003, however, the rate of net federal saving rose, but the net national saving rate was little changed because the net private saving rate fell. Although federal saving and national saving do not move in lockstep, there is generally a close relationship between changes in federal saving and changes in national saving. Put simply, the more the federal government saves, the more the nation tends to save as a whole.

Figure 4.

From another perspective, the elevated level of the nation’s current-account deficit has been driven by the willingness of foreign investors to provide capital to the United States. In other words, the nation’s rate of domestic investment is possible, given the rate of domestic saving, only because foreign entities have been willing to invest significant sums in U.S. assets and securities. From this perspective, inflows of capital from abroad affect the current account by raising the exchange value of the dollar and asset prices in the United States. The strong dollar encourages purchases of imports by U.S. residents and discourages purchases of U.S. exports by the rest of the world. Higher asset prices and correspondingly lower interest rates encourage consumption and investment.

The willingness of foreign investors to buy U.S. debts and assets reflects the attractiveness of the United States as a destination for international investment because of its stable political environment, developed legal institutions, deep and liquid capital market, and strong banking and financial system, among other advantages. Moreover, because the U.S. dollar is the major medium of international transactions, it is less susceptible to extreme and sudden depreciation.⁷ Indeed, the longevity of the large U.S. current-account deficit can be viewed as reflecting a sequence of events that caused demand for U.S. assets to grow even faster than the supply. Between 1997 and 2000, a host of developments—financial globalization, a succession of financial crises (the 1997–1998 Asian crisis, Russia’s default of 1998, and the Brazilian real crisis of 1999), and weaker economic

growth in other industrial countries than in the United States—all added to the demand for U.S. assets.⁸ By propelling the dollar and U.S. asset prices higher, those developments contributed to widening the current-account deficit.

A significant share of the nation’s overall external financing has been from foreign governments in recent years, as suggested by the trends in foreign ownership of U.S. government debt (see Table 2). In 2006, for example, net official inflows (purchases of claims on the United States by foreign governments net of purchases of claims on foreign entities by the U.S. government) were $448 billion, more than half of the $811 billion current-account deficit. Net official inflows also have grown rapidly in the past few years; in 2000, for example, net official inflows were only $42 billion. Almost all official purchases of U.S. assets were made by a handful of Asian governments, particularly China, which did so in order to keep its currency from appreciating outside of the band specified by its managed exchange rate policy. The Japanese government was also actively making purchases to keep the yen from rising before the spring of 2004.

Table 2.

The Unsustainability of the Current-Account Deficit

Regardless of whether its financing is provided by foreign governments or foreign private investors, the large U.S. current-account deficit, analysts generally agree, cannot be sustained indefinitely at its present high level relative to GDP. The United States—like any other country—cannot continue accumulating debt at a rate faster than its ability to repay it. If policy actions or other economic developments do not reduce the current-account deficit, at some point foreign investors will become less willing to keep adding to their holdings of U.S. assets.

To be sure, net U.S. international assets have changed little relative to GDP in recent years despite the large current-account deficit, but that situation is unlikely to continue over the long run. Movements in asset prices and in the exchange rate have raised the dollar value of U.S.-owned foreign securities and direct investments overseas by more than that of U.S. securities and investments held by foreign investors, offsetting the consequence of the current-account deficit. However, such favorable effects of valuation cannot be relied on in the long term, and sooner or later net U.S. international assets will begin to fall rapidly relative to GDP if the large U.S. current-account deficit persists.

A persistently large current-account deficit will, over time, make foreign investors less willing to provide low-cost financing for it. To date, foreign demand for dollar assets has not yet weakened significantly, in part because the dollar is still the major international reserve currency. However, once investors accumulate enough dollar assets to facilitate international transactions and to meet their other needs for holding reserves, they are likely to slow down their purchases of dollar assets for those purposes and increasingly will buy or sell dollar assets on the basis of the expected returns. For example, the Chinese government announced in March this year that it would establish an investment agency to more “profitably” and “efficiently” manage a portion of its foreign reserves, which exceeded $1.2 trillion in the first quarter of this year.⁹ Thus, to the extent that investors and governments believe that the U.S. current-account deficit will cause the dollar to depreciate, which reduces the expected return on dollar assets, the demand for dollar assets will fall.

Once foreign demand for U.S. assets begins to grow more slowly than the supply, there will be growing downward pressure on the dollar and U.S. asset prices. A lower dollar raises the prices of imports and reduces U.S. residents’ purchasing power at home and abroad, and lower asset prices make U.S. residents poorer. As a result, U.S. residents will be less able and willing to borrow and spend, thereby lowering the current-account deficit; the exchange rate and asset price adjustments, in other words, will facilitate the reduction in the current-account deficit. As long as foreign demand for dollar assets does not drop too suddenly, the adjustment in the current account will be a gradual one. In that case, growth of the U.S. economy is likely to remain on track. The gradual rise in exports and decline in imports will entail more production and employment in sectors that export and sectors that compete with imports, helping to offset the negative effects of the gradual adjustment in asset prices, interest rates, and the prices of imports.

How bumpy the adjustment of the U.S. current account will be thus depends on what happens to foreign demand for U.S. assets. If short-term factors boost the growth in the demand for U.S. assets above the growth in supply, the U.S. current-account deficit may temporarily widen further. However, it seems implausible that foreign demand for U.S. assets will be boosted repeatedly by short-term factors. Once long-term downward pressures on demand begin to outweigh temporary supports for dollar assets, they will push down the dollar and those asset prices, facilitating the decline of the current-account deficit.¹⁰

Various factors may mitigate the risk of the type of sudden collapse in foreign financing that would be associated with a relatively rapid adjustment of the current account. For example, the unique role of the U.S. dollar as the world’s main reserve currency should help to reduce the probability of a sudden stop of foreign financing, at least in the near future (although some analysts have warned that the dollar’s role as a primary reserve currency cannot be taken for granted over the long run). Furthermore, nearly all U.S. international liabilities are denominated in dollars, and about two thirds of U.S. holdings of assets abroad are equity assets, denominated in host countries’ currencies. Therefore, a large depreciation of the dollar would lower net U.S. liabilities to foreign investors not only by lowering net imports but also by boosting the dollar value of U.S. assets abroad. Consequently, the depreciation would not necessarily feed on itself and become a full blown dollar crisis, unlike the effects of a sharp drop in the currency of a country with a large amount of debt denominated in foreign currencies.¹¹

Thus, the more likely scenario appears to be a gradual adjustment, in which the current account falls gradually over time.¹² Nonetheless, given the likelihood of a continued decline in the United States’ net international assets as a percentage of GDP, a risk remains that adjustments in the foreign exchange rate and the current account will occur more rapidly than anticipated and that the effects of a rapid adjustment on the economy will be much more severe than with a gradual adjustment. That risk probably increases as the nation’s net international assets fall as a percentage of GDP.

Policy Options

Policymakers cannot directly control the current-account deficit, but they can influence the factors that determine it. Because the current account is equal to the difference between national saving and investment in the United States, policies that influence saving or investment will affect it. Although the current-account deficit could be improved by reducing investment, that outcome would be undesirable. With less investment, the U.S. capital stock would grow more slowly, which would reduce the growth of productivity and real wages over time. Therefore, the more desirable options for reducing the current-account deficit are those that would raise national saving.

Focusing on national saving may be particularly important in light of the economic and budgetary outlook in the United States over the next several decades. Rising federal health care costs, in particular, will place mounting pressure on federal spending, and if revenues remain at their current shares of GDP, the federal budget deficit is projected to grow rapidly, which could substantially reduce national saving.¹³ Over the past four decades, Medicare’s and Medicaid’s costs per beneficiary have increased about 2.5 percentage points faster per year than has per capita GDP. If those costs continued growing at the same rate over the next four decades, federal spending on those two programs alone would rise from 4.5 percent of GDP today to about 20 percent by 2050 (see Figure 5). Indeed, the rate at which health care costs grow relative to income is the most important determinant of the long-term fiscal balance; it exerts a significantly larger influence on the budget over the long term than other commonly cited factors, such as the aging of the population.¹⁴

Figure 5.

National saving can be increased in a number of ways that could involve higher government saving and/or higher private saving. Raising government saving through deficit reduction is one of the most reliable ways through which policymakers could boost national saving. That goal could be achieved through higher taxes, lower spending, or both.¹⁵ Given the nature of the nation’s long-term fiscal challenge, controlling the growth of federal health care costs seems a key component of deficit reduction over the next several decades. A variety of evidence suggests that opportunities exist to constrain health care costs both in the public programs and in the overall health care system without adverse health consequences, although capturing those opportunities to reduce costs without harming health outcomes involves many challenges.

National saving could also be increased through higher private saving. In evaluating policies to raise private saving, it is important to include their effects on government saving. For example, general tax incentives for private saving financed through higher budget deficits might not generate enough additional private saving to offset the higher budget deficits. Consequently, even if such policies increased private saving, they might not raise national saving. By contrast, deficit-neutral policies that encouraged private saving would work to raise national saving (because the increase in private saving would not be offset by a reduction in government saving).

Various options for raising private saving in such a manner have been proposed—for example, establishing automatic aspects for 401(k) and similar savings plans. Currently, many such plans leave it up to the employee to choose whether to participate, how much to contribute, which investment vehicle offered by the employer to select, and when to pull the funds out of the plan and in what form. Workers are thus confronted with a series of financial decisions, each of which involves risk and a certain degree of financial expertise. Many workers shy away from those decisions and simply do not make them, and the result is often a lack of participation. Research has suggested that participation and contribution levels can be substantially affected by changing the defaults at each of those points of decision. Indeed, one of the strongest empirical findings from behavioral economics is that automatic enrollment—that is, enrolling workers in a plan unless they opt out, as opposed to requiring them to sign up in order to participate—boosts the rate of participation substantially.¹⁶ Legislation enacted last year makes it easier for corporations to offer 401(k)-type plans with automatic enrollment and other automatic features, and researchers have proposed ways of expanding the same logic to individual retirement accounts.¹⁷ If such proposals were financed in a deficit-neutral manner, so that any gains to private saving were not offset by decreases in government saving, they could increase national saving. However, even if they were implemented in that manner, they would probably generate only a fraction of the saving needed to close the current-account deficit.

However it is accomplished, achieving a higher level of national saving also entails drawbacks. In the end, policies that raise national saving have one thing in common: They reduce consumption of goods and services and/or leisure. What makes the policies different is how they affect specific households and how they affect the economy at large. Therefore, choosing the appropriate saving policy inevitably involves balancing the economic effects of alternative policies with their distributional consequences.

Despite those various trade-offs and however it is accomplished, encouraging higher national saving probably represents the most effective step that policy- makers can take to facilitate the necessary reduction in the current-account deficit and reduce the risk of a severe economic disruption in foreign financing.