This Selected Issues paper on the United States analyzes the measures of potential output, natural rate of unemployment, and capacity utilization. Traditionally, measures of resource utilization have been used as indicators for the potential build-up of inflation pressures, and hence as guides for the formulation of macroeconomic policy. The paper highlights that the most commonly used indicators of resource utilization in the United States are the output gap, the employment gap, and capacity utilization in industry. The paper also analyzes the wage and price determination and productivity trends in the United States.

Abstract

This Selected Issues paper on the United States analyzes the measures of potential output, natural rate of unemployment, and capacity utilization. Traditionally, measures of resource utilization have been used as indicators for the potential build-up of inflation pressures, and hence as guides for the formulation of macroeconomic policy. The paper highlights that the most commonly used indicators of resource utilization in the United States are the output gap, the employment gap, and capacity utilization in industry. The paper also analyzes the wage and price determination and productivity trends in the United States.

IX. Dollarization and the Implications for U.S. Seigniorage and Macroeconomic Policy1

1. A significant degree of dollarization has been evident in Latin America, and elsewhere, for many years and recently Argentina suggested that full dollarization—i.e., replacement of the peso by the dollar as legal tender—might be a desirable alternative to its currency board.2 Should Argentina eventually proceed with full dollarization, other Latin American countries might follow, thereby stimulating a surge in U.S. dollars circulating abroad. Evidence also suggests that significant dollarization has occurred in many transition economies.3 On the other hand, the creation of the euro and European Monetary Union introduced a strong potential competitor to the dollar. This paper addresses the question of how important foreign demand for U.S. dollars is as a source of seigniorage and budgetary revenue, as well as other possible implications for the U.S. economy of a move toward full dollarization elsewhere in the Western Hemisphere.

2. Although foreign dollar holdings as a percent of U.S. GDP have increased steadily since the mid-1970s, the importance of seigniorage in the U.S. fiscal accounts has declined as a share of GDP, from nearly ½ percent in 1980 to just under ¼ percent of GDP in 1998. Of this amount, it is estimated that from 50 percent to 70 percent is from foreign holdings of U.S. dollars. Although variations in the net dollar flows abroad affect the measured monetary aggregates, the practice of the Federal Reserve in monitoring and targeting the federal funds rate as its principal short-term monetary instrument avoids any real macroeconomic effects on the U.S. economy from variations in overseas dollar holdings. Broadly, the implications of possible afiill dollarization” in one or more countries in the Americas would likely be positive for the United States if it is successful in increasing the dollarizing countries’ integration into the regional and global economies, and, by lowering risk premiums, increases investment and sustainable growth. In contrast, foil dollarization that was not backed by appropriate policies, especially those promoting fiscal stability and labor market flexibility, could increase the vulnerability of the dollarizing country to adverse shocks and, could thereby lower the dollarizing countries longer-term prospects, which would also not be to the advantage of the United States. The U.S. authorities have indicated that they would not consider extending the regulatory and supervisory responsibilities of the Federal Reserve or expanding the Federal Reserve’s lender-of-last resort function. While they have not explicitly ruled out possible seigniorage-sharing arrangements, such an approach would require Congressional approval.

A. Trends in Foreign Holdings of U.S. Dollars

3. The term “dollarization” has been used in a variety of ways, and occasionally has been defined broadly to include the total value of dollars held or circulating abroad, total dollar deposits in non-U.S. banks, plus dollar deposits held by foreigners in the U.S. banking system. In this paper, because the focus is on seigniorage from dollarization, a narrower definition is used; namely, holdings of U.S. dollar currency abroad. Estimates of either the stock or flow value of such holdings are subject to a high degree of uncertainty.4 The Federal Reserve publishes a series on both the level and flow of dollars held outside the United States based on net wholesale shipments abroad reported by large commercial bank-note brokers, primarily large commercial banks, Such shipments fail to capture net dollar transfers abroad through smaller commercial channels and private channels, including clandestine shipments related to illegal activities. Thus, this series is likely to underestimate the actual flow of dollars abroad.

4. Using a variety of direct and indirect information, Porter and Judson (1996) concluded that between 55 percent and 70 percent of the total U.S. currency stock outstanding was held abroad in 1995.5 This compares to an estimate of about 45 percent in 1995 based on the net wholesale shipments data. But all of the estimates suggest the same broad trends. A review of the available time-series data based on net wholesale currency shipments indicates that the rate of real dollarization (the percentage change in inflationadjusted dollars held abroad) has moved through two peak episodes during the last thirty years (Figure 1). The first occurred during the latter part of the 1970s through the early 1980s, and the second took place during the early-to-mid-1990s. The pace of dollarization appears to have slowed somewhat in the last several years, but remains high by historical standards.

Figure 1.
Figure 1.

United States: Real Dollarization, 1965–97

(Annualized quarterly growth in the stock of inflation-adjusted dollars held abroad)

Citation: IMF Staff Country Reports 1999, 101; 10.5089/9781451839579.002.A009

Sources: Board of Governors of the Federal Reserve System, Flow of Funds z-series data tables, Table F204; and Fund Staff estimates.

B. Dollarization and U.S. Seigniorage

5. When the Federal Reserve, or any central bank, places currency into circulation, it exchanges domestic currency units for interest-bearing securities. In accounting terms, an asset swap of equal value has occurred. However, the Federal Reserve has exchanged a noninterest-bearing asset (dollars) for an interest-bearing asset (U.S. government securities). The capacity to do this, using currency manufactured for a fraction of its market value, leads to seigniorage. Seigniorage can be thought of in either stock or flow terms. The central bank’s stock seigniorage from a given exchange of dollars for government securities equals the full value of the exchange less a relatively small expense for manufacturing the currency.

6. Central bank revenue accruing from its holdings of government securities is the flow value of seigniorage, and this is the measure that enters the Federal Reserve’s yearly cashflow statement. The flow concept is particularly relevant to assessing the significance of seigniorage to the federal government budget. Because Federal Reserve profits (roughly flow seigniorage net of central bank operating expenses) are transferred to the U.S. Treasury, higher seigniorage improves the fiscal outlook. Figure 2 displays transfers to the U.S. Treasury from the Federal Reserve system since 1965. In fiscal years 1997 and 1998, for example, the U.S. Federal Reserve transferred to the U.S. Treasury $20.7 billion and $17.8 billion, respectively (i.e., an average of about ¼ percent of GDP and a little over 1 percent of total federal revenues).6 Based on the estimates of Porter and Judson and the conservative series published by the Federal Reserve, it is reasonable to conclude that from 50 percent to 70 percent of total seigniorage in recent years has been attributable to foreign holdings of U.S. dollars.7 So average seigniorage earnings from abroad in these two years was probably in the range of $10-$ 13½ billion.

Figure 2.
Figure 2.

United States: Seignorage, 1965–97

Payments to the Treasury

(Percent of FY GDP)

Citation: IMF Staff Country Reports 1999, 101; 10.5089/9781451839579.002.A009

Sources: Board of Governors of the Federal Reserve System, Annual Reports, various issues; and Fund staff estimates.

7. Despite periods of relatively rapid dollarization in the early to mid-1980s and 1990s, the significance of seigniorage in the United States relative to GDP and as a percentage of total federal government revenues, has generally been declining since the early 1980s (Figure 2), Total seigniorage, of course, depends both on the total amount of currency in circulation (at home and abroad) and interest rates on government securities. The explanation for the declining significance of seigniorage is twofold. The structure of interest rates on government securities generally has drifted downward since the early 1980s. At the same time, there has been a secular decline in dollars to GDP held in the United States associated with increased credit card usage and the emergence of automated teller machines. This has partially offset the continued increase in U.S. dollar holdings abroad (Figure 3). Since 1990, the stock of dollars outstanding has risen by just over 1 percentage point of GDP, owing to the strength of the increase in foreign holdings (Figure 4). If these trends continue, U.S. seigniorage will increasingly be attributable to dollarization throughout the world.

Figure 3.
Figure 3.

United States: U.S. Dollars Outstanding, 1965–98

(Percent of GDP)

Citation: IMF Staff Country Reports 1999, 101; 10.5089/9781451839579.002.A009

Sources: Board of Governors of the Federal Reserve System, Flow of Funds z-series data tables, Table F204; and Fund Staff estimates.
Figure 4.
Figure 4.

United States: Share of the Increase in Total Dollars Outstanding Flowing Abroad, 1965–97

Citation: IMF Staff Country Reports 1999, 101; 10.5089/9781451839579.002.A009

Sources; Board of Governors of the Federal Reserve System, Flow of Funds z-series data tables; and Fund staff estimates.

C. Implications of DoUarization for the Conduct of U.S. Macroeconomic Policy

8. There are a number of possible benefits to the United States from foreign holdings of U.S. dollars. The discussion in Economic Report of the President (February 1999) identifies a number of possible benefits associated with the international currency role of the U.S. dollar including: (i) the “power and prestige” that might be associated with having an international currency; (ii) the possibility of increased business for U.S. banks and other financial institutions; and (iii) a degree of convenience for U.S. resident importers, exporters, borrowers, and lenders. However, seigniorage is possibly the most significant, and certainly the most quantifiable, benefit of dollarization for the United States. Thus, one risk associated with dollarization is that it brings with it the potential for a marked shift in the demand for dollars overseas and the associated loss of seigniorage.8 However, even a very large net inflow of dollars over a short period of time would produce a rather trivial effect for total seigniorage and the U.S. federal budget.9

9. Do shifts in overseas holdings of dollars have real macroeconomic effects on the U.S. economy beyond the seigniorage earnings? In particular, do such shifts complicate the conduct of monetary policy? Net international dollar flows directly affect liquidity in the United States and sudden shifts in net dollar flows might be expected to impart an added degree of short-term volatility to the monetary aggregates and to other indicators of liquidity. However, as with any liquidity shock, whatever the source, if the Federal Reserve can respond in a timely manner, real effects would be prevented. Indeed, under current monetary policy procedures in which the central bank conducts daily open-market operations to achieve a target federal funds rate, the Federal Reserve would respond automatically to such liquidity shocks. Suppose, for example, that the demand for dollars held by the public in Argentina were to increase. To accommodate the increased demand for U.S. dollars, the Argentine central bank would have to sell dollar reserve assets—U.S. government bonds. Even if this transaction were to take place in the private U.S. credit markets, rather than approaching the U.S. Federal Reserve directly, the effect would be the same. Initially, dollar deposits in the U.S. banking system would begin to contract as bonds were offered for sale. At that point, with bank deposits contracting, there would be a shortage of bank reserves and this would put upward pressure on the federal funds rate. The Federal Reserve, however, would respond automatically to this pressure by buying bonds in order to create new bank deposits sufficient to fully offset the pressure on the federal funds rate, The Federal Reserve’s open-market purchases would be sufficient to accommodate the increased demand for U.S. dollars held by the Argentine public.

10. The process, of course, also operates in reverse and does not depend on the specific institutional arrangements associated with the Argentine currency board. Suppose, for example, there was a sudden reversal of preferences and economic agents abroad wished to reduce their holdings of dollars. This would generate a net increase in bank deposits in the United States as foreign-dollar holders, or agents acting on their behalf, made preparations to exchange dollars for U.S. goods, services, or assets. The initial surge in bank deposits would immediately increase reserves in the U.S. banking system and would thus increase the supply of federal funds, putting downward pressure on the federal funds rate. This downward pressure would be met by a withdrawal of liquidity from the domestic banking system through an equivalent open-market sale of bonds by the Fed. Thus, the effect of the net inflow of dollars on domestic liquidity would be quickly neutralized and the seigniorage that had been associated with the repatriated dollars would be lost,

11. These examples indicate that the monetary aggregates (which include currency in circulation both at home and abroad) are affected by international currency flows, even though the effects on domestic liquidity are simultaneously offset by the Federal Reserve. Thus, variations in net international currency outflows, which rose sharply in the early 1990s, would help to explain the breakdown between the growth in the monetary aggregates and nominal GDP. For this reason, it has been suggested by some that currency should be excluded from measures of the U.S. money supply if monetary aggregates are to be useful in guiding the conduct of monetary policy (Sprenkle, 1993).

12. Net international dollar flows began being recorded in the U.S. international accounts in July 1997. Under the new BOP accounting convention, an estimate of net international dollar flows is included in the capital account—net dollar outflows (inflows) are a credit (debit) in the capital account. Balancing items could show up on either the current or capital account. For example, in the case of a decline in overseas holdings, dollars could flow back to the United States through the purchase of U.S. goods, services, or assets. Unless the full amount of the repatriated currency is used to acquire U.S. assets, the composition of the balance of payments would shift toward an improved current account and a worsened capital account.

D. Implications of “Full Dollarization” for the United States10

13. Argentina has raised the idea of moving from its existing currency board arrangement to full dollarization in which the U.S, dollar would become the legal tender, and all peso currency notes in circulation would be replaced by dollars. There has also been a revival of the debate about full dollarization in El Salvador, although the new administration has indicated that it would first give consideration to moving toward a currency board. For a country considering full dollarization, the main benefits would be the potential for narrowing sovereign interest rate spreads by eliminating exchange risk and—less quantifiable but probably more important in the long run—the benefits of fuller integration into the global economy and the elimination of future currency crises. The costs would include potential loss of seigniorage revenues, the loss of a more unrestrained lender of last resort capability, and the loss of the option to devalue in the face of major shocks. For countries that had not already moved to the stage of a currency board, there would also be some additional loss of monetary discretion, as interest rates would become fully linked to the monetary policy cycle of the United States; however, this is a price already paid by countries, like Argentina, with a currency board.

14. But what would be the implications for the United States? In the long term, the main effects—whether positive or negative—are likely to stem from the impact on growth in the dollarizing trading partners. If dollarization successfully increases these countries’ integration into the regional and global economies and, by lowering risk premiums, increases investment and sustainable growth, the United States will also indirectly benefit. However, if domestic policies, especially fiscal and labor market policies, are not fully consistent with the requirements of full dollarization, the result could be lower growth in the dollarizing economies, with consequently adverse consequences for the United States. In these latter circumstances especially, political tensions could also arise if there were to be a marked divergence between the United States monetary policy stance and the near-term cyclical requirements of the dollarized partner.

15. In addition to these considerations, the effects on the United States would depend in large part on the way in which full dollarization were implemented. In principle, there appear to be three broad options: (i) a gradual move toward a full multilateral monetary union between the United States and interested regional partners; (ii) a bilateral monetary arrangement between the United States and individual dollarizing countries; and (iii) unilateral dollarization. Because the United States is a large, diversified, and highly open economy, and because the U.S. dollar is already widely used in international transactions involving the United States—thus minimizing existing trade frictions that might be associated with currency fluctuations—a full monetary union would likely produce relatively small trade-based benefits for the United States. However, the costs would also be relatively small. A full monetary union would impose costs on the United States by requiring it to relinquish monetary policy autonomy and to accept a framework based on regional, rather than domestic, economic developments and objectives. But the economic size of the United States is such that, even with a full monetary union in which the monetary authority’s objectives were defined in terms of the overall union, the United States component would predominate—much more so than Germany does in Europe (see box). For example, in such a union Argentina would have a weight, by GDP, equivalent to that of, say, Florida or Ohio. In practice, the United States authorities have indicated that they would not be prepared to alter the focus of United States monetary policy on serving domestic interests as part of any dollarization initiative. Bayoumi and Eichengreen (1994) find that the supply shocks across countries in the Americas have, on balance, been both large and negatively correlated with those in the United States, making the region relatively unsuited to monetary union—although such estimates cannot take into account the fact that the appropriate theoretical prerequisites for a common currency area are, to some extent, endogenous since the degree of integration and hence the nature of shocks is likely to change once the monetary union is formed.

Box. How Dominant Would the United States be in any Regional Monetary Arrangement?

If the United States were to enter into a monetary union with a number of its regional partners—which is obviously unlikely in the foreseeable future—its economic weight would dominate that union to a far greater extent than any country in the European Monetary Union.

As a practical matter, therefore, monetary policy in such a union would be dominated to a significant extent by domestic economic considerations in the United States.

article image
Sources: WEO data; and Bureau of Economic Analysis, U.S. Department of Commerce.

Based on 1997 GDP (at prevailing exchange rates) and estimates of Gross State Product The sample includes the United States, Canada, and the ten largest economies in Latin America (Argentina, Brazil, Chile, Colombia, Ecuador, Guatemala, Mexico, Peru, Uruguay, and Venezuela).

16. The impact on the United States of any bilateral monetary treaty with, say, Argentina would largely depend on the scope of such a treaty. The U.S. authorities have already indicated that they would not consider extending the regulatory and supervisory responsibilities of the Federal Reserve or expanding the Federal Reserve’s lender-of-last resort obligations.11 In practice, therefore, the most significant difference between a bilateral treaty and unilateral dollarization would revolve around possible arrangements for sharing of seigniorage. Since full dollarization involves an expansion in overseas dollar holdings, even a full remittance of the additional seigniorage to the dollarizing countries would leave the United States no worse off, in budgetary terms, than prior to the dollarization. As an indication of the magnitudes involved, the total amount of seigniorage associated with full dollarization in Argentina would have been about US$¾ billion a year in 1998 (about 0.01 percent of United States GDP).12 At present, there is only one country—South Africa-—that shares seigniorage earnings with its smaller neighbors that use its national currency (i.e., with Namibia and Lesotho as part of the Common Monetary Area (CMA) agreement).13 However, a number of possible arrangements have been suggested.14 Some observers in Argentina have suggested that the seigniorage earnings could be earmarked as backing for a facility—-with private or public lenders—to provide liquidity support in the event of difficulties in Argentina’s banking sector. A recent staff report from the Joint Economic Committee of the U.S. Congress (1999) suggested a specific formula for sharing seigniorage whereby a dollarizing country’s share would be linked to its share of the total dollar monetary base, adjusted for the costs of operating the Federal Reserve system.

List of References

  • Bogetic, Zeljko, 1999, “Official or ‘Full’ Dollarization: Current Experiences and Issues,” International Monetary Fund, mimeograph.

    • Search Google Scholar
    • Export Citation
  • Porter, Richard D. and Judson, Ruth A. 1996, “The Location of U.S. Currency: How Much is Abroad?” Federal Reserve Bulletin, Vol. 82, (October).

    • Search Google Scholar
    • Export Citation
  • Sahay, Ratna and Vegh, Carlos A. 1996, “Dollarization in Transition Economies: Evidence and Policy Implications” in Mizen Paul and Pentecost Eric (eds.), The Macroeconomics of International Currencies: Theory, Policy, and Evidence, Edward Elgar Press.

    • Search Google Scholar
    • Export Citation
  • Savastano, Miguel A., 1996, “Dollarization in Latin America: Recent Evidence and Policy Issues” in Mizen Paul and Pentecost Eric (eds.), The Macroeconomics of International Currencies: Theory, Policy, and Evidence, Edward Elgar Press.

    • Search Google Scholar
    • Export Citation
  • Sprenkle, Case M., 1993, “The Case of the Missing Currency” Journal of Economic Perspectives, Vol. 7(4).

  • Summers, Lawrence H., 1999, Text delivered to the Senate Banking Committee Subcommittee on Economic Policy and Subcommittee on International Trade and Finance, (April 22).

    • Search Google Scholar
    • Export Citation
  • United States Senate Joint Economic Committee, 1999, “Encouraging Official Dollarization in Emerging Markets” (April).

1

Prepared by Michael Leidy.

2

Panama adopted the U.S. dollar as legal tender shortly after independence in 1904. See Savastano (1996) for a discussion of dollarization in Latin America.

4

See Porter and Judson (1996) for a review of the various ways in which the quantity of dollars held abroad can be estimated and for a sense of the uncertainty associated with such estimates.

5

By comparison, work at the Deutsche Bundesbank suggests that between 30 percent and 40 percent of deutsche marks are held abroad (Porter and Judson, 1996).

6

These amounts are the so-called “statutory transfers.” The Omnibus Budget Reconciliation Act of 1993 requires that surplus Federal Reserve Bank earnings be transferred from the regional Federal Reserve banks to the Board, and then to the U.S. Treasury. These earnings are principally from holdings of U.S. Treasury securities. Federal Reserve holdings of U.S. Treasury and federal agency securities in 1998 were valued at $473 billion.

7

This range also encompasses the U.S. Administration’s view that “Foreign holdings of U.S. currency are conservatively estimated at 60 percent of the total in circulation.” (Economic Report of the President, February 1999).

8

The most likely factor on the horizon that might eventually lead to a degree of dedollarization is the emergence of the euro. Of the national currencies it replaced, the mark had the most significant overseas holdings. The euro could emerge over the years as a strong competitor to the dollar, but significant shifts in currency holdings are not likely to take place suddenly on a large scale.

9

For example, even a 75 percent reduction in total world dollarization, assuming that the current foreign share of dollars outstanding is 70 percent, would have increased the federal budget deficit by only about 0.1 percent of GDP in FY 1997.

10

This section focuses on the possible implications for the United States of “full dollarization.” The potential costs and benefits for Argentina, or any other country, are discussed in detail in the forthcoming Board seminar paper on full dollarization.

11

Deputy Treasury Secretary Summers (1999).

12

The stock of peso currency in circulation in Argentina in 1998 was equivalent to about $13.5 billion. This would yield $709 million in additional seigniorage for the United States assuming a 5.25 percent average yield on United States government securities.

13

See Bogetic (1999) for details.

14

Any provision calling upon the United States to share seigniorage would require an appropriation from Congress. Present arrangements with Panama and other countries using the dollar as legal tender do not include any provisions for seigniorage-sharing.

United States: Selected Issues
Author: International Monetary Fund
  • View in gallery

    United States: Real Dollarization, 1965–97

    (Annualized quarterly growth in the stock of inflation-adjusted dollars held abroad)

  • View in gallery

    United States: Seignorage, 1965–97

    Payments to the Treasury

    (Percent of FY GDP)

  • View in gallery

    United States: U.S. Dollars Outstanding, 1965–98

    (Percent of GDP)

  • View in gallery

    United States: Share of the Increase in Total Dollars Outstanding Flowing Abroad, 1965–97