|
The
World Dollar Standard and the East Asian Exchange Rate Dilemma-Part
One Ronald I. MCKINNON (Editor's Note: This two-part article was originally prepared as a testimony before the British House of Lords. We thank Professor McKinnon for permitting Perspectives to publish it. The second part is to follow in the June 2002 issue of Perspectives.) Lord Peston asked that I testify on "globalization:" what are its ramifications-both positive and negative-and whether it is truly a new phenomenon or just a new label. Are our international institutions, such as the International Monetary Fund and World Trade Organization, adequate to ensure that positive aspects outweigh the negative? Finally, I was charged with answering Lord Peston's question "What is it about globalization that makes people so uneasy?" The enhanced hegemony of the United States is a prime source of international uneasiness in the new millennium-just as British military and financial hegemony made other countries uneasy with the spread of freer international trade in the 19th century. In today's military terms, there is just one superpower that sends gunboats- i.e., read aircraft carriers-to keep the peace in far away places, at least where its vital interests are concerned. There is also the invasive crass commercialism of multinational firms, mainly American, that non-Americans see as threats to their traditional way of life-as when French farmers set fire to MacDonald's hamburger stands. Other countries, particularly regimes that force their people into subservience through a blinkered religion, see foreign influences undermining national cultures. However, my submission to the Economic Affairs Committee will approach the problem of American global hegemony quite differently-which at first glance might seem like an arcane exercise in monetary economics. In the absence of a common international money (such as gold in the 19th century), the ever-widening ambit of international trade and finance today accentuates an entirely natural asymmetry among national currencies. A strong central money (or key currency) becomes dominant-as the U.S dollar now dominates on a worldwide scale outside of Europe, and as the old deutsche mark dominated within Europe before its monetary unification with the advent of the euro. (In the 19th century, Britain was also resented as the world's dominant creditor country that kept the rest of the world somewhat in thrall to the London capital market. But because Britain was then on the gold standard more or less on a par with the other industrial countries, it had much less autonomy in monetary matters than does the United States in today's world of "fiat" national monies.) We live in an inherently asymmetrical, and perhaps unfair, world because there can be only one central money for facilitating international exchange. Inevitably, this leaves most countries' currencies on the periphery of that central money where countries, particularly developing ones, on this monetary periphery have domestic financial systems that are naturally more fragile. They live with the ever-present threat of a currency crisis, i.e., a run from their peripheral money into the central one, the American dollar. Indeed, managing foreign exchange and financial policy is more difficult on the periphery than at the center. It is easier to be the U.S. Secretary of the Treasury than to be the Korean, or Argentinian, or Turkish, Minister of Finance! One important aspect of this asymmetry is the nature of currency risk in the foreign exchanges. The U.S. economy is by far the biggest debtor to the rest of the world-something like $2.5 trillion of net indebtedness, which continues to increase with the current American trade deficit. But nobody thinks that the dollar could really be attacked-or that there could be a currency crisis in the ordinary sense. Insofar as American banks, insurance companies, and so on receive foreign funds as the counterpart of America's trade deficit, this buildup of liabilities to foreigners is entirely denominated in U.S. dollars. So American banks have dollar-denominated liabilities both to foreigners and to domestic nationals, and they make dollar-denominated loans-largely to American firms and households. Because of this absence of foreign exchange exposure, American financial institutions can absorb this huge capital inflow without currency risk. There are other risks, but they aren't associated with fluctuations in the dollar's exchange rate against other currencies. However, if smaller debtor economies on the periphery of the dollar standard-such as Korea, Thailand, or any in Latin America-absorb foreign capital, typically the debts are denominated in another country's currency, i.e., mainly the U.S. dollar but also the yen or the euro on occasion. The genesis of the 1997-98 crisis was the huge short-term inflow of capital into East Asian economies, but denominated in dollars or yen. This meant their banks and financial institutions were at risk if there were any exchange rate fluctuations. In particular, any devaluation made repaying these external dollar obligations from earnings on domestic assets denominated in won, or baht, or pesos much more difficult. In contrast, today many people in the U.S. think that the dollar is too strong and that U.S. exporters would benefit from a devaluation of the dollar that reduces their costs relative to those of international competitors. And a one-time dollar devaluation would not hinder American financial institutions from paying off their dollar-denominated foreign debts-or from rolling them over indefinitely. Part I of my analysis of the monetary consequences of globalization provides a historical perspective on how the world dollar standard has evolved since World War II-with special concern for developing countries and emerging markets on its periphery. Then, Part II, which is to follow as a separate article, focuses on East Asia. Specifically, I link what I call "the East Asian exchange rate dilemma"-including the current plight of Japan-to how the dollar standard now works. (My companion submission to the Economic Affairs Committee, "Optimum Currency Areas and the European Experience", focuses on the great success of EMU in eliminating currency asymmetries within continental Western Europe.) Part I: The World Dollar Standard in Historical Perspective How did this asymmetrical position of the dollar become established in the world economy? After World War II, the U.S. had the world's only intact financial system. There were inflation, currency controls, and so on in Europe, as well as in Japan and most developing countries. Thus, in open foreign exchange markets, the dollar naturally became the world's vehicle currency for (private) inter-bank transacting and the intervention currency that governments used for stabilizing their exchange rates. Under the Bretton Woods agreement of 1945, every country pegged to the dollar, and the U.S. did not have a formal exchange rate policy, except for the residual tie to gold. This was quite natural given the history of the situation. The U.S. had the only open capital market, so countries could easily build up their dollar reserves and have a liquid market in which to buy and sell them. Similarly, private corporations in other countries all built up dollar reserves as well because their own currencies had exchange controls. Because of this accident of history, the U.S. dollar became the intermediary currency in international exchange between any pair of "peripheral" monies. The Dollar as Facilitator of International Exchange But why does the dollar continue with this facilitating function even when most other industrial countries-such as Japan and those in Europe-no longer have exchange controls? A little algebra helps explain continued dollar predominance. Suppose you have N, say 150, currencies in the world economy. The markets, themselves, would always pick one currency to facilitate international exchange. The reason for that is a big economy of markets. If we think of world of N countries with independent national monies, then just from your basic high school probability theory, the total number of country pairs in the system is the combination of N things taken two at a time (NC2). If foreign exchange dealers tried to trade across each pair, say, Swedish crowns against Australian dollars, or Korean won against Japanese yen, it would turn out that there would be a huge number of different foreign exchange markets. With 150 national currencies in the world (N = 150), and you tried to trade each pair, there would be 11,175 foreign exchange markets! It is expensive for any bank to set up a foreign exchange trading desk. Thus, rather than trading all pairs of currencies bilaterally, in practice just one currency, the Nth, is chosen as the central vehicle currency. Then all trading and exchange takes place first against the vehicle currency before going to the others. By having all currency trading against that one currency, you can reduce the number of markets in the system to N-1. Thus, with 150 countries, we need to have just 149 foreign exchange markets-instead of 11,175. Unlike the Bretton Woods system where all countries set official dollar parities, this result doesn't depend on any formal agreement among governments. In private markets today, choosing one currency like the dollar to be the intermediary currency is the most natural way of economizing on foreign exchange transacting. But history is important. If one country starts off providing the central money, as the U.S. in the late 1940s did, then it becomes a natural monopoly because of the economies of scale. The more countries that deal in dollars, the cheaper it is for everybody to deal in dollars. If you're a Japanese importer of Swedish Volvos and you want to pay for the Volvos, you first get your bank to convert your yen into dollars on the open market, then use the dollars to buy Swedish crowns. Volvo corporation receives the Swedish crowns and the importer gets the Volvos. However, the dollar is the intermediary currency. Using the standard textbook classification of the roles of money, Box 1 summarizes our paradigm of the dollar's central role in facilitating of international exchange. For both the private and government sectors, the dollar performs as medium of exchange, store of value, unit of account, and standard of deferred payment for international transacting on current and capital account-and has so from 1945 into the new millennium. It is a slight generalization of a similar table presented by Peter Kenen in 1983, but it remains as valid today as then.
Second in Box 1, the dollar is an international store of value. Corporations and some individuals hold dollar bank accounts in London, Singapore and other "offshore" banking centers-as well as in the U.S. itself. For governments, international reserves are mainly in dollars-largely U.S, Treasury bonds: Korea has $95 billion, Japan almost $400 billion, China nearly $200 billion, and so on. As a matter of fact, almost half of U.S. Treasury bonds outstanding are held by foreign central banks. Third in Box 1, the dollar serves as a unit of account for much of international trade. Trade in primary commodities shows a strong pattern of using the dollar as the main currency of invoice. Exports of homogeneous primary products such as oil, wheat, and copper all tend to be invoiced in dollars, with worldwide price formation in a centralized exchange. Spot trading, but particularly forward contracting, is concentrated at these centralized exchanges-which are usually in American cities such as Chicago and New York, although dollar-denominated commodity exchanges do exist in London and elsewhere. Invoicing patterns for exports of manufactured goods are more complex. Major industrial countries with strong currencies tend to invoice their exports in their home currencies. Before European Monetary Union, more than 75 percent German exports had been invoiced in marks, more than 50 percent of French exports invoiced in francs, and so on. But these illustrative ratios were dominated by intra-European trade. With the advent of the European Monetary Union, how much continental European countries will invoice their exports outside of Europe in euros remains unknown. Within East Asia, however, foreign trade is invoiced mainly in dollars: Korean trade with Thailand is typically dollar invoiced. Even Japanese trade with other East Asian countries is invoiced more in dollars than in yen. Outside of Europe, the prevalence of dollar invoicing is also true in other parts of the world. For example, intra Latin American exports are almost entirely dollar invoiced. For pricing manufactures, more than pure invoicing is involved. Exporters everywhere outside of Europe typically opt to quote selling prices for their products in dollars, and then keep these dollar prices fairly constant in industrial catalogs and other published price lists. In effect, they price to the world market-and not just to the American one-in dollar terms. Thus national central banks aiming to stabilize the international purchasing power of their currencies, often opt-either formally or informally-to peg against the dollar, and thus against the huge sticky-priced mass of internationally traded goods that it represents. Fourth in Box 1, if we think of a standard of deferred payment-which is also a traditional role of money-private and sovereign bonds in international markets are largely denominated in U.S. dollars, though some are now in euros. In international bond markets, U.S Treasuries are taken as the bench-mark or "risk-free" asset. That is, dollar-denominated sovereign bonds issued by emerging markets over the world have their credit ratings (by Moody's, Standard and Poor's, or Fitch) measured relative to U.S Treasuries. Thus, risk premia in interest rates on these bonds are typically quoted as so many percentage points over U.S. Treasuries. The Dollar as Nominal Anchor Beyond facilitating international exchange, the dollar has a second and complementary international function. Foreign monetary authorities may better anchor their own domestic price levels by choosing to peg, officially or unofficially, to the dollar. By opting to keep their dollar exchange rates stable, foreign governments are essentially opting to harmonize-without always succeeding-their monetary policies with that of the United States. This monetary harmonization has two avenues: (1) international commodity arbitrage-the arbitrage avenue, and (2) the signaling avenue where other central banks take their cue from actions of the U.S. Federal Reserve Bank. The arbitrage avenue arises naturally out of the dollar's facilitating role in international finance. Because international trade in goods and services is largely dollar invoiced (including trade between countries outside of the United States), international arbitrage in the markets for goods and services through a fixed dollar exchange rate can be a powerful device to anchor any one country's domestic price level. Putting the matter more negatively, if other countries fail to prevent their dollar exchange rates from fluctuating, the degree of pass-through of these exchange rate fluctuations into their domestic prices is (ultimately) very high. (The one big exception would be countries in the large euro area-whose domestic price levels are fairly well insulated from fluctuations in the euro's exchange rate against the dollar.) Asymmetrically, because both American imports and exports are invoiced in dollars, America's own domestic price level is relatively insulated from fluctuations in the dollar's exchange rate. More generally in the world at large, the dollar prices of internationally traded commodities are relatively invariant to fluctuations in the dollar's value against other currencies. So, as the Nth country in the system, the U.S. alone can carry out an independent monetary policy to target its own domestic price level without being much disturbed by exchange rate fluctuations. For the other N-1 countries, however, direct international commodity arbitrage through a fixed exchange rate can help stabilize their own internal price levels. In securing monetary harmonization with the United States, the signaling avenue can also be important. If any one national government resists upward pressure on its currency in the foreign exchanges, the resulting increase in its official dollar reserves signals the need for domestic monetary expansion-and vice versa. The national central bank can even takes its cue directly from what the Fed is doing. For example, the Bank of Canada typically changes its own discount rate (inter-bank lending rate) relatively quickly in response to changes in the U.S. Federal Funds rate. However, for the dollar to function successfully as nominal anchor, two important conditions must be satisfied: (1)
The American price level, as measured by a broad index of
tradable goods prices, is stable and expected to remain so;
and In the history of the postwar dollar standard, these two conditions were satisfied in some periods-but not so in others. Indeed, in contrast to the dollar's ongoing robustness as the facilitator of international exchange under either fixed or floating exchange rates, its function as nominal anchor has continually metamorphosed. High Bretton Woods, 1950 to 1968 From the 1950s through 1968, the U.S. price level for tradable goods prices-as measured by the U.S. wholesale price index-was stable. Also interest rates on dollar assets were low and stable because of the absence of expected inflation. So, under the old Bretton Woods par value system, all other countries willingly declared dollar parities-and kept their market exchange rates within a narrow band of 2 percent around these central parities, which were seldom changed. During this period of "high" Bretton Woods, IMF member countries could use price stability in the center country as an anchor for their own domestic price levels. But more than just the behavior of the center country was involved in this anchoring process. Because virtually all the major industrial countries were on the same fixed exchange rate regime, the "world" price level was more secure. Precipitate devaluations (or appreciations) of any one country, which could impart deflationary pressure to a neighboring one, were avoided. In addition, potentially inflationary national macroeconomic shocks were dampened. The inertia or "stickiness" in each country's price level was greater because all of them were committed to, and bound together under, a common monetary standard-albeit one ultimately dollar based. During this high Bretton Woods regime, even the American price level itself was more stable because of the generally fixed exchange rates. In the short and medium terms, the center country could benefit from commodity arbitrage with neighboring countries across the fixed exchange rates to dampen potentially inflationary shocks originating at home. In the end, however, the system could not survive persistent inflationary pressure in the center country-as we shall see. Finally,
nominal interest rates in the industrial countries were low
and remarkably stable in the 1950s and 1960s. Until the very
late 1960s, the common rate of price inflation was so low
that ordinary Fisher effects in interest rates were largely
absent. In these immediate postwar decades, the perceived
continued stability in exchange rates meant that cross-country
interest differentials remained modest-despite the presence
of capital controls in most of the industrial countries. This
commitment to fixed dollar parities by the industrial countries
finally collapsed in early 1973. However, the common monetary
anchor undergirded that era's famously high real economic
growth-not matched in the industrial world in any sustained
way before or since. Losing the Anchor 1968-73: The Advent of Floating Exchange Rates With hindsight, the old fixed rate dollar standard began to unravel in the late 1960s as WPI inflation in the United States-the center country-began to escalate toward 3 percent per year. Other countries-particularly Germany-became unwilling to maintain their old dollar parity and import even moderate inflationary pressure. The deutsche mark was revalued upward in 1969. More importantly, the United States was then hampered by the Keynesian belief (as encapsulated in the so-called Phillips curve) that disinflation would permanently increase domestic unemployment. So largely for doctrinal reasons, the center country refused to embark on a serious program of disinflation. But the ongoing inflation reduced America's industrial competitiveness. Worried about America's declining foreign trade position, President Nixon in August 1971 closed the vestigial "gold window": America's formal commitment under the old Bretton Woods articles to formally fix the dollar's value in terms of gold. Simultaneously, Nixon imposed an across-the-board tariff of 10 percent on American imports of manufactures, and insisted that the tariff would not be removed until all the other industrial countries appreciated their currencies against the dollar. They all appreciated between 10 and 20 percent before re-establishing their new "Smithsonian" dollar parities in December 1971. However, because the center country continued to inflate, the Smithsonian dollar parities were destined to fail. In February 1973, the industrial countries gave up on their dollar parities and moved to no-par floating. In the 1970s into the 1980s in the United States, high and variable price inflation coupled with high and volatile nominal interest rates largely eroded the dollar's usefulness as nominal anchor. In most developing countries as well as many industrial ones, inflation also increased sharply. Many industrial countries were now quite willing to have their currencies appreciate against the dollar to better insulate themselves from what had become a maelstrom of variable inflation rates worldwide. (Europeans were induced to look for a new center currency as anchor-and tried to rebuild monetary stability around the deutsche mark. This effort culminated with the successful advent of the euro in the late 1990s.) The collective effect of this worldwide monetary instability on world productivity growth was catastrophic. Without a common anchor for domestic price levels and exchange rates, productivity in the industrial world and its periphery-except for the East Asian "tigers"-slowed dramatically after 1973 through to the early 1990s. Paradise Regained in the 1990s? But from the early 1990s into the new millennium, there was a return to price stability in the United States-with U.S. interest rates becoming moderate to low once more. Thus, the dollar has again become attractive as an international anchor currency, and as the predominant reserve asset worldwide. After the dollar's decline as a reserve asset in the inflationary 1970s and 1980s, the dollar's share in official foreign exchange reserves has greatly increased over the last decade. The dollar rose from 51.3 percent of official holdings of foreign exchange (of members of the International Monetary Fund) in 1991 to 68.2 percent in 2001. And if one assumed a pro rata share of "unspecified currencies" to be dollars, the dollar's current share in international reserves seems well over 75 percent. Surprisingly, the advent of the euro has not reduced the dollar's predominance in international reserve holdings. The share of euros in official foreign exchange reserves in 1999 and 2000 was no greater than was the sum of the old legacy currencies-mark, francs, and guilders-before the advent of the euro on January 1, 1999. Although euro has been very successful for securing regional monetary integration in Europe, the dollar remains king in international finance worldwide. However, in the new millennium, this stronger form of the international dollar standard differs from High Bretton Woods of the 1950s and 1960s in at least two important respects: (1)
In non-crisis periods, most governments in developing economies
stabilize their exchange rates against the dollar but without
declaring official dollar parities. And such informal pegging
is also "soft" in the sense that many exchange rates
drift. Let
us discuss soft pegging and the encroachment problem in turn. In their landmark study of 155 country exchange rate regimes using monthly data, Guillermo Calvo and Carmen Reinhart show that the only truly floating exchange rates are the euro, dollar, yen, and possibly the pound sterling, against each other. Month-to-month variance in these industrial countries' exchange rates is high-and variance in short-term interest rates is low: short-run shifts in cross-currency portfolio preferences are mainly absorbed by exchange rate changes-while their central banks target short-term interest rates as an instrument of domestic monetary policy. In contrast, in developing or emerging-market economies, Calvo and Reinhart show that their monetary policies are arranged so that monthly variance in their exchange rates against some key currency-either the dollar or the euro-is low, but that monthly variance in their interest rates is much higher than in the core industrial countries. Except for an Eastern European fringe of countries keying on the euro, the others key on the dollar. The main shock absorber for cross-currency shifts in international asset preferences is changes in their domestic interest rates-except for those developing countries with effective capital controls. This surprising difference between the core industrial economies at the "center" and emerging-market economies on the "periphery" is even more pronounced at higher frequencies of observation. By accepting higher volatility in domestic short-term interest rates, monetary authorities in emerging markets generally succeed in keeping their dollar exchange rates relatively constant on a day-to-day or week-to-week basis. However, at low frequencies, e.g., quarter-to-quarter, these soft pegs sometimes drift; and, in major crises, even short-term exchange rate stabilization may be impossible. This new regime of informal i.e., undeclared, dollar pegs for countries on the periphery of the United States differs from High Bretton Woods with its officially fixed dollar parities. In East Asia outside of Japan, for example, all the countries are dollar peggers to a greater or lesser degree. But only Hong Kong with its currency board admits to an official dollar parity of HK$ 7.8 for one American dollar. The others all claim to be "independently floating", or a "managed float", or pegged to a "currency basket". Although the Chinese call their regime a "managed float", the RMB's exchange rate of 8.3 yuan to the dollar has hardly moved since 1994. The others' dollar pegs may drift a bit more when measured at low frequencies, but the variance in their dollar exchange rates is an order of magnitude less than that in the yen/dollar exchange rate. Negligence of the International Monetary Fund Why this reticence of governments in emerging markets in East Asia and elsewhere to admit to keying on the dollar-or to go further and declare official dollar parities? The reasons are both political and economic. On the political side, the asymmetry among national monies-with a center and a periphery-is simply too impolitic to admit. Nationalists in any peripheral country would get restless if their government admitted, by declaring an official dollar parity, that it was in thrall to the United States. De jure, the original Bretton Woods Agreement appeared to treat all its member countries symmetrically. Under Article IV of the 1945 Agreement, all members were obligated to declare an official parity for their exchange rate against gold or any currency tied to gold. In the event, only the United States adopted a very limited form of a gold peg-whereas all the others chose to peg to the dollar as the Nth currency (as described above). Nevertheless, in the 1950s and 1960s, the Bretton Woods Articles provided an acceptable political fig leaf for disguising what was really a dollar standard. But now the IMF's parity obligation for membership exists no more; it was blown apart by the American inflation of the 1970s. On the economic side, the modern reluctance of any one government to declare an official dollar parity appears too risky precisely because neighboring countries have not done so. If Country A (say, Argentina) declared an official dollar parity, and then a its close neighbor Country B (say, Brazil) allowed its currency to depreciate against the dollar, Country A could lose competitiveness and be badly hurt. Better for A not to commit itself formally to a particular dollar exchange rate to begin with in case it may want to depreciate in response to a surprise depreciation by B. Hence A dare not commit if B, C, D...... have not committed-and vice versa. In effect, there is a need for collective action-as in 1945-to re-institute a more general system of dollar parities to prevent beggar-thy-neighbor devaluations. But the old collective agreement under high Bretton Woods was undermined by the American inflation of the 1970s into the 1980s. With no stable anchor currency, maintenance of the old regime of exchange parities became impossible. Now, although the American price level has now been quite stable for almost a decade, the IMF has not attempted to orchestrate a collective return to a parity regime. Whence the prevalence of soft dollar pegging where governments, forced to act individually, are unwilling to commit themselves to anything harder. The IMF's Article VIII-the commitment of member countries to work toward current account convertibility, i.e., to remove all restrictions on making or receiving payments from importing or exporting or repatriating interest and dividends, was equally important for the success of high Bretton Woods-and retains its crucial importance today. But, in the 1950s and 1960s, the obligation of member countries to liberalize exchange controls stopped with Article VIII. Because of the bad experience with "hot' money flows in the 1930s, the peripheral countries around the United States all retained some degree of control over international capital movements-particularly short-term financial flows. The industrial countries of Western Europe retained capital controls well into the 1970s-and Japan into the early 1980s. Indeed, the IMF's articles required any member country receiving funds under a Fund program to impose capital controls if there was any danger of capital flight. In summary, the IMF's policies today suffer from major sins of omission and of commission. On the omission side, it has failed to promote regional exchange rate stabilization (where feasible) by encouraging the restoration of official exchange rate parities-as if the beggar-thy-neighbor exchange rate devaluations of the 1930s had been forgotten. Apart from outright dollarization, the IMF has even leaned on individual developing countries to flex their exchange rates as if the effect of such changes on neighboring countries did not matter. For its sin of commission, the IMF has actively encouraged peripheral countries to jettison their capital controls too soon in the process of liberalization-not recognizing the natural asymmetry between a strong center and naturally weaker periphery. (Although within the last year or two there are signs that the IMF may be repenting.) Consequently, dollar encroachment on the monies of developing countries and emerging markets in domestic uses is more pronounced than need be. The Problem of Dollar Encroachment This central role of the dollar in international finance today has a darker side: the potential displacement of national monies for domestic uses-displacement that is particularly marked in the Latin American context. Box 2 summarizes how the U.S. dollar might encroach (has encroached) on the natural domains of national monies as medium of exchange, store of value, unit of account, and standard of deferred payment within the country in question. In countries with a history of high and variable price inflation, the dollar encroaches on the national monetary domains in all four dimensions. But outside of this inflationary extreme, encroachment is still a problem. To be sure this dollar encroachment is not now a problem in the industrial economies, although it was a potential problem in the aftermath of World War II when European and Japanese currencies suffered from a complete loss of confidence. Most countries in Western Europe, as well as Japan, retained capital controls well into the 1970s-in large part to protect the domains of their domestic currencies. But step-by-step European unification, culminating in the late 1990s with the adoption of the euro, ended any lingering problem of dollar encroachment in Europe. This huge new, but highly credible, euro-based regime can operate on a stand-alone basis with perhaps the world's largest market for long-term bonds. But for countries outside of Europe in the new millennium, let us consider the problem of dollar encroachment in the context of each of the basic domestic functions of money-as laid out in Box 2-in turn. As medium of exchange as per Box 2, the dollar now circulates widely as hand-to-hand currency throughout Latin America, Africa, and many part of the former Soviet Union. In several Latin American countries, dollar bank accounts (interest-bearing and some checking) have even been legalized. This parallel circulation means that comprehensive capital controls, designed to prevent switching between the domestic money and dollars, are impossible to enforce. (But mild reserve requirements or taxes on foreign borrowing, as in Chile until recently, may still be feasible.)
Why have Latin American monetary authorities and several elsewhere allowed such invasive parallel circulation in dollars, where the demand for the domestic monetary base erodes and becomes quite unstable, to develop? First, many governments, with short time horizons of their own, want to attract emigrant remittances to the home country. So they offer domestic dollar deposits to nationals returning money to the country. (Even if Mexico's banking system does not now offer dollar-linked bank accounts, Mexico's long border with the United States with heavy two-way migration makes holding of interest-bearing dollar bank accounts just across the border very easy.) Second, where records of illegal export earnings don't exist for very important export products, such as narcotics, the national government can neither tax them nor force conversion of dollar export proceeds back into its domestic currency. Better to keep at least some of the dollar proceeds from the coca trade in banks within the country by offering attractive domestic deposit facilities in dollars. Last, but not least, is the long history in almost all Latin American countries of persistent financial instability: high inflation, temporary stabilizations, currency crashes, renewed inflation, and so on. Holders of naked cash balances in the domestic currency have been heavily taxed in the past. Thus, the precautionary motive for holding at least some dollar balances, at home or abroad, is strong. Similar relatively large dollar holdings are commonplace in much of Africa and in the disintegrated fragments of the old Soviet Union-including Russia itself. But the internal circulation of dollars in parallel to domestic currencies is not a general phenomenon. Virtually all the economies of East Asia provide counter examples. By and large, they did not have the same turbulent history of inflation and currency attacks so common in Latin America in the postwar. Even in those economies-Indonesia, Korea, Malaysia, Philippines, and Thailand-whose currencies were attacked in the great crisis of 1997-98, the internal circulation of U.S dollars was negligible before the attacks began and (with the possible exception of Indonesia) and is negligible today. These crisis economies-as well as the non crisis ones of China, Hong Kong, Singapore, and Taiwan-all had what looked like sustainable, if informal, fixes for their dollar exchange rates before 1997 and after 1998. However, as a store of value as per Box 2, interest-bearing dollar assets dominate domestic assets of the same term to maturity in Asia as well as in Latin America and other developing countries-unless protected by effective capital controls (as in China). A political or economic crisis in any one of the developing countries on this periphery of the dollar standard generates pressure from domestic nationals to fly into interest-bearing dollar assets as a safe haven. Even in East Asia (except for Japan), firms and households will only willingly hold domestic bonds or interest-bearing deposits if they bear a real rate of return higher than those on dollar bonds at an equivalent term to maturity. In effect, a substantial risk premium must be paid on term deposits (or bonds) in domestic currency compared to term deposits (or bonds) denominated in dollars-and this risk premium is typically much greater at long term than at short term. Indeed, the risk premium on long-term bonds denominated in domestic currency may be so great that an open market at the long-end of the maturity spectrum usually doesn't exist. How to measure this risk premium, i.e., distinguish it from the expected annualized depreciation (or appreciation) of the domestic currency, is a tricky econometric problem. Moreover, within developing economies, interest rates are highly variable-both in time series and across countries. Before the 1997 currency attacks began in Thailand, the relevant risk premia on three-month deposits in the East Asian debtor economies averaged about 4 percentage points, whereas in Latin America they averaged closer to 5 to 6 percentage points, above those on benchmark dollar assets. In the financial markets, unit of account and standard of deferred payment in Box 2 are closely related concepts, and refer to money's role as a numéraire in domestic contracts: the former is more of short-term concept whereas the latter is longer term. For longer term private debt contracts within Latin American countries, the dollar is commonly used as the standard of deferred payment even when the domestic currency is used as the means of settlement. The presumption is that dollar keeps its real purchasing value through time better, and that one can get instantaneous exchange rate quotes on the value of the dollar in domestic currency when the contract matures. Correspondingly, private debt contracts are seldom linked to domestic price indexes-such as the WPI or CPI-in part because of doubts over the statistical reliability of such indexes and because of lags in collecting price data. Even
with the dollar as numéraire for domestic private and
many sovereign bond issues, such bond issues are usually short
term-or have a floating interest rate set by the yield on
short-term (30-day) assets. Dollar predominance in the international
long-term bond markets-where U.S. Treasuries are considered
to be the world's "risk-free" asset-provides a competing
asset that inhibits the issue of long-term bonds, particularly
those issued by the private sector in developing countries.
The absence of a firm long-term exchange rate parity that
keeps the purchasing power of domestic bonds fairly constant
in terms of the world's risk free asset, i.e., U.S. Treasuries,
significantly hinders markets in domestic long-term bonds
in the peripheral countries. (The author is the William D. Eberle Professor of Economics at the Department of Economics of Stanford University.) |
