Forecasting exchange rates is an inexact business. As Alan Greenspan, the chairman of America’s Federal Reserve, once said, the activity “has a success rate no better than that of forecasting the outcome of a coin toss.” Recent years have borne this out: most currency forecasters would actually have done better if they had simply tossed a coin — at least they would have been half right. Yet over the next few years it seems an excellent bet that there will be a large drop in the dollar.
Since mid-October the dollar has fallen by around 7 percent against the other main currencies, hitting a new all-time low against the euro and a five-year low against the yen. The dollar has lost a total of 35 percent against the euro since early 2002; but it has fallen by a more modest 17 percent against a broad basket of currencies, including the Chinese yuan, which is pegged to the greenback. The dollar wobbled badly this week, having fallen for five successive days after Mr. Greenspan said that America’s current-account deficit was unsustainable because foreigners would eventually lose their appetite for more dollar-denominated assets.
Mr. Greenspan may not be the only central banker to have become bearish on the dollar. Markets have been rattled by concerns that foreign central banks might reduce their holdings of American Treasury bonds. Last week, officials at the central banks of both Russia and Indonesia said that their banks were considering reducing the share of dollars in their reserves. Even more alarming were reports that China’s central bank, the second-biggest holder (after Japan) of foreign-exchange reserves, may have trimmed its purchases of American Treasury bonds.
This combination of events has led some economists to ponder the once unthinkable: might the dollar lose its reserve-currency status? Over the past 2,000 years, the leading international currency has changed many times, from the Roman denarius via the Byzantine solidus to the Dutch guilder and then to sterling. The dollar has been the dominant reserve currency for more than 60 years, delivering big economic benefits for America, which can pay for imports and borrow in domestic currency and at low interest costs.
The dollar’s share of global foreign-exchange reserves has already fallen from 80 percent in the mid-1970s to around 65 percent today. And yet does the dollar really risk losing its status as the world’s main currency? The same question was asked in the early 1990s after the dollar’s previous long slide, but the dollar’s pre-eminence survived. Then, however, there was no alternative to the dollar. Today the euro exists, and could yet emerge as a rival to the greenback.
The requirements of a reserve currency are a large economy, open and deep financial markets, low inflation and confidence in the value of the currency. At current exchange rates the euro area’s economy is not that much smaller than America’s; the euro area is also the world’s biggest exporter; and since the creation of the single currency, European financial markets have become deeper and more liquid. It is true that the euro area has had slower real GDP growth than America. But in dollar terms the euro area’s economic weight has actually grown relative to America’s over the past five years.
Where the dollar has failed is as a store of value. Since 1960 the dollar has fallen by around two-thirds against the euro (using Germany’s currency as a proxy before 1999) and the yen (see chart 1). The euro area, unlike America, is a net creditor. Never before has the guardian of the world’s main reserve currency been its biggest net debtor. And a debtor may be tempted to use devaluation to reduce its external deficit — hardly a desirable property for a reserve currency.
Those bearish on the dollar are asking why investors will want to hold the assets of a country that has, by its own actions, jeopardized its reserve-currency position. And, they point out, without the intervention of central banks, which have been huge net buyers of dollars, the dollar would already be lower. If those same central banks were to begin to sell some of their $2.3 trillion dollar assets, then there would be a risk of a collapse in the dollar. However you look at it, America is likely to find it increasingly hard to finance its huge current-account deficit.
The deficit is at the heart of this issue. Various economists have put forward at least four arguments why the deficit does not matter and the dollar’s reserve status is safe. First, the deficit is a sign of America’s economic might, not a symptom of weakness. Second, sluggish demand overseas is a big cause of the deficit, so it is reversible. Third, the deficit exists largely because of multinationals’ overseas subsidiaries. And fourth, central-bank demand for dollars creates, in effect, a stable economic system. It is not difficult to demolish each argument in turn.
Why the Deficit Matters
Start with the first argument, which has been favored by America’s Treasury. Foreigners want to invest in America, it is claimed, because it offers higher returns than Europe or Japan; and if America runs a capital-account surplus, it must by definition run a current-account deficit. There may have been some truth to this argument in the late 1990s, when America enjoyed large net inflows of direct and equity investment, but over the past year or so, there has actually been a net outflow from America of such long-term investment. Moreover, in the past few years America has had lower returns on foreign direct investment, equities and bonds than Europe or Japan.
The current-account deficit is now being financed by foreign central banks and short-term money. In the year to mid-2004, foreign central banks financed as much as three-fifths of America’s deficit. The recent purchase of reserves by central banks is unprecedented. Global foreign-exchange reserves (65 percent, remember, are denominated in dollars) have risen by $1 trillion in just 18 months. The previous addition of $1 trillion to official reserves took a decade. These purchases of dollars have nothing to do with the prospective returns in America, but are aimed at holding down the currencies of the purchasing countries.
Worse still, in recent years capital inflows into America have been financing not productive investment (which would boost future income) but a consumer-spending binge and a growing budget deficit. A current-account deficit that reflects a lack of saving is hardly a sign of strength.
What about the second argument, that sluggish demand in the rest of the world is to blame for America’s external deficit? If only Europe and Asia would save less, spend more and so import more from America, it is argued, the deficit would simply vanish. Martin Barnes, an economist at the Bank Credit Analyst, a Canadian investment-research firm, reckons that this is much exaggerated. (“Re-assessing the Dollar Outlook” by Martin Barnes, The Bank Credit Analyst, December 2004.) In 2001, when domestic demand did grow slightly faster in Europe and Japan than in America, America’s deficit barely budged.
The problem is that America’s imports are 50 percent bigger than its exports, so if exports and imports simply grow at the same pace, the trade deficit automatically widens. If imports rise by, say 10 percent, then exports need to grow by 15 percent just to prevent the deficit from widening. This means that while stronger foreign demand would undoubtedly help, it would be virtually impossible for America to reduce its deficit significantly through stronger exports alone. Li Ruogu, the deputy governor of the People’s Bank of China, said last week that America should put its own house in order — i.e., save more — and stop blaming others for its problems. He was right.
The third argument is that fretting about the current-account deficit is outmoded because a large slice of the deficit reflects transactions between American multinationals and their foreign subsidiaries. Thus, it is claimed, importing an IBM computer from China is not the same as importing a Toshiba from Japan. Outsourcing by American firms boosts their profits. The problem with this argument, as Mr. Barnes points out, is that the total trade between multinationals and their foreign subsidiaries still creates a deficit even allowing for the return of profits and dividends, and this gap must still be financed by borrowing from abroad.
Last, but not least, last summer’s favorite explanation of why America’s deficit is not a problem is the notion that the world now enjoys the equivalent of the Bretton Woods system (the system of fixed exchange rates after the second world war), in which Asian governments happily buy the Treasury bonds that finance America’s deficit in order to maintain cheap currencies to support their own export-led growth. In turn, Asia’s purchases of bonds hold down interest rates in America, and so support consumer spending and imports. This cycle, it has been argued, could last another decade.
One big difference is that under the original Bretton Woods system America ran a current-account surplus and the value of the dollar was officially pegged to gold. No wonder, perhaps, that today’s “system” is already starting to creak as some Asian central banks start to worry about the value of their dollar reserves. To sustain the current arrangement, they will have to keep buying more and more dollars as America’s current-account deficit widens. Asian central banks are already exposed to enormous potential losses in local-currency terms should their currencies appreciate against the dollar. It would be prudent for them to diversify their reserves, but that could send the dollar tumbling. Larry Summers, a Treasury secretary under President Clinton, calls this the “balance of financial terror”: in effect, America relies on the costs to Asian central banks of not financing its deficit as assurance that financing will continue indefinitely.
For almost two decades, economists have worried about America’s current-account deficit and predicted a plunge in the dollar and a hard landing for the economy. The dollar did indeed fall sharply in the late 1980s, but with few ill effects on the economy. So why worry more now? One good reason is that the current-account deficit, currently running at close to 6 percent of GDP, is almost twice as big as at its peak in the late 1980s, and on current policies it will keep widening. Second, in the 1980s America was still a net foreign creditor. Today it has net foreign liabilities and these are expected to reach $3.3 trillion, or 28 percent of GDP, by the end of 2004 (see chart 2).
Some economies, such as Australia and New Zealand, have built up bigger debt ratios without obvious adverse economic consequences, but they are small countries so their current-account deficits absorb only a tiny fraction of global saving. This year alone, America’s new borrowing from abroad will mop up a massive 75 percent of the world’s surplus saving.
So far America’s hefty debt has not been a burden on its economy, mainly because it has pulled off an extraordinary trick. Although it is a large net debtor, it does not have to make net payments of interest and dividends to the rest of the world. Instead, America still enjoys a net inflow of investment income because it earns a higher average return on its foreign assets than it pays on its liabilities. Returns on foreign direct investment and equities are higher abroad than at home, and America has benefited from unusually low interest rates on its borrowing in recent years. Unlike in previous periods of dollar decline, bond yields have remained low — largely thanks to those huge purchases by foreign central banks. But as interest rates rise in future and net foreign debt mounts, America’s net investment income is likely to turn negative, probably next year. Not only will that swell its current-account deficit, but it will also exert an increasing drag on the economy.
America has enjoyed another huge advantage in its ability to borrow in its own currency. A normal debtor country, such as Argentina, has to borrow in foreign currency, so while a devaluation will help to reduce its trade deficit, it will also increase the local currency value of its debt. In contrast, foreign creditors carry the currency risk on America’s $11 trillion-worth of gross liabilities. Its net foreign investment position actually improves as the dollar declines, because this boosts the dollar value of overseas assets. This makes devaluation an attractive option for America.
The dollar’s position as the world’s main reserve currency allows it to attract finance on exceptionally favorable terms. However, this is a mixed blessing. It encourages America to borrow excessively, which increases the eventual cost of adjustment. The issue is not whether America can afford to take on more debt, but whether the rising debt burden will make investors less willing to finance future deficits at current exchange and interest rates.
A recent paper by Nouriel Roubini, of New York University, and Brad Setser, of Oxford University, estimates that, if the real trade-weighted value of the dollar remains close to its average in 1990-2003 (slightly above current levels) and there is no change in domestic policy, America’s current-account deficit would rise to 8 percent of GDP in 2008, and its net debt would increase to over 50 percent of GDP. (“The U.S. as a Net Debtor: The Sustainability of the U.S. External Imbalances”.) In practice, such levels are unlikely to be reached because private investors would be unwilling to finance debts of that size without much higher interest rates and/or a lower dollar, both of which would help to shrink the current-account deficit.
Despite its recent drop, the dollar is far from cheap. After adjusting for inflation differentials, the dollar’s real trade-weighted value against a broad basket of currencies is close to its average level over the past 30 years. Although it has barely fallen against most emerging-market currencies, the greenback is already below most estimates of its “fair value” against the euro. But that should be no surprise. Typically, a currency needs to undershoot its fair value by a wide margin in order to reduce a country’s large external deficit. The real broad trade-weighted dollar has so far fallen by only 15 percent since early 2002, compared with a drop of 34 percent from its peak in 1985 (see chart 3). Yet America’s current-account deficit is much bigger today than in the 1980s, so the dollar is likely to fall more sharply. Some economists reckon that it needs to fall by at least another 30 percent. That would imply a rate of over $1.80 for one euro, compared with today’s $1.33.
The less the dollar falls against emerging-market currencies, such as the Chinese yuan, the more it is likely to drop against the euro. China accounts for one-quarter of America’s total trade deficit. Speculation has mounted in recent weeks that the yuan will soon be revalued against the dollar. But Beijing has indicated that it will not be rushed into changing its exchange rate, especially if pressured by America.
In any case, the current-account deficit cannot be corrected by a fall in the dollar alone: domestic saving also needs to rise. The best way would be for the government to cut its budget deficit. That would reduce America’s need to borrow from abroad, and so mitigate the fall in the dollar and rise in bond yields that will otherwise be demanded by investors. If combined with stronger growth abroad, then the current-account deficit could slowly shrink. America’s growth would be depressed by tax increases or spending cuts, but there would be no need for recession. If, on the other hand, the government fails to cut its budget deficit, the dollar will fall more sharply and bond yields will rise. America’s housing bubble might then burst and consumer spending would certainly slow sharply. That combination would reduce the external deficit, but only at the cost of a deep recession.
A History Lesson
In 1913, at the height of its empire, Britain was the world’s biggest creditor. Within 40 years, after two costly world wars and economic mismanagement, it became a net debtor and the dollar usurped sterling’s role. Dislodging an incumbent currency can take years. Sterling maintained a central international role for at least half a century after America’s GDP overtook Britain’s at the end of the 19th century. But it did eventually lose that status.
If America continues on its current profligate path, the dollar is likely to suffer a similar fate. But in future no one currency, such as the euro, is likely to take over. Instead, the world might drift towards a multiple reserve-currency system shared among the dollar, the euro and the yen (or indeed the yuan at some time in the future). That still implies a big drop in the long-term share of dollar assets in central banks’ vaults and private portfolios. A slow, steady shift out of dollars could perhaps be handled. But if America continues to show such neglect of its own currency, then a fast-falling dollar and rising American interest rates would result. It will be how far and how fast the dollar falls that determines the future for America’s economy and the world’s. Not even Mr. Greenspan can forecast that.