On the morning of September 22, 2000, the Federal Reserve Bank of New York's foreign exchange desk began buying euros for the first time since the currency came into existence 21 months earlier. Meanwhile, central bank officials in Frankfurt, London, Tokyo, and Ottawa (Canada) did the same. Among them, they bought between €4 and €6 billion. The Fed alone bought €1.5 billion for $1.34 billion. This was the Fed's second foreign exchange intervention in five years.¹

The whole operation took about two hours; when it was over, the central banks announced what they had done. As the ECB wrote in its press release, they did it because of "shared concern about the potential implications of recent movements in the euro exchange rate for the world economy." Since its inception on January 1, 1999, the euro had fallen steadily from $1.18 to $0.85, a decline of more than 25 percent. Though its low value had made exports cheap, bolstering the foreign sales of European-made products, it had also forced up the prices of imports. ECB officials, charged with maintaining price stability, found the high price of imports particularly troubling since they really did not want to raise interest rates just to bolster the value of their currency (which is what they would have had to do). Experts debated whether the euro should be worth $1.10 or $1, but all agreed that $0.85 was too low.

The coordinated intervention in the foreign exchange market made headlines around the world. The euro did appreciate briefly, rising to $0.89 on the day of the intervention, but by mid-October, it had returned to $0.85. The action may have been dramatic, but it wasn't effective. To understand why the intervention didn't work—and why the Fed almost never engages in foreign currency transactions—we need to examine the mechanics of how a central bank manages its country's exchange rate.

Both the United States and Europe are huge and largely self-contained economies. Exports and imports account for 10 to 15 percent of GDP. For the most part, these economies produce what they consume and invest, so on most days policymakers at the Fed and the ECB are justified in concentrating on the domestic economy and letting their exchange rates take care of themselves. But in small countries, central banks do not have that luxury. These countries are much more exposed to what goes on in the rest of the world, so changes in their exchange rates can have a dramatic impact on them.

Argentina provides an interesting example of how external and domestic factors interact in the making of monetary policy. Over the years, Argentina has suffered from severe inflation. During the 1970s, inflation averaged about 100 percent, meaning that prices doubled every year, while the economy grew about 3 percent a year. By 1989, inflation had climbed to more than 2,000 percent per year and the price level was 60 billion times what it had been 20 years before. Needless to say, growth fell. In 1990, real GDP was below its 1973 level and Argentina's economy was at a standstill.

The cause of such high inflation is often a combination of failed fiscal policy and failed monetary policy. Politicians want to spend too much, so they lean on central bankers to print more money. To discipline policymakers, in 1991 Argentineans implemented a mechanism called a currency board, which had two important attributes. First, Argentina's central bank, the Banco Central de la Republica Argentina, guaranteed that it would exchange Argentinean pesos for U.S. dollars on a one-forone basis; it fixed its exchange rate. Second, the central bank was required to hold dollar assets equal to its domestic currency liabilities, again at a one-to-one exchange rate. For every peso note that was issued and every peso in commercial bank reserves that it created, the Central Bank of Argentina had to hold one U.S. dollar.

The results were almost miraculous. Inflation fell immediately; after a few years, it had completely disappeared. But as we will see later, the victory didn't last. By early January 2002, the currency board had collapsed, GDP had fallen by a quarter, and inflation had risen to over 30 percent.

The examples of the ECB and Argentina suggest a connection between domestic monetary policy and exchange rate policy. To avoid raising domestic interest rates, the ECB organized a coordinated intervention to shore up the value of the euro. To control the inflationary impulses of fiscal and monetary authorities, Argentina fixed its exchange rate to the dollar. If exchange rate policy is inseparable from interest-rate policy, we have left something essential out of our analysis by ignoring cross-border transactions. To rectify the omission, we turn now to a discussion of exchange-rate regimes. Why is a country's exchange rate linked to its domestic monetary policy? Are there circumstances when exchange-rate stabilization becomes the overriding objective of central bankers? If so, should they try to fix the rate at which their currency can be exchanged for some other currency? Should a country even consider giving up its currency entirely?

¹In an attempt to stem the fall in the value of the Japanese yen, on June 17, 1998, U.S. and Japanese authorities both exchanged dollars for yen. U.S. authorities bought $633 million worth of the Japanese currency.