How is sterilised intervention supposed to affect exchange rates? It does so by altering the relative amounts of securities of different currency denomination held by the public. When a central bank engages in sterilised intervention, it buys or sells a foreign currency and sells or buys an equivalent amount of domestic securities. Suppose the Federal Reserve buys D-marks for dollars and sells an equal amount of U.S. Treasury securities from its portfolio. The effect is to reduce the world private sector’s holdings of D-mark assets and increase its holdings of dollar assets. Assume, in accordance with the portfolio balance approach to exchange-rate determination, (Isard, 1995) that the public is not indifferent as to the composition of its assets. In other words, a risk premium exists, and D-marks and dollars are not regarded by the public as perfect substitutes. The result will be an increase in the value of the D-mark relative to the dollar.

Whether or not intervention in the foreign exchange markets has significant effects on exchange rates has been the subject of much research and little consensus. One of the difficulties is that economists have not succeeded in systematically explaining movements of exchange rates. As Hali Edison put it in her survey of the literature, “Failing to find a statistically reliable relation between exchange rates and intervention is no different from failing to find a statistically and quantitatively significant relation between exchange rates and other economic variables such as interest rates.” (1993, pp. 5) In her conclusions she notes that “exchange-market participants appear to believe that central-bank intervention is important and… they therefore react to news of intervention.” But the effects are not long lasting. (1993, pp. 55)

Along similar lines but a bit more positive are the findings in the study by Kathryn Dominguez and Jeffrey Frankel. They argue that the prime concern is whether foreign exchange traders respond to intervention by revising their predictions of future exchange rates. If they do, they also change present exchange rates. It follows that for intervention to be effective; it has to be known to the markets. Sometimes intervention signals a change in monetary policy, but that is not necessary for it to affect exchange rates, at least in the short run. In particular, intervention can burst a speculative bubble such as the one that pushed the dollar to new highs in 1984–85. These authors regard intervention as more potent than the Jurgensen report did. (Dominguez and Frankel, 1993)

A more negative view emerges from the analysis of Maurice Obstfeld, who concluded that “the portfolio effects of pure intervention have generally been elusive enough that intervention cannot be regarded as a macroeconomic policy tool in its own right, with an impact somehow independent of short-term decisions on monetary and fiscal policy.” (Obstfeld, 1996, pp. 746) Mark Taylor’s literature survey concludes as follows: “Overall, therefore, the evidence on the effectiveness of official intervention is unclear, although some recent studies do suggest a significant link.” (Taylor, 1995, pp. 37)

Effects Of The Strong Dollar

The U.S. current-account deficit remained small and changed rather little from 1980 to 1982 (table 1) as the effects of the recession and the appreciation of the dollar offset each other. Over the next three years, the deficit rose to $124.5 billion—3 percent of GDP. Although the dollar turned down in early 1985, the deficit did not peak until 1987, owing to the well-known J-curve: An initial effect of exchange-rate depreciation is to raise the domestic cost of imports. Until the volume of imports responds to this price rise, the domestic value of imports increases following a depreciation.

Table 1: Current Account Balances ($ billions)
1980
1982
1985
1987
1989
1990
United States
2.3
-11.4
-124.0
-168.1
-104.2
-91.9
Germany
-13.7
4.9
17.6
46.4
56.7
48.1
Japan
-10.8
6.9
49.2
87.0
57.0
35.9
The enlargement of the current-account deficit from 1982 to 1985 was accounted for mainly by a slowdown in exports. (Solomon, 1985) Merchandise exports rose only $4.7 billion—little more than 2 percent—in three years. In real terms, exports fell by more than 12 percent from 1980 to 1983 and in 1986 were less than 3 percent above the 1980 level. Both the dollar value and the real value of U.S. imports increased by almost 50 percent from 1982 to 1985. Much of the import expansion was a product of the economic recovery from the recession of 1982, as real GDP grew by 15.2 percent, or an average of 4.8 percent per year. Gross domestic purchases increased 18.7 percent in real terms from 1982 to 1985 while imports rose 49.1 percent.

If the income elasticity of demand for imports is assumed to be 2.5 in a period of cyclical recovery (Stevens et al, 1984) —in other words, imports normally rise 2.5 times as fast as GDP during a recovery from recession—imports would have been expected to grow by just over 38 percent if the dollar had remained stable. Thus, something like one-fourth of the increase in imports from 1982 to 1985 can be attributed to the appreciation of the dollar.

The rapid increase in America’s imports in 1982–85 had a positive effect on the growth of its trade partners. This would have happened even in the absence of an appreciating dollar. But the appreciation, by leading to an increase in imports one-third greater than if the dollar had not risen in value, meant that the United States was exerting a larger influence on aggregate demand in the rest of the world. While Europeans were complaining about the strong dollar, what was being ignored was the boost to European economies from the substantial increase in exports to the United States. Three OECD economists estimated that about one-third of the growth in Europe in 1983 and 1984 could be attributed, directly or indirectly, to the expansion of American imports. (Koromzay, Llewellyn, and Potter, 1987)

;

Effects Of Capital Flows On Recipient Countries

The growing supply of funds flowing to emerging markets showed up, in the period from 1995 through mid-1997, in a distinct narrowing of the spread of interest rates on emerging market debt over the yield on U. S. Treasury obligations of the same maturities. Those spreads turned up again as the east Asian crisis worsened in the second half of 1997. (Cline and Barnes, 1997)

It is significant that a substantial portion of the capital inflows were, in effect, loaned back to the industrial countries as the recipient nations accumulated foreign exchange reserves. The increase in the reserves of all developing countries from the end of 1989 to the end of 1996 came to just over $500 billion, which is more than half of the cumulative inflow shown in table 5.1. Among the larger recipients of capital, the proportion that was added to reserves varied somewhat. In the case of Brazil, two-thirds of the inflows showed up in reserves in 1990–96; in Thailand, 59 percent; in China, about 42 percent; and in Malaysia, 32 percent.

As might be expected under these conditions, domestic investment increased in developing countries. In eastern and southern Asia, it grew from 24.1 percent of GDP in 1985 to 31.9 percent in 1995. In Latin America as a whole, investment went up much less relative to GDP—from 19.1 percent in 1985 to 19.7 percent in 1995, but there were large differences among countries in that region. (World Economic and Social Survey, 1997)

Along with the higher investment came larger current-account deficits, easily financed by the inflow of capital. For all developing countries, the current-account deficit increased from near zero in 1990 to $76 billion in 1996, and was accounted for mainly by nations in Asia and Latin America.

Economic growth accelerated in most regions of the developing world. For all developing countries, GDP growth rose from 4.1 percent in 1990 to 6.4 percent in 1996. Asia was the fastest-growing region—about 9 percent per year in 1992–96. China’s economy expanded at double-digit rates in 1992–95 and 9.6 percent in 1996, owing largely to rapid productivity growth. (Khan, 1997) In the developing countries of the Western Hemisphere, growth picked up from 1.1 percent in 1990 to 5 percent in 1994 but was then affected by the Mexican crisis. The IMF projected that GDP growth in Latin America would be 5.2 percent in 1997 and 3.5 percent in 1998.

A United Nations study found that between 1991 and 1996 the number of developing countries with rising per capita GDP increased from fifty-four to seventy-six; those seventy-six nations accounted for 96 percent of the population in all developing countries. (United Nations, 1997)

While these were favourable developments, problems and risks also arose from the viewpoint of the recipient countries. The two principal—and related—problems were that the inflows would be inflationary and that they would cause a real appreciation of exchange rates. The real appreciation could come about either as the result of inflation greater than the depreciation, if any, of the nominal exchange rate or of an upward movement of the nominal exchange rate in response to incoming flows of capital. Related risks were that the current-account deficits would become too large and unsustainable and that the banking systems in the recipient countries would be vulnerable to crisis. In the early 1980s, it was the banks in the lending countries that were vulnerable. In the mid-1990s, apart from Japan, it was banks in some of the countries to which capital was flowing, including China, Korea, and Thailand, that faced serious problems.

These problems and risks led to a number of policy reactions. In most countries to which large amounts of capital moved, sterilised intervention was used as a means of preventing or limiting increases in bank reserves and monetary and credit expansion. Such market intervention helps to explain the increase in foreign exchange reserves, to which attention was called above. Sterilization was carried out in a variety of ways: open market sales of securities, central bank borrowing from commercial banks, and shifts of deposits to the central bank.

One of the problems with sterilised intervention is that it tends to maintain or raise domestic interest rates, thereby attracting additional capital inflows. Sterilization was “scaled back” in Chile, Colombia, Indonesia, and Malaysia “as it became clear that high domestic interest rates were attracting more short-term inflows and were changing the composition of inflows toward the short end.” (International Capital Markets, 1995, 12) That helps to explain the use of other measures.

Some countries adopted specific policies, sometimes only temporarily, designed to limit inflows. (Montiel, 1996) There was a large variety of such “capital controls.” (International Capital Markets, 1995) Mexico and Malaysia, among others, used quantitative controls on types of capital inflow or foreign liabilities of banks or sales of securities abroad. Others, including Chile and Colombia, used taxation to discourage inflows along with reserve requirements on banks’ foreign liabilities. Brazil imposed a tax—a so-called Tobin tax – on some types of foreign exchange transactions. Another technique was to widen the band for exchange rate variation so as to increase the risks involved in foreign borrowing. A number of countries liberalized both imports and capital outflows by residents, a policy that had been underway in any event as an aspect of deregulation and liberalization of financial markets.

Where inflation takes hold, an option is to tighten fiscal policy. This was done in Thailand, Chile, and Malaysia. But, as Calvo, Leiderman, and Reinhart point out, the effect of fiscal tightening is likely to be stronger if it is thought to be temporary. “If it is seen as permanent, individuals may perceive a rise in lifetime disposable income and increase their borrowing to finance higher spending—thus partially offsetting the effect of the cut in public expenditure.” (Calvo, Leiderman, and Reinhart, 1996)

Significant amounts of the incoming capital to emerging markets went through their banking systems. In Malaysia, for example, the foreign liabilities of commercial banks increased from 7 to 19 percent of GDP between 1990 and 1993. In Mexico, those liabilities rose from 8 percent of GDP in 1991 to 13 percent in 1994, and in Thailand, from 4 percent in 1988 to 20 percent in 1994. (International Capital Markets, 1995) The result was potential credit risk and exchange rate risk. Whether such risks became actually dependent on the quality of bank supervision and regulation.

;