The association between money growth and inflation is evidence for one of the principal monetarist propositions originally explained by Milton Friedman: sustained money growth in excess of the growth of output produces inflation; to end inflation or produce deflation, money growth must fall below the growth of output. It is noteworthy that one country with low or negative money growth, Ethiopia, reports a falling price level despite a long civil war and periodic famines. Consumer prices reported for 1987 were below the level reached two years earlier.

What is true across countries also is true over time in a particular country. Inflation will be sustained if the rate of money growth far exceeds the rate of output. To end inflation, money growth must be reduced permanently. Countries as diverse as Chile, Israel, Brazil, Argentina, Italy, Japan, Turkey, and the United States, to name only a few, have increased or reduced inflation at different times by speeding up or reducing the rate of money growth. In some countries the changes in money growth and inflation have ranged over hundreds or thousands of percentage points. In others the range has been narrower.

Recent decades provide many examples. In the years since World War II, almost all countries experienced inflation. Average rates of inflation differ markedly, however, both from country to country and over time within a country. For example, comparing five-year averages for the United States shows that for 1960 to 1964, the growth of money (currency and checking deposits) remained close to the average growth of output, 2.5 to 3 percent. Inflation, measured by the deflator for total output, averaged 1.6 percent. The Federal Reserve increased money growth from 1965 to 1969 to help finance government spending for the Vietnam War and for the War on Poverty. Inflation increased. By the late seventies money growth was nearly 7 percent a year on average and inflation reached an 8 percent average. At that rate prices doubled in less than a decade. Money growth slowed and remained low after the middle eighties. In the five years ending in 1991, inflation and money growth were back at the levels of 1965 to 1969.  This illustrates that the relation, while generally reliable, is not mechanical. In the years 1985 to 1989, inflation fell even though average money growth remained high. Explanations for this differ. What is most important is that such exceptions can occur; money growth in excess of output growth is a necessary but not a sufficient condition for inflation.

A second monetarist proposition is that when inflation is expected to be high, interest rates on the open market are high and the foreign-exchange value of a currency falls relative to more stable currencies. These monetarist claims have been validated across countries and over time. Interest rates in 1989 reached 8,000 percent a year in Argentina and Yugoslavia, an almost twentyfold increase in one year. Between 1985 and 1990 the Argentine australes depreciated against the dollar from 0.80 to 1, to 6,000 to 1. In the same period the Yugoslav dinar went from 0.03 to 10.6, and Brazilian currency (under various names) fell from 0.01 to 177.

In each of these countries, as in others experiencing rapid inflation, sustained high growth of money was followed by a flight from money that left the currency worthless. Government's efforts to use price controls in order to hide these effects of past inflation and of anticipations of future inflation may succeed for a time, but they do not succeed permanently. Although inflation may not be reflected fully in official measures, black market or open-market rates on unofficial markets tell a more correct story.

Rising money growth and rising inflation after 1964 (see table 1) brought the Bretton Woods system of fixed exchange rates to an end. The dollar depreciated against major currencies by 20 percent (based on the Federal Reserve's index) between August 1971 and March 1973. Continued inflation during the seventies contributed to the further 12 percent depreciation of the dollar by the end of the decade. After 1980, disinflationary policy contributed to the appreciation of the dollar; the Federal Reserve index reached 143 percent of its 1973 value before falling again during the period of more rapid money growth from 1985 to 1987 (see table 1).

Again, there is not a one-to-one relation between inflation and currency depreciation. Other factors—such as growth of defense spending, government purchases, tax rates, productivity growth at home and abroad, and foreign decisions—affect currency values. But sustained inflation induces depreciation, and disinflation induces appreciation, as monetarist theory implies.

When inflation increases, output often rises for a time above its trend rate. Reductions of inflation have the opposite effect; output falls or grows at less than trend rate. These temporary changes in the growth rate of output illustrate a third monetarist proposition: the first effects of changes in money growth are on output; later, the rate of inflation changes. The synchronous reduction in money growth in most of the industrial countries at the beginning of the 1980s produced a severe downturn in many of these countries. The size and duration of the downturn differed from country to country. The United States experienced a sharp contraction of real output; output fell by 2.5 percent in 1982 and the unemployment rate rose above 10 percent. Germany and much of Europe experienced a much longer recession; unemployment rates in France and Italy rose annually from 1981 to 1986 and were between 10 percent and 11 percent at the end of the period, while Germany's unemployment rate reached a peak above 9 percent in 1985. Japan escaped with only a modest reduction in the growth rate of output.

Not all recessions are caused by monetary change, but many are. During the past thirty years in the United States, money growth declined markedly from its previous trend in 1960, 1966, 1969, 1974, 1979, and 1989. In each instance the growth of output fell in the same year or the succeeding year, and recessions occurred in many of these years. Other countries show a similar association between reductions in money growth and reductions in the growth of output. For example, Germany slowed its money growth from a 10.4 percent average rate in 1977 to 1979, to a 1.8 percent rate in 1980 and 1981. Real output fell in 1981 and 1982. Later, inflation fell from a peak rate of 4.8 percent in 1980 to between 2 and 3 percent in the middle of the decade.

Similarly, Britain reduced its money growth from an average of 14 percent for the 1976-79 period, to a 6 percent average rate for 1979 to 1982. Output fell in 1980 and 1981. The recession in these years was the longest and deepest of the postwar years. By 1983, output was rising, and inflation had been brought from about 15 percent to 4 to 5 percent. With lower inflation, market interest rates declined and the pound sterling appreciated in value.

These monetarist propositions about inflation, interest rates, exchange rates, and output are now widely accepted by academic economists and policymakers. Many central bankers have adopted targets or guidelines for money growth. Conversations with central bank governors these days find them more alert to the risks of inflation, more conscious of the costs of slowing inflation once the inflation has become widely anticipated, and more aware of the long-term relation between money growth and inflation.

Contrast the responses of the United States and Japanese central banks to the oil shocks in 1973 to 1974 and in 1979 to 1980. Between 1973 and 1975 U.S. and Japanese money (currency plus checking deposits) rose by 10 percent and 29 percent, respectively, and consumer prices rose by 20 percent and 35 percent. In the 1979-81 period the relative positions reversed. The U.S. money stock increased by 14 percent and consumer prices rose 25 percent; in Japan money and prices rose by 5 percent and 13 percent, respectively. A lesson learned from these different approaches to the common experience, and the analyses of that experience, is that oil shocks can change the price level, but if money growth remains moderate, the surge in prices will be temporary and short-lived. In 1982, Japanese prices rose by 2.7 percent, and by the middle of the decade prices were stable.

Experience during the war over Kuwait showed again that maintaining relatively slow money growth (in the United States, Britain, and Japan, for example) assured that the one-time oil price increase had a short-lived, temporary effect on measures of inflation. Monetarists have emphasized the distinction between one-time price-level changes and the sustained rates of change that are properly called inflation.

Academic and professional opinion has now accepted several of the monetarist propositions that many once regarded as wrongheaded or even heretical. Central bankers in leading countries, including the United States, no longer offer a laundry list of important objectives. They now most often describe their principal task as the maintenance of price stability. Countries like Italy, France, and Britain, with a history of inflationary policy, tie their currencies to the German mark to borrow credibility from the successful, low-inflation policies of the Bundesbank. And the Bundesbank sets a target for the growth rate for the money stock that it achieves much of the time. Just as important, consumers and producers believe that the directors of the Bundesbank will not persistently exceed their monetary target.

Keynesians and Monetarists

The Keynesian tradition gave government the responsibility for stabilizing an unruly economy. Keynesians developed the notion of a fiscal/monetary mix to control spending and the balance of payments simultaneously. Judicious, well-timed changes in taxes and government spending were to be balanced against propitious changes in money to control the economy. The famous Phillips curve trade-off supposedly gave economists a tool for choosing between inflation and unemployment. If the choice didn't work out as intended, Keynesians relied on informal price and wage controls, jawboning (threats), and guideposts to improve the trade-off. Under flexible exchange rates they urged international policy coordination and selective exchange-market intervention to manage the global economy. In these and other ways they presented economists as engineers who adjust the controls and, when necessary, design new controls to maintain just the right mix of policies.

To know when and how much to adjust policies, Keynesian economists developed forecasting models. Some had hundreds of equations. On large-scale computers the models could simulate possible policy changes to predict their effect and more closely adjust the mix of policy actions.

Monetarists have always been critical of these models and their use in policy. They favor stable policy rules that reduce variability and uncertainty for private decision makers. They argue that government serves the economy best by enhancing stability and acting predictably, not by trying to engineer carefully timed changes in policy actions. Monetarists saw such efforts as frequently destabilizing (that is, doing the opposite of what they were supposed to do).

The attempt to apply Keynesian policies, notably in the United States and Britain, produced alternating periods of rising inflation and rising unemployment, not the finely adjusted trade-off that the Keynesians sought. As inflation increased during the late sixties and seventies (see table 1), unemployment rose from the 3½ to 4 percent range of the late sixties to the 6 to 7 percent range of the late seventies. Lower inflation in the late eighties was accompanied by lower unemployment rates, about 5½ percent in the last years of the decade.

Instead of a carefully crafted adjustment of domestic output and the balance of payments, Keynesian policies brought the world economy a surge of inflation, unprecedented in peacetime history. Later, increases in oil prices added to the problem of rising prices, but the oil price increases were themselves a reaction, at least in part, to the surge in the world price level.

Forecasting proved a weak foundation for policy actions. The best forecasts of spending, output, prices, and inflation proved to be unreliable. Systematic studies of forecasting accuracy show that on average forecasters have been unable to distinguish between booms and recessions a quarter or a year ahead, so they are as likely to mislead as to benefit policymakers. The records of the Federal Reserve that have become available show that during the period of rising inflation, annual inflation was always underpredicted. When inflation fell in the eighties, the Federal Reserve persistently predicted too high an inflation rate. A vast amount of research has shown that econometric models cannot accurately forecast interest rates and exchange rates. This research concludes that changes in interest rates and exchange rates are caused mainly by unforeseen changes in policy and in the economy.

Inflation put an end to the Bretton Woods system of fixed but adjustable exchange rates. The Bretton Woods system of fixed exchange rates required all countries to accept the inflationary consequences of U.S. economic policy. Once the system ended, countries were free to adopt independent policies. Many did just that. Of particular interest are the policies of Japan, Germany, and Switzerland. These countries undertook to lower inflation by gradual but persistent reductions in money growth. Later, several European countries adopted medium-term fiscal strategies. And although countries did not call their actions "rules" and did not always follow their rules, the general approach is much closer to the monetarist prescription for policies based on rules than to Keynesian activist intervention.

Nowhere was the change more apparent than in Britain in the eighties. A medium-term fiscal plan designed to achieve gradually lower tax rates, persistent reductions in money growth, and an end to exchange controls and wage-price guidelines were Margaret Thatcher's main macroeconomic reforms. These reforms produced a revival of growth and confidence. In the eighties, for the first time in many decades, Britain's economy outperformed most other industrial economies. Not all of the British reforms were monetarist prescriptions, but the shift toward rules or medium-term strategies and the reduction in money growth and inflation were key parts of the policy.

Later, the monetary policy was changed. Instead of controlling domestic money growth to maintain domestic price stability, the chancellor of the Exchequer, Nigel Lawson, told the Bank of England to control the exchange rate against the German mark and other European currencies.

I believe this was a poor choice. It illustrates that a fixed exchange rate does not prevent inflation at home if there is high money growth in the country (Germany in this case) to which the currency is fixed. From 1985 to 1988, the growth of German money (currency plus checking deposits) averaged 9.5 percent a year. The result for Britain was higher money growth followed by booming demand and higher inflation, then by a deflationary policy and a recession. Trying to keep the pound level with the mark left Britain with the highest interest rates and inflation among major countries. The spending boom, the return of inflation and high interest rates, and later, the onset of recession show the familiar monetarist associations of money growth with inflation and high interest rates, unanticipated increases in money growth with booms, and unanticipated reductions of money growth with recessions. In 1989 and 1990 Germany reduced its money growth rate to 6.5 percent and 4.5 percent. Britain's money growth fell sharply from 11 percent in 1988 and 14 percent in 1989 to 7.5 percent in 1990. As monetarist theory predicts, unemployment rates rose from a low of 7 percent to more than 10 percent by 1991.

Why the Skepticism?

Although monetarism is as alive and well as ever, considerable skepticism and contrary opinions can be found.

There are two factors behind the skepticism. First, the Federal Reserve's "monetarist" experiment in the early eighties is generally described as a failure. The presumed reasons for the alleged failure differ, but prominent among them is a relatively large increase in the demand for money in 1982. Second, the critics and the monetarists have very different policy agendas. The critics see government policy action as a way of removing instability caused by unruly private behavior. They have long advocated activist policies to control spending.

When taxes and spending proved to be less flexible and their influence on output and prices less potent than Keynesians (and other activists) believed, many of the advocates of activist policy shifted attention to monetary policy. They hoped to use changes in money, credit, and interest rates to fine-tune the economy. Some monetarists may have encouraged this behavior by making short-term forecasts (that often proved wide of the mark) and by overstating what monetarism could deliver. Monetary relations are a basis for policy rules, not for short-run policy activism.

Leading monetarists were very critical of the Federal Reserve's experiment at the time. They pointed out that the Fed made very few of the technical changes needed to make the experiment a success. Further, using measures of the money stock and estimates of the demand for money to predict income or spending proved to be inaccurate and misleading in 1981 and 1982.

Short-term monetarist forecasts went awry as a result. Monetarists did not predict the rapid fall in inflation after 1982 or the magnitude of the decline in output in 1982. The same can be said, however, for all other systematic efforts to forecast the economy. The Congressional Budget Office (CBO), for example, substantially underestimated the recession of 1982 and the decline in inflation in 1983. In February 1982 it predicted a 0.1 percent decline in real (inflation-adjusted) gross national product for 1982. The actual change was a decline of 2.5 percent. CBO also forecast that inflation, as measured by the GNP deflator, would be 7.5 percent in 1982 and 7.3 percent in 1983. The actual inflation was 6.4 percent and 3.9 percent.

The lesson to be learned is that economics does not deliver tight forecasts of economic variables. Economists' forecasts are probably the best forecasts available. But they are not good enough to form a reliable basis for setting policies designed to stabilize the economy in the short run. An adaptive, monetarist rule that adjusts to reflect past experience reduces some of these problems. The adaptive rule calls for money growth to adjust to an average of recent changes in the growth rates of output and the demand for money. An adaptive rule of this kind would not eliminate all fluctuations. But it would do a substantially better job of stabilizing the economy and avoiding inflation than policies based on forecasts. Some countries have learned that lesson—the monetarist lesson. They have low inflation, strong currencies, and greater stability. Unfortunately ours is not yet one of them.