The Fed's Confession

When it comes to admitting error, central bankers tend to emulate Benjamin Disraeli, who famously said "never complain, never explain." So it was nothing short of astonishing last week for Richard Fisher, President of the Federal Reserve Bank of Dallas, to confess in public that the Fed had blundered by keeping monetary policy too easy for too long in 2003 and 2004.

Speaking to bankers in New York, Mr. Fisher issued a mea maxima culpa that deserves wide attention: "In retrospect, the real fed funds rate turned out to be lower than what was deemed appropriate at the time and was held lower longer than it should have been." As the nearby chart shows, a Fed worried about possible "deflation" moved the overnight interest rate it charges banks in June 2003 to an extraordinarily low 1% and kept it there for another year.

Mr. Fisher blamed this mistake on "poor data" that underestimated inflation, leading "to a policy action that amplified speculative activity in the housing and other markets. Today, as anybody not from the former planet of Pluto knows, the housing market is undergoing a substantial correction and inflicting real costs to millions of homeowners across the country." In other words, the Fed itself is responsible for the current housing bust because its over-easy policy created a real estate frenzy that was bound to end once the Fed tried to regain control over inflation expectations.

From: Wall Street Journal

Why Money Matters, by Milton Friedman

The third of three episodes in a major natural experiment in monetary policy that started more than 80 years ago is just now coming to an end. The experiment consists in observing the effect on the economy and the stock market of the monetary policies followed during, and after, three very similar periods of rapid economic growth in response to rapid technological change: to wit, the booms of the 1920s in the U.S., the '80s in Japan, and the '90s in the U.S.

The prosperous '20s in the U.S. were followed by the most severe economic contraction in its history. In our "Monetary History" (1963), Anna Schwartz and I attributed the severity of the contraction to a monetary policy that permitted the quantity of money to decline by one-third from 1929 to 1933. Since 1963, two episodes have occurred that are almost mirror images of the U.S. economy in the '20s: the '80s in Japan, and the '90s in the U.S. All three episodes were marked by a long period of rapid economic growth, sparked by rapid technological change and the emergence of new industries, and accompanied by a stock market boom that terminated in a crash. Monetary policy played a role in these booms, but only a supporting role. Technological change appears to have been the major player.

These three episodes provide the equivalent of a controlled experiment to test our hypothesis about what we termed the Great Contraction. In this experiment, the quantity of money is the counterpart of the experimenter's input. The performance of the economy and the level of the stock market are the counterpart of the experimenter's output... The three boom episodes all occurred in developed private enterprise market economies, involved in international finance and trade, and with similar monetary systems, including a central bank with power to control the quantity of money. This is the counterpart of the controlled conditions of the experimenter's laboratory.

The Money Supply: In addition, history has provided a close counterpart to the kind of variation in input that our hypothetical experimenter might have deliberately chosen. As Fig. 1 shows, monetary policy ... was very similar in the three boom periods, and very different in the three post boom periods, with settings that might be described as low, medium, high.

To measure the quantity of money, I use M2 in the U.S. and the conceptually equivalent M2 plus certificates of deposit in Japan. To express the data for the two countries and the widely separated periods in comparable units, I use as an index of the money stock the ratio of the quantity of money to its average value for the six years prior to the cycle peak... Finally, the data are plotted to align the dates at the cycle peak.

Fig. 1 shows a striking contrast between the period before the cycle peak and the period after the cycle peak. There are some differences before the peak -- money growth is slowest on the average for the earlier U.S. episode, fastest for Japan -- but the differences are small and there is reasonably steady money growth in all three episodes. The contrast with the period after the cycle peak could hardly be greater. Money supply declines sharply after the cycle peak in the first episode, goes from stable to rising mildly in the second, and rises steadily and sharply in the third. Our hypothetical experimenter planned his experiment well.

The GDP: The results of the third episode of this natural experiment are now all in. Fig. 2 shows how GDP in nominal terms (dollars or yen in current prices) behaved during the boom and post boom periods. I use nominal GDP rather than real GDP because M2 is also a nominal magnitude. How changes in nominal GDP are divided between prices and output is an important question but one that is not directly relevant to this experiment...

As in Fig. 1, there is a striking contrast between the boom and the post-boom periods: roughly similar growth during the booms, widely variable growth during the post-boom. Both before and after the cycle peak, nominal GDP growth paralleled monetary growth. During the boom, money and nominal GDP grew most rapidly in Japan, most slowly in the first U.S. episode, and at an intermediate rate in the second U.S. episode...

After the cycle peak, money fell sharply in the first episode and so did nominal GDP; money growth stagnated in the second episode and so did GDP; money grew at a rapid rate in the third episode and, after a brief lag (corresponding to the mild 2001 recession) so did GDP. ...

The Stock Market: The peak of the stock market, as measured by S&P's index, coincided with the cycle peak in the first episode, both occurring in the third quarter of 1929... However, that was not the case in the later episodes. ... Accordingly, Fig. 3 plots the data to align the series at the stock market peak.

The near identity of the three stock market series during the boom is truly remarkable. ... Of more interest for our purpose is what happened after the peak. For a year after, the three stock-price series fell in tandem, responding to the inner dynamics of a collapsing bubble. Then, the differences in monetary policy began to have an effect. Beginning in late 1930, the S&P index started falling away from the others under the influence of a collapsing money stock. For another year and a half, the other two indexes move in tandem. Then the much more expansive policy of the Fed in the '90s than of the Bank of Japan in the '80s takes effect and pulls the S&P 500 away from the Nikkei, which stabilizes in response to the passive monetary policy of the Bank of Japan...

The results of this natural experiment are clear, at least for major ups and downs: What happens to the quantity of money has a determinative effect on what happens to national income and to stock prices. The results strongly support Anna Schwartz's and my 1963 conjecture about the role of monetary policy in the Great Contraction. They also support the view that monetary policy deserves much credit for the mildness of the recession that followed the collapse of the U.S. boom in late 2000.