Money in the Classical Model

So far we have focused on real shocks to the economy, such as productivity shocks

and changes in government purchases. However, many macroeconomists believe

that nominal shocks—shocks to money supply and money demand—also affect the classical model.

Monetary Policy and the Economy

Monetary policy refers to the central bank's decisions about how much money to supply to the economy (Chapter 7). Recall that the central bank (the Federal Reserve in the United States) can control the money supply through open market operations, in which it sells government bonds to the public in exchange for money (to reduce the money supply) or uses newly created money to buy bonds from the public (to increase the money supply).

In Chapter 9 we examined the effects of changes in the money supply, using the IS-LM model (Fig. 9.9) and the AD-AS model (Fig. 9.14). With both models we found that, after prices fully adjust, changes in the money supply are neutral: A change in the nominal money supply, M, causes the price level, P, to change pro­portionally, but a change in the money supply has no effect on real variables, such as output, employment, or the real interest rate. Our analysis left open the possi­bility that a change in the money supply would affect real variables, such as out­put, in the short run before prices had a chance to adjust. However, because classical economists believe that the price adjustment process is rapid, they view the period of time during which the price level is fixed—and money is not neutral—to be too short to matter. That is, for practical purposes, they view money as neutral for any relevant time horizon.

Monetary Nonneutrality and Reverse Causation

The prediction that money is neutral is a striking result of the classical model, but it seems inconsistent with the business cycle fact that money is a leading, procyclical variable.

²⁴Bernanke, Ben S. “The Jobless Recovery.” Remarks at the Global Economic and Investment Outlook Conference, Carnegie Mellon University, November 6, 2003. Available on-line at www.federalreserve.gov/boarddocs/speeches/2003/200311062/default.htm.

Some classical economists have responded to these questions by pointing out that, although increases in the money supply tend to precede expansions in output, this fact doesn't necessarily prove that economic expansions are caused by those increases. After all, just because people put storm windows on their houses before winter begins doesn't mean that winter is caused by putting on storm windows. Rather, people put storm windows on their houses because they know that winter is coming.

Many classical economists, including RBC economists in particular, argue that the link between money growth and economic expansion is like the link between putting on storm windows and the onset of winter, a relationship they call reverse causation. Specifically, reverse causation means that expected future increases in output cause increases in the current money supply and that expected future decreases in output cause decreases in the current money supply, rather than the other way around. Reverse causation explains how money could be a procyclical and leading variable even if the classical model is correct and changes in the money supply are neutral and have no real effects.^[184]

Reverse causation might arise in any of several ways. One possibility (which you are asked to explore in more detail in Analytical Problem 5 at the end of the chapter) is based on the idea that money demand depends on both expected future output and current output. Suppose that a firm's managers expect business to pick up considerably in the next few quarters.

Now suppose that the Fed observes this increase in the demand for money. If the Fed does nothing, leaving the money supply unchanged, the increase in money demand will cause the equilibrium value of the price level to fall. As one of the Fed's objectives is stable prices, it won't like this outcome; to keep prices stable, instead of doing nothing, the Fed should provide enough extra money to the economy to meet the higher money demand. But if the Fed does so, the money supply will rise in advance of the increase in output, consistent with the business cycle fact—even though money is neutral.

Undoubtedly, reverse causation explains at least some of the tendency of money to lead output. However, this explanation doesn't rule out the possibility that changes in the money supply also sometimes cause changes in output so that money is nonneutral. That is, a combination of reverse causation and monetary nonneutrality could account for the procyclical behavior of money.

The Nonneutrality of Money: Additional Evidence

Because of reverse causation, the leading and procyclical behavior of money can't by itself establish that money is nonneutral. To settle the issue of whether money is neutral, we need additional evidence. One useful source is a historical analysis of monetary policy. The classic study is Milton Friedman and Anna J. Schwartz's, A Monetary History of the United States, 1867-1960.^[185] Using a variety of sources, including Federal Reserve policy statements and the journals and correspondence of monetary policymakers, Friedman and Schwartz carefully described and ana­lyzed the causes of money supply fluctuations and the interrelation of money and other economic variables.

Throughout the near-century examined in detail we have found that:

1. Changes in the behavior of the money stock have been closely associated with changes in economic activity, [nominal] income, and prices.

2. The interrelation between monetary and economic change has been highly stable.

3. Monetary changes have often had an independent origin; they have not been simply a reflection of changes in economic activity.

The first two conclusions restate the basic business cycle fact that money is procyclical. The third conclusion states that reverse causation can't explain the entire relationship between money and real income or output. Friedman and Schwartz focused on historical episodes in which changes in the supply of money were not (they argued) responses to macroeconomic conditions but instead resulted from other factors such as gold discoveries (which affected money supplies under the gold standard), changes in monetary institutions, or changes in the leadership of the Federal Reserve. In the majority of these cases, "independent" changes in money growth were followed by changes in the same direction in real output. This evidence suggests that money isn't neutral.

Christina Romer and David Romer^[186] of the University of California at Berkeley reviewed and updated the Friedman-Schwartz analysis. Although they disputed some of Friedman and Schwartz's interpretations, they generally agreed with the conclusion that money isn't neutral. In particular, they argued that since 1960 half a dozen additional episodes of monetary nonneutrality have occurred. Probably the most famous one occurred in 1979, when Federal Reserve Chairman Paul Volcker announced that money supply procedures would change and that the money growth rate would be reduced to fight inflation. A minor recession in 1980 and a severe downturn in 1981-1982 followed Volcker's change in monetary policy. An economic boom followed relaxation of the Fed's anti-inflationary monetary policy in 1982.

Because of the Friedman-Schwartz evidence and episodes such as the 1979-1982 Volcker policy (and a similar experience in Great Britain at the same time), most economists now believe that money is not neutral. If we accept that evidence, contrary to the prediction of the classical model, we are left with two choices: Either we must adopt a different framework for macroeconomic analy­sis, or we must modify the classical model. In Section 10.5 we take the second approach and consider how monetary nonneutrality can be explained in a clas­sical model.

10.5