Keynes’s ideas fundamentally changed the way economists think about business cycles. They did not, however, go unquestioned. In the wake of the success of government expenditures in ending the Great Depression, Keynesian economics faced a new series of challenges. As a result, by the 1980s the consensus of macroeconomists retreated somewhat from the strong version of Keynesianism that prevailed in the 1950s. In particular, many economists began to suggest limits to the effectiveness of macroeconomic policy activism.

Many macroeconomists agree with Keynes’s view that during a depression monetary policy would be relatively ineffective. We met this phenomenon in Chapter 31 in what we called the liquidity trap, a situation in which monetary policy is ineffective because the interest rate is down against the zero bound. Remember that at the zero bound, interest rates cannot be pushed down any further to revive the economy, so expansionary monetary policy has no effect. In the 1930s, when Keynes wrote, interest rates were, in fact, very close to 0%.

When the era of near-0% interest rates came to an end after World War II and the economy had recovered, the pendulum had swung so far in favor of Keynesian economics that many economists continued to emphasize fiscal policy and downplay the usefulness of monetary policy. Eventually, however, the pendulum swung partly back as macroeconomists eventually reassessed the importance of monetary policy.

A key milestone in this reassessment was the 1963 publication of A Monetary History of the United States, 1867–1960 by Milton Friedman, of the University of Chicago, and Anna Schwartz, of the National Bureau of Economic Research. Friedman and Schwartz showed that business cycles had historically been associated with fluctuations in the money supply. In particular, the money supply fell sharply during the onset of the Great Depression. As we mentioned in the opening story, Friedman and Schwartz persuaded many, though not all, economists that the Great Depression could have been avoided if the Federal Reserve had acted to prevent that monetary contraction by increasing the monetary base. They persuaded most economists that monetary policy should play a key role in economic management.

The revival of interest in monetary policy was significant because it suggested that the burden of managing the economy could be shifted away from fiscal policy—meaning that economic management could largely be taken out of the hands of politicians. This feature was attractive to many because fiscal policy necessarily involves political choices. If the government tries to stimulate the economy by cutting taxes, it must decide whose taxes will be cut. If it tries to stimulate the economy with government spending, it must decide what to spend the money on. As a result, management of the economy would often be bogged down by the political process if fiscal policy were the only tool available.

Monetary policy, in contrast, does not involve such political choices: when the central bank cuts interest rates to fight a recession, it cuts everyone’s interest rate at the same time. So a shift from relying on fiscal policy to relying on monetary policy makes macroeconomics a more technical, less political undertaking. In fact, as we learned in Chapter 29, monetary policy in most major economies is set by an independent central bank that is insulated from the political process.

After the publication of A Monetary History, Milton Friedman led a movement that sought to eliminate all forms of macroeconomic policy activism—fiscal and monetary. Instead, he asserted that the best way to manage the economy was with non-activist or nondiscretionary monetary policy. Monetarism asserts that GDP will grow steadily if the money supply grows steadily. According to the monetarist policy prescription, the central bank should target a constant rate of growth of the money supply, such as 3% per year, and maintain that target regardless of any fluctuations in the economy.

It’s important to realize that monetarism retained many Keynesian ideas. Like Keynes, Friedman asserted that the short run is important and that short-run changes in aggregate demand affect aggregate output as well as aggregate prices. Like Keynes, he argued that macroeconomic policy should have been much more expansionary during the Great Depression, even if he believed that only monetary policy was needed.

Monetarists argued, however, that in most cases activist macroeconomic policy to smooth out the business cycle actually makes things worse. In Chapter 28 we described how lags can cause problems for discretionary fiscal policy. For example, a government that tries to respond to recessions by increasing spending sometimes finds that by the time it realizes that a recession is underway, takes action, and gets results, the recession is over, and the spending increase feeds a boom instead of fighting a slump. According to monetarists, discretionary monetary policy, changes in the interest rate or the money supply by the central bank in order to stabilize the economy, faces similar problems and can easily make the economy less stable.

Friedman also argued that if the central bank followed his advice, adopting a non-activist monetary policy and refusing to change the money supply in response to fluctuations in the economy, fiscal policy would be much less effective than Keynesians believed due to crowding out—when government spending crowds out private investment spending. In Chapter 25 we analyzed how this can occur: government spending leads to deficits, which drive up interest rates and reduce investment spending. Friedman and others pointed out that if the money supply is held fixed while the government pursues an expansionary fiscal policy, crowding out will occur as the interest rate rises, limiting the effect of the fiscal expansion on aggregate demand.

Figure 33-4 illustrates this argument. Panel (a) shows aggregate output and the aggregate price level. AD₁ is the initial aggregate demand curve and SRAS is the short-run aggregate supply curve. At the initial equilibrium, E₁, the level of aggregate output is Y₁ and the aggregate price level is P₁. Panel (b) shows the money market. MS is the money supply curve and MD₁ is the initial money demand curve, so the initial interest rate is r₁.

Now suppose the government increases purchases of goods and services. We know that this will shift the AD curve rightward, as illustrated by the shift from AD₁ to AD₂, and that aggregate output will rise, from Y₁ to Y₂, and the aggregate price level will rise, from P₁ to P₂. Both the rise in aggregate output and the rise in the aggregate price level will, however, increase the demand for money, shifting the money demand curve rightward from MD₁ to MD₂. This drives up the equilibrium interest rate to r₂. Friedman’s point was that this rise in the interest rate reduces investment spending, partially offsetting the initial rise in government spending. As a result, the rightward shift of the AD curve is smaller than the multiplier analysis in Chapter 28 indicated. And Friedman argued that with a constant rate of increase in the money supply, the multiplier is so small that there’s not much point in using fiscal policy, even in a depressed economy.

As we’ve already noted, Friedman didn’t favor activist monetary policy either, arguing that the problems of time lags that limit the ability of discretionary fiscal policy to stabilize the economy also apply to discretionary monetary policy. Friedman’s solution was to make monetary policy nondiscretionary, to put it on “autopilot.” The central bank, he argued, should follow a monetary policy rule, a formula that determines its actions and leaves it relatively little discretion. During the 1960s and 1970s, most monetarists favored a monetary policy rule of slow, steady growth in the money supply.

Underlying this view was the concept of the velocity of money, the ratio of nominal GDP to the money supply. Velocity is a measure of the number of times the average dollar bill in the economy turns over per year between buyers and sellers (e.g., I tip the Starbucks barista a dollar, she uses it to buy lunch, and so on). This concept gives rise to the velocity equation, which relates nominal GDP to the money supply:

Monetarists believed, with considerable historical justification, that the velocity of money, V, was stable in the short run and changes only slowly in the long run. As a result, they claimed, steady growth in M, the money supply, by the central bank would ensure steady growth in spending, and therefore in GDP.

Monetarism strongly influenced U.S. monetary policy in the late 1970s and early 1980s as the Fed tried to keep the rate of growth in the money supply constant. It quickly became clear, however, that this didn’t ensure steady growth in the economy: the velocity of money wasn’t stable enough for such a simple policy rule to work. Figure 33-5 shows how events eventually undermined the monetarists’ view. The figure shows the velocity of money, as measured by the ratio of nominal GDP to M1, from 1960 through mid-2014. As you can see, until 1980 velocity followed a fairly smooth, seemingly predictable trend. After the Fed began to adopt monetarist ideas in the late 1970s and early 1980s, however, the velocity of money began moving erratically—probably due to financial market innovations such as the greater use of credit cards.

Consequently, traditional monetarists—those who believe that GDP will grow steadily if the money supply grows steadily—are hard to find among today’s macroeconomists. As we’ll see shortly, however, the concern that originally motivated the monetarists—that too much discretionary monetary policy can actually destabilize the economy—has become widely accepted.

The problem of time lags in the implementation of activist macroeconomic policy was not the only criticism leveled at Keynesian economics. Another serious concern arose over its effect on inflation. During the 1940s and 1950s, many Keynesian economists believed that expansionary fiscal policy could be used to achieve full employment on a permanent basis. By the 1960s, however, many economists realized that persistently expansionary policies could cause problems with inflation. Yet they still believed that governments could choose to keep unemployment low if they were willing to accept higher inflation.

In 1968, however, Milton Friedman and Edmund Phelps of Columbia University, working independently, argued that there isn’t actually a long-run trade-off between unemployment and inflation. According to Friedman and Phelps’s natural rate hypothesis, any attempt to keep unemployment below a minimum level would lead not just to inflation but to ever-rising inflation. We described the logic of this argument in Chapter 31; the important point to recognize here is that if true, the natural rate hypothesis implies that Keynesian policies can’t accomplish as much as macroeconomists previously believed. Because the government can’t keep unemployment below the natural rate, its task is not to keep unemployment low but to keep it stable around the natural rate—to prevent large fluctuations in unemployment above or below the natural rate.

And the natural rate hypothesis was, in fact, accepted by most economists after the 1970s. The Friedman–Phelps hypothesis made a strong prediction: that the apparent trade-off between unemployment and inflation would not survive an extended period of rising prices. Once inflation was embedded into the public’s expectations, it would continue even in the face of high unemployment. Sure enough, that’s exactly what happened in the 1970s. This accurate prediction was one of the triumphs of macroeconomic analysis. It convinced the great majority of economists that the natural rate hypothesis was correct, although some macroeconomists believe that at very low rates of inflation or deflation the hypothesis doesn’t work.

One final challenge to Keynesian economics focused not on the validity of the economic analysis but on its political consequences. A number of economists and political scientists pointed out that activist macroeconomic policy lends itself to political manipulation.

Statistical evidence suggests that election results tend to be determined largely by the state of the economy in the months just before the election. In the United States, if the economy is growing rapidly and the unemployment rate is falling in the six months or so before Election Day, the incumbent party tends to be reelected even if the economy performed poorly in the preceding three years. This creates an obvious temptation to abuse activist macroeconomic policy: pump up the economy in an election year, and pay the price in higher inflation and/or higher unemployment later. The consequence will be unnecessary instability in the economy, a political business cycle caused by the use of macroeconomic policy to serve political ends.

An often-cited example is the combination of expansionary fiscal and monetary policy that led to rapid growth in the U.S. economy just before the 1972 election and a sharp acceleration in inflation after the election. Kenneth Rogoff, a highly respected macroeconomist who served as chief economist at the International Monetary Fund, has proclaimed Richard Nixon, the president at the time, “the all-time hero of political business cycles.”

As we’ve learned, one way to avoid a political business cycle is to place monetary policy in the hands of an independent central bank, insulated from political pressure. The political business cycle is also a reason to limit the use of discretionary fiscal policy to extreme circumstances like a liquidity trap.