In Chapter 13 we saw how fiscal policy can be used to stabilize the economy. Now we will see how monetary policy—changes in the money supply, and the interest rate—can play the same role.

In Chapter 12 we learned that monetary policy shifts the aggregate demand curve. We can now explain how that works: through the effect of monetary policy on the interest rate.

Figure 15-7 illustrates the process. Suppose, first, that the Federal Reserve wants to reduce interest rates, so it expands the money supply. As you can see in the top portion of the figure, a lower interest rate, in turn, will lead, other things equal, to more investment spending. This will in turn lead to higher consumer spending, through the multiplier process, and to an increase in aggregate output demanded. In the end, the total quantity of goods and services demanded at any given aggregate price level rises when the quantity of money increases, and the AD curve shifts to the right. Monetary policy that increases the demand for goods and services is known as expansionary monetary policy.

Suppose, alternatively, that the Federal Reserve wants to increase interest rates, so it contracts the money supply. You can see this process illustrated in the bottom portion of the diagram. Contraction of the money supply leads to a higher interest rate. The higher interest rate leads to lower investment spending, then to lower consumer spending, and then to a decrease in aggregate output demanded. So the total quantity of goods and services demanded falls when the money supply is reduced, and the AD curve shifts to the left. Monetary policy that decreases the demand for goods and services is called contractionary monetary policy.

How does the Fed decide whether to use expansionary or contractionary monetary policy? And how does it decide how much is enough? In Chapter 6 we learned that policy makers try to fight recessions, as well as try to ensure price stability: low (though usually not zero) inflation. Actual monetary policy reflects a combination of these goals.

In general, the Federal Reserve and other central banks tend to engage in expansionary monetary policy when actual real GDP is below potential output. Panel (a) of Figure 15-8 shows the U.S. output gap, which we defined in Chapter 12 as the percentage difference between actual real GDP and potential output, versus the federal funds rate since 1985. (Recall that the output gap is positive when actual real GDP exceeds potential output.) As you can see, the Fed has tended to raise interest rates when the output gap is rising—that is, when the economy is developing an inflationary gap—and cut rates when the output gap is falling.

The big exception was the late 1990s, when the Fed left rates steady for several years even as the economy developed a positive output gap (which went along with a low unemployment rate). One reason the Fed was willing to keep interest rates low in the late 1990s was that inflation was low.

Panel (b) of Figure 15-8 compares the inflation rate, measured as the rate of change in consumer prices excluding food and energy, with the federal funds rate. You can see how low inflation during the mid-1990s, the early 2000s, and the late 2000s helped encourage loose monetary policy in the late 1990s, in 2002–2003, and again beginning in 2008.

In 1993 Stanford economist John Taylor suggested that monetary policy should follow a simple rule that takes into account concerns about both the business cycle and inflation. He also suggested that actual monetary policy often looks as if the Federal Reserve was, in fact, more or less following the proposed rule. A Taylor rule for monetary policy is a rule for setting interest rates that takes into account the inflation rate and the output gap or, in some cases, the unemployment rate.

A widely cited example of a Taylor rule is a relationship among Fed policy, inflation, and unemployment estimated by economists at the Federal Reserve Bank of San Francisco. These economists found that between 1988 and 2008 the Fed’s behavior was well summarized by the following Taylor rule:

Federal funds rate = 2.07 + 1.28 × inflation rate − 1.95 × unemployment gap

where the inflation rate was measured by the change over the previous year in consumer prices excluding food and energy, and the unemployment gap was the difference between the actual unemployment rate and Congressional Budget Office estimates of the natural rate of unemployment.

Figure 15-9 compares the federal funds rate predicted by this rule with the actual federal funds rate from 1985 to mid-2014. As you can see, the Fed’s decisions were quite close to those predicted by this particular Taylor rule from 1988 through the end of 2008. We’ll talk about what happened after 2008 shortly.

Until January 2012, the Fed did not explicitly commit itself to achieving a particular inflation rate. However, in January 2012, the Fed announced that it would set its policy to maintain an approximately 2% inflation rate per year. With that statement, the Fed joined a number of other central banks that have explicit inflation targets. So rather than using a Taylor rule to set monetary policy, they instead announce the inflation rate that they want to achieve—the inflation target—and set policy in an attempt to hit that target. This method of setting monetary policy, called inflation targeting, involves having the central bank announce the inflation rate it is trying to achieve and set policy in an attempt to hit that target. The central bank of New Zealand, which was the first country to adopt inflation targeting, specified a range for that target of 1% to 3%.

Other central banks commit themselves to achieving a specific number. For example, the Bank of England has committed to keeping inflation at 2%. In practice, there doesn’t seem to be much difference between these versions: central banks with a target range for inflation seem to aim for the middle of that range, and central banks with a fixed target tend to give themselves considerable wiggle room.

One major difference between inflation targeting and the Taylor rule method is that inflation targeting is forward-looking rather than backward-looking. That is, the Taylor rule method adjusts monetary policy in response to past inflation, but inflation targeting is based on a forecast of future inflation.

Advocates of inflation targeting argue that it has two key advantages over a Taylor rule: transparency and accountability. First, economic uncertainty is reduced because the central bank’s plan is transparent: the public knows the objective of an inflation-targeting central bank. Second, the central bank’s success can be judged by seeing how closely actual inflation rates have matched the inflation target, making central bankers accountable.

Critics of inflation targeting argue that it’s too restrictive because there are times when other concerns—like the stability of the financial system—should take priority over achieving any particular inflation rate. Indeed, in late 2007 and early 2008 the Fed cut interest rates much more than either a Taylor rule or inflation targeting would have dictated because it feared that turmoil in the financial markets would lead to a major recession. (In fact, it did.)

Many American macroeconomists have had positive things to say about inflation targeting—including Ben Bernanke (the previous chairman of the Fed). And in January 2012 the Fed declared that what it means by the “price stability” it seeks is 2% inflation, although there was no explicit commitment about when this inflation rate would be achieved.

As Figure 15-9 shows, a Taylor rule based on inflation and the unemployment rate does a good job of predicting Federal Reserve policy from 1988 through 2008. After that, however, things go awry, and for a simple reason: with very high unemployment and low inflation, the same Taylor rule called for an interest rate less than zero, which isn’t possible.

Why aren’t negative interest rates possible? Because people always have the alternative of holding cash, which offers a zero interest rate. Nobody would ever buy a bond yielding an interest rate less than zero because holding cash would be a better alternative.

The fact that interest rates can’t go below zero—called the zero lower bound for interest rates—sets limits to the power of monetary policy. In 2009 and 2010, inflation was low and the economy was operating far below potential, so the Federal Reserve wanted to increase aggregate demand. Yet the normal way it does this—open-market purchases of short-term government debt to expand the money supply—had run out of room to operate because short-term interest rates were already at or near zero.

In November 2010 the Fed began an attempt to circumvent this problem, which went by the somewhat obscure name “quantitative easing.” Instead of purchasing only short-term government debt, it began buying longer-term government debt—five-year or six-year bonds, rather than three-month Treasury bills. As we have already pointed out, long-term interest rates don’t exactly follow short-term rates. At the time the Fed began this program, short-term rates were near zero, but rates on longer-term bonds were between 2% and 3%. The Fed hoped that direct purchases of these longer-term bonds would drive down interest rates on long-term debt, exerting an expansionary effect on the economy.

Later the Fed expanded the program further, also purchasing mortgage-backed securities, which normally offer somewhat higher rates than U.S. government debt. Here too the hope was that these rates could be driven down, with an expansionary effect on the economy. As with ordinary open-market operations, quantitative easing was undertaken by the Federal Reserve Bank of New York.

Was this policy effective? The Federal Reserve believes that it helped the economy. However, the pace of recovery remained disappointingly slow.

So yes, the Fed gets what it wants.