Thursday, November 3, 2011

Schlaes contra Nominal GDP Targeting

Amity Shlaes criticized Nominal GDP targeting. She focuses on the Goldman-Sachs proposal for a 4.5 percent growth path. While she generally has the proposal right, she seems unable to distinguish between a proposal for a growth path for nominal GDP and a target for a growth rate for real GDP.

Everyone wants stronger growth. By “4.5 percent growth,” however, we generally mean real growth. Under NGDP targeting, it’s possible for the Fed to get a growth rate of 4.5 percent, of which 3.5 percentage points are inflation and only 1 percentage point real. That hardly accords with the spirit of the line in the Fed statute about increased production.

The goal of the proposal is to target to volume of spending on goods and services in the economy. In equilibrium, a 4.5 percent rate of spending, when combined with the historic 3 percent in the productive capacity of the economy would generate a 1.5 percent inflation rate. According to the CBO, the U.S. entered a productivity slowdown five years ago, and it is projected to continue for the next decade. Presumably, that is why the Goldman Sachs proposal is for 4.5 percent growth. Potential output may grow only 2.5 percent, and so a 4.5 percent increase in spending on output will result in 2 percent inflation. In equilibrium, the Goldman Sachs proposal will generate the inflation rate that the Fed appears to be targeting.

My preference is a 3 percent growth path for nominal GDP. This is equal to the long run trend growth in potential output and so should result in zero inflation on average. If potential output has really grown slowly for the last few years and that continues, the result would be low inflation, close to 1/2 percent. The best way to address the problem is supply-side reforms.

While the long run consequences of a 5 percent, 4.5 percent, or 3 percent growth path for nominal GDP is not going to be inflationary, the Goldman-Sachs proposal is not for 4.5 percent nominal GDP growth from its current value. It is rather to move to a 4.5 percent growth path starting from the end of 2007. This involves moving at least part of the way to the growth path for nominal GDP that the economy was one between 1984 to 2007. The proposed growth rate for the next two years is closer to 10 percent.

Is 10 percent nominal growth unrealistic? Is it inherently inflationary? The growth rate of nominal GDP was over 10 percent for most of 1983 during the Reagan-Volcker recovery from the recession of 1982. The inflation rates during those years was about 3 percent. But did not result in progressively higher inflation.

Shlaes claims that economists like Taylor and Friedman would oppose this rule. While Friedman did advocated slow, stable growth for the quantity of money, he claimed that velocity was very stable. If velocity is stable, then a stable growth path for the quantity of money is the same thing as a stable growth path for nominal GDP. What would happen to the price level and inflation due to supply shocks, like bad harvests or an increase in the price of oil, or even a persistent productivity slowdown are exactly the same as with a nominal GDP target. (Friedman's proposal for 3 percent money growth with stable velocity is the same thing as a 3 percent nominal GDP target.)

The difference between a money supply rule and a nominal GDP rule, is that the nominal GDP rule changes the quantity of money to accommodate changes in the demand to hold money. In other words, the quantity of money changes to offset changes in velocity. Friedman believed that the Fed could offset changes in the money multiplier and so keep the quantity of money on target. I certainly will grant that adjusting the quantity of money to offset changes in velocity is more difficult than simply offsetting changes in the money multiplier. But with large changes in velocity, there is little choice. If velocity was stable, the targeting nominal GDP would be the same thing as targeting the growth rate of the money supply.

Shlaes asked Taylor what he thought about NGDP targeting and described his preferred reform. According to Shlaes, he favors "not widening the Fed’s growth mandate, but rather removing it." In other words, he favors targeting inflation only.

Generally, the "Taylor rule," has the Fed manipulate short term interest rates based on past or expected future inflation rates and the past or expected future output gap between real output and potential output. Taylor favors dropping the output gap portion of "his" rule.

The "rule" was initially an empirical regularity--what central bankers actually did during the Great Moderation. It seemed to be quite effective. Unfortunately, it has proven ineffective. Changes in short run interest rates have not proven able to close output gaps while keeping inflation under control. Inflation is under control, but the output gap is not closing. Taylor's "solution" is to keep his rule and just ignore the output gap.

The Market Monetarist perspective is that the Fed is responsible for adjusting the quantity of money to meet the demand to hold money. A failure to do so results in undesirable changes in nominal expenditure. With nominal expenditure 14 percent below the trend of the Great Moderation, it is clear that the "Taylor rule" that worked so well for 20 years has failed. It is time for something new. Targeting the growth path of nominal GDP, and letting the price level change based upon changes in productivity, is a better approach.

1 comment:

  1. We need to be very careful with not making any guessing with GDP rates because it can cause huge problem because it’s not easy to get out of bad trade if it’s done without planning. I always go with planning and due to strong money management, I am always able to work it out and even in difficult situation, I am able to handle it nicely. The biggest benefit for me comes with OctaFX broker using their 50% bonus on deposit scheme which can be used too.