For the last few years, the claim that an increase in economic growth leads to an increase in inflation and that decreased growth reduces inflation has been a mantra. That is the conventional wisdom in Washington and, to a lesser extent, on Wall Street. On October 12, for example, Federal Reserve Board governor Laurence Meyer stated, “Tightening monetary policy slows spending growth, opens up some slack temporarily in labor and product markets, and allows the slack to reduce inflation.” Yet, taken literally, that claim cannot be true. All other things being equal, an increase in economic growth must cause inflation to drop.

Here’s why. The seat-of-the-pants explanation of inflation is that it is caused by too much money “chasing” too few goods. It follows that the more goods that are produced, the lower the prices of goods. This connection between the level of production and the level of prices also holds for the rate of change of production (that is, the rate of economic growth) and the rate of change of prices (that is, the inflation rate).

Some simple arithmetic will help clarify. Start with the famous equation of exchange, MV = Pq, where M is the money supply; V is the velocity of money, that is, the speed at which money circulates; P is the price level; and q is the real output of the economy. If the growth rate of the economy increases, that is, if the growth rate of q increases, then, if the growth rates of M and V are held constant, the growth rate of the price level must fall. Since the growth rate of the price level is just another term for the inflation rate, the inflation rate must fall. An increase in the rate of economic growth means more goods for money to “chase,” which puts downward pressure on the inflation rate. Assume, for illustrative purposes, that the money supply grows at 6 percent a year and velocity is constant. Then, if annual economic growth is 3 percent, inflation must be 3 percent. (Actually, inflation must be 2.9 percent, which is approximately 3 percent). If, however, economic growth rises to 4 percent, inflation falls to 2 percent. (Actually, it falls to 1.9 percent.)

That higher economic growth must reduce inflation is straightforward. Why, then, does a man as brilliant as Federal Reserve chairman Alan Greenspan not get it? Actually, he does. He just never says things simply and straightforwardly when he can be complicated and obtuse instead. In 1995 Chairman Greenspan testified before the Senate Committee on Finance: “One factor in judging the inflationary risks in the economy is the potential for expansion of our productive capacity. If ‘potential GDP’ is growing rapidly, actual output can also continue to grow rapidly without intensifying pressures on resources.” Translation: if growth is higher, inflation is lower.

Can Productivity Grow?

Of course, it’s hard to know at any given time what the potential growth rate of an economy is. Various macro economists, especially those trained in the Keynesian tradition, may tell us that the upper limit on an economy’s long-term growth rate is 2 or 2.5 percent. But if you push them for their reasons, the best they can do is to tell you that the economy’s growth rate equals the sum of employment growth and productivity growth, that employment growth can’t be much more than 1 percent in the long run, and that productivity can’t be expected to grow by much more than about 1.25 percent a year. The first statement is necessarily true: an economy’s growth rate does equal the growth of employment plus the growth rate of output per worker (productivity). The second statement is probably true, but it depends on the unstated, and crucial, assumption that limits on immigration to the United States are not progressively relaxed: allowing the number of workers admitted into the United States to increase by a million a year, for example, would increase the U.S. labor force by about 0.8 percent.

The third statement is the most controversial. People who believe that productivity growth will be limited point to the past: between 1973 and 1993, for example, measured productivity—output per man-hour—in the United States increased by an annual average of 1.23 percent. But that past says little about the future. Even in the recent past, between 1993 and 1998, productivity growth averaged 1.46 percent, above what the “growth skeptics” thought was the upper limit. This higher productivity growth is one of the reasons that real gross domestic product has grown by an average of 3.4 percent, much higher than the 2 to 2.5 percent that was anticipated by the growth skeptics. (The other, more important reason for higher growth is that employment grew by 2.56 percent a year between 1993 and 1998.)

There is strong reason to believe that productivity can grow by 1.5 to 2 percent a year. Analysts who have studied the information technology (IT) industries can point to many ways in which the Internet, computers, and software will be able to increase productivity further. Some have compared the role of the IT industries in the current U.S. economy to the economic role of electricity toward the end of the 19th century before electricity had been completely integrated into the economy. If they’re right, then we can expect an explosion of growth as the IT industries are further integrated into the economy. Moreover, exciting advances in biotechnology and artificial intelligence will likely lead to further productivity improvements over the next 10 to 20 years. Stock prices have risen greatly over the last 5 years because the market has increasingly come to anticipate large productivity increases, some of which will be captured by the firms that create them.

What Greenspan Can Know

Of course, I could be wrong about future growth and the growth skeptics could be right. Why does it matter? This gets us back to the connection between economic growth and inflation. The reason it matters to Greenspan is that it affects his decision and that of his colleagues on the Federal Open Market Committee about how quickly to increase the money supply. Greenspan has made clear that his main goal as Fed chairman is to prevent inflation from increasing and gradually bring it down to zero. In the 12 years of his chairmanship of the Federal Reserve, from 1987 to the present, inflation has averaged 3.2 percent, and in the last 8 years it has averaged 2.6 percent. But because there is a lag in the effect of monetary policy on the economy, when Greenspan and the Fed control the growth of the money supply by buying and selling bonds, they affect only future inflation. That’s why Greenspan would like to know how fast the economy can grow. If the economy’s future growth rate is, say, one percentage point higher than previously thought, then, for a given inflation target, the Fed should buy bonds to increase the growth rate of the money supply by an additional percentage point. If, on the other hand, the economy’s future growth rate is one percentage point lower than previously thought, then the Fed should sell bonds to reduce the growth rate of the money supply by an additional percentage point. Unfortunately, no Federal Reserve chairman, no matter how brilliant, and no one else, for that matter, can know for certain the economy’s future growth rate. That’s why it makes sense for Federal Reserve officials to pay attention to indicators of future growth and of future inflation.

But what does not make sense is for the Federal Reserve chairman to pay much attention to three of the main economic variables that he talks about in his speeches and testimony: the recent inflation rate, the recent unemployment rate, and the recent economic growth rate. Those variables all have two big limitations. First, they are, necessarily, known only after the fact. The inflation rate and unemployment rate that are reported early every month are the rates for the previous month; the economic growth rate reported every three months is the government’s estimated growth rate for the previous quarter, and the government’s estimates are often wrong—sometimes wildly wrong. Second, the recent inflation rate, unemployment rate, and economic growth rate, to the extent they are affected by monetary policy, are the result of past monetary policy. They aren’t much help in figuring out how to run monetary policy now and in the future. What Greenspan and the Fed care about are the future values of those three variables, especially the inflation rate, and those values will be the result of recent and future monetary policy.

Consider a concrete example. Let’s say that, in a hypothetical October, the unemployment report from the federal government’s Bureau of Labor Statistics shows that the unemployment rate fell from 4.5 to 4.2 percent in September and the Department of Commerce’s October report shows that during the summer quarter the annualized growth rate of real GDP rose from 2.5 to 3.2 percent. Most commentators, seeing those data, report that the economy is heating up. But they don’t know that. All they know is that the economy did heat up. The data reported tell them literally nothing about what the economy is doing now, let alone what it will do in the future. Moreover, to the extent the increase in growth and decrease in unemployment were due to monetary policy, the monetary policy that affected them was implemented sometime earlier in the year or even in the previous year.

Why, then, doesn’t the Federal Reserve chairman stop looking at those data and, instead, settle on a constant growth rate of the money supply, as monetarist Milton Friedman has recommended for decades? Friedman has argued that by causing the money supply to grow by, say, 3 percent a year, you can have zero inflation because the economy will grow by 2 to 3 percent a year, and the resulting inflation rate will be somewhere between zero and 1 percent. There are two main problems with a constant-money-supply-growth rule. First, as has already been stated, you don’t know that the economy will grow by 2 to 3 percent. It could grow by only 1 percent or by 4 percent. Second, Friedman’s reasoning assumes that the velocity of money, which is inversely related to the holding of real cash balances, is constant. That assumption has been badly wrong in the last 20 years—velocity has bounced around a lot. If, to take a concrete example, the money supply grew by 3 percent but velocity fell by 4 percent, then nominal GDP, which is Pq in the equation of exchange, would fall by 1 percent. If that happened, but the economy’s real growth rate were 3 percent, then we would have 2 percent deflation. This is not necessarily bad. There would be a problem if the decline in velocity were unanticipated and prices didn’t fall to keep the economy in equilibrium, in which case real output would fall and a recession would occur. The instability of the velocity of money is the main reason that few economists are willing to advocate a constant-money-growth rule anymore.

Prices Are Indicators

So it doesn’t make much sense to focus on past unemployment, past economic growth, or past inflation, but simply ignoring the data and implementing Friedman’s constant-money-growth rule doesn’t seem like a good idea either. So what should a Fed chairman do? Because he cares about the future, he should pay the most attention to data that are indicators of future inflation. And the best data are those generated by the market, reflecting, as they do, the combined information and wisdom of millions of market participants. Even on the question of whether stock prices are too high, an issue on which he has at times doubted the collective wisdom of the market, Greenspan has expressed appropriate humility about market prices. In 1999 he testified before the Joint Economic Committee: “To spot a bubble in advance requires a judgment that hundreds of thousands of informed investors have it all wrong. Betting against markets is usually precarious at best.”

Similarly, the price of gold has often been a good indicator of future inflation: a large increase in the price of gold often indicates that the market has come to expect inflation in the future, and a large decrease often precedes a decline in inflation. Even more tight ought to be the relation between expected inflation and the spread between the yields on ordinary Treasury bonds and inflation-indexed Treasury bonds of the same maturity. Both the price of gold and this yield spread are the result of the judgments of market participants about the future, not the past. For that reason, they are far more valuable than knowledge of last month’s unemployment rate.

The idea that higher growth causes higher inflation has had one main pernicious effect on government policy. One of the latest arguments for not deregulating and not cutting government spending or taxes is that, even if such policies do increase growth, Greenspan, fearing that growth causes inflation, will raise interest rates (by reducing the growth of the money supply) and choke off increased growth. So why bother, goes the argument, undertaking policies that will increase the economy’s potential growth rate? This is patently absurd. Greenspan does seem to worry that high economic growth indicates that the money supply has grown too much. But he also pays attention to taxes, spending, and regulatory policy and understands clearly the effect they can have, for good or ill, on the economy’s growth potential. If cuts in taxes, spending, and regulations raised the economy’s growth potential, Greenspan would recognize that fact and would probably not try to use monetary policy to rein in growth.

One last point: this whole discussion has taken as given that the government, through the Federal Reserve system, controls the money supply. But as Nobel-prize-winning economist Friedrich Hayek pointed out over 20 years ago, there is a case for denationalizing money, that is, getting the government out of the business of providing money and allowing competing money supplies. University of Georgia economists George Selgin and Lawrence White, and other economists, have made persuasive cases for getting the government out of the money business. But that is another story.

The bottom line is that an increase in economic growth must, all else being equal, cause a decline in inflation.

This article originally appeared in the November/December 1999 edition of Cato Policy Report.