How and Why the Fed Must Change in Its Second Century
This is an edited and condensed version of Allan Meltzer's remarks on December 14, 2012, at the Cleveland Fed’s conference Current Policy Under the Lens of Economic History.
The original Federal Reserve of 1914 had very little authority; it was bound by the gold standard and the real bills doctrine. It acted mainly when banks came to it, not when it went to the market. It was prevented by law from financing the Treasury.
But the Fed soon circumvented that by developing open market operations so that it could buy the government’s bonds on the day of the sale. It had a series of semi-autonomous Reserve Banks with independent directors. Are the directors still independent? The Reserve Banks are certainly not semi-autonomous. All of those restrictions are gone. With crises, power has moved to the Board of Governors of the Federal Reserve System. From the very first, the argument was that the Board was political and the Reserve Banks were market oriented. But the Board’s power has increased enormously, and the Reserve Banks’ power has decreased commensurately.
The Federal Reserve is engaged in debt management, credit allocation, and fiscal actions, especially in the mortgage market. Why are we doing all these things now? Maybe it’s because fiscal policy is broken; that is, given the deficits, it’s really hard to ask for much more spending. So the Congress, or parts of it, has taken the idea that they can use credit allocation to do that. If you’re going to use the banks to allocate credit to the poor and disadvantaged, who will give credit to keep the economy growing, to take risks for investment? Is that going to be done somewhere, or do we not care about that?
A look back
We’ve been here before. In wartime, in post-war debt management, in policy coordination, and in activities in the 1960s. Former Chairman William McChesney Martin’s view was that the Fed was independent within the government. He explained many times what that meant. It meant that the Congress and the president adopted a budget and the Fed helped finance it. That was policy coordination. He didn’t like policy coordination, but he followed it almost until the very end of his term in office.
Fed history shows many errors. There was the Great Depression, the Great Inflation, and the contributions to the recent crisis. A large literature points out the acquiescence or support of many other failures. There were other errors—failure to raise the Regulation Q (bank interest rate) ceilings, which tore up a good part of the investment of the mortgage banking system; and a failure to distinguish real and nominal interest rates.
In the 1970s, the Fed failed to make the distinction between real and nominal rates. This failure led the Fed to believe that interest rates were very high, when in fact real interest rates were mostly negative. Milton Friedman, in his 1968 address to the American Economic Association, observed that there were differences between monetary and real variables and it was important to keep that in mind. But the Fed was slow to act on Friedman’s advice and through the 1970s it kept trying to improve the economy by inflating. In the 1970s, the Phillips curve underestimated inflation 16 quarters in a row.
Then came Chairman Paul Volcker, who offered the anti-Phillips curve. Publicly, internally, in testimony, and over and over, Volcker said the way to lower unemployment is to lower expected inflation. That lesson has been lost again.
Volcker pointed out that in the 1970s, inflation and unemployment declined together. He said people were presuming there was a tradeoff between inflation and unemployment. But he looked at the data and saw that the unemployment rate went up as the inflation rate went up, and he believed the unemployment rate would come down as the inflation rate came down. He preached that message over and over again to his staff and to anyone within hearing range. And he was right. That is what happened.
Once again, the Fed is failing to distinguish between real and nominal problems. One of the real problems is the uncertainty that comes from the fact that businesses cannot discount their cash flow. How can you discount your cash flow if you don’t know what the tax rate is going to be, or how to estimate health care costs, energy costs, or labor costs. So what are businesses doing? To the extent that they invest, much of the investment goes into labor-saving capital—computers and robotics. Productivity is good, and that’s what keeps the inflation rate down. It isn’t the unemployment rate—inflation rate tradeoff. It’s the fact that we have high—or at least reasonable—productivity growth in most of the quarters. As long as that’s true, inflation will be kept under control. When it isn’t true, inflation will break out.
In 100 years of Federal Reserve policy, there are two long runs of success. One is 1923 to 1928. Another is 1985 to 2003. There are no comparable periods under discretionary policy.
Now, excess reserves are $2.9 trillion. They sit idle on banks’ balance sheets. Along with that, corporate balance sheets have hundreds of billions of dollars of idle money. What can anyone believe is the value of adding more excess reserves? It’s going to create problems for the future; those problems are real, not monetary.
In 100 years of Federal Reserve policy, there are two long runs of success. One is 1923 to 1928. Another is 1985 to 2003. There are no comparable periods under discretionary policy. Following rules, or quasi-rules, the Fed managed to achieve what it never achieved by discretion for any long period of time. That is, low inflation and relatively stable growth, punctuated by mild recessions.
Why is that so? One of the main Fed mistakes that we see very visibly now is excessive concentration on near-term events and on unreliable forecasts. There’s not much agreement among the members about the forecasts. But more important than lack of agreement is a neglect of longer-term consequences. From 1926 until 1986, you do not find in the minutes, except for the Volcker period, a statement that says, "if we take this action today, where will be a year from now?" It’s all about where we will be next quarter. Should we raise the funds rate an eighth or a quarter? Should we lower it? About such things, there can be lots of discussion. But that discussion almost never says, where will we be a year from now?
One thing that is certainly true of US and probably other countries’ data is that it’s full of temporary, random changes that are very hard to distinguish from the underlying trends at the time they occur. You have to wait to see what the permanent effects are going to be. Operating on these short-term measures is usually not a good idea, yet the Fed does that all the time, including the present time.
To escape the consequences of its errors, the Fed needs to adopt a rule or a quasi-rule. There’s not going to be a rule that is going to work under all circumstances. Deviations from the rule have to be permitted—under restrained circumstances. You don’t give the Fed the unlimited, unrestrained authority that it has now; you make it subject to a rule. There are some who would prefer a rule for price stability. But a rule that had a dual mandate might be operational, provided they would follow it, and it would be more readily accepted in the Congress.
There would be two reports if there were deviations. One would explain why they deviated. The other would offer their resignations.
There would be two reports if there were deviations. One would explain why they deviated. The other would offer their resignations. The administration could accept the resignations or it could accept the explanation. This would close the gap between accountability and responsibility.
How do we get back to stability? People at the Fed say not to worry about inflation because all they have to do is raise the interest rate. That’s correct but too simple. First, there’s always a political reaction to an increase in interest rates. But more importantly, at the current moment we have a huge debt with an average maturity which is under five years. Even worse, 28 percent of the debt is under one year; 40 percent is less than two years. And that’s just the Treasury debt. It doesn’t include Fannie Mae, Freddie Mac, and a whole variety of other institutions. Within two years, the budget deficit—just as a result of raising interest rates—by 3 percentage points, that is, back to some normal rate—increases by at least $36 billion for each 1 percentage point increase in the interest rate. Three points—more than $100 billion. Think about the political repercussions of that in a Congress that is having difficulty agreeing on any cuts in the budget.
Finally, there is international coordination. An international monetary stability pact: major currencies—the ECB, the dollar, the yen, and perhaps the renminbi—should accept a common inflation rate of zero to 2 percent. There would be no enforcement mechanism; there would just be commitment. Anyone who wished to have what no one could have at the moment—that is, stable exchange rates and low inflation—could voluntarily peg to one or more of those currencies. They get the public good that has been missing since the breakdown of the Bretton Woods system. There would be no meetings, no coordination actions. Like the gold standard, there would be market enforcement.
Would that be perfect? Hardly. Would it be better? Probably so. It permits changes in relative productivity to alter real exchange rates. The dollar, the ECB, the yen—they would not have a fixed exchange rate. They would have a fixed inflation rate so that real exchange rates would be allowed to change in the face of productivity and other changes. It gives countries the opportunity to choose both low inflation and stable exchange rates. What about us? What do we get from it? We get the benefit of low inflation and stable exchange rates with all those countries that choose to peg to the 0 to 2 percent inflation rate countries. That proposal restores badly needed discipline to the monetary system.
The case for a rule to bind the Fed
A monetary rule is important for many reasons. But one of the most important is this: From 1789, the beginning of the federal system, to 1930, the United States ran budget surpluses in two-thirds of all non-war years. Since 1930, only two presidents—Eisenhower and Clinton—achieved back-to-back surpluses. We won’t go back to small government, but we can go back to more stability. Our future growth depends on it. The presence of a monetary rule enhances the effectiveness of a fiscal rule that balances the budget.
We need a rule that gives people in a democratic country what they most want: low inflation and greater economic stability. That’s where we must go in the second century. That’s where the Fed has failed itself, failed us, and failed the world.