Main Content

Stephen Cecchetti on monetary policy

Monetary Policy is Not the Force that it Used to be

August 6, 2003

Has Alan Greenspan lost his touch? The US economy seems to be floundering, unemployment is higher than it has been in nearly a decade, the dollar has tanked and inflation seems perilously low. Add to this the abrupt upward move in long-term interest rates and even a casual observer will wonder whether the Federal Reserve chairman has placed too much trust in monetary policy.

Monetary policy, conventional or otherwise, cannot solve all our problems. Textbooks make it sound simple. When central banks change interest rates, it affects real capital investment. Lower interest rates make it profitable for companies to invest in the more numerous projects that have lower rates of return. It is a nice story, but unfortunately it does not work. No matter how you look at it, capital investment is not very interest rate sensitive. That is a fact, not a theory.

However, lowering interest rates does have a demonstrable impact on growth and inflation. It raises both. That is another fact. So how are we to understand it? The answer is that there are at least two other channels through which monetary policy actions are transmitted to the economy.

First, there is the exchange rate; and second, there is bank lending. Exchange rates do what you would expect. A depreciation caused by a monetary policy easing makes exports cheaper and imports more expensive. The result is both higher inflation and higher output. But depending on the economic environment, any particular policy change may not move exchange rates in the intended manner. We saw this in recent years when European Central Bank tightening was often accompanied by euro depreciation.

The solution to the mystery of why monetary policy has any impact lies in its effect on bank lending. Even in the most advanced economies, large numbers of companies depend on banks for financing; and monetary policy affects bank loan supply. Easier policy allows banks to increase the size of their balance sheets, increasing the quantity of loans. This raises investment, and business activity goes up with it.

What happened to money in all of this? The answer is that monetary policy does not have much to do with money. The central bank controls the size of its balance sheet exactly, not the quantity of money. The banking system creates money using central bank money as a raw material input.

But private sector bankers decide on their own how much money to make. The link between what central bankers control and the amount of money in the economy is just too unpredictable to be useful. And, importantly, monetary policymakers do not care, because money growth and inflation are virtually unrelated when inflation is low.

Returning to what does work, there is hard evidence that bank lending is quantitatively the most important channel through which monetary policy affects the economy. If you find that somehow reassuring, you should not. It means that monetary policy affects the economy expressly because there is a large set of companies that depend on banks for financing, and that policy is influencing the supply of loans to these borrowers.

The trouble is that financial innovation has created a world in which most bank assets are marketable, essentially giving borrowers, who otherwise would not have it, direct access to financial markets. And individuals have access to financing that circumvents banks through securitised mortgages, car loans, credit card debt and so on.

Today's bankers are loan brokers. Virtually no borrower is tied to a particular source of funds.

What this means is that changes in the cost and availability of bank finance have much less impact on business investment decisions than they once did. It is expectations of revenue growth that determine investment in today's world. That is, a company's rate of return calculation is much more sensitive to changes in expectations about future business conditions than it is to changes in the cost of financing.

So Mr Greenspan and his colleagues have a problem. Their primary tool just does not look as powerful as it once was. And what is true for conventional interest rate policy is also true for unconventional policies.

Fed officials have been clear about the steps they plan to take should the economy fail to respond to current policy settings. First, they will commit themselves to keeping short-term interest rates low for an extended period to force long-term interest rates down. If that does not work, they will push short-term interest rates to zero. And last, they will initiate a programme of rapid balance sheet expansion.

But will these measures succeed? As noted earlier, interest rate cuts do not encourage investment the way textbooks say they should, so driving long-term interest rates down may not be all that effective. And just because the central bank expands its balance sheet, there is no guarantee that commercial banks will expand theirs. Commercial bank lending and business borrowing require optimism about economic prospects.

To be clear, I still believe that policy can be effective. I have no doubt that the Bank of Japan could solve Japan's low growth and deflation problem by committing itself to increasing the size of the monetary base at a significant rate until Japanese nominal gross domestic product started increasing at 4 per cent a year. But the crucial aspect of such a policy would be coupling the public commitment to a specific operational policy with an observable goal for performance. The Federal Open Markets Committee could do the same thing by simply announcing a numerical inflation objective and saying it would keep interest rates low until that objective was reached. This improved communication would anchor long-term inflation objectives and stabilise long-term interest rates.

It looks to me as if the fear of deflation has led the Fed to raise its tolerance for inflation and that the implicit target is now between 2 and 3 per cent on the consumer price index. If this is true, Mr Greenspan and his FOMC colleagues should tell us. It is in their interest to do it. Confusion about this is what has caused the fluctuation in bond prices and the dismay over policy effectiveness.

With expectations of inflation at 2 per cent, for example, wages and prices would begin rising accordingly. And, most importantly, investments in new factories and equipment could be made with the knowledge that revenues were likely to grow at those rates. Believing something will happen is at least halfway to making it happen.

The writer is Professor of Economics at Brandeis University, Massachusetts, and a research associate at the National Bureau of Economic Research.