How the US imbalances can be corrected

Gilles Saint-Paul

06 December 2007



There is agreement among many analysts that the Fed should pursue a low interest rates policy in order to prevent the US credit crisis from degenerating into a recession. On what grounds are we told that? The bottom line is that monetary policy is supposed to fine-tune the economy by targeting inflation and the output gap. Thus, monetary policy is supposed to become tighter when there are fears of inflation, and looser when there are fears of a recession and no sign of inflation. Consequently, the fed’s recent moves to lower interest rates seem perfectly orthodox.

This focus on macroeconomic aggregates ignores any other effect that interest rates can have on the economy. It totally ignores that interest rates are a price which affects many allocative decisions and has important distributive consequences. In 2001, the Fed engaged in a policy of drastic reduction of interest rates, for fear that the conjunction between the end of the so-called “Internet bubble” and the attacks of September 11 would drive the US economy into a recession. These considerations were compounded by the increasingly popular view that inflation was no longer a problem. The strong expansion of the late 1990s had been accompanied with little inflationary pressures and there were fears that the deflationary experience of Japan might hit the United States.

The result of these policies is that the US was in a regime of very low real interest rates. From 2002 to 2004, the federal funds rate did not exceed some 1.5 %, while inflation moved from 1.6 % to 2.7 % during that period. Thus short-term real interest rates were clearly negative. As for longer maturities, some real rates fell to 1.5 %. Many would argue that this was the right thing to do; GDP stayed at its potential level, or below it, and the incipient increase in unemployment was reversed.

The problem is that low interest rates not only stimulate the economy, they do plenty of other things. In other words, focusing only on GDP has costs and may generate mounting problems—the low rates policy makes a current recession better, but the next one may be worse.

One reason why the US economy is less inflation-prone than in the past is that a bigger share of any increase in domestic demand is absorbed by imports: the economy is more open than it used to be. Thus, instead of having “overheating” because demand is greater than supply, the gap between the two is filled by trade deficits. Hence, low rates stimulated consumer spending and the trade balance deteriorated by two percentage points of GDP. The US is rapidly accumulating foreign debt and that may lead to a brutal correction with a sharp drop in consumer spending and a large depreciation of the real exchange rate. In fact, that correction may have already begun. Yet the Fed is not supposed to look at the net foreign asset position of the US economy, even though both its deterioration and rising inflation are the symptom of the same problem – excess domestic demand.

The other issue is asset prices. When interest rates are very low, and expected to remain so, asset prices can be very high. In fact, when interest rates fall below the growth rate, assets become impossible to price. Consider, for example, a share that pays a dividend which grows at 5 % a year. With a 2% interest rate, it is profitable to buy that asset regardless of its price, because I only need to hold it for a sufficiently long time for the dividends to eventually exceed the interest payments. So the price of the asset is in principle infinite. In fact, people do not live forever, so they will have to sell the asset back at some point; but one can show that any change in markets' expectations about that future price can be validated by a corresponding change in the current price—so, the current price can be anything.

In particular, low interest rates may start asset bubbles. One mechanism is as follows. As the price starts rising due to lower interest rates, irrational speculators start buying the asset on the grounds that the price increases are going to continue. That fuels the price increase which may eventually develop into a bubble where all speculators, including the rational ones, pay a high price for the asset because they expect the price to be even higher in the future. So one by-product of the fall in interest rates is that real house prices started to go up very quickly.

To summarise, the low interest rate policy led to a wrong intertemporal price of consumption – consumption was too cheap today relative to the future – which led to excess spending and trade deficits. It also led to a mis-pricing of housing, which led to excess residential investment and excess borrowing by households. That is the price that was paid to make the 2001-2002 slowdown milder.

These imbalances have to be corrected. In principle, consumer spending can be brought down without the economy having to go through a recession, provided there is a sharp real depreciation of the US dollar, which would shift the structure of demand away from domestic spending and in favour of exports. On the other hand, the correction in house prices is likely to be contractionary. Some consumers have borrowed against the capital gains they made on their house, to purchase, for example,  a second house or consumer durables. They are going to cut their consumption since they are more likely to become insolvent. As the collateral value of their houses falls, consumers will get less credit; hence a further drop in consumption. Furthermore, the securities backed by mortgages, subprime or otherwise, have been used as collateral by financial institutions; that collateral is worth less, thus reducing credit between those institutions. As a consequence, they will have more trouble lending to firms, so that investment will also be hit. The housing bubble has jeopardised the financial sector both because people have borrowed to hold it and because institutions have used the corresponding securities as collateral.

Because of this gloomy scenario, the Fed has been under pressure to cut rates. The problem is that such a policy is likely to perpetuate the current imbalances. Indirectly, it amounts to bailing out the poor loans and poor investment decisions made by many banks and households in the last five years. The bail-out comes at the expense of savers and new entrants in the housing market. The signal sent by the Fed is that it is sound to join any market fad or bubble provided enough people do so, because one will be rescued by low interest rates once things turn sour. Worse, the more people join, the greater the lobby in favour of an eventual bail-out.

All this suggests that the US has to go through a recession in order to get the required correction in house prices and consumer spending. Instead of pre-emptively cutting rates, the Fed should signal that it will not do so unless there are signs of severe trouble (and there are no such signs yet since the latest news on the unemployment front are good) and decide how much of a fall in GDP growth it is willing to go through before intervening. As an analogy, one may remember the Volcker deflation. It triggered a sharp recession which was after all short-lived and bought the US the end of high inflation.



Topics:  Monetary policy

Tags:  interest rates, Federal Reserve, imbalances

Toulouse School of Economics and Scientific Adviser to the Economic Studies Directorate at the French Ministry of the Environment. CEPR Research Fellow