Originally published in THE WALL STREET JOURNAL

Wednesday, May 13, 1998

Don't Raise Rates Now

By MARTIN FELDSTEIN
Board of Contributors

"The Fed cannot base an increase in interest rates on a stream of good news: rising real wages, higher share prices, lower unemployment and favorable consumer sentiment."

The recent drop in the unemployment rate to a breathtakingly low 4.3% strengthens the hand of those who are calling for the Federal Reserve to raise interest rates at the May 19 meeting of its Open Market Committee. But with prices falling, the time has not come for a rate increase.

The case for raising rates now is clear enough. The tight labor market, reflected in the low

unemployment rate, has been pushing up wage rates at an increasingly rapid pace since the unemployment rate dropped below 1994s average of 6%. Wages rose by 2.8% in 1995, 3.4% in 1996 and 3.9% in 1997. These statistics argue strongly against the view that we are in a "new economy" in which the old relation between low unemployment and accelerating wage increases no longer exists. And its not just wages: Even when fringe benefits and payroll taxes are taken into account, total compensation rose at an increasing pace: by 2.6% in 1995, 3.1% in 1996 and 3.4% in 1997.

Virtuous Circle

But the case for increasing interest rates cannot be based on rising wages when price inflation has actually been declining. The consumer price index rose 2.5% in 1995, 2.3% in 1996 and 1.7% in 1997. The broader gross domestic product price deflator rose at an annual rate of less than 1 % in the first quarter of this year. The price index for domestic purchases showed no increase at all, for the first time in nearly half a century. And producer prices actually declined during the past year.

The seeming paradox of accelerating wage increases and declining price inflation is explained by two factors: lower import costs and rising productivity. Import prices were 6% lower at the end of 1997 than two years earlier, in contrast to the 3% rise in import prices over 1994-95. The lower import prices reflected the decline in the world oil price and the higher value of the dollar relative to foreign currencies. The fall in import prices fed directly into lower prices to American consumers and put pressure on American firms to keep prices down in order to compete with cheaper imports

The impact of the rise in productivity was even greater. Productivity has increased at an annual rate of 1.8% in the past two years, up sharply from the rate of less than 0.3% a year between 1992 and 1995. Since higher productivity means more real output per employee hour, the greater productivity rise translates directly into lower increases in the unit labor costs that are the key cost determinant of product prices. Thanks to the faster productivity gains, unit labor costs rose more slowly in 1996 and 1997 than they did in 1995, despite the faster rise in hourly compensation rates in the more recent years.

The decreases in consumer price inflation in the past few years may be the start of a virtuous circle in which lower price inflation leads to lower money wage increases which then bring price inflation down further. In the first quarter of 1998, compensation rose at an annual rate of 3.3%, down from the 3.8% rate in 1997, despite the tighter labor market. The most likely explanation of this favorable shift in wages was the slower rise in consumer prices (an annual rate of less than 1% in the first quarter). With consumer prices rising more slowly than before, the real wage increase (the difference between the rise in nominal money wage rates and the rise in consumer prices) was actually greater despite the slower rise in money wage rates. Higher real wages are the natural response to tighter labor markets and, with falling inflation, are compatible with a lower rate of increase for money wages.

So while the very low unemployment rate might continue to push up wages at a rate that leads to higher price inflation, there is a real chance that this will not happen in the foreseeable future. High employment wont fuel inflation if the recent rate of productivity gains continues and if the virtuous price-wage circle keeps real wages rising faster through dropping inflation rather than through greater increases of money wages.

The outlook is uncertain. The first-quarter productivity figures reverted to earlier sluggishness. Future import prices may accelerate domestic inflation when oil prices rise and the dollar declines. But these are reasons for vigilance, not for pushing up interest rates now.

Despite its recent strength, economic growth could slow significantly because of weaker export demand and rapidly rising imports. Employment growth in 1998 is already slower than it was in the second half of 1997. The recent drop in unemployment reflected lower labor force participation rather than a surge in employment.

A further reason to leave the Fed funds rate unchanged now is that it is already at a relatively high real level. With a nominal funds rate of 5.5%, the real rate is at least 3.5%. When the Fed last began a significant increase in the funds rate back in 1994, the real rate was less than 1%. And while the nominal funds rate has been raised by only 25 basis points in the past 18 months, the real rate has increased by much more because of the decline in inflation.

Raising the Fed funds rate now would impose unnecessary risks on the economy. Even a small rate increase would be interpreted as the first of a series of hikes, causing the entire yield curve to rise. That could slow an economy that is already losing momentum, push down share prices in a nervous market, destabilize the delicate Latin American financial markets and cause the overvalued dollar to rise even further.

'Ahead of the Curve

The argument that the Federal Reserve needs to act now to "be ahead of the curve" is unconvincing after three years in which being ahead of the curve would have meant raising interest rates while inflation was actually falling. If inflation does turn around, it is likely to rise very slowly, giving the Fed ample time to respond. Confidence in the Feds long-term anti-inflationary stance, reflected in the remarkably low inflation expectations of less than 2% implicit in the Treasury Departments inflation-protected 10-year bonds, is further reason for the Fed to recognize that there is no urgency in raising interest rates.

The Federal Reserve should emphasize to the public its commitment to price stability. It should continue the course of monetary policy that has brought inflation down to 2% from 4% a decade ago. But the Fed cannot base an increase in interest rates on a stream of good news: rising real wages, higher share prices, lower unemployment and favorable consumer sentiment. It has to wait until there is clear evidence that consumer prices are rising. And then it will have to act decisively.

Mr. Feldstein is a professor of economics at Harvard University and the head of the National Bureau of Economic Research.