Setting Monetary Policy in the New Economy

Thank you for your invitation. And thank you for not withdrawing it after the slowdown caused our New Economy to lose some of its luster. It has caused some people to question whether a New Economy ever existed and if so, whether it will survive.

I think the New Economy was real and is real, although its benefits will be diminished during the slowdown. Most attention in the U.S. is now focused on how long the slowdown will last, whether it might yet tip over into recession, and how soon and how rapid the recovery is likely to be. I don't know the answers to those questions, and I can't speculate on them today because I'm still in the blackout period from Tuesday's FOMC meeting. The press release from that meeting is still current. You can find it on the Dallas Fed's web site at www.dallasfed.org.

I understand that my invitation was prompted by a talk I gave last September to the National Association of Business Economics. Among other things, I chided the audience of economists for their overly cautious forecasts. At the end of 1999, Business Week listed the forecasts of 50 prominent economists and ranked their GDP forecasts from the highest to the lowest. The most optimistic forecast was 4 percent, which amazed me since growth had averaged more than 4 percent for four years and more than 7 percent in the previous two quarters. Economists had consistently underestimated growth, and I thought this would be the biggest miss of all.

But I was wrong. After three quarters of 8.3, 4.8 and 5.6 percent, we got 2.2, 1 and a preliminary estimate of 2 percent for first-quarter 2001. The economy hit an air pocket on its glide path to a soft landing, and I owe those "overly cautious" economists an apology.

What happened? Had the Fed been too heavy-handed? We had begun tapping lightly on the brakes on June 30, 1999, and continued through May 2000, for a cumulative increase of 1.75 percentage points in the target federal funds rate, our interbank rate. That seems modest compared with the 3-percentage-point increase in 1994 that successfully achieved a soft landing.

The impact of monetary policy was heightened by high oil and natural gas prices and the deflation of the tech stock bubble that began in the spring of 2000. A growing fiscal drag may have contributed as well. Even so, more than a year after the tightening began, there was little discernable impact on the economy. I did warn in a couple of speeches that monetary tightening is a bit like drinking vodka: It sneaks up on you. (I neglected to mention that my brand is Absolut.) Since the abrupt slowdown came at year-end, I believe the uncertainty of the presidential election outcome in November and December probably also played a role.

The Fed responded to the year-end weakness with a 50-basis-point rate cut on January 3, almost a month before our next regularly scheduled meeting. We did 50 basis points again on January 31, March 20 and April 18, the latter also before the next scheduled meeting. At Tuesday's meeting, we did an additional half percent.

My assigned topic today is "Setting Monetary Policy in the New Economy." I'll use our experience as the context and leave it to you to identify any parallels for Sweden, if, indeed, there are any. I wouldn't presume to give you advice.

I suppose that appropriate monetary policy in the New Economy is not different in most respects from appropriate monetary policy in the Old Economy. The main problem in our case, however, was not knowing for sure that the paradigm was shifting. For a long time the doubters still argued that nothing fundamental had changed. They related our improved performance to good luck on a number of fronts, what economists call positive supply shocks. Believing that the changes were not structural or likely to be permanent, there was a lot of pressure to rein in what was regarded by many as unsustainably rapid growth and unsustainably low unemployment rates.

The argument was that since monetary policy works with a lag, you have to be preemptive and fight inflation before it becomes visible. After all, you know it's coming because growth this high and unemployment this low has been inflationary in the past.

The Phillips curve, which purports to show an inverse relationship between inflation and unemployment, was invoked as a reason to tighten. The low unemployment would surely lead to higher inflation down the road. A similar concept was the NAIRU—the non-accelerating inflation rate of unemployment—which held that below a certain level of unemployment, inflation will accelerate. We entered the 1990s with a consensus among economists that the NAIRU was around 6 percent. Any unemployment below 6 percent would cause inflation to accelerate.

A related concept is the potential growth rate, or the so-called speed limit that critics thought policymakers were enforcing. If economic growth exceeds the potential growth rate—or the speed limit—inflation will accelerate.

You get the potential growth rate by adding the rate of productivity growth—the growth in output per hour worked—to the rate of employment growth—the growth in the number of hours worked. From the early 1970s to the early 1990s, productivity growth was stagnant, rising just over 1 percent per year for two decades. (Those numbers have since been revised upward to around 1.4 percent per year, rather than the originally estimated 1.1 percent per year.) Add to that figure a similar growth rate in the number of hours worked by an expanding employment base, and you get a "potential" growth rate of 2 to 2.5 percent. Growth faster than that, assuming unemployment no higher than around 6 percent, would cause inflation to accelerate.

Toward the end of 1995, the U.S. economy began to get a second wind, more energy. In the second half of the decade, growth regularly exceeded the presumed speed limit by substantial margins. Unemployment fell below 6 percent and then below 5 percent, before settling down around 4 percent, where it's been for a couple of years. Yet for most of this period, inflation declined rather than increased. Many thought this was too good to be true and urged tightening of policy to head off the inflation always lurking just around the corner. The problem was that inflation had not made its appearance yet.

The more sophisticated economists, especially those from prestigious universities that don't have good football teams—American-style football, that is—urged tightening based on past economic relationships that were built into current econometric models. Those models—including the Fed's—were consistently predicting lower growth and higher inflation than we were getting. I joked that the way I forecast is to take the models' results and subtract a percentage point from expected inflation and add it to expected real growth. It worked pretty well.

Not being a very sophisticated economist, and especially not being an econometrician, I followed the advice of Yogi Berra, an old baseball player and manager of unusual wisdom. Yogi pointed out that you can observe a lot just by watching. I also followed the lead of comedian Richard Pryor, who once famously asked, "Who are you going to believe? Me or your own lying eyes?" The models were telling us one thing and our eyes were telling us something else. As for an economy too good to be true, I agree with Mae West, who said, "Too much of a good thing is just about right."

The circumstances posed a dilemma for monetary policy. The uncertainty argued for wait and see, but wait and see conflicts with the need to be preemptive. If you wait and see inflation actually building, it's too late to deal with it without risking a recession. To avoid that, you do need to have indicators of impending inflation as an early warning signal.

The pressure to tighten preemptively drew support from the successful soft landing engineered in 1994, which probably prolonged the expansion significantly. To me, however, there were two important differences between 1994 and the late 1990s. For one thing, we had held short-term interest rates to around 3 percent for almost a year and a half going into 1994. With inflation at about 3 percent as well, real short-term interest rates were effectively zero—clearly not appropriate after growth had become vigorous.

The other difference is that throughout 1994, sensitive commodity prices rose rapidly, clearly a threat to overall inflation. In 1999, real short-term interest rates were near their historical norms of around 3 percent and commodity prices were just coming out of a prolonged decline.

In the late 1990s, the leading indicators of inflation were real growth and employment, not financial indicators, which were rather benign. It didn't feel right to me to tighten policy because the real economy was strong. It seemed like a vote against prosperity, especially since leading financial indicators of inflation were not present.

As it turned out, monetary policy forbearance prior to mid-1999 worked out very well. By not trying to enforce a speed limit that was no longer relevant, we got good growth and much lower unemployment over a prolonged period. Collateral benefits included the swing in the budget from deficit to surplus, successful welfare reform, a decline in minority unemployment to the lowest levels on record and even a drop in crime. However, it was a risk, and we could have been wrong.

When the FOMC voted to tighten in June of 1999, as I said in my dissenting statement, I just wanted to test the growth limits of the New Economy further.

Inflation eventually did begin to creep up, but it was mainly because of higher oil and gas prices. Core inflation went up only marginally and only back to levels that existed prior to the depressing effect of the East Asian crisis.

So to summarize so far, the main problem is knowing whether you really do have a new economic paradigm or if you are just having good luck. Waiting to find out runs the risk of failing to be preemptive. Not waiting risks choking off growth unnecessarily.

The more than doubling of productivity growth in the latter part of the decade was driven by technology and led to the higher overall growth rates and lower unemployment. That this faster growth was less inflationary than expected resulted from a combination of the inherently deflationary attributes of technology combined with increasing global competition. Pricing power was hard to maintain locally when competitors could produce anywhere in the world and sell to anyone in the world. The new technology—especially the Internet—and the New Economy generally are kinder to consumers than to producers. Consumers get to participate. Producers have to participate or lose out. That's all right. Economies are for consumers.

Let me give you some unconventional food for thought. The conventional wisdom has it that inflation was depressed in the 1990s despite rapid growth. I believe that inflation was repressed largely because of rapid growth. We are so accustomed to thinking of inflation being caused by too much money chasing too few goods that we forget that more goods—that is, faster output—can be just as helpful in curbing inflation as lower money growth. Increasing supply is as helpful as containing demand.

Economists usually ignore that because they are used to thinking of the supply side as a given and concentrate on adjusting demand to the predetermined supply. But in a decade that saw the collapse of Soviet-style communism and hard-core socialism, new countries entering the world trading system for the first time, worldwide privatization and deregulation, and the lowering of trade and investment barriers, it's not reasonable to treat the supply side as a constant. Obviously, inflation is determined by the relative growth of supply and demand. Vigorous supply-side growth can be very useful and should not be ignored by policymakers.

Unfortunately, if I'm right about that, there may be a downside risk as well. If inflation is depressed in part by rapid supply-side growth, inflation might rise as that growth declines in a slowdown or recession. The conventional wisdom is that slowing economies bring inflation down, but not if supply slows faster than demand. It's something to think about. The key is to deregulate and invigorate the supply side of the economy even while holding demand in check with monetary policy.

Another issue in the new environment is how to gauge the stance of monetary policy. In other words, how easy or tight is it? Looking at money growth is the most intuitive way to me. But the monetary aggregates have become unreliable as their velocities have been made unstable and/or unpredictable by financial changes that impact money demand.

We use interest rates as the gauge these days, but only because it seems to be the least-worst way rather than the most-best way. Targeting interest rates can lead to procyclical policies or other policy mistakes if you misjudge whether a given rate change is due mainly to demand changes or to supply changes.

I've already explained why I don't like to use GDP and unemployment as indicators of pending inflationary pressures. So what do I use? I still keep an eye on the monetary aggregates for information, but which of the aggregates? Recent behavior of the monetary base suggests that policy is somewhat tight, while the broader measures of the money supply suggest otherwise.

The prices of gold and some other commodities suggest firmness, as does the foreign exchange value of the dollar. On the other hand, our term structure of interest rates (the yield curve) has become more normal recently, suggesting some easing. Recent changes in interest rates certainly suggest ease, compared with where they were, but short-term market rates are below the rates the Fed influences more directly. I guess you pay your money and take your chances.

The appropriate level of interest rate in the New Economy is not just a matter of stabilization policy. There is an argument that says faster productivity growth requires higher interest rates to help encourage saving and discourage overinvestment. That presumably would be part of the cure for our overdependence on foreign savings to finance domestic investment through the capital inflow counterpart to our current account deficit.

The sudden deceleration we experienced late last year seems to have resulted from old-fashioned overinvestment rather than any sudden decline in consumer demand or an overtightening of policy to curb inflation. When the Fed first started tightening in mid-1999 and well into 2000, many people told me the move would not touch the high-flying dot-coms since they don't get their financing from banks but from the venture capital and IPO markets. A policy designed to curb unsustainable growth driven by New Economy firms would only hurt Old Economy firms with traditional financing relationships. Instead, the opposite seemed to happen.

As it turned out, however, the dot-coms were hurt via the nontraditional route of the stock market and the drying up of venture capital, while many established firms are adopting new technology and thriving. The New Economy appears to be in transition from "The early bird gets the worm" phase to "The second mouse gets the cheese" phase.

In any case, the greater importance of the stock market, both in financing new businesses and in consumer portfolios, creates dilemmas for monetary policy not present in the Old Economy. We don't want to target stock prices or otherwise try to influence them. On the other hand, they do play an increasing role in determining both business and consumer behavior. There is a large moral hazard if investors get the idea that they will be bailed out—the nonexistent, but famous anyway, Greenspan put. There is no Greenspan put. There's not even a McTeer put.

It is interesting to think about the chairman's famous rhetorical question about the possibility of irrational exuberance and the reaction to it. That was way back in 1996. The implicit warning was ignored, but many of those who ignored it and created a tech stock bubble look elsewhere than themselves to assign blame.

I'll stop here and try to answer some questions. I hope our recent experiences conducting policy under rapidly changing circumstances offer some useful food for thought. Perhaps you can benefit from our success as well as our mistakes.