Here's a link to all the Capital Comments columns.
 

President Bush just nominated Alan Greenspan for a fifth term as head of the Federal Reserve.  That’s our central bank, the institution that moves interest rates up or down to try to balance inflation, unemployment and economic growth.  Greenspan has been the Fed head since 1987.  If he serves another four years, he’ll be the longest serving chairman in history.

That’s it, though.  Unless Congress changes the rules, Greenspan won’t be allowed another term.  And he’s 78 years old.  You can tell he loves his job, but even he might want to retire someday.  Within a few years we’ll have to get by without guidance from our long time monetary guru. 

The main interest rate that the Fed influences is called the federal funds rate.  It’s the interest rate banks charge each other for overnight loans.  When the Fed wants to slow down the economy to fight inflation, it raises the federal funds rate.  When the Fed wants to speed things up to ward off recession, it cuts the rate. 

There are formulas that try to predict what the Fed will do with the federal funds rate.  They’re called “Taylor rules”, after their inventor, an economist named John Taylor.  The simplest Taylor rule I’ve found was done by Gregory Mankiw, who is now President Bush’s chief economist.  It works like this.

Take the inflation rate for the last twelve months or so, and subtract the current unemployment rate.  That’s an index of the economy’s condition.  Get the calculator and multiply the index by 1.4, then add 8.5.  Why 1.4 and 8.5?  With those two numbers your result will match the interest rate choices the Fed has made since Greenspan arrived, within a couple of points in most years.   In 1996, to pick a year, the inflation rate was 3% and the unemployment rate was 5.4%.  The difference is -2.4%, times 1.4 is -3.4%, plus 8.5 is 5.1%.  The federal funds rate was actually 5.3% that year.   That’s pretty good, for economics.

The Taylor rule works because it reflects the Fed’s mission.  The index goes up when inflation increases or unemployment decreases.  That’s when the Fed should raise interest rates to slow the economy down.  The index goes down when inflation decreases or unemployment increases.  That’s when the Fed should cut interest rates to get the economy moving. 

If the Taylor rule works so well, why do we need Greenspan at all?  Why not pop the equation into a computer and let it make the Fed’s decisions?  Some economists support something like that.  I think, though, that we need some smart human beings in the decision mix.  The Taylor rule is accurate only to plus or minus two points or so, which means there’s room for the humans at the Fed to make choices.  And sometimes there are events like stock market crashes or international currency crises that require quick action before they affect unemployment or inflation. 

The inflation rate for the last twelve months has been 2.3%, and the unemployment rate was 5.6% in April.  Do the math and you get 3.9%—way higher than the actual 1% federal funds rate we’ve got now.  That’s farther from the Taylor equation than Greenspan has ever been.  One reason is that recent inflation partly reflects oil and milk price hikes.  Those prices have more to do with Middle East politics and mad cow disease than with demand and supply in the general economy.  And the Fed’s researchers suspect that the unemployment rate is measured too low, because the estimate of the labor force is too high.  Inflation may be lower, and unemployment higher, than the numbers say.  The actual federal funds rate reflects that.

Still, the Taylor equation tells us that Greenspan’s Fed probably will raise rates soon.  Many think that will happen at the Fed’s next policy meeting, at the end of June. 

Greenspan will leave the Fed by the end of 2008*, but the next chair can use the Taylor rule to get a pretty good idea of what he would recommend in most circumstances.  We can’t keep the man forever, but maybe we can bottle his monetary policy skills.

*Since this column was written, Greenspan has announced that he will leave in January 2006.