In December 1967, Milton Friedman delivered the presidential address at the annual meeting of the American Economic Association. In it, he outlined the things monetary policy can and can’t do.

Friedman, along with co-author Anna Schwartz, had already rewritten history. Their 1963 book, “A Monetary History of the United States,” placed the blame for the Great Depression squarely at the feet of the Federal Reserve for allowing the monetary base to contract by one-third between 1929 and 1933. At the AEA conference, he talked about how their work had influenced the thinking on the effectiveness of monetary policy: from obsolete -- “pushing on a string” was the favored Keynesian diagnosis of the 1930s -- to potent once again.

Then he got to the gist of his argument. The two things monetary policy can’t do are peg a real interest rate or the unemployment rate for more than a very limited period. Coincidentally, these happen to be “the two main unattainable tasks” most likely assigned to monetary policy, Friedman said.

Fast forward 46 years. The Fed, confronted with stubbornly high unemployment, is moving in the direction of doing just that. In December, policy makers said the benchmark rate would remain at 0 to 0.25 percent at least until the unemployment rate dips below 6.5 percent.

Full employment is one of the Fed’s two mandates (the other is stable prices). It’s also a qualitative term: What constitutes “full” in today’s environment may differ from “full” in the past.

Economists agree that the U.S. economy is operating well below its potential: Many people who would like a job can’t find one. What they don’t know is how many of the long-term jobless are unemployed because they lack the skills employers need. Such structural employment may be an impediment to achieving full employment, which the Fed thinks is 5 percent to 6 percent in the long run.

Just as there is some natural, unobservable rate of interest that keeps the economy growing at its potential in perpetuity, there is also a natural rate of unemployment, Friedman explained. Monetary policy can fool Mother Nature for a while, increasing employment at the expense of higher inflation. But ultimately all it has to show for its effort is higher inflation.

Back then, Friedman recommended steady money-supply growth as to promote economic stability. With velocity, or the rate at which money turns over, no longer stable, a strict money target isn’t viable. That’s why many economists have turned to nominal income as a substitute.

Using an interest rate or the unemployment rate to determine the stance of monetary policy is akin to a space vehicle fixated on the wrong star, Friedman said. Even with a sophisticated guidance system, “the space vehicle will go astray.”

(Caroline Baum is a Bloomberg View columnist. Follow her on Twitter.)