The Fed Needs Some Guidance

The Fed doesn't know when it will raise rates, and letting people think otherwise is confusing.

A Taylor rule could clarify Yellen's words.

Photographer: Andrew Harrer/Bloomberg

Every central bank tries to steer expectations about monetary policy. If businesses and consumers expect the Federal Reserve to let inflation surge out of control, the chances are fair that inflation will surge out of control. If they expect the Fed to keep inflation on track, inflation is more likely to stay on track. Shaping expectations is nine-tenths of the Fed's job.

That's why economists are so preoccupied with "forward guidance" -- the message Fed policy makers send about their intentions. The Fed's thinking on this just changed, and in a good way. It ought to change a bit more.

QuickTake Forward Guidance

Lately the Fed had promised to be "patient" about raising interest rates. Analysts invested the word with enormous significance. With unemployment at 5.5 percent and falling, a bump in interest rates was moving closer. The new statement was widely expected to discard "patient" -- and investors were ready to take this as a sign that rates would rise in June.

The new statement did drop "patient" -- but it also stressed that the Fed hadn't decided when to raise rates. In fact, the Fed noted, recent data points to a slowdown in growth: Investors pushed shares higher because they concluded rates would stay low for longer.

Why is this better forward guidance? Because the Fed is no longer suggesting it has a schedule. The simple fact is, it doesn't know when it will raise rates, and letting people think otherwise is confusing. Fed Chair Janet Yellen emphasized in her press conference yesterday that everything depends on the data.

That's good, but it wouldn't hurt to stress the point still further. The Fed could do this in two ways. First, stop publishing the interest-rate forecasts of its policy makers. The so-called dot plot suggests a schedule that doesn't exist, conveys almost no useful information and is apt to be misunderstood by analysts intent on finding deep significance (preferably where there isn't any). It's noise more than signal, and better dropped.

Second, analyze and present the relevant data more methodically. In this, explicit use of the "Taylor rule" would be helpful. This suggestion makes the Fed recoil, and with reason. Yet the idea is worth thinking about.

The Taylor rule says "set the short-term interest rate equal to 1.5 times the inflation rate plus 0.5 times the output gap [the difference between actual and potential output] plus 1." Stanford University's John Taylor first suggested it as a description of what the Fed was actually doing, even if it wasn't aware. To begin with, the rule wasn't intended as a prescription, much less a binding one. Over time, it fit the data pretty well, but far from perfectly. Taylor began arguing that monetary policy would have worked better if the rule had been followed more faithfully. 

Republicans in Congress want the Fed to design and adopt a Taylor-type rule and then explain subsequent deviations. Yellen says chaining the Fed to any such rule would be a mistake. Eric Beinhocker and David Hendry argued on Bloomberg View that it would be crazy to "replace [the Fed's policy makers] with a fixed mechanical rule."

Beinhocker and Hendry appear to be attacking a proposal that nobody has advanced. And the idea isn't, as Yellen says, to "chain" the Fed to anything. The suggestion is that the Fed could rewrite or depart from its rule as it saw fit. Its only obligation would be to say why.  

Actually I do think it would be a mistake -- and a huge one -- to put the Fed under any obligation that gave Congress more voice in monetary policy than it already has. The risk of infecting the Fed with Congress's flailing incapacity just doesn't bear thinking about. But this needn't stop the Fed from using the Taylor rule on its own initiative as a way to organize its thinking and present it to the public.

At the moment Taylor's own version of the rule calls for a short-term interest rate of around 1.5 percent (this assumes an output gap of roughly 3 percent). If the Fed were paying attention to the rule, it would have to explain why it was holding rates at zero. That wouldn't be hard under present circumstances: The Fed would explain that the economy has more spare capacity than the standard measures suggest.

Doesn't this make the rule uselessly vague? No, because precision isn't the point. Explaining policy in terms of the rule would put the focus more squarely on data about inflation and unemployment -- added protection from commitment to real or imaginary schedules. It would underline the importance of special circumstances, as and when they apply, and show their weight in the Fed's calculation. It would make plain that policy remains loose, even as rates start to rise, until they reach the "normal" level recommended by the rule. And by imposing more structure on the discussion, it would lean against shape-shifting discretion and ad hoc explanations of policy.

In short, it would be an aid to thought and exposition. Implicitly, Taylor-rule considerations are bound to influence monetary policy, as Yellen has acknowledged. Making its role a bit more constraining and a lot more explicit would be a good thing.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Clive Crook at

    To contact the editor on this story:
    James Gibney at

    Before it's here, it's on the Bloomberg Terminal.