Nobody Knows the “Correct” Fed Policy

Jerome Powell has only been Fed chairman since last February, but he has already been initiated into the worst part of the job. Whatever the Fed does, some complain that it didn’t act sooner and/or do enough, while others complain that it acted too soon and/or did too much. The sniping also extends to every word he did or didn’t say and when he did or didn’t say it. And even the president who appointed him frequently treats him like his 2020 scapegoat-in-waiting.

When they focus on particular failings by the Fed, critics may well be correct in their objections. But they typically go farther and propose “solutions” which reflect a presumption that they know what the proper Fed policy is for whatever economic problems they see. Unfortunately, not only might their vision of the problems be faulty or incomplete, or both, but there is a huge gap between identifying monetary policy issues and knowing the right answers to them — between Fed snipers and Fed saviors. The fact is that if we rely on discretionary monetary policy, nobody knows precisely what monetary course the Fed should follow.

­The first problem facing successful discretionary stabilization or manipulation of the economy is ­the long and variable time lags between monetary policy changes and their effects on the economy (most often, but not always, 6 to 18 months). In addition, the lag before output and employment response tends to be shorter than the lag before prices respond, so that solving an employment problem can create a lagged inflation problem or solving an inflation problem could create a lagged employment problem. But even apart from that “solutions causing new problems” issue (or how much of monetary-policy-induced investment is really malinvestment, and the future problems that will impose), the different and varying lags mean that for Fed policy to effectively stabilize or stimulate the economy, it must focus on uncertain fu­ture output and inflation patterns, not present circumstances.

That problem is compounded by the notorious difficulty of accurately forecasting turning points in the economy, when it is most essential to read the tea leaves correctly. When things are stable, there is little which could improve that stability, so there is little need to accurately forecast when little or nothing should be done. But responding appropriately to sharp turning points in the economy requires that the right thing be done, to the right extent, at the right time. Yet earlier this year, Simon Kennedy and Peter Coy cited not just the “widespread failure to forecast America’s Great Recession,” but a marked pattern of unreliability over time and across countries. Other indicators have similar problems. For example, downturns in the most-watched forecasting tool, the index of leading indicators, have preceded recessions by half a year to two-and-a-half years since the 1950s, but did a poor job of predicting their magnitudes, and have even predicted recessions that did not happen on multiple occasions. Absent highly reliable, commonly accepted forecasts, deciding what to do and when to do it is a far cry from a reliable science. This is even more so when every other country’s future developments and policy changes (e.g., Brexit, slowing international growth and negative interest rates in some countries) must also be factored into decisions.

Active monetary policy is made even more difficult by the continually changing torrent of economic data the Fed must rely on to judge the state of the economy. That torrent includes everything from new claims for unemployment and purchasing managers’ surveys to raw materials prices and new home sales, each released at a different time, and reams of international data, as well. But every piece of data has limitations in both timeliness and accuracy (limitations that are virtually unknown to the public). Further, the data typically sends conflicting signals about the state of the economy (e.g., overall inflation and “core” inflation may be acting differently, or employment growth can be slowing while unemployment rates are falling). When no statistic is precisely right, deciding which combination of conflicting pieces of evidence provides the most accurate read this time is a task beyond our knowledge, made even harder by multiple potential interpretations of “what it all means.”

The difficulty in conducting active monetary policy “correctly” is compounded still further by the need to accurately predict how, and how quickly, the public’s expectations will respond to any policy change. The Fed needs to anticipate how its actions and its members’ statements will alter expectations about both the present and future state of the economy and the path of future policy. It also needs to take account of what will then happen to expectations as each subsequent piece of economic news, good or bad, emerges afterward. In doing so, the Fed is hampered by the fact that today’s expectations will not change exactly as in the past, because the situation is never exactly the same and because people have adapted to some extent from what has happened in the past. But no one knows how much current reactions will differ from earlier episodes.

There is always the potential for valid criticisms of active monetary policy, particularly about the necessity, timing, magnitude, and mechanisms of policy changes (including innovations, such as the introduction on interest payments on bank reserves and their effects), and how clearly the Fed’s intentions are communicated. However, in the blurry and con­stantly changing real world of conflicting economic indicators and forecasts, competing interests and policy surprises, activist policymakers do not know exactly what the “right” policy is today. Neither do their activist critics, however. Yet this should not come as a surprise. As Austrian economists have recognized for decades, central planners cannot know enough to reliably achieve efficiency or whatever else they intend, and monetary policy is a form of central planning. In fact, it is an unusually pernicious form, whose interventions undermine the accuracy of particularly important prices  — interest rates — and the ability of savers and investors to productively coordinate their behavior, which teaches us not to trust our current monetary policy saviors to live up to that billing, or that every monetary policy sniper qualifies as a monetary policy savior.

Gary M. Galles is a professor of economics at Pepperdine University. He is the author of The Apostle of Peace: The Radical Mind of Leonard Read.