You are here

Federal Reserve policy is becoming less predictable

The Federal Reserve this month is widely expected to raise interest rates for the eighth time in the current hiking cycle that began in December 2015.

[WASHINGTON] The Federal Reserve this month is widely expected to raise interest rates for the eighth time in the current hiking cycle that began in December 2015. Get ready for tensions within the Fed to spill over into the public as monetary policy moves closer to estimates of the neutral rate.

The Fed staff will likely push harder for policy makers to follow a model-based approach with fairly hawkish implications that would result in rates rising beyond what is considered a neutral level. Although Fed Chairman Jerome Powell doesn't look married to the Fed's models, he hasn't provided a great deal of alternative guidance. That means policy will become less predictable in 2019.

Consider the Fed's basic theoretical framework as exemplified in the "dot plot." In that chart of the likely path of rates, central bankers estimate values of key exogenous policy variables, such as the longer-run values for unemployment, that provide a gravitational pull for the economy. Over time, the economy should settle into equilibrium, with actual values the same as the estimated natural values. For example, the Fed estimates - not picks - the natural rate of unemployment. As such, altering policy rates will impact the time it takes to reach equilibrium, but not the level of equilibrium.

The Fed, however, can choose the rate of inflation. In these models, the success or failure of policy ultimately falls on the deviation of the inflation rate from target because that is the only thing under control. The impact of policy on unemployment two years out during a "normal" economic period like now is small by comparison. Consequently, central bankers should not become distracted by things such as uncertainty over inflation persistence or the natural rate of unemployment. See recent Fed staff research here and here as well as related news reporting here and here.

Under the current projections as embodied by the median of the dot plot estimates, policy rates need to rise above equilibrium rates to guide the economy into equilibrium. That's what the model says. But this underweights the risk of recession from raising rates above neutral. Recessions are inherently difficult to predict in the Feds' models. Recessions are just simply very rare events. Econometric models aren't good at predicting rare events.

Your feedback is important to us

Tell us what you think. Email us at

The upshot is that being married to the Fed's models will generate a biased view of the appropriate policy path. During normal economic periods, market participants should consider that view as hawkish in the sense that they imply a heightened concern about inflation and are excessively complacent about the risk that Fed action (or inaction) triggers a recession.

But does Powell share a commitment to the Fed staffs' models? That's where the internal tension lies. One take is that the Fed staff research noted above influenced Mr Powell's recent Jackson Hole speech. In that speech, he discussed the challenges of making policy when key variables such as the natural rate of unemployment are uncertain. Combined with this research, the implication is that the Fed is about to take a hawkish turn.

Perhaps Mr Powell, who is not a trained macroeconomist, is not married to the models and sees their biased outcomes. Such a person would see that the interesting things that make or break policy - such as a financial crisis - happen outside the context of the models.

Indeed, Mr Powell's Jackson Hole recent speech was about the importance of judgement over models. That is where policy succeeds or fails. The models are something of a sideshow. As I wrote recently, this means that the Mr Powell does not appear committed to any firm policy guidelines, leaving me with an unsettling feeling that he plans to continue hiking until something breaks and a recession occurs.

Alternatively, Mr Powell may be less complacent about the risk of recession than his staff, and hence more dovish, leading him to push for an extended pause when policy hits neutral. It would be helpful for Mr Powell to lay down some more markers for policy guidance, but it doesn't appear to be his style.

Watch for tension between Mr Powell, the Fed staff and his more model-focused colleagues. It would yield an increasingly chaotic environment for market participants as Fed-speak becomes more discordant. Mr Powell's favorable view of relying on judgement over models may be the right choice - assuming of course that his judgement turns out to be as good as his view of former Chair Alan Greenspan's judgement. But it leaves us guessing as to what his judgement is telling him to do from one day to the next.


BT is now on Telegram!

For daily updates on weekdays and specially selected content for the weekend. Subscribe to