Englander: Is The Fed (Or Any Central Bank) Still Credible?

The Chicago Fed’s recent "Conference on Monetary Policy Strategy,Tools,and Communications Practices" was designed to address various questions (and doubts) that academics and economists have raised, and also to listen to those supposed practitioners' perspectives.

However, as Standard Chartered's Steve Englander details in the note below, the conference will have very little impact on market expectations and do little to allay investor concerns about the ineffectiveness of monetary policy near the zero bound.

Critically, Englander says, the problems that the papers at this conference address are not the problems that worry market participants.

Instead, the questions they should have asked are more straightforward to put forth - and more difficult to answer honestly for The Fed.

Are the Fed’s models broad enough?

The basic model includes a Phillip’s curve, a zero lower bound and a policy rule. It is meant to optimise policy outcomes over a long period of time using alternative rules and choose the one with the best set of outcomes based on some objective function. This is a simplified version of something that requires considerable technical expertise.

In this modelling framework, nothing can go wrong. Some policy rules work better than others, but all eventually meet the targets. There is no banking sector that stops lending when rates are super-low, no investors who go too far out the risk-return curve, and no financial markets that stop trading because the safest assets are concentrated in the Fed’s balance sheet.

Such features are very difficult to model analytically. They require parametrisation that reflects empirical regularities, which are very hard to come by. However, investor concerns are not whether price-level targeting beats inflation averaging, but whether either will solve the problem of the zero lower bound.

Is the Fed or any central bank credible enough?

To his credit, Fed Chair Powell has admitted: “In models, great confidence in central bankers is achieved by assumption.”

The credibility assumption requires the model to be correct and for investors to have confidence that the central bank can stick to it.

If the model is not correct, investors must choose between believing that the central bank will follow a doomed policy or that it will change its policy.

As far as we are aware, there is no model that examines how credibility is achieved when the underlying behavioural model goes wrong, but models cannot capture everything.

Recent surprises on the Phillips curve slope and monetary policy effectiveness could easily lead households and investors to question the practicality of the central bank’s objectives.

What are the risks of locking in policy over the long run?

Do we really want one cohort of central bankers to bind their successors, setting aside whether a central bank can credibly make such a commitment? The European Central Bank (ECB) seems reluctant to guarantee the current rates structure much beyond ECB President Draghi’s tenure. We have questioned whether there is a desire or practical ability to lock in Fed policy for an extended period.

In academic models, commitment is seen as the best long-term strategy; deviating from the commitment is considered bad. These assumptions imply that tying one’s hands is the best strategy. However, if there is a 50% probability that it is optimal to deviate from the commitment, for example, because the subsequent shock would be much harsher than the initial one, adhering to the commitment would no longer be valuable. In fact, the commitment outcome could be worse than outcomes in an ad hoc policy-making framework. In such a scenario, the market’s commitment calculus could incorporate the probability that the central bank will correctly deviate from its commitment, but the commitment signal would be watered down greatly.

Central banks tend to hedge forward guidance by making the policy commitment conditional – e.g., pledging to raise rates extremely slowly if inflation and activity evolve as expected. For example, In July 2009 the Bank of Canada (BoC) wrote: “Conditional on the outlook for inflation, the target overnight rate can be expected to remain at its current level until the end of the second quarter of 2010”. The BoC backed away from this guidance in March 2010 and raised rates in May. No one can accuse it of lying, but it wouldn’t be accused of telling the truth either.

A formal commitment that is adhered to irrespective of the economic circumstances runs the risk of limiting policy options when flexibility is needed. The greater the conditionality the more limited the commitment. Central banks often imply that their forward guidance is more binding than it is, but forward guidance is not much better than a forecast.

Are all shocks small shocks?

The six-month high for 10Y Treasury (UST) yields is 2.75%, the high for 2Y UST rates is 2.55%. Current yields are 2.13% for the 10Y and 1.88% for the 2Y. We recently lowered our year-end US Treasury forecasts to 1.75% for 10Y and 1.65% for 2Y UST maturities.

The Fed clearly has room to ease in response to a small or moderate shock to the US economy. In unemployment rate terms, we are confident it can ease enough with rates and QE to offset a 1ppt upward move in the unemployment rate. It may be able to deal with a 1.5ppt shock, but we doubt they have enough tools to deal with a 2.5ppt or higher shock, however it modifies its monetary policy framework.

Since 1969 there has only been no downturn in which the unemployment rate went up less than 2ppt and only one in which the run-up was not much more than 2ppt. The Fed’s current approach in the face of a big shock would be to make a big commitment and hope that it is credible enough to convince businesses and households to start spending. As noted, even Fed Chair Powell is not convinced that this will work.

Investors seem reasonably confident that the Fed could respond to a modest recession and that they could top up the stimulus tank with promises policy rates ‘lower for longer’. They are very sceptical that the Fed and other central banks have enough in the tank to deal with a big downward shock. The lingering question is what the central bank does when the shock is so big that stimulus resources are exhausted. In a model you can commit to keeping rates low and QE steaming for an extended period, but maintaining such a heavy foot on the gas for a long time without side effects in practice seems untenable.

Fiscal policy under lock and key

Our scan of the papers at the Chicago Fed conference did not show any paper that mentioned a role for fiscal policy. Fed Chairs have been reluctant to give anything but the vaguest of advice on fiscal policy and have indicated that they see fiscal policy as an input into monetary policy but not as a topic requiring any Fed decision. Investors might find this monetary policy approach more credible if there were an acknowledged role for fiscal policy in the event of a big shock. Rapid fiscal stimulus in a downturn could raise inflation expectations, real rate expectations, and speed recovery. Many countries’ institutional framework does not have fiscal policy as a reliable partner to monetary policy in times of economic stress. However fiscal stimulus would likely be much more effective near the zero bound than additional monetary stimulus and would probably enhance the effectiveness of monetary policy. It is useful to differentiate between the pace of recovery that can be expected based on monetary policy alone and with a fiscal kick.

There may be some fear that leaving room for fiscal strategies opens the door to Modern Monetary Theory (MMT), in which macroeconomic policy is driven by the fiscal stance and monetary policy plays a secondary role. We are somewhat sceptical of MMT (see Modern monetary theory for conservatives and Modern monetary theory and asset markets), but find it hard to see how fiscal policy would not be a major driver of cyclical recovery in the event of a major shock.

Finally, Englander points out an issue not discussed directly in the note above is the possibility that monetary policy is less effective than it used to be. But the Fed didn’t discuss it either at their Chicago Conference, so we are in good company.

With 2s and 10s both less than 2.25%, H2 had better be a corker in terms of growth.

By lowering rates so dramatically, markets have already priced in a fair amount of the monetary stimulus the Fed had in its armory last Fall. 

We argue that the Fed likely does not have the tools to deal with a major negative shock. If monetary policy is ineffective, they may not have the tools to do with a modest one either.

The evidence is mixed but worth examining:

  • The last decade of low rates and the expansion of the Fed’s balance sheet did not accomplish nearly as much as we (and the Fed) would have expected.

  • But tightening seems to have been very effective in terms of restraining demand at least abroad and maybe domestically.

We are not comfortable with unexplained asymmetries, but neither should the Fed be.