Should central banks publish their interest rate forecasts?

By: Stefan Gerlach, University of Frankfurt

 

Whether or not a central bank publishes its internal interest rate forecasts has become the new litmus test of best practice in monetary policy. Leaders in this area are the Reserve Bank of New Zealand, Norges Bank and the Swedish Riksbank.

 

The arguments for publishing interest rate projections are compelling: by releasing such forecasts central banks can impact on expectations of future interest rates held by firms and households and affect long-term interest rates. Since these are much more important determinants of the demand for goods and services than the very short-term interest rates central banks control directly, this increases the effectiveness of monetary policy, particularly in economies with large and deep financial markets.

 

A high level of transparency also makes it possible to hold the central bank accountable which makes for better policy: the ability of parliaments, academics and journalists to review the facts and point to shortcomings in policy provides good incentives for policy makers to do their very best when setting interest rates.

 

Furthermore, releasing forecasts of future interest rates promotes financial stability. One wonders whether the recent tensions in US subprime mortgage markets would have been so severe if mortgage applicants as a matter of course had seen a fan chart for the likely future course of US monetary policy.  

 

But there are also arguments against publishing interest rates paths. In particular, if the central bank’s forecasts were much less accurate than the public’s forecasts (and were published without a measure of the uncertainty associated with them), then releasing interest rate paths would only add noise. But that that would be the case seems rather hypothetical – the central bank would at least appear to be in an advantaged position in predicting its own future behavior. 

 

Another argument against publishing interest rate paths is that some central banks set policy one MPC meeting at a time. Since they do not discuss potential future policy decisions, they have in fact little information about future policy to reveal. But then again, the modus operandi of MPCs could presumably be changed and policy be discussed in terms of interest rate paths.

 

While the battle lines are drawn, somewhat surprisingly the debate has not focused on just how much information interest rate paths published by central banks contain. Doing so is instructive.

 

The graphs below come from recent reports published by the Norges Bank (top) and the Swedish Riksbank (bottom). They tell an interesting story: the repo rate in a year from now will be, with 90% probability, somewhere between 4% and 7.75% in Norway and between 3.25% and 5.50% in Sweden. Even broader confidence bands apply at longer horizons.

 

 

 

While releasing interest rate projections is surely attractive in theory, one is hard pressed to believe that, in practice, releasing these forecasts materially improves our understanding of monetary policy in Norway and Sweden. The confidence bands are simply too wide, reflecting the fact that central banks in fact don’t have firm ideas of how monetary policy will evolve in the near future. Before the level of uncertainty can be reduced, the benefits will remain largely academic.

 

Stefan Gerlach

Professor of Monetary Economics

Institute for Monetary and Financial Stability

University of Frankfurt

---