Central bank promises: do markets believe them?

Long gone is the era when markets were left guessing about what the federal funds target actually was. Now we have a Federal Reserve which has not only told us about their current policy rate, but what they expect the rate to be two years from now.

At the forefront of these efforts for altogether more transparent monetary policy have been the select group of central banks that publish a projected path for interest rates.

Advocates say the projections give central banks greater influence over markets’ expectations, which in turn enhances the transmission of monetary policy. The case for projections, then, is similar to that for conditional commitments such as the Fed’s.

But do they actually work?

A report commissioned by Sweden’s parliament into the performance of the Riksbank – among this select group (which also includes the Czech National Bank, Norges Bank and the Reserve Bank of New Zealand) – suggests not. At least not past a year ahead. 

The Riksbank publishes a repo rate path – such as the one below – six times a year, at the same time as it sets rates.

But since introducing the projected paths in 2007, the Riksbank has frequently got it spectacularly wrong (as have markets). See the charts below, taken from the report:

Given the inaccuracy of the Riksbank’s forecasts, it is perhaps not surprising that market expectations have failed to adjust to the degree that advocates for (and those against) projected paths had assumed. Market yields – represented by the red and yellow lines – do not follow the projected path past more than a year ahead:

As the report notes, it is too early to draw any firm conclusions from this given the nature of events in recent years.

What do we make of this? Perhaps not much as yet. The period over which central banks have been experimenting with this procedure is short and recently interrupted by extreme shocks.

However, it appears the period has fed market scepticism of the projected paths:

One of the economists in a commercial bank told us that he viewed the Riksbank’s projections for the next couple of updates ahead as informative, for the more distant future mainly as an attempt to influence expectations without any real informational content.

If the projections are to become more credible, the report argues that the Riksbank must show more “self doubt”. It has two suggestions how.

First, the central bank should place greater emphasis on uncertainty. However, along with the uncertainty bands, the Riksbank already cautions that the projected path is “a forecast, not a promise”. What more could the central bank do to demonstrate uncertainty without this backfiring and markets viewing the forecasts with greater scepticism?

The second is that the projections incorporate the market yield curve on the grounds that the market is likely to be a better predictor of the future:

Beyond a couple of quarters ahead no single forecaster, including the central bank, has, or possibly can ever have, much clear information on likely future developments.  In such circumstances the ‘wisdom of markets’, bringing together multiple participants, using many different models, other sources of information, and differing subjective probabilities, constrained by the need to put their money where their mouth is, may well be greater than the accuracy of central bank forecasters.

One of the report’s co-authors, Charles Goodhart, has long favoured projections based on market expectations over those based on the central bank’s forecasts. However, the market yield curves for June and October of 2007 indicate they were even poorer predictors of the path of interest rates than the central bank. There is little here to indicate such projections would offer any marked improvement.

The Riksbank appears to have already ruled this recommendation out. This from Stefan Ingves, its governor, on Wednesday:

If we use market expectations as the starting point, it may be difficult to determine how the interest rate is connected to other developments in the macro economy. Our own forecast has the advantage that it is produced together with the forecasts for GDP, inflation, employment and so on.

The evidence suggests markets are less moved by central banks’ commitments on the future path of interest rates than the Riksbank and the Fed might hope. This is not to say that central banks are wrong to make such commitments – they are more transparent than other options, such as projected paths based on market expectations. More likely is that the projections’ influence is limited because markets recognise, after the turmoil of recent years, just how difficult it is to attach credibility to any forecast of more than a short while ahead.