Forward interest rates are interesting inflation indicators since they may give information about market expectations of the expected path of inflation. The aim of Paul Söderling in Discussion Paper No. 1313 is therefore to evaluate this approach and to address the following question: If we observe an increase in the nominal forward interest rate, what is the best guess of the increase in inflation expectations? The answer depends on variances of and correlations between inflation expectations and real interest rates. For example, if they are completely uncorrelated and the real interest rate is constant, then inflation expectations move one-for-one with the nominal interest rate. Conversely, if real interest rates are more volatile than inflation expectations the increase in the nominal interest rate probably has little to do with changes in inflation expectations.

The results for US data show that real interest rates are far from constant, but less volatile than inflation expectations, and that there is a tendency for inflation expectations and real interest rates to move into opposite directions, which has to be taken into account. These results are obtained in a framework where inflation expectations and expected real interest rates are assumed to have a joint normal distribution and the real interest rate is constant. The aim of the paper is to determine empirically the coefficient of the nominal interest rate, and to compare the optimal rule with the common practice of assuming a constant real interest rate. The main finding is that assuming a constant interest rate is wrong, but happens to give the best guess of inflation expectations.