Interest rate forecasts: a pathology

This is the first of three prospective papers examining how well forecasters can predict the future time path of short-term interest rates. Most prior work has been done using US data; in this exercise we use forecasts made for New Zealand (NZ) by the Reserve Bank of New Zealand (RBNZ), and those derived from money market yield curves in the UK. In this first exercise we broadly replicate recent US findings for NZ and UK, to show that such forecasts in NZ and UK have been excellent for the immediate forthcoming quarter, reasonable for the next quarter and useless thereafter. Moreover, when ex post errors are assessed depending on whether interest rates have been upwards, or downwards, trending, they are shown to have been biased and, apparently, inefficient. In the second paper we shall examine whether (NZ and UK) forecasts for inflation exhibit the same syndromes, and whether errors in inflation forecasts can help to explain errors in interest rate forecasts. In the third paper we shall set out an hypothesis to explain those findings, and examine whether the apparent ex post forecast inefficiencies may still be consistent with ex ante forecast efficiency. Even if the forecasts may be ex ante efficient, their negligible ex post forecasting ability suggests that, beyond a six months’ horizon from the forecast date, they would be better replaced by a simple ‘no-change thereafter’ assumption.