Do interest-rate forecasters herd? International evidence

Abstract

Results of earlier empirical research on whether survey data on forecasts of interest rates exhibit signs of forecaster herding are mixed. We reconsider the question of forecaster herding using a large international data set of interest-rate forecasts. We do not find much evidence of forecaster herding. Rather, forecasters seem to anti-herd, that is, they seem to intentionally scatter their interest-rate forecasts around the consensus forecast.

Keywords:

• interest rates
• forecasting
• (anti-)herding

JEL Classification:

• E43
• E47
• E44

Acknowledgements

We are grateful to the European Central Bank for providing parts of the data.

Notes

1. ¹ Forecasters work with the private sector in the respective country. The survey participants are professional economists working for universities and financial institutions such as international economic research institutes, investment and commercial banks. Further information concerning the survey can be found at www.consensuseconomics.com.

2. ² We found strong evidence of forecaster anti-herding when we used the contemporaneous consensus forecast. We also found strong evidence of forecaster anti-herding when we used the forward rate to measure the consensus forecast. Results are not reported for the sake of brevity, but are available upon request.

3. ³ It should be noted that the number of forecasts available for the Philippines is rather small. The interpretation of the empirical results for the Philippines, thus, should not be stretched too far.

4. ⁴ Results (not reported but available upon request) for most other countries confirm the significantly negative correlation between forecaster anti-herding and interest rate volatility.