Treasury's Forecasting Performance: A Head-to-Head Comparison (WP 06/10)

Abstract

Work on assessing Treasury’s forecasting performance to date has focussed on comparisons against consensus forecasts. This study compares Treasury’s GDP and CPI forecast performance against individual private sector forecasters as well as major public sector institutions such as the IMF, OECD and the Reserve Bank of New Zealand. The head-to-head comparison makes it possible to assess Treasury’s forecasting performance relative to its peers. When compared across all evaluation periods covering 1996-2005, Treasury’s GDP forecast performance was ranked in the middle at seventh out of 16. The large forecast error for the 1998 year had a material impact on Treasury’s overall forecast performance. Treasury’s CPI forecast performance was not as good, placing tenth out of 12. Large forecast errors for the 1998-2000 period accounted for the poor CPI forecast performance. Treasury’s overall forecast performance was better when evaluating only the current year Budget forecasts, placing fourth for GDP and sixth for CPI. This suggests that Treasury is better at forecasting the current year than the year ahead. Consistent with international studies, no single forecaster consistently outperforms the Consensus, with Treasury beating the Mean 30% of the time for GDP and Consensus 33% of the time for CPI. All forecasters find it difficult to pick recessions and turning points. Large forecasting groups generally have a poorer forecasting record on average.

Acknowledgements

We would like to thank Satish Ranchhod for supplying us with historical Reserve Bank of New Zealand forecasts.  We are also grateful to Yuong Ha and to the participants of Treasury’s Macro Reading Group for providing helpful comments and suggestions.

Disclaimer

The views, opinions, findings, and conclusions or recommendations expressed in this Working Paper are strictly those of the author(s).  They do not necessarily reflect the views of the New Zealand Treasury. The Treasury takes no responsibility for any errors or omissions in, or for the correctness of, the information contained in these working papers. The paper is presented not as policy, but with a view to inform and stimulate wider debate.