Fiscal forecasting bias

If you’ve read this blog much you’ll know how obsessed Matt is with independent fiscal forecasts. He has variously called for an independent body to cost parties’ election proposals — Peter Dunne’s with him on that one — and independent bodies to set tax rates and maintain medium-term fiscal balance. The two have been jointly brought to the fore by the arguments about Treasury’s PREFU economic forecasts and the ability of each major party to balance their budget. Each party has accused the other of wild optimism but I haven’t yet noticed people taking pot-shots at the PREFU forecasts themselves. But can we really trust the government’s forecasts?

A recent paper on just that topic (ht. Robin Hanson) has, conveniently for us, investigated whether governments’ economic forecasts tended to be systematically biased in any particular way. It turns out, slightly unsurprisingly, that they are almost always optimistic and more so in the medium run than the short run. That, in turn, leads to over-spending by governments and constant budget deficits relative to forecasts:

Over-optimism in predicting growth appears linked to over-optimism in predicting budget balances. On average, the upward bias in growth forecasts is 0.4 percent when looking one year ahead, 1.1 percent at the two-year horizon, and 1.8 percent at three years.

So that’s bad news for governments and bad news for Matt’s budget-balancing agency, you might think; but not so fast! It turns out that, first, New Zealand’s growth forecasts are, on average, pessimistic. So maybe trusting The Treasury isn’t quite as dangerous as David Cunliffe suggests. Secondly, the one country in which Frankel detected no consistent bias was Chile. He attributes that to the existence of an independent forecasting body that is insulated from political pressure and manned by economists. That certainly seems to confirm Matt’s priors about the worth of an independent body, too.