The RBA just released a wonderful RDP, which looks at historical forecast accuracy of the RBA’s SMP projections.
I love this RDP because it shows how difficult a job forecasting (in the real world) really is.
The GDP fan chart says it all — the 90% confidence interval interval on a 3% GDP growth forecast for two years ahead is ~500bps wide.
The confidence intervals in Figure 3 strike many observers as wide, particularly for GDP growth. In other words, our estimates of uncertainty are surprisingly high.
Initial impressions presumably reflect comparisons with subjective estimates of uncertainty. Psychological studies find that subjective estimates of uncertainty are regularly too low, often by large margins. People have a systematic bias towards overconfidence.
So next time you hear two ‘experts’ argue over a 100bps difference in their GDP forecasts, ask them if the forecasts are statistically different. If they claim that they are, they probably aren’t experts (or perhaps they really are both just that much better than everyone else).