They’re rare but creating a superforecaster, or cultivating some of the attributes, might be possible.
They may be rare, but superforecasters exist and corporate treasury executives should be on the look out for these talented individuals, whose forecasts far outperform control groups and even CIA analysts with access to classified information.
Does this mean the end of sketchy cash forecasts? Maybe. And although they are few and far between, there is a chance superforecasters can be created, provided a person is of average intelligence, level headed and works well with others. And even if a person doesn’t turn into a superforecaster, their forecasting abilities could improve with the right cultivation.
Recently, Michael Mauboussin, head of global financial strategies at Credit Suisse, told members of the NeuGroup’s Assistant Treasurers’ Group of Thirty (AT30) about a 21-year study conducted by Wharton School of Business professor Philip Tetlock, who tested 300 “experts” in academia and elsewhere on their social, economic and political forecasts. It found that the accuracy rate of their forecasts to be little better than chance, and in fact the media’s talking heads turned out to be some of the worst predictors.
In 2011 Professor Tetlock and several other academics joined with the US intelligence community in testing the forecasting abilities of 3,000 people from “off the street,” and they found that 2% were supeforecasters, who made consistently good forecasts way beyond what chance would dictate.
In the context of corporate finance, Mr. Mauboussin pointed to “key” attributes of superforecasting. One is personality, and superforecasters share the trait of clearly understanding what they know and what they don’t. Where most people are overconfident about their understanding of an issue and so their forecasts tend to be wildly off, superforecasters more carefully calibrate their level of understanding into their forecasts.
In addition, when they confront an occurrence that doesn’t match their world view, unlike most they actually adapt their view. And they are also “actively open minded,” Mauboussin said, and “not only willing to see other points of view but actively seek them.”
They may have an average intelligence quotient (IQ), but their rationality quotient (RQ) excels, and that’s good news for organizations such as treasury, because unlike IQ, RQ can be cultivated. By providing feedback and training, companies can improve their employees RQ and thus their forecasting abilities, if perhaps never to the level of a superforecaster.
Another attribute of superforecasters is that while they work well on their own, their forecasting becomes even better in teams. Larger teams can bring more skill sets to the table, but that’s offset by greater challenges coordinating the effort. Mr. Mauboussin said the ideal team size is between four and six members, and smaller is better than larger. In addition, diverse teams in terms of social and cognitive categories, such as race and gender, and education and functional knowledge, tend to generate more effective forecasts. Members should contribute equally to the discussion, and teams with women tend to outperform those with men.
Identifying superforecasters tends to take several months, and there’s no sure way to identify them in the hiring process, although personality tests may provide a clue. Mr. Mauboussin said the Comprehensive Assessment of Rational Thinking may be one of the more effective tests to screen for superforecasters.
Mr. Mauboussin concluded his presentation with the following summary:
- Foresight is a real and measurable skill
- Superforecasters are actively-open minded, intellectually humble, numerate, thoughtful updaters, and hard working
- Teams can be better than individuals, but only under the right conditions
- Integrating base rates can sharpen forecasts
- Getting good answers is different than asking good questions