The CFA Society Toronto annual forecast dinner drew more than 900 attendees to celebrate the Society’s 80th anniversary. The gala event, renamed the Annual Investment Dinner, took a new direction in its program this year. Instead of focusing only on specific forecasts, it explored the art and science of forecasting and the philosophy and process behind investment decision-making.

The keynote speech was given by Wharton professor Philip E. Tetlock, whose book Superforecasting: The Art and Science of Prediction is widely regarded as one of the most important studies of the skill of prediction. It draws on many years of research and the results of a 20-year-long, U.S. government-funded forecasting competition or “tournament,” entitled The Good Judgment Project. The project dealt with a wide range of subjects including geopolitical, economic, and financial.

“Superforecasting is not a gift but a skill that can be taught, practised, and improved.”

Tetlock and his colleagues have concluded that some people, whom he calls “superforecasters,” can make much better predictions than others and, importantly, that their foresight is not a gift but a skill that can be taught, practised, and improved.

Superforecasters, he argued, are clever, on average, but by no means geniuses. Most important, Tetlock maintains, is mental attitude. There are two categories of people: those whose understanding of the world depends on one or two big ideas and those who think the world is too complicated to reduce to simple terms, he explained. Superforecasters are drawn exclusively from the latter category.

Humility in the face of a complex world makes superforecasters subtle thinkers, he said. They’re generally comfortable with numbers and statistical concepts, but they’re not statisticians or builders of complex mathematical models. They break seemingly intractable problems into manageable sub-problems and balance inside views with outside views.

Superforecasters have an appetite for information, a willingness to revise their estimates based on new information, and the ability to draw on information from many diverse sources, he continued. Their mindset is most important: a mixture of determination, self-reflection, and a willingness to learn from mistakes. The best forecasters, he maintains, are less interested in whether they’re right or wrong than in why they’re right or wrong. They’re always looking for ways to improve their performance.

“Would-be superforecasters should keep score of their performance, analyze what went right and wrong, seek inside and outside opinions, and continually challenge prevailing views.”

Tetlock’s advice to would-be superforecasters is to eschew “vague-verbiage” forecasts, keep score of their own performance, analyze what went right and what went wrong, look for errors behind past mistakes (while being wary of hindsight biases), and continually challenge their own thinking and that of others.

He contends that politics and human affairs are not complete mysteries. Instead, they’re like weather forecasting, where short-term predictions are possible and can be reasonably accurate.

“Forgetting the inevitable cyclicality of markets is a common and costly mistake.”

Making Investment Decisions

Insights into the philosophy and process behind investment decision-making came from two fireside chats, moderated by Barry Ritholtz, founder and chief investment officer of Ritholtz Wealth Management and a columnist for Bloomberg View and the Washington Post. In the first, Howard Marks, co-founder of Oaktree Capital, said that for individual and institutional investors alike, emotion is the enemy of good judgment on matters concerning valuation and investment cycles.

He cited Warren Buffett’s investments in Goldman Sachs and Wells Fargo in the wake of the financial crisis as examples of unemotional decision-making.

Marks observed that forgetting the inevitable cyclicality of markets is a common and costly mistake. Investors should have a clearly defined creed and should identify the sectors in which they believe they have an edge over the market, he said, noting that the macro variables such as economic activity, interest rates, currencies, and commodities are so well analyzed that it’s highly unlikely investors will find a competitive edge in analyzing them. Instead, he recommended seeking an edge in micro areas: lesser known and/or misunderstood industries and companies and their securities. He said the extent of media coverage and public investor interest should be regarded as contrary indicators.

Investors should find their competitive edge, regard the extent of media coverage and public investor interest as contrary indicators, and realize that controlling risk is more important than making the highest return.

To facilitate access to capital when markets are most attractively priced, Marks said investment managers should seek pre-commitments (i.e., pools of available funds they could draw down when depressed markets offer the best investment opportunities) from their clients. And, noting that the future is largely unpredictable, he said investors should beware forecasters who “don’t know or, worse, don’t know that they don’t know.”

The number one job of investment managers is not to make a very high return but to control risk, he said.

For the next one to three years, he noted that U.S. markets offer the best prospects, although valuations aren’t particularly attractive, while for the next 10, emerging markets offer the greatest potential returns, albeit with the prospect of more price volatility. However, he recommended that social stability and a strong rule of law should be requirements for investing in emerging markets.

The biggest mistake in investment management lies in failing to distinguish between being right on fundamentals and being right on prices.

The second chat featured Michael Mauboussin, head of global financial strategies at Credit Suisse and an adjunct professor of finance at Columbia Business School since 1993. Mauboussin said the biggest mistake in investment management lies in failing to distinguish between being right on fundamentals and being right on prices. Failing to judge correctly what is priced into market prices at any point in time and what will be priced into future market prices leads to poor performance, even if forecasts of the fundamentals prove correct. He emphasized that it’s changes in expected values that move markets and that developing an expected value mindset is vital to investment success. Investment managers, he said, must be aware of the heuristic biases that can lead them to make errors or fabricate rationales, and, most important, they must also ensure their processes are continuously scrutinized to uncover such biases.

Polling Station

Master of ceremonies Catherine Murray, host and anchor of Business News Network, conducted audience polls throughout the evening on a range of topical questions.

Asking audience members about their own forecasting success, 59 percent said their own forecasting ability was no better than average, and 21 percent felt it was only slightly better than average. One forecaster, however, received recognition for abilities that unequivocally belong in the superforecasting category: Rosemary Formusa of State Street Global Advisors won the prize for the most accurate forecasts of key economic variables made at last year’s event. Despite considerable price volatility over the last 12 months, her predictions for the S&P/TSX Composite Index, the Canadian-U.S. dollar exchange rate, and the price of gold all came within one percent of the actual figures.