Expert Forecast vs. Dart-Throwing Monkeys

expindustry_dart-throwing-monkey.png

“People who spend their time, and earn their living, studying a particular topic produce poorer predictions than dart-throwing monkeys who would have distributed their choices evenly over the options.”

Daniel Kahneman, Nobel Prize in Economics and author of Thinking, Fast and Slow

Far from the celebrated homo economicus, which is a strong assumption in most economic thinking, human beings are often irrational. We are consistently and predictably biased in a number of ways: anchoring, loss aversion, availability heuristic, overconfidence, planning fallacy, etc.

For instance, our capacity to remember the past far outweighs our capacity to imagine the future. So when we (experts, analysts, planners…) make forecasts, we tend to reproduce the past instead of genuinely envisioning the future.

“If economists could make reliable predictions about the economic world, far more of them would be rich by now.”

Nassim Nicholas Taleb, Distinguished Professor of Risk Engineering at New York University and author of Fooled By Randomness

Another bias that affects our forecasts is loss aversion. We are disproportionally resistant to a prospect of loss compared to an equivalent prospect of gain.

Some Examples #

Example 1. Should we trust market forecasts?

Hint: not always

The IMF solemnly advised the nations of Europe coming out of the financial crisis to raise taxes and wind back government spending. And it had modeled that, for every dollar they cut their budgets; their economic growth would suffer just 50¢. Its forecasts said so. But in 2013, in a working paper published by the IMF [ref. WP 2013/01], it revealed that for every dollar those nations cut their budgets, their economies crumpled around $1.50.

So why should companies base their market analysis on GDP forecasts when the forecasts themselves may be questionable?

Example 2. Can we manage production changes?

Hint: it looks tricky

In the aerospace industry, companies seem to be consistently biased in their planning. They are over-optimistic: consistently, more units are planned than actually produced; and ramp-down averse: they take larger risks to avoid a loss than to secure a gain, so although they under-estimate both production ramp-up and ramp-down, they are much more averse to ramp-downs [ref. 013-002-001].

So why should a company detect the next production ramp-down when it completely missed the last one?

Example 3. Can product development programs be delivered on time?

Hint: it seems exceptional

Study after study [ref. HSE 12853] seems to confirm an optimistic bias and planning fallacy in the organizational context. Comparative quantitative analyses reveal that project managers routinely perceive the future as inherently easier than the past. We fall for best case scenarios; fail to draw upon the experience of others; have a tendency for power bias (high hierarchical positions underestimate the duration and oversimplify the task more); etc.

So why should a company act as if the next program will be delivered on time when the last one was not?

Great News #

The GREAT news in all of this is that people are predictably irrational and not just randomly crazy! So the simple fact of knowing and recognizing these cognitive limits and biases can give progressive companies a decisive competitive advantage over their competitors.

Author: E. Dib #
References #
Further Reading #
 
19
Kudos
 
19
Kudos

Now read this

Emilio Dib

Emilio’s papers on exp(industry). Antifragility and the conservatism of large organisations # Why? # A look at commercial passenger jet airliners: Strategic considerations # Is it a better economic choice to drive an electric car, today?... Continue →