Strategic vs parametric thinking

Brad De Long makes an important point when he says that rational thinking depends upon context, and we must always ask: “are we playing some kind of game against nature, or are we playing against another mind?” If it’s the former, then standard parametric thinking is appropriate (max U subject to constraints). If it’s the latter then we are in the realm of game theory.

You might think the distinction obvious. Many people, however, fail to make it. As Meir Statman says:

Think of trading stocks, bonds and other investments as a trading race. Traders who commit framing errors frame the trading race as between them and the market…Traders possessing human-behaviour and financial-facts knowledge frame trading correctly as against traders on the other side of the trades. (Finance for Normal People, p40)

An especially egregious example of this is the IPO anomaly, the tendency for newly-floated shares to do badly in the three years (pdf) after they come to market. One reason for this is that investors think they are playing a game against nature when in fact they are playing against another mind – those on the other side of the trade. They fail to ask: “if this is such a good investment, why are its owners selling it?”

David Navon calls this mistake the egocentric framing error – the tendency to think only from our own point of view and to neglect how others are thinking. Bad chess or poker players, for example, consider only their own strategies without asking what their opponents are planning. In traffic jams we often try to switch lanes without getting any further because we fail to see that others are behaving just like us. And bad negotiators consider only what they want without thinking about their counterparty’s strategy: as Roger Fisher and William Ury show in Getting to Yes (pdf) good negotiators put themselves in the shows of those on the other side of the table. One reason (of many) why we’ve ended up with a bad Brexit deal is that the government failed to do this.

There are other examples of egocentric framing, described in Jerry Muller’s The Tyranny of Metrics. Managers and politicians who set targets without seeing that they can be gamed or misused are committing this mistake. They believe they are in a parametric environment when in fact they’re in a strategic one. They’re attributing agency only to themselves and not to others.

Perhaps this point broadens. Maybe there are particular types of people who are especially prone to egocentric framing. Those who are traditionally the subjects rather than objects of history have historically had less need to treat others as agents with their own strategies – especially if those others have been cowed into submission, Men have traditionally had little need to think about gender, or the posh about class, or white people about race. They haven’t therefore had the need to ask David Mitchell’s question: “are we the baddies?” Those on the dirtier end of these sticks haven’t had such luxury. As Robert Burns wrote:  

O wad some Power the giftie gie us 
To see oursels as ithers see us! 
It wad frae mony a blunder free us, 
An' foolish notion.

Comments

[csissoko]

Wasn't this situation precisely what the traditional jobber-broker structure of the Stock Exchange designed to deal with? The game theory of the market was such that an outsider would just get scammed. The job of the broker was to understand the game theory underlying the market and play it in the interests of his client.

[AllanW]

It's a shame your otherwise interesting article is one paragraph too long.

When you make such an egregious mistake as to state;

"Men have traditionally had little need to think about gender, or the posh about class, or white people about race."

you lose whatever credibility you had built up to that point. Humans constantly think about gender, the wealthy constantly think about their position atop the pile and those 'beneath' them, and all people are constantly aware of racial differences.

[Nick Drew]

The entire enterprise of "forecasting prices" is based on the fallacy you describe.

Unsurprsingly, the track record of all forecasting is absolutely risible

And yet it is an entire industry; people pay good money for the output; and (many of) the producers are highly educated people who claim to "believe" in what they are doing, Nash Equilibrium and all

The cognitive dissonance and doublethink involved when they are invited to look back at their previous forecasts is so great, in polite circles it's considered inappropriate ever to do so. (All you get is: errrr, we've recalibrated the model since then)

[GrueBleen]

Well, not a bad summary S&M, though I do think that when discussing 'strategy' you might at least mention the Keynesian 'beauty contest selection' question.

Otherwise, surely if men try to appreciate women, or the rich appreciate the poor, or whites appreciate 'coloureds', then they are just committing 'cultural (mis)appropriation' ?

[Mat Weldon]

Are you intentionally channeling Robert Aumann's 1987 paper "Correlated Equilibrium as an Expression of Bayesian Rationality" (Sorry can't figure out how to link on my phone) or by accident?

It not only discusses exactly the distinction between Bayesian and game-theoretic decision making, but it also begins with the Burn quite you ended with!

[Mat Weldon]

*Burns quote*!