jump to navigation

Hoops, Paydays, and Epistemology April 12, 2009

Posted by Dwight Furrow in Culture, Current Events, Dwight Furrow's Posts, Philosophy.
Tags: , , , , , ,

Cross-posted at Reviving the Left 

The excessive risk-taking by banks and investment firms that caused the economic meltdown was in part the result of the skewed incentive structure of their compensation packages. Because their compensation was based on how well their company’s stock did over the relatively short term, they had every incentive to take on excessive, long-term risk if it would boost their short-term profits.

What was bad for the health of the firm was good for the stock trader or manager who pocketed immediate benefits.

NPR’s Scott Horsley, referencing a recent NY Times article by Michael Lewis on NBA star Michael Battier, compared Wall Street’s approach to compensation to compensation based on individual statistics in the NBA. Players often take poor shots in order to pad their stats, even though passing the ball would be better for the team, and many players virtually ignore defense, an essential part of the game, because it is more difficult to represent in statistical measures. If a player’s compensation is dependent on individual statistics only, they are rewarded for actions that often hurt the team.

These skewed incentives indicate a disturbing trend in contemporary society. We tend to form beliefs around data that is pervasive only because it is easy to acquire. It is easy to count and assign individual responsibility for baskets or sales minus expenses. These are convenient ways of keeping score.

However, the fact that a bit of data is easy to gather does not mean that it is providing a comprehensive, accurate measure of the health of the firm (or a basketball team). Stuff that isn’t easy to measure is not part of the calculation.

So why do hard-working, serious people take the easy way out when trying to measure performance? Isn’t it obvious to management (whether in sports or business) that the way the measure performance can be incomplete?

In a recent post, I argued that our economic problems were the result of an epistemological crisis. Our Wall St. wizards created lots of investment vehicles they didn’t understand and could not analyze.

But this tendency to form beliefs and hence policies around easily accessible data suggests another epistemological dimension to this crisis that explains why we take the easy way out.

Jerry Z. Muller writes:

From the point of view of top management, the diversity of operations means that executives were managing assets and services with which they have little familiarity. This has led to the spread of pseudo-objectivity: the search for standardized measures of achievement across large and disparate organizations. Its implicit premises were these: that information which is numerically measurable is the only sort of knowledge necessary; that numerical data can substitute for other forms of inquiry; and that numerical acumen can substitute for practical knowledge about the underlying assets and services. A good deal of our current economic travails can be traced to this increasing valuation of purportedly objective criteria, so denoted because they can be expressed and manipulated in mathematical form by people who may be skilled at such manipulation but who lack “concrete” knowledge or experience of the things being made or traded.

This is a theme that I emphasize in Reviving the Left. The pursuit of objectivity, and the resulting abstract representations, often leave out crucial data we need in order to act well, especially where human beings are concerned. Sensitive, situational knowledge that sometimes involves feelings and intuition is essential for guiding human action, but it is not well represented by mathematical models. Yet the “cult of accountability” is fast spreading throughout society. From finance and basketball to education and health care, every activity is being measured by a metric that may not capture crucial components of the activity—with sometimes disastrous consequences.



1. Ian Duckles - April 13, 2009

I definitely agree with the major claim here, but I think it is also important to include the fact that humans generally do a pretty bad job of assessing long-term risk, and a big problem here is that we rely too much on feeling and intuition. My favorite example of this phenomena concerns that increase in vehicle fatalities after 9-11-01.
Now certainly, we have an expectation that the financial folk should be able to do a better job of this than the man on the street. After all, that is supposedly what they are being compensated for. This inability to accurately assess risk does account to some degree for the desire for nice metrics that can summarize risk. Because it is so hard for us to do this, we desire a nice little number that does our thinking for us and tells us whether to invest or not.
I don’t think that the solution is necessarily a move away from metrics, instead, I think it makes more sense to only have simpler investment vehicles where the value of the investment is tied more directly to some actual thing like a company rather than these complex sliced-and-diced investment vehicles that have a very tenuous connection to the underlying asset that one is supposed to be investing in. In this way, it would be easier for individuals to see and understand what they were investing in, and thus have a better sense of what the metric is a measure of.

2. Dwight Furrow - April 14, 2009


I agree. Metrics are important but only if we know precisely what they are measuring and only when we have a good sense of when they apply and when they do not–judgment is always important. And of course you are right. The transparancy of the practice being measured is crucial in exercising good judgment.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: