The Undercover Economist in the Dragons' Den

In his regular column for the Financial Times (UK), Tim Hartford -- the author of The Undercover Economist -- discusses a BBC reality show, Dragons' Den, that combines American Idol with venture capital investing. The article, The Undercover Economist: Shot down in flames (Sep. 15, 2006), analyzes the show -- particularly the dynamics between the hopeful entrepreneurs and the venture capitalist judges -- by applying auction theory (and related topics like the 'winner's curse') from economics and game theory. I found the observations about the information revelation that takes place in auctions -- which Dragons' Den essentially is -- to be particularly interesting:

The den is, in fact, an auction room by another name. Dragons bid against each other, and against the unknown outside offers that the business may receive, and they should be aware that every bid reveals information to the other dragons and the entrepreneur. This is the point of an auction: the seller gives buyers an incentive to reveal, through their bids, what they know about the prize’s value. The auction also, rather neatly, collects money on that basis.

I don't know if you can watch the show in the US (either on BBC America or PBS). Considering the fact that many American reality shows had there start in the UK or in Continental Europe (including American Idol), I wouldn't be shocked if some version of this show was adapted for the US.

New Yorker on Neuroeconomics

The New Yorker has an article in its latest issue on neuroeconomics, Mind Games (Sep. 11, 2006). I just heard an interview on NPR with the author of the piece, John Cassidy, and it sounded interesting.

Efficient Market Hypothesis as Law

The Economist has a fascinating article, Dismal science, dismal sentence (Sep. 7,2006), on how the efficient market hypothesis -- a guiding principle for many financial economists -- has been applied in the US legal system via the Supreme Court decision in Basic Inc. v. Levinson. In that decision, the Supreme Court: (1) enshrined the efficient market hypothesis (at least in its weak version) as the framework within which federal judges and juries should view the workings of the financial markets; and (2) where the actions of defendants are found to deviate from the efficient market hypothesis in securities law and securities regulation cases (such as in insider trading cases), the defendant will be deemed to have committed a 'fraud on the market' by distorting the informational value of securities prices.

There seems to be two intellectual objections to the 'fraud on the market' principle adopted by the the US federal courts and the securities regulators (e.g., the Securities and Exchange Commission). The first type of objection is that, assuming the efficient market hypothesis should be used as the guiding framework to analyze the workings of the financial markets, the courts are misapplying it by not distinguishing between the issue of relying on the distorted pricing information and the issue of what type of damages or other punishments should be associated with the distorting behavior. This argument is laid out in Bradford Cornell's paper, Market Efficiency, Crashes and Securities Litigation.

The second type of objection is that the markets aren't really that efficient any way; i.e., the efficient market hypothesis (even in a relatively weak form) shouldn't be used as the analytical framework by the courts. I should note that, even if this second type of objection should be the better economics (although not if you're from a Chicago School mindset), it is not necessarily good law since it seems to me that insiders that misuse insider information to illegallly profit should be found guilty and/or liable for committing a fraud on both the markets as well as the existing shareholders.

To paraphrase James Buchanan, a Nobel prize winning economist, law and economics don't always mix.

Math Against Terrorism

A week from now, we will be commemorating the fifth anniversary of 9/11. I can still remember that morning. As I was groggily waking up in my apartment in Cambridge, Massachusetts, I was shocked to hear on my alarm clock radio that the World Trade Center had been hit by two airliners. (It shook me up even more because I had seriously thought about flying to California around that time ... so it was possible that I might have found myself on one of those planes from Boston's Logan Airport to the West Coast.) So, I think it's appropriate to write this brief blog posting on how math is being used to help in the fight against terrorism.

The best news article I've read on how mathematics is being employed in the war on terrorism is Wired News: Can Math Help in Terror War?. The article mentions the two ways that mathematicians (and I will include computer scientists, economists, mathematical psychologists, et al., in that label for the purposes of this blog post) take in applying mathematics to counter-terrorism.

The first path mathematicians take in analyzing terrorism is game theory and/or similar but alternative approaches that stem from applications in the social sciences (economics, psychology, etc.). One 'similar but alternative' approach is reflexive theory (a topic I have wroted about previously). The basic theme common to this branch of thinking is the application of mathematics to the strategic thinking of terrorists.

The second road taken by applied mathematicians in the War on Terrorism is based more on the organizational structure of terrorism (obviously, there is some overlap between strategy and organization). The methodologies approached to getting a clearer picture of the organizational structure of terrorists groups, like Al Qaeda, include Social Network Analysis and Formal Concept Analysis. These and other approaches basically boils down to drawing a 'map' of of the structure of terrorist group and/or its cells based on available data.

The assumption that underlies the attempts to uncover the organizational structure of terrorist groups is that terrorist groups are organized as, what mathematicians call, a partially ordered set (or a poset) as opposed to a totally ordered set (without getting into the technical details, think trees and/or lattices when thinking of partially ordered sets (or posets), and think of a straight line when thinking of totally ordered sets). The advantages of trying to map the partially ordered structure of terrorists groups is that: (a) it is more realistic to think there are overlapping ('horizontal') links between terrorists interspersed within any top-down ('vertical') hierarchy, and (b) this realism would, hopefully, help us to pinpoint terrorists (and/or their cells) and what importance they have within the overall organizational structure.

There are problems with taking either of the two paths -- analyzing terrorist strategy or their organization -- to battling terrorism via math. [Some of these problems are highlighted in a New York Times op-ed piece by mathematician, Jonathan Farley, The N.S.A.'s Math Problem (May 16, 2006).]

One of the problems with looking at terrorist strategy is that terrorists adapt their strategies as conditions change. For example, when it became more difficult to bring in knives and box cutters on airplanes, Al Qaeda came up with the strategy to bring hard-to-detect liquid explosives onto aircraft. Another problem with trying to unravel terrorist strategy is that strategy almost always (except for highly idealized games like chess) involves a random (or stochastic) component. In trying to unmask the strategy of terrorists, think poker instead of chess.

Trying to uncover the structure of terrorist organizations also has limitations. For example, using the methods that are currently available to us, one person who might be mis-identified (due to all of the empirical 'links' between this person and known terrorists) as a 'terrorist' would be Secretary of State Condeleeza Rice!

To be fair, the Center for Computational Analysis of Social and Organizational Systems (CASOS) at Carnegie Mellon has succeeded in identifying the changing leadership of Hamas in the past. So, better data and better methods would probably lead to a more robust method of mapping terrorist networks, and hopefully, preventing another tragedy that woke all of us up to the dangers of terrorism on that fateful morning five years ago.

Revenge of the Long Tail

There is a new Hollywood drama that is gaining a lot of attention lately starring Tom Cruise, but it's not a movie ... but if it was a movie, it might be called the 'Revenge of the Long Tail.' Sumner Redstone, the chairman of Viacom, all but ran Tom Cruise -- one of the biggest stars in Hollywood -- out of Paramount Studio's lot. A recent New York Times article, A Big Star May Not a Profitable Movie Make (Aug. 28, 2006), argues that -- from both an economics and statistical perspective -- Sumner might have been, in the article's words, "crazy like a fox."

According to the research cited in the article, there seems to be little or no significant correlation between Hollywood stars and the success of the films they star in. For example, Art Devany -- an economist who has specialized in analyzing Hollywood -- conducted research (with W. David Walls) that showed "only seven actors and actresses — Tom Hanks, Michelle Pfeiffer, Sandra Bullock, Jodie Foster, Jim Carrey, Barbra Streisand and Robin Williams — had a positive impact on the box office, mostly in the first few weeks of a film’s release. ... No other star had any statistically significant impact at all." (One could argue that ascribing success to those seven actors may be confusing correlation with causation.)

What this incident between Mr. Cruise and Mr. Redstone may represent is the realization by the business side of the entertainment industry that the huge salaries and perks demanded by the stars aren't quantitatively justified. This possible reversal of fortunes for Hollywood's star system may represent what I would have called (in a previous blog) overcoming the Tyranny of the Power Law -- where a 'Long Tail' based strategy is embraced over a more elitist strategy.