Effective Tournaments

"Arguing in My Spare Time" No. 2.20

by Arnold Kling

May not be redistributed commercially without the author's permission.
Index

"I’m going to do it differently next time," Rob Main said to me last week during lunch. "I’m not going to work hard, risk all of my own money, and build real software. When I start my next business, I’m just going to pick a topic, put up a web site, get someone else to pay for it, lose tons of money, and go public after nine months."

Rob, another Web pioneer who recently sold his business, was describing the way that newer entrepreneurs seem to have started later and become wealthier much faster. Apart from whatever personal jealousy may be involved in our assessment, there are some valid concerns about the process generating the Net nouveau riche.

It seems reasonable that in the business world we would want winners to be selected in effective tournaments. This means a process that preserves businesses that implement good ideas while winnowing out businesses with flaws in their concept or execution. In fact, I would argue that a major challenge for the economy today is to come up with the best tournaments for selecting new opportunities to pursue on the basis of the Internet.

The most effective tournaments in which I have participated were for Othello (a board game somewhat like Chess or Go). When I played in the U.S. national championships, the first step was to qualify. The best veteran players qualified on the basis of their national rating, a highly reliable score that reflected their cumulative performance in other tournaments. It took me about four years to reach the point where I could qualify for the nationals on the basis of my rating. Regional qualifying tournaments created opportunities for newer upstarts, but in practice the regional winners rarely had success at the Nationals.

The U.S. Othello championship tournament itself was a two-day event. The first day was a preliminary round, which eliminated all but four contestants. Although there were years when some top seeds failed to advance, the final round always included at least three of the best players. Then, on the second day, a double round-robin was used to determine a champion. The winner’s achievement always was deserved.

Another effective tournament in which I participated was the academic tournament. In that tournament, the winners become professors with distinguished publication records and tenure at prestigious universities. The rest of us fall off somewhere along the way.

In academics, the process for selecting the winners is long and involved. Obtaining a Ph.D. requires hard work and demonstrated mastery of a subject. Building a publication record requires obtaining favorable reviews from peers on an anonymous basis. Gaining tenure requires support from very demanding colleagues in a highly competitive environment.

Academicians frequently grumble about the success of some of their colleagues. In that sense, they question the results of their tournament in ways that Othello players do not. However, the overall results of the academic tournament appear to me to be highly defensible. In general, it is fair to say that professors with impressive publication records and tenure at prestigious universities are more accomplished in their disciplines than those who lack such trophies.

The tournament for choosing CEO’s of large, established corporations probably is less effective than the academic tournament. It appears to me that idiosyncratic personal connections, timing, and luck play a big role in determining who gets to be a major CEO. By idiosyncratic personal connections, I mean relationships, such as country-club memberships, that do not affect the ability of the CEO to run the business profitably. Relationships with customers or suppliers are pertinent, rather than idiosyncratic.

After describing these examples, let me articulate a necessary condition for an effective tournament:

To be effective, a tournament must behave like an experiment with repeatable results.

For example, an Othello tournament is effective if the winner and other top players are the same that would result if the tournament were repeated many times. The U.S. Othello championships would meet this test.

In the academic tournament, my guess is that if we started everyone over again as twenty-one year-old graduate students, the ultimate outcome probably would be very similar to what has occurred. Perhaps a few professors would come out much better and a few would wind up much worse, but for the most part the results are repeatable.

In the traditional corporate environment, my guess is that if we started everyone over again with their first job after completing their formal education, only a small percentage of people who are CEO’s today would be able to repeat their success. Instead, a different set of lucky individuals would manage to navigate their way to the top.

However, the CEO tournament is a model of efficiency compared with the Internet IPO tournament. The latter might be termed the McKinsey Circus.

I only have come across a handful of strategic recommendations from McKinsey. Some of these came while I was working for a large company, and others I have read in articles and books. While my encounters do not constitute a large enough sample to provide statistically overwhelming evidence, there is an unmistakable pattern. No matter what the question, McKinsey always ends up giving the same answer:

"Sacrifice short-run profitability to go after market share."

This seems to be the advice regardless of whether the business is traditional or contemporary, whether it is established or start-up, or whether the supporting research and analysis has any bearing on the question of profitability vs. market share.

The IPO market has become a McKinsey Circus because of the ability of some venture capital firms to play the role of P.T. Barnum. Finding suckers who are willing to assign phenomenal market capitalizations to money-losing startups, the VC’s have turned McKinsey business planning into a cash machine.

The traditional venture capital process would seem to be a fairly reliable tournament. In that process, venture capital was needed to take a business to the point of profitability before the execution of the "exit strategy," which is either to be sold to another company or to go public. In the traditional model, companies that are unable to achieve profitability while under the venture capitalist’s wing are winnowed out.

What the P.T. Barnum method accomplishes is that it skips over the step of achieving profitability. Another way of putting this is that the last stage of venture capital funding comes from the investing public. The public is taking on the task of funding a company during its unprofitable start-up phase. While the venture capital process today enjoys an image of genius and wealth-creation, the reality is that our venture capitalists are nothing but fools selling to greater fools.

The rationale for sacrificing short-term profitability for market share is that supposedly it will result in long-term profitability. The investing public does not appear to be interested in testing that hypothesis. Instead, as long as a company can convince investors that it is following the McKinsey strategy, it will be a hit when it goes public.

In short, the companies that are favored in the Internet tournament are those that can convince the public of their ability to execute the strategy of losing money in order to gain market share. This does not strike me as a tournament with repeatable results. How likely is it that if we created Internet life in a new petri dish that investors would come to value market share in online book-selling or in outrageous patent infringement lawsuits (the latter being Priceline.com’s primary arena of market leadership)?

Moreover, one wonders whether we will be able to say that this tournament leads to repeatable instances of companies that achieve the tremendous long-term levels of profitability that are the promised rewards of following the McKinsey business plan. One wonders whether even one of these companies will reach that point. My guess is that for the Internet economy to perform the important task of separating winners from losers in the future, we are going to have to come up with a more effective tournament.