Tuesday, September 28, 2010

The Ultimate Niche Business

While on a business trip near Dallas, TX, I ran across what may be the ultimate niche business:

So if you are a Hun, and you are in the Dallas / Fort Worth metropolitan area, and you need a tailor, accept no substitutes.

Sunday, September 26, 2010

Warrent Buffett versus The Efficient Market Hypothesis

My wife Jacqueline got her MBA from the University of Chicago, where the efficient market hypothesis (EMH) was religion. EMH says that in the long run, no one can beat the stock market because it is too efficient. An investor might beat the market for a year, or maybe several years, but that’s just luck.

This was the standard view at Chicago during Jacqueline’s time there, and it was well supported by data: The long-term performance of professional fund managers against market indexes was notoriously poor.

But there was always the standard challenge, “What about Warren Buffett?”

That challenge had its own standard reply: If you have enough investors, somebody like Buffett is inevitable—just like if you have enough people flipping coins, you’ll get someone who tosses twenty heads in a row.

It made sense at the time. However, this year Jacqueline was surprised to read about a debate held in 1984—well before her time at Chicago—that pitted an EMH advocate, University of Rochester Professor Michael Jensen, versus Buffett. The episode is recounted in Sebatian Mallaby’s More Money Than God.

Jensen presented the standard argument that Buffett was a random fluke. In response, Buffett continued Jensen’s line of thinking. He said that if 225 million people (then roughly the population of the United States) were in a coin-flipping contest to get the most heads in a row, 215 people should achieve 20 heads in a row by chance. Buffett then said that if those 215 people were randomly distributed, he would agree that chance was merely at work. But if a significant number of those people all had something rare in common—say, a specific coin-flipping technique—that would be a different situation. As Mallaby summarizes, “If you found that a rare cancer was common in a particular village, you would not put that down to chance. You would analyze the water.”

Buffett went on to argue that the few investors who beat the market over the long term were not randomly distributed. As evidence, he cited nine fund mangers, including himself, whose value-investing strategies descended from those of investor Ben Graham. As Mallaby writes:

Buffett insisted that he had not cherry-picked his examples; he was reporting the results of all Graham-Newman alumni for whom there were records and all the fund managers whom he had won over to the value-investing method. Without any exceptions, and without copying one another’s stock choices, each of Ben Graham’s heirs had beaten the market. Could this be simple fortune?

Buffett later converted his debate comments into an article, which has its own Wikipedia page. The page is perhaps most notable for its “Rebuttals” section, which suggests that academia has ignored, rather than rebutted, Buffett’s argument: “A 2004 search of 23,000 papers on economics revealed only 20 references to any publication by Buffett.”

Of course, EMH is about a model. A model can be wrong in some respects but right (or at least useful) in many other ways. Buffett’s challenge may have exposed some wrongness in EMH, as have more recent critiques from behavioral and complexity economics. Moreover, the financial crisis that started in 2007 further undermined EMH as a safe assumption. But a widely accepted, “less wrong” alternative to EMH has not emerged.

What has emerged is a questioning of standard views. Such questioning always existed—even at Chicago—in rarefied theory classes. But I suspect today’s equivalents of Jacqueline’s MBA classes have a little less of Jensen’s perspective and a little more of Buffett’s skepticism.

Thursday, September 23, 2010

America the Start-Up

I was reading William Goetzmann’s Beyond the Revolution: A History of American Thought from Paine to Pragmatism, when I came across this:

Thanks to [the American revolutionary thinkers’] efforts, certain Enlightenment qualities or habits of mind became traditional American values. Among these were a reverence for principles, particularly individual liberty, a dedication to reason and the rational solution, a belief in order and at the same time constant change, a talent for practicality and down-to-earth political organization, a faith in learning, a sense of world responsibility and mission, and perhaps most important of all, an extreme and sensitive receptivity to new ideas, and a confidence in intellect.

Don’t those values seem similar to the best traits of great start-up companies?

Sunday, September 12, 2010

Grading Teachers Based on Their Students’ Test Scores

In a tour de force of investigative journalism, The Los Angeles Times analyzed thousands of teachers’ performance over seven years, based on their students’ progress on standardized tests. Here are some key findings, quoted from the excellent lead article of a series:

  • Highly effective teachers routinely propel students from below grade level to advanced in a single year. There is a substantial gap at year’s end between students whose teachers were in the top 10% in effectiveness and the bottom 10%. The fortunate students ranked 17 percentile points higher in English and 25 points higher in math.
  • Some students landed in the classrooms of the poorest-performing instructors year after year — a potentially devastating setback that the district could have avoided. Over the period analyzed, more than 8,000 students got such a math or English teacher at least twice in a row.
  • Contrary to popular belief, the best teachers were not concentrated in schools in the most affluent neighborhoods, nor were the weakest instructors bunched in poor areas. Rather, these teachers were scattered throughout the district. The quality of instruction typically varied far more within a school than between schools.
  • Although many parents fixate on picking the right school for their child, it matters far more which teacher the child gets. Teachers had three times as much influence on students’ academic development as the school they attend. Yet parents have no access to objective information about individual instructors, and they often have little say in which teacher their child gets.
  • Many of the factors commonly assumed to be important to teachers’ effectiveness were not. Although teachers are paid more for experience, education and training, none of this had much bearing on whether they improved their students’ performance.

The analytical technique used is called a value-added analysis, explained in the article this way:

In essence, a student’s past performance on tests is used to project his or her future results. The difference between the prediction and the student’s actual performance after a year is the “value” that the teacher added or subtracted.

For example, if a third-grade student ranked in the 60th percentile among all district third-graders, he would be expected to rank similarly in fourth grade. If he fell to the 40th percentile, it would suggest that his teacher had not been very effective, at least for him. If he sprang into the 80th percentile, his teacher would appear to have been highly effective.

Any single student’s performance in a given year could be due to other factors — a child’s attention could suffer during a divorce, for example. But when the performance of dozens of a teacher’s students is averaged — often over several years — the value-added score becomes more reliable, statisticians say.

The Times quoted experts who said students’ test performance should not be the only way a teacher is evaluated, especially in high-stakes decisions like firing. Academic and government supporters of value-added analysis in education, including the Obama Administration, suggest that it comprise half a teacher’s evaluation.

Nevertheless, the president of the local teachers’ union responded, “You’re leading people in a dangerous direction, making it seem like you can judge the quality of a teacher by...a test.” The teachers’ union launched a boycott of The Times, asking the union’s members to cancel their subscriptions.

Bad response. Better would have been to respond like two low-scoring teachers did when interviewed by The Times. They said they want to use the data to help them improve. The Los Angeles Unified school district has always had the data, but never used it for teacher feedback.

That said, there are legitimate questions to be asked when teachers are rated largely by students’ scores on standardized tests. I know too many good teachers who say it forces teaching to the test, which leads to standardized, least-common-denominator teaching—not what comes to mind when you think “great teacher.” And of course no one wins if we create a generation of ace test-takers who lack any deeper understanding or motivation.

Acknowledging those concerns, I’d still say the lesson of The Los Angeles Times investigation is that its type of teacher-targeted analysis has merit. The questions are how to maximize such an analysis’ fairness, how to integrate it into a larger evaluation of a teacher, and how to avoid undesirable side effects like teaching only to the test.

In other words, we can argue about issues like how to choose the data, how to analyze it, and what decisions it should affect. But please let’s have those arguments rather than the one about whether test data should be used at all. If you ever had doubts, The Times series should put those to rest.