"Kee" Points with Jim Kee, Ph.D.

  • Investing and “Bad Science”
  • Avoid Investing Like a Pigeon

Investing and “Bad Science"

The Wall Street Journal had an interesting article over the weekend, “Too Much Academic Science is Bad Science.” The article discussed results of an experiment (from Tulane University) on hormone disrupting chemicals in paper receipts that was published in Science magazine. Subsequent efforts to replicate the results of the experiment have failed. This “replication crisis” as it is called, is said to plague over half of the claims made in scientific papers; that is, they fail to replicate when retested, meaning the results or conclusions are not really valid. The article got my attention because the same result of non-replicability has been found in financial research. I’ve mentioned this before, but Duke University’s Campbell Harvey, President of the American Finance Association, has argued that 27%-53% of “factors” in published finance studies are false discoveries. Factors are characteristics of assets that are believed to explain the risk and return profiles of those assets. Early studies identified company size (i.e. small caps) and valuation ratios (i.e. price-to-book) as key factors that explained or described stock returns. Subsequently published work has identified hundreds of these so-called factors, which has led to the proliferation of “factor-based” mutual funds, ETFs, and variously labeled “smart funds.” Putting it all together, the academic work (i.e. Harvey mentioned above) is increasingly suggesting that many of these funds are nonsense. At STMM, we have been a critic of this endless proliferation of funds and other “products” for years. It is gratifying to see these concerns validated by the highest levels of academic research.

Avoid Investing Like a Pigeon

Staying with Campbell Harvey for a minute, one of his more recent presentations (2017), “The Search for Repeatable Performance,” discussed a number of interesting errors in reasoning. One of them came from an experiment in 1947 by psychologist B.F. Skinner. Skinner put pigeons in a cage where food was delivered at regular intervals, i.e. the feeding time had nothing to do with the behavior of the birds. But the birds associated their behavior with food delivery anyway. One bird would tilt its head back while another would turn counter-clockwise, both expecting those actions to bring them food. This describes well a lot of the behavior that I have observed for over a decade working as a portfolio consultant to analysts and portfolio managers at some of the world’s largest institutions. They would each have certain rituals or research “tricks of the trade” (so they thought) that they felt were responsible for their results. In the end, it was probably just the result of what is known as random clustering, like flipping a coin and getting 7 heads in a row. That’s why maverick stock pickers or “hot hands” rarely have the same kind of consistency that firms like ours have. They tend to favor narrow or concentrated strategies that grow out of interpreting market movements during certain periods as confirmation of their actions (like pigeons turning counter-clockwise). But a lot of times these two (i.e. their actions and the performance delivered by the market) were probably unrelated. By staying rigorously within a discipline — like always owning value and growth, and controlling sector exposure — rather than relying solely on stock selection, we reduce the chances of misinterpreting our results and what led to them. And that produces consistency.”