John Rekenthaler, CFA, Vice President of Research for Morningstar, muses on the difficulties of implementing traditional economic models, with the help of the example of the Fed's ex-chairman, Alan Greenspan, who bypassed reliance on a single model in favour of a more fluid interest in multiple perspectives.
In college, everybody knew that Stevie was the smartest guy around. He wowed them in the honors Shakespeare seminar, beat the math majors at their own game (eventually picking up a math degree of his own, just for the fun of it), and was outright brilliant in his chosen field, economics. By the time he was a senior he had outstripped the undergraduate curriculum and was taking graduate courses under the tutelage of the previous year's Nobel Laureate.
And then, he quit. Abandoned all plans of becoming an economist, deciding instead between law school (Harvard, naturally) or an advanced degree in computer science (MIT, naturally). I asked him, why? He had been so excited about economics, since the first day as a callow, pimpled freshman.
It was because he no longer believed. He said the theories were beautiful, the math was precise, but as he advanced further in the subject he had come to the conclusion that the foundation was unsound. He no longer regarded economics as an accurate description of human behavior; rather, he regarded it as a lovely, elaborate, and fictional construction. He was thoroughly crushed, as only the young and idealistic can be. He had lost his faith.
I thought about Stevie when reading Paul Krugman's article in The New York Times Magazine, How Did Economists Get It All Wrong? Forget Krugman's politics, or even his positions within various economic debates. The point is, the article makes amply clear just how prescient Stevie was. Across the board in economics, the traditional models have been shaken by the appearance of this anomaly and that exception. These oddities exist because of this model's simplifying assumption or that model's useful postulate--neither of which turned out to be, ahem, true. To correct those assumptions or postulates means mucking up the model with complexities to the extent where it no longer can be manipulated, even with the techiques of higher mathematics, or if it can, then the output cannot readily be interpreted.
Which is why in his (mostly successful) 20 years at the Fed (we'll pass over his misunderstanding of the housing bubble), Alan Greenspan bypassed relying on a big model, and instead cobbled together a plan by picking one insight here, a different viewpoint there, a third perspective in another place, eventually assembling these items without an official instruction manual. Bric-a-brac economics.
In Greenspan's view, the task of determining the appropriate Federal funds rate could not be handed over to a unified theory. Instead, it needed several theories--signals, really, "theories" being too grand a word--then human judgment to assemble that information.
The same holds true for that specialised sub-branch of economics, investing. For the most part, the investment managers who make the real money do so on a case-by-case basis. Traditional stock-picking managers, such as Peter Lynch or Warren Buffett, operate on general principles and a whole lot of gut (Buffett in particular on his read of the opposing party's principles). Most hedge fund managers are also generally situational rather than mathematically driven--for example, Julian Robertson or George Soros. Successful commodity traders typically build systems that seek to take advantage of observed datapoints, as opposed to being based on defensible theory. So too for arbitrageurs.
The message for us, as mutual-fund investors, is analogous. I say, let's not bother building models--or ratings systems--that attempt find the perfect fund. Instead, let's use general principles, judgment, and a dash of personal opinion to form portfolios that are built on a sound basis, and which suit our own tastes. General principles being: the fund carries low costs; the sponsoring fund company has a good record of stewardship; the fund company has demonstrated expertise in the fund's particular investment arena; and the asset class is not obviously "hot" (i.e., gold today). This construction won't be pretty. It will be bric-a-brac, not an investment castle. But when the sands shift, the bric-a-brac has the better chance of remaining upright.
John Rekenthaler is Morningstar's Vice President of Research. He views his job as "helping investors to avoid common mistakes." In his 18 years at Morningstar, John has worked as a fund analyst, editorial director, and research director; in addition, he has earned an Masters in Business Administration from the University of Chicago and has become a Chartered Financial Analyst. Says John, "Investment science is useful when regarded as a tool, not an answer. The best investors are not those who master high-level math, but rather those who master their emotions and make sound decisions when others are losing their bearings."
