I’ve recently become intrigued by the writings of Nicholas Taleb. They resonate with me because his criticism of mathematical modeling of markets–i.e., the central activity of “econometrics” and the basis of so much Wall Street activity–has long been a theme of the Austrian School of Economics. The same skepticism of predictability that leads the Austrians to reject socialism also casts a shadow over so much of the noise-making activity of Wall Street.
Stressing instead, the unpredictability of markets, the role of idiosyncratic (and changing) individual preferences, and the heroic role of the entrepreneur and experience-driven gambles about unmet demand, the Austrians have long condemned the idea that the government, central banks, or even individual investment houses could reliably predict market movements in advance. There is a huge “X Factor,” namely, human action.
Taleb stresses that our social scientific hypotheses and data sets are likely misleading. We craft comforting narratives from past events, even though these explanations may have little predictive value and can lead to positively dangerous “false positives” going forward. We are, in essence, stumbling around, looking for answers, with various baffles and blinders.
Like Taleb’s insights into Black Swans, the nice thing about the Austrian School is that it does not promise utopia. Instead, it promises the least systemic damage because an authentic free market system–particularly one divorced from central banking and fiat money–is the least systematized. There may be longer and shorter chains of production, misallocations of capital, and supply shocks, but these problems are dispersed and rooted in the unavoidable uncertainties of market activity. Unlike the “modern” world of central banking based on fiat currency, temporarily correct but disastrous false signals cannot arise without the harmful pressures of central banking coupled with fiat currency. Your loss is my gain in a true free market system, and speculators tend to cancel one another out without the false signal of inflationary monetary policy. Wealth creation might be slower than in the present world of monteary-policy-driven economic bubbles, but so too are the crashes less frequent and less severe.
Murray Rothbard was fond of pointing to the relatively short-lived 19th Century panics in contrast to the decade-long Great Depression and the repeated multi-year-post-war recessions.
One would have thought the econometricians would have retreated after the failure of Long Term Capital Management in the late 90s, but no such luck. They returned with the same bravado and cock-sure certainty again, enabled by the same kind of monetary policy that presumes to ”create growth.” Of course, no one can manage the economy, not least the government who, unlike private actors, is concerned equally with social goals and concentrated public choice players as it is with profits and wealth creation.
I was happy to see some criticism of the economists counseling government intervention in today’s Wall Street Journal. Yet this criticism will likely be drowned out by calls for intervention. So long as Wall Street’s economists embrace the wrong kind of mathematics in an attempt to predict the unpredictable–with extreme confidence, in fact, backed up by the wrong kinds of statistical reasoning–then the theoretical framework supporting central planning and government intervention will remain.
3 Dec 2008 at 3:55 pm
So in other words a fellow who recognizes the importance of Black Swans? Bravo.
3 Dec 2008 at 6:08 pm
The problem is exacerbated because we have an interdependent, global economy but also have numerous central banks pursuing what they perceive as their own national interests. This results in all sorts of unhealthy, artificial imbalances that eventually have to collapse.
I remain convinced that the largest root cause of this recession is the trade imbalance with Asia, fueled significantly by China’s undervalued fiat currency, and the asset bubble that resulted from their reinvestment of cash surpluses back into the West.
3 Dec 2008 at 7:43 pm
Manley wrote: “I remain convinced that the largest root cause of this recession is the trade imbalance with Asia, fueled significantly by China’s undervalued fiat currency, and the asset bubble that resulted from their reinvestment of cash surpluses back into the West.”
This is the long-winded way of saying what you avoided saying: Americans consume too much.
3 Dec 2008 at 7:50 pm
Quantitative/statistical economics has been inevitable ever since Black-Scholes. The notion that value can be quantitatively (read: rationally) determined mathematically, let to atrocities like the Value at Risk model (VaR), which is standard MBA-fare now for how portfolio risk should be measured. Never mind that VaR totally failed and got us into our current mess, as well as other quantitative models like mortgage securitization default modelling, etc.
Unlike Taleb, I’m not entirely dismissive of Gaussian mathematics as a lens through which you can see the world and predict the future. I certainly have no better, rational, internally consistent, and theoretically justifiable suggestion for how people should make decisions under uncertainty.
That said, risk managers surely ought not to rely so much on quantitative models, and they ought to employ a qualitative “common sense” test. With the benefit of hindsight, the qualitative flaws in the quantitative models are quite apparent. Why wasnt this forseeable?
3 Dec 2008 at 9:49 pm
It’s like the newbie Lt. who is lost but insists: this river is not on the map!!!
Numbers have the appearance of certitude and precision, and people like this b/c it makes for nice charts on powerpoints.
I think also Americans consume too much is a bit simplisitc. We’re not collectively retarded. But so long as wealth continued to grow as it did, people continued to consume more sure that a bonus check, refi, or something else would allow them out of the hole. We’re all tightening now, we the same basic people.
3 Dec 2008 at 11:25 pm
Whenever a person borrows too much, there are two people at fault; the person who borrowed more than he should have, and the person who lent him more than he should have. Especially if the lender is lending so that the borrower will buy more of his stuff.
Speaking of numbers, it’s amazing how anytime a number is thrown into an excel sheet or onto a powerpoint it automatically assumes some sort of authority. Many people tend to forget another MBA rule, GIGO – garbage in, garbage out.
I think there is another MBA rule that was forgotten at our peril. All value is based on future cash flow. So all these quants that came up with amazingly complex mortgage backed securities didn’t seem to notice the very simple fact that home prices had no relation to buyer’s incomes or potential rental revenue.
4 Dec 2008 at 1:16 pm
In a different life, when I used to work more regularly with numbers as a management consultant, there was a rule about how many significant digits could be presented to clients. The rule basically said that any significant digits trailing the decimal point were discouraged, because it conveyed a sense of false precision. Rounded numbers and ranges were preferred to whole number answers, i.e. “losses could be $100-110mm/year” rather than “losses could be $115mm/year”
5 Dec 2008 at 7:11 pm
While the article is interesting and I agree with it, the root cause of the current down turn and the Great Depression are not part of the normal business cycle. The genesis of the current recession lay in the “Sub-Prime” mortgage. In and of themselves the sub-primes were not too great a problem, as they would have been localized. But, when Sub-Prime, read unsecured paper, loans were bundled as securities the infection became system wide. Incidently, there is aname for selling unsecured paper as securities, it is called fraud. The twenties had their bogas holding companies.
Too soon old, too late wise!
5 Dec 2008 at 9:03 pm
How do you model “human action”? Too many unknown variables. I am not a mathemetician, but it would seem more sensible to use qualitative or subjective methods as the primary means of predicting future market movements and quantititative methods as a check.