Levitt’s Honesty About Economics
posted by Deven Desai
Should we, or can we, stop pretending we know what lies ahead or should we see what we can do when we know that predictability is an illusion? Steven Levitt of Freakonmics fame did a recent interview on NPR about the current economy that points to this question. The part that jumped out at me was his statement about macroeconomics at around 3:40 into the interview. Levitt compares macroeconomics to the weather in that both are complex systems that may be predicted in the near term but are “terrible” for long term predictions. Levitt is quite honest when he states that “economists are really stabbing in the dark a lot when we try and say what will happen. I think you wouldn’t want to believe any economist’s, individual economist’s, forecast about what will happen in the future but what we are better at is thinking about the past and what’s happened in the past and how we’ve got to where we are.” That idea reminds me of Nassim Taleb’s work such as Black Swans (although there is a small paradox about the prediction that predictions will fail).
John Cassidy of Portfolio.com has a recent article with some chilling thoughts about the future of the economy. The most stark point for me was the comparison to Japan:
Yet it isn’t entirely clear what [Bernanke] should do. During Japan’s “lost decade” of the 1990s, it discovered that navigating a deflationary real estate and credit bust is far from easy, no matter what you do. To be sure, the Japanese government made some initial mistakes in following an inflexible monetary policy and allowing banks to hide losses. Eventually, though, it did most of the things that outsiders such as Bernanke had recommended, like recapitalizing the banks at the taxpayers’ expense and experimenting with new monetary rules. Even then, prices kept falling, and the economy barely managed to expand.
As for the Japanese stock market, I can hardly bear to talk of it. In 1989, when I first visited Tokyo, the Nikkei was approaching 39,000. Ten years later, when I returned, it was languishing at about 14,000. Today, it is below 10,000. Before shifting what remains of your retirement fund into an S&P 500 index fund to take advantage of a coming rebound, you might want to take a look at the Nikkei chart. It is quite a sight.
Nate Oman’s post on and the comments about the bailout have some thoughts on the specific comparison, but here I am suggesting that a larger issue may be in play. As Levitt offers we may have to change the perspective that the future will be rosy because this is a cyclical matter. Levitt then posits that the question returns to whether we make goods and services for which others will pay. (see also Nate’s post on CDS and the comments for more on this idea). Levitt notes that the idea is to think of long term ways to make America more productive and not focus on the short term boosts that may become entitlements. He notes that infrastructure investment that impacts long term productivity is a good way to proceed. Now that sounds like a prediction. But I think the idea is not a prediction. It is a prescription or guideline that may embrace unpredictability. More on that idea soon.
December 21, 2008 at 2:23 pm
Posted in: Economic Analysis of Law
Print This Post
Responses (2)
Frank - December 21, 2008 at 2:43 pm
We also need to start questioning the policies of no-holds-barred globalization that led to the hollowing out of the US manufacturing base. As both David Cay Johnston and Louis Uchitelle have shown, massive lay-offs have not merely been permitted by our public policy, but encouraged by it. As economist Alan Blinder has predicted, tens of millions of jobs could be lost:
http://economistsview.typepad.com/economistsview/2007/05/alan_blinder_fr.html
And before we just blithely accept that eventuality, we might want to think about the social disorder it could unleash.
A.J. Sutter - December 21, 2008 at 11:59 pm
Greetings from Tokyo. Compared to you folks in the US, we are doing just fine, thank you. We don’t have any major industries collapsing here. The local financial institutions, other than the subs of US companies like Lehman, didn’t get heavily into sub-prime instruments. In fact, they are buying American institutions on the cheap. The nation is aghast at the notion that 30,000 jobs are expected to be lost by next March, whereas US companies axe that many people without batting an eyelash. (The statistics may be understated, but the country will probably lose fewer jobs this fiscal year than Citibank alone is cutting.) Sure, the yen is up compared to the dollar and euro, which hits exporters. But Japan’s dependence on exports is less than that of the UK, France, Australia, Russia and Germany, to say nothing of China, Korea and Singapore. (Moreover, it’s nice for me, since I get paid in yen and and pay some expenses in dollars and euro.) And there’s some feeling in the air that this crisis is an opportunity for Japan to re-orient itself toward a new way of thinking, less dependent on the US. In other words, you should be so lucky as to be here right now.
I don’t see what’s so revelatory about Levitt’s remark. It shouldn’t be surprising that economics isn’t so predictive. If you delve into the history of ideas of mainstream economics, you’ll discover that its scientific trappings don’t have any empirical foundation whatsoever. At the micro level, this has ben documented in, among other places, Philip Mirowski’s More Heat than Light and Machine Dreams, and even, though this is only a very qualified recommendation, the first few chapters of Eric Beinhofer’s The Origins of Wealth (please see my 3-star review on Amazon for an explanation of the qualifications). In finance, most of Taleb is derivative from 1960s Benoit Mandelbrot and, for that matter, from many elementary probability theory textbooks. For the notion that you can’t derive macroeconomics from the usual mainstream micro foundations, see the 1970s-vintage Sonnenschein-Mantel-Debreu theorem, and D. Saari’s extension of it from 1995. At the macro level, Hyman Minsky’s critiques date from the 1980s and 1990s. All these critiques have been around for years, unheeded. At best, Levitt is just catching up. More people should do the same.
One shouldn’t even need the history of economics to make one wonder about the validity of economists’ predictions. Even assuming that there may be certain “laws,” it’s difficult to know whether the conditions precedent for their applicability obtain in any particular situation. History doesn’t take place in controlled lab conditions. In the 1990s, the yen wasn’t the world’s reserve currency, China wasn’t such a major economic power, and America’s powers of moral and political suasion had not yet been weakened to the point we find today. All of those factors distinguish the current American predicament from Japan’s back then and from any other situation in the past, and there may be other factors we can’t know are pertinent but with hindsight.
The challenge for policy-makers is that they have to do something. Prescriptions or guidelines is all they’ve ever been able to act on. But recently those guidelines are all too often based on theories whose authority falls somewhere between that of astrology and the medical theory of humours. Acknowledging that we know less than we think we do is a good first step. (And the biggest challenge will be when new environmental disasters start popping up on the nightly news as often as financial ones have done this autumn; that will make our musings about economics seem very trivial indeed.)
Leave a Reply