Reflections on the financial crisis, Neo-Liberalism, and mathematical economics

The financial crisis has cast renewed critical attention on the discipline of economics. In particular, the crisis is taken as evidence of the bankruptcy of Neo-Liberalism, often associated with Milton Friedman and the so-called "Chicago school of Economics," and the over-reliance of formal mathematical techniques in evaluating risk models for financial products.

In opposition to Keynesian economics and policies, Milton Friedman succeeded in becoming the face of economics in the second half of the twentieth century while advocating deregulation and free-markets. Nevertheless, Friedman insisted that his policy advocacy was founded on value-neutral science. He often likened the technical policy expertise of the economist with the medical advice of a physician. This argument got intertwined with cold war politics in 1976 when Friedman’s Nobel price celebrations got caught up in protests over how economists trained at The University of Chicago turned out to be the economic policy masterminds behind General Pinochet’s military dictatorship in Chile. The so-called ‘Chicago Boys’ self-consciously and explicitly relied on Friedman’s arguments about economic experts being like medical experts to justify their involvement in the dictatorship. (Of course, some also appealed to national duty and/or anti-Communist sentiment/arguments.) One of Friedman’s foremost critics, Orlando Letelier (a former minister under the deposed President Allende), was assassinated by Chile ’s secret police in Washington, DC, a few weeks before the Nobel ceremony. By the 1980s Chile became the laboratory for neo-Liberal policies. (In practice, this involved a lot of trial and error, but the "Washington consensus" tended to ignore this.)

The story of neo-Liberalism is rich with ironies and not only because the advocate of small government finds himself on the side of despotic dictatorship. For example, it turns out that most of the Chilean ‘Chicago Boys’ had been protégés of Friedman’s then Chicago-colleague, Arnold Harberger, not Friedman. Harberger was an early and important advocate of mathematical economics, especially welfare economics (as developed by Friedman’s rival, Paul Samuelson). By contrast, Friedman, who (together with LJ Savage) helped develop some of the mathematical tools behind a field known as portfolio theory (which forms the basis of many of the models implicated in the financial crisis), was also one of the early leading critics of econometrics and the reliance on mathematical models (such as Harberger’s) in economics. In a further twist, portfolio theory only got off the ground only after a reinterpretation of Friedman’s mathematical result by Harry Markowitz (who also won a Nobel Prize for it). One of Friedman's contributions to statistics is the Friedman rank test which one uses when you don't believe in normality. One of the technical problems with standard risk models is of course the assumption of normality. It turns out Friedman got credit for lots of ideas he opposed.

But ironies aside, while populist commentators attack financial bonuses and business-spokespeople attack regulatory failures, we ought to encourage politicians to reform the system so that we end up with structure in which private financial incentives is reasonably aligned with long-term public welfare. Meanwhile, many economists are re-examining their assumptions: many place their faith in making economics more realistic psychologically (by drawing on neuroscience and behavioral psychology). In all of this, everybody is relying on experts who are trained into thinking that when evaluating evidence values can be kept at arm’s length.

But, of course, as long as experts do not model their own role and influence on the ways models are applied in the real world and can shield themselves from potential ill effects, we are all rats in the social scientific laboratory. So, we should encourage social institutions (within education, financial regulation, bureaucracy) in which we make sure that some of the risks (in reputation and financial costs) that come from using various mathematical techniques are also felt by the folks (professors, business schools, consultants, regulators, financial analysts, etc) that promote the use of these.