I continue the discussion of Frédéric Lordon’s Jusqu’à quand, which contains an explanation of the practices that made the 2008 crash inevitable that has yet to be translated into English although it is more specific and complete than Margot Robbie in a bubble bath or celebrity chef Anthony Bourdain making fish stew. Here is a quotation that grapples directly with the most dangerous of the many illusions promoted by the manufacturers of mathematical models, namely that the objectivity of such models can be separated not only from the empirical observations needed to confirm and perfect them but also from need for a conceptual framework in which the question of the model’s objectivity can be raised. Lordon argues that quantitative (probabilistic) models of finance are meaningless if there is no reason to believe that finance is governed by natural laws:
Obviously the most devoted partisans of quantitative finance will argue that any imperfection is transitory and remind us how the development of science is progressive but irresistible; even if today’s models still make a few mistakes, there is no problem that won’t eventually give way to hard work and research. There are nevertheless reasons to think that this optimism will come up against a fundamental and insurmountable obstacle, rooted in the very question of knowing whether it is possible, and if so how far, to grasp financial risk through the calculus of probabilities. Transcribing risk in the language of probability is such a common practice that it is never called into question. The modelers, who consider the question trivial, are thus barely aware of the — absolutely non-trivial — theoretical choices they undertake when they make the hypothesis that the prices of financial assets are governed by this or that probabilistic law. … Of all the tools offered by the calculus of probability, the so-called “Gaussian” law is the one most commonly used … because it’s the simplest to manipulate. But Gaussian laws have the unfortunate property of considering large price variations as events of minuscule probability… even though they are frequently observed in financial reality. [Thus there is a competition to find the most realistic law…] But the more frenetic the search for the “right hypothesis” becomes, the more one loses sight of the essential point…: the belief that one will someday find the “right probability density” amounts to the belief that there are natural laws of finance. This belief can be given the name of “objective probabilism” because it presupposes that there objectively exists a certain law of probability — “we’ll find it in the end” — that governs the price of assets.
Lordon finds more credible an “alternative approach,” which he associates with the name of André Orléan, according to which
“the” probability density doesn’t fall from the sky of “objective natures” but is rather the endogenous product of the interaction of financial operatives. … It’s the radical modification of the configuration of interactions between operatives, expressed notably by the brutal variation of the degree of heterogeneity (or homogeneity) of behaviors, that produces the regime shift, and the qualitative transformation of the relevant probability density.
Students of dynamical systems will recognize the similarity to the language of René Thom’s catastrophe theory. Lordon continues:
If one is absolutely set on maintaining the notion of law to describe the phenomena of finance — and the observation can be extended to all the phenomena of economics — one must bear in mind that the laws in question are not natural and invariant but rather temporary, variable, and contingent. If one wishes, one can therefore preserve the general probabilistic framework but only after amending it substantially… where quantitative finance believes firmly in an objective probabilism, what one could even call a transcendent probabilism, there is in reality only an immanent probabilism: the laws of probability do not fall to earth from a heaven of ideas, they emerge from below, shaped by the concrete interactions of agents — another way to rediscover that God doesn’t exist.
The last few words notwithstanding, here we see Lordon on the road to Spinozism. We will not follow him there, but rather draw the conclusion that, if “There is no alternative,” the formula associated with Margaret Thatcher, it’s because the decision-makers, the Powerful Beings, as they are called in MWA, have contrived to make alternatives impossible. It’s certainly not because alternatives are mathematically unthinkable.
Comments 0