The subprime crisis: a human complex system phenomenon

I’ll be speaking at UCLA on Saturday March 8th, 2008 at the
Human Complex Systems – UCLA Four Campus Complexity Conference,
UCLA Haines Hall 352. The conference starts at 9 AM, my own talk is at 2:30 PM.

If it looks like proposing an entirely new paradigm for financial studies, that’s because that’s precisely what it is. Hope you can join!

The subprime crisis: a human complex system phenomenon

Explanations of the subprime crisis typically combine partial explanations, illustrations, “speaking” analogies, etc. treading at different levels: from the “elementary particle” level where the consumer and the financial trader are acting, to that of the “field” level where entities such as “market distrust” or “credit crunch” are being invoked as observables. Understanding is assumed to derive unproblematically from such an impressionistic portrait where intuition is expected to fill the gaps of an overall explanation.

What is presented here is what aims to be instead a total explanation of the subprime crisis, connecting in an integrated whole the “particle” and “field” levels of the financial system which provides the economy with its bloodstream. The mechanics of the financial instruments involved in the process (Asset-Backed Securities; Collateralized Debt Obligations; Asset-Backed Commercial Paper and Credit-Default Swaps) is first presented: their anatomy and their physiology where the circulation of cash flows underlines their analogy with hydraulic systems regulated by control structures. Is then added to the picture, the human agents’ interaction with them, their representations of these instruments’ behavior – or lack of it – as models and their failed as well as successful attempts, based on these models, at correcting what they observe as the unintended consequences of these instruments.

Human models are shown to imply in most cases unwarranted assumptions about the feasibility of accurate forecasting. Adequate models typically favor homeostasis as they suggest ways for implementing corrective behavior or negative feedback while inadequate models typically encourage “herd behavior” or positive feedback leading to catastrophes. Positive feedback is however shown to be the leading dynamics of some core financial processes such as speculative pricing (i.e. pricing as an intrinsic dynamics severed from fundamentals); leverage (providing a multiplier to chances of gains and of losses) and derivatives (allowing to replicate the chances of gains and losses of an underlying instrument into a new “synthetic” one).

Crises within human institutions derive often from an incomplete understanding of the processes at work. The paper has therefore the pragmatic aim of providing means for countering future disasters within the financial system.