This entry is a day late (with apologies). One of the most important (Hayekian) arguments for using markets and prices is an epistemic one: the market process as a discovery mechanism. Prices aggregate and convey information to dispersed actors and promote behaviors that are otherwise not even available. (I use the plural "prices" because I am skeptical about the law of one price.) Now generally such an argument is trotted out in debates between pro-government vs pro-market types (by the pro-market types). But here I have been arguing that, this traditional framing skews our attention so that we ignore important questions. This week I want to reflect a bit more on financial markets (not as I have been doing thus far in this blog from the vantage point of modern finance theory); for all their social benefits, financial markets are complex systems that are prone to occasional disasters (i.e., crashes). I take this as an empirical fact (without making any claims about why this is so). These crashes harm lots of innocent bystanders. So it is worth reflecting on how to prevent these without damaging the core functions and benefits of markets (this may be impossible, of course). But to get to this, we need to get clear on the nature of the problem. (Promise: at the end of this blog I offer a remedy.)
After reading some of my previous blogs on uncertainty/risk/markets, Michael Krigsman (a technology guru) called my attention to a brilliant, short paper, "How Complex Systems Fail" by Richard Cook, that is really just 18 very smart bullet points (number 16 is the title of this entry). In accord with my general stance, the paper presupposes that we live in an uncertain world. (This is most clear in item 10: "After accidents, the overt failure often appears to have been inevitable and the practitioner’s actions as blunders or deliberate willful disregard of certain impending failure. But all practitioner actions are actually gambles, that is, acts that take place in the face of uncertain outcomes." [I tend to avoid the language of gambles because I want to prevent the conflation of uncertainty with with randomness, but I think Cook's point stil stands.]) Moreover, Cook's view is resolutely anti-monocausal in accounting for systemic failure (see his seventh point).
As is widely recognized, one striking feature of contemporary financial markets is that the costs of failure are not (always) borne by participants in it. In fact, gains have been privatized and (non trivial) costs have been socialized. So, these costs are not reflected in the prices that matter to participants within the system. (In terms of discovery process; they remain invisible.) Now this much is generally studied under the rubric of moral hazard. But Cook helps us see something more significant: currently it is *not* part of the job description of the market participants to "work to keep the system within the boundaries of tolerable performance." By contrast in well functioning complex systems, "These activities [to keep the system within tolerable bounds--ES] are, for the most part, part of normal operations and superficially straightforward. But because system operations are never trouble free, human practitioner adaptations to changing conditions actually create safety from moment to moment." (17 [I like that Cook makes this an art and not an exact science--it is practical know-how].) In our real world financial markets this guarding function is outsourced to regulators, compliance departments, the financial press, and risk managers etc?). Yet, these are generally working with "Post-accident remedies" (of earlier crashes) and "These end-of-the-chain measures do little to reduce the likelihood of further accidents." (point 15). Meanwhile, it takes no genius to recognize that system failure(s) are also opportunities for huge gains, so some market participants are on the look-out to work their way around the new, lagging rules that the system monitors put in place. (All of this reinforces Cook's first point that complex systems are fundamentally hazardous.)
Cook's position amounts to this: a complex system in which monitoring of the health of the system is not part of the normal functioning of the most important (and well trained) actors in the system, is fundamentally unstable. So, what reform of the financial markets must encourage is a set of practices and norms (an ethos) in which market stability is monitored within the system and that its (relative) stability becomes one of the fundamental aims of market participants. How to allign practices, incentives, and institutions to achieve this (and not create another gigantic opportunity for risk-free rent-seeking of some privileged class), is no small matter. How to do this without losing the major epistemic advantages of markets is a serious conceptual problem. I see little evidence, alas, that anybody is trying to figure it out.