Friday, October 24, 2014

Hayek, Auto-poetic Social Order, and Critical Realism: Critical Realist Social Ontology as a Theoretical Tool Set for Economists

I have heard some say, “if you can’t measure it, it doesn’t exist.” At first look, the statement seems innocuous. If you cannot measure something, how do you know that it exists? Empirical-realism attempts to measure the world directly through the five senses. It requires some “objective” measurement. Not all causation can be measured. Some structures in society only evidence themselves through the imprint they leave behind. These imprints appear, to the careful observer, in the form of networks, norms, and other patterns of organization that cannot always receive neat, numerical measurement. The world does not fit well onto the Cartesian plane as Cartesian reasoning implies causality through correlation, not rules. A critical realist social ontology suggests causality from rules and structure. We must seek different forms of confirmation than are typically sought within modern economics if we seek evidence that our theory conforms to reality. By adopting this toolset in parallel to existing tools may also provide further improvement of equilibrium modeling and econometrics.

If we search for the subject of our study by constantly breaking down objects, both material and immaterial, into their constituent parts, where does the search stop? What is the source of the parameters that we estimate? We are measuring chance occurrences and assuming them guided by some force represented by the average. I don’t mean to pick on econometricians here. Their job of providing additional data to the study of history is indispensable! Theory cannot be limited by the domain of econometric testing. The desire of some individuals to engage in a more modern and, sadly, more mundane form of sorcery, is another issue entirely. Fitting a line or curve to past data is not the same thing as extrapolating an estimation. Attempting to predict the future by these methods assumes that underlying parameters are constant. More than likely, parameters don’t govern reality. Parameters measured ex post provide information – though the information provided is not exhaustive – about the processes of the underlying system of interest. As long as individuals using these methods do not undermine theory by arguing that their tool set is a substitute for it, this is not a problem. It is critical to the promotion of good economics that economists remain cognizant of something approaching a pure theory as a source of inspiration for modeling and to ensure that a theory misrepresents as little as possible the content it is used to present.

What we search for is not just the unseen, but the unrecorded. We are missing the information not captured by conventional models. “If you can’t measure it, it doesn’t exist!” But some entities are beyond the ability of humans to model. Our reality is remarkably consistent and can conform, in piecemeal form, to equilibrium theory. The reason for the piecemeal nature of equilibrium theory is that it is not suited to describe complexity. Twentieth and early twenty-first century empirical work has spoken wonders to what the equilibrium analysis was able to produce. However, new methods of data interpretation are available to economists, and economic theorizing must stay ahead of these changes. 

At its core, economic theory points the way to entities not yet easily, if at all, measurable. Theory posits laws that govern the objects that we observe. Modern theory gravitates toward theories whose laws govern human action through statistically observable relationships. Individuals make decisions according to the calculus of optimization. This does not leave much room for actual decision making. Equilibrium theory posits the world as a closed system and assumes that outside of that closed system, no laws exist that govern the order. Hand waving is the far less than perfect backfall to which most economists resort when this constraint becomes a problem. Thus, the dependence on exogenous shocks to initiate change within the economic system is perpetuated through a combination of equilibrium theorizing and narration. This is not a useless approach, just a limited one.

Thankfully, this problem does not hold true for the whole of economic theory. During the 1950s and early 1960s, Hayek developed a robust program into the study of complexity that he observed in society’s institutions. His development of this program also produced a challenge to modern economists. As he argued in “The Pretense of Knowledge”,

It seems to me that this failure of the economists to guide policy more successfully is closely connected with their propensity to imitate as closely as possible the procedures of the brilliantly successful physical sciences – an attempt which in our field may lead to outright error. It is an approach which has come to be described as the “scientistic” attitude – an attitude which as I defined it some thirty years ago, “is decidedly unscientific in the true sense of the word., since it involves a mechanical and uncritical application of habits of thought to fields different from those in which they have been formed [emphasis author]."

At first look this is the classic critique that correlation does not imply causation. Upon further reading, we see that Hayek presents the problem from a novel light.

While in the physical sciences it is generally assumed, probably with good reason, that any important factor which determines the observed events will itself be directly observable and measurable, in the study of such complex phenomena as the market, which depend on the actions of many individuals, all the circumstances which will determine the outcome of a process, for reasons which I shall explain later, will hardly ever be fully known or measurable.

Neoclassical economics, as it has come to be accepted among most modern economists, functions well when the world is “smooth, continuous, and twice differentiable,” to steal a phrase from Richard Wagner. To state it more clearly, causation is assumed to be consistent, continuous, and linear or log-linear. In these models, process comes to be represented – inappropriately as a means for deep theorizing – by parameters. But what use is a parameter if it is constantly fluctuating? While some have brilliantly addressed the problem by using models with fat tails and other more nuanced implements, the underlying process remains masked if we do not take off our econometrician hats. It is a mistaken method. When outcomes are generated by complex process, causation ceases to be identifiable by equilibrium analysis. Simultaneous events are subject to interdependency that prevents causation of the individual pieces of a system from being uniquely represented. Causation must be observed from a systems level. To observe features of the market system that are not currently measurable, we must investigate process in the system through the use of types.

Austrian economists would do well to be aware of, if not adopt to a greater or lesser extent, a critical realist social ontology. Fleetwood references 5 types of deep structures: structures, mechanisms, rules, powers, relations. These concepts have close relation to institutions consistently emphasized by Austrian economists. But they also include something that is not emphasized as greatly. Internal rule sets have not played a prominent role in Austrian economics in the way that it does in social ontology. The application of internal rule sets expands the scope of investigation to include not only formal and informal institutions, but the orientation of expectations through the employment of mental structures shared in common within particular networks. This provides a decent enough of a model for free individuals in the system to form coalescent plans over time.

Now we are ready to consider briefly some deep structures and their relation to Austrian analysis. The most fundamental aspect of society, I believe, are rules. These include both external rules that set bounds on allowable behavior and internal rules that guide decision-making. Informal norms arise through the interaction of rule sets within a population. Economic agents set out each period (I do not mean period in the homogeneous sense but as a period where there is planning followed by action) to accomplish some activity. It may be to buy a good or to do something as mundane as taking a nap. The ability of an agent to accomplish any activity requires that the agents internal and external rule sets allow for the special space necessary in the social system for actors to form expectation and attempt to make that expectation a reality. We see demonstrated that the other terms fall out of the analysis. Structures, mechanisms, powers, and relations all owe their existence to the interaction of rule sets. Fleetwood argues that “Owing to the existence of a set of deep structures in the form of social rules of conduct [emphasis author], a high degree of compatibility of actions and consequences is ensured, where incompatibility appears [emphasis mine] to be the more likely outcome.” Rules tie together deep structures that bring are responsible for order in society.

Norms, a type of informal rule, arise in order to tie together each individual’s internal rule set. Individual rules are allowed to interact. Rules develop that contain the information made available by a tortuous struggle of trial and error. Especially innovative rules that guide internal and external organizations tend to be adopted while failing rules lack the ability to survive absent a strong incentive for distortion. This is not to suggest that we suffer no loss from failing to adopt rule sets that may generally improve welfare. At least in a free society, if the benefits of a form of organization can be recognized by just one entrepreneur, that entrepreneur has the opportunity to also make the new organization a realization. Continuation of a bad rule set typically requires that the rule is not that costly to those practicing it or that the mode of conduct supported by a rule or rules be implemented through force. The latter case is associated with myriad complications that arise by limiting the span of human behavior in a way that prevents the information generated by that behavior from even coming into existence! That is, the more ways human behavior is restricted by force, the less adaptive the system becomes. Having expressed my point, I leave it to the reader to decide to what extent this approach is compatible with economics: Austrian and otherwise. I am certain that CR perspective will be a dominant method for the creation and discovery of the theory of auto-poetic orders.

* This post is originally for an informal reading group.
**I am currently investigating the CR literature, so my own ignorance may be obvious those more knowledgeable. If you find that I have incorrectly represented any concept, please let me know in the comments.

No comments:

Post a Comment