(Des)quilibrio general y macroeconomía

En su último post, David Glasner hace un comentario a la última aportación de Roger Farmer. No me interesa aquí el conjunto del post, sino estos párrafos en concreto, que dan una idea de la falacia de La macro actual, al hacernos creer que está fundada en el Equilibrio General:

This is an important point, so I will try to restate it within a wider context. Modern macroeconomics, of which Roger’s model is one of the more interesting examples, flatters itself by claiming to be grounded in the secure microfoundations of the Arrow-Debreu-McKenzie general equilibrium model. But the great achievement of the ADM model was to show the logical possibility of an equilibrium of the independently formulated, optimizing plans of an unlimited number of economic agents producing and trading an unlimited number of commodities over an unlimited number of time periods.
To prove the mutual consistency of such a decentralized decision-making process coordinated by a system of equilibrium prices was a remarkable intellectual achievement. Modern macroeconomics deceptively trades on the prestige of this achievement in claiming to be founded on the ADM general-equilibrium model; the claim is at best misleading, because modern macroeconomics collapses the multiplicity of goods, services, and assets into a single non-durable commodity, so that the only relevant plan the agents in the modern macromodel are called upon to make is a decision about how much to spend in the current period given a shared utility function and a shared production technology for the single output. In the process, all the hard work performed by the ADM general-equilibrium model in explaining how a system of competitive prices could achieve an equilibrium of the complex independent — but interdependent — intertemporal plans of a multitude of decision-makers is effectively discarded and disregarded.
This approach to macroeconomics is not microfounded, but its opposite. The approach relies on the assumption that all but a very small set of microeconomic issues are irrelevant to macroeconomics. Now it is legitimate for macroeconomics to disregard many microeconomic issues, but the assumption that there is continuous microeconomic coordination, apart from the handful of potential imperfections on which modern macroeconomics chooses to focus is not legitimate. In particular, to collapse the entire economy into a single output, implies that all the separate markets encompassed by an actual economy are in equilibrium and that the equilibrium is maintained over time. For that equilibrium to be maintained over time, agents must formulate correct expectations of all the individual relative prices that prevail in those markets over time. The ADM model sidestepped that expectational problem by assuming that a full set of current and forward markets exists in the initial period and that all the agents participating in the economy are present and endowed with wealth enabling them to trade in the initial period. Under those rather demanding assumptions, if an equilibrium price vector covering all current and future markets is arrived at, the optimizing agents will formulate a set of mutually consistent optimal plans conditional on that vector of equilibrium prices so that all the optimal plans can and will be carried out as time happily unfolds for as long as the agents continue in their blissful existence.
However, without a complete set of current and forward markets, achieving the full equilibrium of the ADM model requires that agents formulate consistent expectations of the future prices that will be realized only over the course of time not in the initial period. Roy Radner who extended the ADM model to accommodate the case of incomplete markets called such a sequential equilibrium, an equilibrium of plans, prices and expectations. The sequential equilibrium described by Radner has the property that expectations are rational, but the assumption of rational expectations for all future prices over a sequence of future time periods is so unbelievably outlandish as an approximation to reality — sort of like the assumption that it could be 76 degrees fahrenheit in Washington DC in February — that to build that assumption into a macroeconomic model is an absurdity of mind-boggling proportions. But that is precisely what modern macroeconomics, in both its Real Business Cycle and New Keynesian incarnations, has done.
If instead of the sequential equilibrium of plans, prices and expectations, one tries to model an economy in which the price expectations of agents can be inconsistent, while prices adjust within any period to clear markets – the method of temporary equilibrium first described by Hicks in Value and Capital – one can begin to develop a richer conception of how a macroeconomic system can be subject to the financial disturbances, and financial crises to which modern macroeconomies are occasionally, if not routinely, vulnerable. But that would require a reorientation, if not a repudiation, of the path on which macroeconomics has been resolutely marching for nigh on forty years. In his 1984 paper “Consistent Temporary Equilibrium,” published in a volume edited by J. P. Fitoussi, C. J. Bliss made a start on developing such a macroeconomic theory.

A mí me parece una magnífica descripción de la Falacia de la Composición cometida por la Macro actual, llamada también el modelo Neo Keynesiano - que como he repetido muchas veces, no tiene nada que ver con Keynes. Keynes es el primer econonomista en denunciar la economía clásica del equilibrio económico logrado espontáneamente a través de los mercados, como algo imposible cuando hay millones de decisiones de millones de agentes que no se han puesto de acuerdo sobre qué es lo que esperan de la evolución del futuro. Keynes bautizó esto como la Falacia de la Composición.

Es más, los modelos de Arrow-Debreu-McKenzie que menciona Glasner, no son una consagración del Equilibrio General, sino justo lo contrario: las condiciones tan absurdas que se tienen que dar para se dieran en la realidad dicho equilibrio.Tan absurdas como que existieran unos mercados de futuros - tantos con bienes hay - en los que en el momento cero todos pudieran tomar sus decisiones en función de los precios que reflejaran esos mercados, de manera que el equilibrio estuviera ya determinado desde el principio. No existen esos mercados, obviamente, por lo que no es posible la coordinación de todas las decisiones individuales continuamente. Por el contrario, que no existan exigiría que los agentes tomen sus decisiones en cada momento sin equivocarse, sólo con la ayuda del mercado presente.

Lo que hacen los neokeynesianos es en realidad dar por su puesto que sí existen, lo cual es como subsumir todos los productos de una economía en un solo bien y todos los millones de individuos en un solo agente, que además es racional y si se equivoca es transitoriamente, y rectifica hacia el punto de equilibrio. Esto es como decir que los millones de individuos "aprenden" de sus errores, y "rectifican" sin volverse a equivocar, por lo que al final todo es estupendo. La incertidumbre keynesiana queda arrumbada al baúl del desván, lo cual es prescindir de una de las grandes aportaciones de la historia de la Economía. Sin esta aportación, al final el desempleo se corrige, tarde o temprano, a través de la acción de los mercados, en los que los individuos van rectificando hasta llegar a La coordinación óptima. Con incertidumbre, en cambio, el desempleo puede ser permanente.

De la concepción optimista de los Neokeynesianos se infiere que el único problema es mantener una tasa de inflación óptima, que permita alcanzar ese óptimo (algo sospechosamente idéntico a Friedman). De la idea keynesiana se infiere, por el contrario, que el problema es que la coordinación de los individuos puede retardarse años, y que una acción del gobierno que la "avive" es necesaria.

Desde los años setenta vivimo sbajo la égide del modelo Neokeynesiano. Casi todos - o todos- los bancos centrales siguen un modelo neokeynesiano.

El lector se puede preguntar por qué le he dedicado un tiempo a estas quizás torpes reflexiones. Es simplemente por mi propio gusto en aclararme a mí mismo después de leer a Glasner, que siempre es un placer.