The Theory of Money and Financial Institutions, Volume 2

By Martin Shubik
The MIT Press, 1999

This is a second volume in a three-volume exposition of Martin Shubik's vision of "mathematical institutional economics" -- a term he coined in 1959 to describe the theoretical underpinnings needed for the construction of an economic dynamics. The goal is to develop a process-oriented theory of money and financial institutions that reconciles micro- and macroeconomics, using as a prime tool the theory of games in strategic and extensive form. The approach involves a search for minimal financial institutions that appear as a logical, technological, and institutional necessity, as part of the "rules of the game." Money and financial institutions are assumed to be the basic elements of the network that transmits the sociopolitical imperatives to the economy.

Volume 1 deals with a one-period approach to economic exchange with money, debt, and bankruptcy. Volume 2 explores the new economic features that arise when we consider multiperiod finite and infinite horizon economies. Volume 3 will consider the specific role of financial institutions and government, and formulae the economic financial control problem linking micro- and macroeconomics.

Martin Shubik is the Seymour H. Knox Professor of Mathematical Institutional Economics at Yale University.

* * * * *

Preface

The staggering growth of computer communication and e-commerce together with the explosion in financial computation and the construction of data banks has highlighted the basic value of information in the economy. These developments have allowed us to see that the true worth of money and credit is in terms of the dynamics of information, communication, evaluation, trust, disclosure, and enforcement.

The value of fiat money, e-money, and other forms of credit is an emergent property of the dynamics of markets guided and assisted by the law, government force, and societal tradition. Volume 1 concentrated on single-period games where much of the financial structure could be examined, but no real dynamics could be illustrated. Here we consider the multiperiod properties of goods; the special role of three-period models (where the needs for a money may be fully cyclical) is considered, followed by an investigation of finite and infinite horizon models.

Volume 3 is projected to be simultaneously far more institutional, yet more abstract. The specific roles of corporations, banks, central banks, insurance, markets, clearinghouses, and other financial institutions are considered. Along with understanding the functions of these institutions is a need to understand the essence of much of economic trade as a mass agent stochastic process with one-on-one encounters occurring possibly without intermediation or, for the most part, intermediated and processed by markets and brokers, where a key abstract property of the intermediary is as an information processing buffer of varying length depending on the nature and complexity of the physical items being produced and exchanged.

In this and the previous volume we set up enough structure to be able to consider the nature of the models needed for the reconciliation of micro- and macroeconomic theorizing. Good macroeconomic advice, I suspect, will continue to depend on ad hoc models limited to clarifying and answering specific current policy questions. A better macroeconomic basis for macroeconomic theory in general cannot come from general equilibrium theory. It must go beyond. It requires a structure where money, financial institutions, and government play active roles in a world with varying time lags, incomplete markets, and stochastic elements. The stress on strategic market games is designed to provide a means to model a closed economy with money, financial institutions, and government regardless of equilibrium. The models may be solved for equilibrium behavior (if it exists), solved for local optimizing behavior, or simulated.

The key feature is that the whole state space is specified, thus a test bed is provided for all types of behavior. There is little evidence of the existence of a unique rational expectations noncooperative equilibrium with normative properties such as Pareto optimality. The macroeconomic models of optimization appear to be more relevant to small agents bent on local optimization in an evolving global environment, where the myopic pursuit of local optimization feeds into and influences the feedback from the overall sociopolitical environment.

The question of what are the right models of local behavior for corporations, consumers, financiers, government agencies, and politicians is subject to empirical answers, which at this stage in the scientific development of economics cannot be easily validated. Instead we settle for an ongoing debate among monetarists, Keynesians, Marxists, rational expectations fundamentalists, behavioral economists, and others. This debate, in and of itself, provides an evolutionary process.

The purpose here has been to provide a previously missing context: an adequate, logically consistent, and complete description of the minimally required simple rules of the game -- a mathematical institutional structure rich enough to carry economic process and precise enough to help to describe the need for and the evolution of minimal economic and financial institutions.