Taylor: Socially Relevant Policy Analysis

Taylor, Lance. (1990). Socially relevant policy analysis : structuralist computable general equilibrium models for the developing world. Cambridge, Mass.: MIT Press.

This edited volume is designed to promote the use and development of structural CGE models. The first chapter by Taylor is of interest to me. In that chapter, Taylor outlines structural CGE models and distinguishes them from neoclassical CGE (or Walrasian) models.

“The emphasis here is on ‘structuralist’ models, which account for only some of the CGEs. Their perspective is broader than that of other, more ‘neoclassical’ constructs. Lustig (1988) nicely summarizes the ideas around which they are build: ‘Structuralist thought considers that structural characteristics (the repetition is appropriate) of the economy are fundamental to its behavior. Among the structural factors are the distribution of income and wealth, tenancy relationships on the land, the type and degree of specialization in foreign trade, the density of chains of production, the degree of concentration in markets, control of the means of production by distinct types of actors (the private sector, the state, or transnational capital), the functioning of financial intermediaries, and penetration of technical advance, as well as sociopolitical factors associated with the extent of organization of the working class and other influential sectors and classes, the geographical and sectoral distribution of the population, and its level of skills” (1).

Some general guides for a structural model of the economy: economic institutions (classes, states, corporations, for example) are not price takers; the flow is typically from investment, exports and fiscal demand (all predetermined) to income distribution, import/exports and savings; the money supply adjusts relatively statically to inflation; financial intermediation is an important driver; imported intermediate goods required to support domestic production; and development is not balanced (2-3).

“How are these views about the economy built into a model? Structuralists’s [sic] practice has five key features. First, they begin their analysis by singling out economically relevant sets of people and institutions and specifying how they fit into available data on income and wealth distributions. In Adelman and Robinson’s (1989) phrase, they create an ‘extended functional distribution’ based on the institutional structure of the economy at hand. Each set of economic actors typically is related to a functional category of the income distribution…or a production sector; it ahs different behavior patterns and partial controls over the system. In practice, a group, a class, or institution may be labeled by specific income flows rather than membership in an identifiable cluster of firms or individuals, but that is a shortcut. The justification for taking it is the belief that actions both of legal persons such as the state, or corporate sector and of real households reflect their positions in the economic system as revealed by their principal income source. They interact according to behavioral rules the modeler specifies, with consequent repercussions on macroeconomic equilibrium” (3).

“Startin from such notions of economic structure and institutions or class runs counter to neoclassical practice. Economists presently in the ‘mainstream’ devote their time to ‘deriving’ macro behavioral functions from stylized optimization exercises that individual persons, firms, and even the government are supposed to undertake. If clever, the analyst can fold the separate optimizations into a Pareto first-, second-, or nth-best equilibrium that can be represented as the extreme value of a social welfare function. A harmonious, welfare-maximizing economy (perhaps with a few ‘distortions’ to be removed) represents the mainstream version…More adventurous neoclassicals may build in rent-seeking bureaucrats and other obstructions, but few go far in trying to bring class and economic power into their models through the rational agent back door” (4).

Second distinction: money flows in terms of prices and monetary terms, not real terms.

Third: “…different prices are under varying degrees of control by distinct groups in the economy…the models herein postulate that power plays a role in price formation” (5).

Fourth: substitution based on price occurs, but it is not core to the models.

Fifth: “…their behavior [the models] depends crucially upon their description of causal linkages in the macroeconomic system” (6).