We wish to acknowledge the help of three referees and of John Pencavel on several earlier drafts, as well as the assistance of the modelers, whose work is referred to in the paper. They corrected our lack of understanding of their work and provided many other helpful comments. Excellent research and bibliographical assistance have been provided by Debbie Fretz, Radwan Shaban, and Janet Stotsky. Helpful comments have been made by Charles Ballard, Michael Boskin, Lans Bovenberg, Sylvester Damus, Harvey Galper, Glenn Harrison, Gordon Lenjosek, Jack Mutti, Serena Ng, T. N. Srinivasan, Charles Stuart, and Eric Toder. The authors also acknowledgefinancial support from the National Bureau of Economic Research, the National Science Foundation, International Business Machines, and the Social Sciences and Humanities Research Council: Ottawa, Canada.

In this paper we outline the computation of general equilibrium in a pure exchange economy via a fixed point decomposition procedure. For general equilibrium models of the required structure, a full equilibrium may be computed through the solution of a sequence of smaller scale 'sub-equilibrium' problems. The text contains a presentation of the methods involved along with a discussion of initial computational experience for some numerical examples. IN THIS PAPER we describe the computation of general equilibrium in a pure exchange economy via a fixed point decomposition procedure similar in spirit to the Dantzig-Wolfe decomposition algorithm for the solution of linear programming problems (Dantzig and Wolfe [2]). The method involves the generation of labels for vertices on a master simplex through the separate solution of subequilibrium problems whose parameters are determined by the coordinates of the vertex on the master simplex. For general equilibrium models of the required structure, it is possible to compute a full equilibrium through the solution of a sequence of smaller scale 'sub-equilibrium' problems. The analogues to the common constraints in the Dantzig-Wolfe procedure are common commodities with common prices, and the block diagonal structure on non-common constraints in Dantzig-Wolfe is replaced by an analogous block diagonal pattern of demands and endowments of agents over non-common goods. A natural application of the method is to international trade models with 'traded' and 'non-traded' goods. Traded goods are common to all countries, non-traded goods are traded only within the country involved. We report execution times for numerical examples using Merrill's [5] algorithm for solution of both full dimensional problems and the same problems by the decomposition procedure. We do not discuss the application of these procedures to economies with production, although it seems likely to us that a similar procedure can be applied if a comparable block diagonal structure characterizes the production set. The economic interpretation we offer for our procedure draws on the partition of the full list of commodities in a general equilibrium model into 'common' goods traded among all agents, and 'non-common' goods traded among a subset of agents. The assignment of non-common goods to agents is represented in a block diagonal pattern of demands and asset ownership by agent. Agents have

This paper numerically investigates the significance of rents for both the welfare costs and the optimal design of commodity taxes using a general-equilibrium model calibrated to 1986 Canadian data. In the data we use, Ricardian rents are concentrated in agriculture and utilities, with market structure rents concentrated in manufacturing. Different types of rents have different implications for the welfare cost of taxes, and hence also for appropriate tax design. Ricardian rents lower the cost of taxes; rents supported by imperfect competition (with no free entry) raise the cost of taxes; rents supported by regulation generate rent-seeking costs, and if taxed improve resource allocation. Model results show a markedly nonuniform optimal tax structure, and a substantial influence of the treatment of rents on the pattern of optimal tax rates by commodity.

A bstract-An eight-region numerical general equilibrium model of global trade is used to investigate the impacts of various geographically discriminatory trade policy arrangements (GDAs) on regional trade and welfare. Results suggest that the important factors determining gains and losses in any GDA are such issues as whether initial levels of protection are asymmetric, the relative sizes of participating regions, and the pattern of trade between participating and non-participating countries. Results also appear to confirm the implication of Wonnacott and Wonnacott (1981) that the gain from reducing a partner's tariff is typically a more important consideration in evaluating potential benefits from a customs union, than the traditional concerns of trade creation and trade diversion.

In this paper we explore the incidence of a sector-specific minimum wage in a two-sector Harris-Todaro model with intersectorally mobile capital. In addition to the output and substitution effects in Harberger's tax incidence analysis, an additional effect reflecting the endogenously generated unemployment arises in the Harris-Todaro case. We also explore the functional incidence issue numerically for Mexico by using general equilibrium computational techniques. In a number of the calculations reported, capital more than bears the income loss from unemployment caused by the sector-specific minimum wage.

Under a progressive income tax, conventional wisdom is that taxing individuals rather than households is preferred from an efficiency point of view. The reason is that secondary workers, whose labor supply elasticity is high, will be taxed at a lower marginal rate than primary workers, whose labor supply elasticity is low. Here, we argue that once household production is taken into account, things are more complicated since tax design should also not distort the input of family members' time in household production. Factor input distortions as well as Ramsey considerations thus need to enter the choice of the tax unit. We provide a numerical example of an economy for which a move from an individual to a household basis in the income tax can be efficiency improving. We then use a general equilibrium model, parameterized using Australian tax rates and data, whose results clearly show that welfare gains can occur under changes from an individual to a household basis for an existing income tax. Our results thus challenge conventional wisdom and suggest that household unit taxation deserves more sympathetic consideration than is currently the case.