We present a general method for computing the set of supergame equilibria in infinitely repeated games with perfect monitoring and public randomization. We present a three-stage algorithm that constructs a convex set containing the set of equilibrium values, constructs another convex set contained in the set of equilibrium values, and produces strategies that support them. We explore the properties of this algorithm by applying it to familiar games.

While taxes may be certain, U.S. tax policy has certainly not been. Furthermore, intrinsic economic risk makes investment decisions risky. Therefore, a serious examination of the effects of tax policy on dynamic economic behavior should consider both sources of uncertainty. This paper presents a simple theoretical and computational model that can analyze both intrinsic risk and uncertain taxation. Furthermore, it will be clear that these techniques will be useful for examining general problems of taxation and risk. When studying the impact of past and/or proposed tax changes, one of two extreme assumptions are usually made: either agents are perfectly aware of future tax policy, a perfect foresight assumption, or they always believe that no change will ever occur, a myopic foresight assumption. These two assumptions yield substantially different views of recent tax experience, as Alan Auerbach and James Hines (1987) demonstrate in a partial-equilibrium context. Both are clearly wrong. The myopic specification assumes that individuals believe at each point in time that the current tax law will surely continue forever, even after they have been hit repeatedly with tax changes. On the other hand, it is absurd to think that in, say, 1977, a significant number of individuals perfectly knew the various tax changes that would occur during the following decade. This paper analyzes a dynamic general equilibrium model wherein taxpayers understand the uncertainty in tax policy when making their deci-

This paper examines the impact of social security on national saving and individual welfare in the presence of realistic capital-market imperfections: market failure in the private provision of annuities and restrictions on borrowing against anticipated future wages. The introduction of social security increases lifetime welfare and reduces national saving if borrowing restrictions are absent. However, the increase in individual welfare is reduced, and in some cases eliminated, when borrowing constraints are taken into consideration. The substantial difference suggests the importance of reexamining the proportional payroll tax finance of social security.

We analyze complex bond portfolios within the framework of a dynamic general equilibrium asset-pricing model. Equilibrium bond portfolios are nonsensical and imply a trading volume that vastly exceeds observed trading volume on financial markets. Instead, portfolios that combine bond ladders with a market portfolio of equity assets are nearly optimal investment strategies. The welfare loss of these simple investment strategies, when compared to the equilibrium portfolio, converges to zero as the length of the bond ladder increases. This article, therefore, provides a rationale for naming bond ladders as a popular bond investment strategy.

It is found that the welfare gain per unit of revenue raised is maximized for an export tariff on technology transfer, followed by an import tariff on goods, with an export tariff on goods the poorest policy alternative. These results are derived within a monopolistic competition model, where the production of any good requires some initial research and development (R&D), and technology transfer occurs when R&D is done in one country for production of goods in the other. An intuitive explanation is presented, based on the public-good nature of R&D and also the elasticity of demand for technologies from firms.

Trading volume of infinitely lived securities, such as equity, is generically zero in Lucas asset pricing models with heterogeneous agents. More generally, the end‐of‐period portfolio of all securities is constant over time and states in the generic economy. General equilibrium restrictions rule out trading of equity after an initial period. This result contrasts the prediction of portfolio allocation analyses that portfolio rebalancing motives produce nontrivial trade volume. Therefore, other causes of trade must be present in asset markets with large trading volume.

We propose a novel methodology for evaluating the accuracy of numerical solutions to dynamic economic models. It consists in constructing a lower bound on the size of approximation errors. A small lower bound on errors is a necessary condition for accuracy: If a lower error bound is unacceptably large, then the actual approximation errors are even larger, and hence, the approximation is inaccurate. Our lower‐bound error analysis is complementary to the conventional upper‐error (worst‐case) bound analysis, which provides a sufficient condition for accuracy. As an illustration of our methodology, we assess approximation in the first‐ and second‐order perturbation solutions for two stylized models: a neoclassical growth model and a new Keynesian model. The errors are small for the former model but unacceptably large for the latter model under some empirically relevant parameterizations.