In this paper we discuss the (Pareto) optimal provision of a public good (desirable or undesirable) in an intergenerational model of an economy, where regeneration is an endogenous decision variable of the households in the economy. We show that in addition to providing desirable public goods and/or levying the well known Pigouvian taxes on polluting industries, a government must subsidize households who are consuming a desirable public good and tax consumers who are consuming an undesirable public good (pollution). In addition, we show that there is a public rate of return, distinct from the market rate of return, that should be used by the government for the evaluation of investments in public goods. This public rate of return is smaller than the market rate of return in a growing economy and larger than the market rate of return in a declining economy. We investigate biases in both the public and market rates of return, as well as in other parameters which characterize the economy, as a result of nonoptimal governmental behavior. We discuss the question of how to aggregate biases due to different public goods. 1. INTRODUCTION THE RENEWED INTEREST in the interrelationship between household behavior and economic growth has recently been emphasized by the contributions of Nerlove [15] and of Razin and Ben Zion [17], both of which offer an intergenerational model of population growth. In this paper we follow the lead of Razin and Ben Zion but add to the economy a public good (desirable or undesirable) and a government whose function is to insure the (Pareto) optimal provision of this public good. We assume a homogeneous population with respect to tastes, abilities, and initial endowments and thus we address only efficiency considerations.

[The existence of pollution externalities calls for government intervention in developing policy measures that will improve social welfare. This paper suggests a method to predict and compare the short-run aggregate output, pollution, and labor input of a competitive industry facing various environmental policies. Assuming that the choice between alternative production technologies can only take place prior to the investment decision, the labor output ratios and the pollution output ratios are fixed in the short run but vary among plants; their distribution is the information used in the aggregation procedure. The performance of different environmental policies--taxes and standards--is examined and compared.]

I. Introduction, 625.—II. Model I—the case of positive social costs of pollution at the urban center when natural absorption exists, 627.—III. Model II—the case of pollution externality in production, 633.—IV. Concluding remarks: integration of the two models, 637.