"Optimum City Size: Some Thoughts on Theory and Policy": Reply

Abstract

I believe the major point made in Professor Blair's critique of my article Optimum City Size: Some Thoughts on Theory and Policy, is that the treatment is limited to that of partial equilibrium analysis (with respect to a particular hypothetical city). This limitation was clearly intended in all but one regard. That is, I chose to overlook the factors of topography, culture, preference for size per se, and other similar factors' because, while they are clearly important, I wanted to focus on what I believe are the undesirable by-products of political subdivisions within a metropolitan area attempting to limit growth because they can borrow city size. This phenomenon appears to be emerging in a large number of suburbs across the nation and is once again pressing the need for a national plan for urbanization. I did not, however, consider directly the issue of a migration policy which Blair's comment has convinced me is critical. Thus he argues that if a city expands to the size at which marginal cost (MC) and marginal benefit (MP) are equated, this may not be a national optimum if individuals elsewhere in the system could migrate and add less to the net disadvantage than they did in the city from which they came. This is, of course, a real possibility. In a nation experiencing population growth and where economic expansion requires a constant rearrangement of the population, some second-best optimum might be a necessity for long periods of time. Hopefully a national plan for urbanization would have a new town policy which might respond to these pressures. In any case, a national or local strategy for urban migration might be critical to achieving an efficient distribution of the population. The basic question, however, is whether Blair's approach of a market for residential openings provides an adequate framework for dealing with the problems raised in my paper. I think not, although it may be possible to combine the two approaches to conceptually arrive at such an objective. To demonstrate the inadequacy of this approach, consider the following case in which all cities in the national system are assumed to be operating at the optimum, defined as Co in Figure 1. Assume further that the urban population of the nation is completely allocated and the national economy is stable. Thus there is no current need for net migration between urban centers. I think Blair would agree that given these conditions, Co would define the optimal size (from a national point of view) for the city. Further, operating at this optimum size would require that the residents of each city define this size as the one at which the