In this paper we present the theoretical model underlying a series of experiments that use cellular automata (CA) simulations to explore the impact of alternative systems of pollution property rights on urban morphology and performance. It is a partial equilibrium model of developer and community behaviour which allows a formal expression of the urban development processes under alternative regulative regimes. These include pure markets; impure markets without government; voluntary agreements on externality solutions; clubs and other near-market mechanisms of supplying quasi-public goods; and rigid zone-planning. In a second paper we describe how the model is embedded in a nondeterministic CA algorithm that yields simulated land-use patterns. Because the simulations are based on behavioural theory rather than ad hoc cell-transitions rules, they also yield meaningful urban performance indicators such as total, average, and marginal private profits and social costs. These permit tests of conventional urban economic theory within an explicit spatial framework.