Smith’s rent gap theory and local real estate dynamics: A multi-agent model

The aim of this paper is to investigate local housing market dynamics by applying an urban spatial model of gentrification based on Smith’s rent gap theory [Smith, N. (1979). Toward a theory of gentrification: a back to the city movement by capital, not people. APA Journal, 538–548]. Smith’s supply side approach explains the emergence of gentrifying neighbourhoods on the basis of investments spent in “large scale renewal projects” which only investors or developers looking for profits are able to carry out. They invest in degraded areas on the base of the gap between the actual rent and the potential rent after rehabilitation (rent gap). Afterwards the process is sustained by an imitative cooperative behaviour of property owners having advantage in investing in their properties since the quality of the nearby buildings, and their rents, arises. In Smith’s thesis home owners, landlords and developers are the leading actors coming into play of gentrification; they behave according to the search of profit or to the enhancement of their property rents, but their decisions to invest or not are conditioned by the quality of the neighbourhood. Space is a founding element in Smith’s approach to gentrification. In this paper, multi-agent systems and cellular automata have been envisaged as offering the most suitable modelling approach to the rent gap theory. A set of behavioural rules for each agent involved (homeowner, landlord, tenant and developer, and the passive agent “property unit”) has been formalized in the model and implemented on a NetLogo platform. A wide range of urban system simulations have been carried on with different parameter values concerning size of the neighbourhood, rent gap threshold, and amount of capital invested. The state of each cell (property unit) at each cycle is defined in terms of rent value and a maintenance level. All the simulations show in the long run an oscillatory behaviour of the system. The study identifies the critical parameter values which affect radical changes in the system evolution.