skip to content

Keynes Fund


Project Summary

Dr. Tiago Cavalcanti - Gray Zones: On the Causes and Consequences of Slums (JHOB).

Slums represent a large proportion of housing markets in developing countries. Although the definition of slums varies depending on the country, slums are always associated with some sort of deprivation such as insecure land tenure, low standards of urban services, and non-durable housing structures. Slums generally occupy marginal unauthorised land, mostly in environmentally vulnerable locations such as around garbage disposal sites, riverbanks, and slopes susceptible to landslides. Slum settlements usually have limited access to clean water, proper sewage treatment, or adequate heating, and therefore are associated with worse quality of life. Even though slums are subject to environmental hazards, constant raids and patrols, and social stigma, hundreds of millions of people worldwide migrate to cities and establish residence in slum areas. According to UN-Habitat (2016), in 2014 30% of the urban population in developing countries lived in slums. This corresponds to about 881 million people living in slums. Despite the large number of slum dwellers around the world, not much is known about the economic incentives associated with the decision to live in informal settlements, or about which public policies are most effective in altering slum formation (Brueckner and Selod, 2009).

In this project, we address this gap and study the determinants of slums by investigating the economic incentives behind squatting. We build a general equilibrium model of a city with heterogeneous agents and housing choices. The model is consistent with the main empirical evidence on slum formation and can quantitatively assess the role of each determinant of slum growth. The model is calibrated and estimated to be consistent with several statistics related to the Brazilian urbanisation process occurring from 1980 to 2000 (with a focus on the city of Sao Paulo). We use the Brazilian case to motivate the analysis and estimate the model, but the theoretical framework can be easily applied to other developing countries. We perform several counterfactual event and policy simulations to disentangle the impacts of specific factors on informal housing growth.

In our model, households are heterogeneous in their labour productivity, and they have to decide between two housing tenure modes (formal or informal housing). In making decisions on tenure modes, households face the following trade-off: if they choose to live in an informal settlement, they avoid paying property taxes and complying with building regulations. However, informal housing is insecure (due to, e.g., the exogenous risks of eviction and demolition by the government or property theft by other residents); therefore, they incur (i) a utility reduction per unit of housing space and (ii) protection costs. We model protection costs as forgone labour income: agents must spend part of their time physically present to protect their informal plots. Apart from making up for lack of police protection, the time costs of being in a slum can include the time required to recover from health problems related to poor urban infrastructure, the time cost of getting potable water, and paying fees to a community leader.

The basic economic mechanism generates two income thresholds, that separate formal from informal housing agents. The first cut-off comes from the opportunity cost of protecting the informal plot. As income grows, protection costs in the form of forgone labour income increase, and therefore households are better off living in the formal housing sector. Thus, more affluent individuals tend to live in formal housing units (e.g., Friedman, Jimenez and Mayo (1988)). The second cut-off is generated by zoning constraints that interfere with households' decisions and indicates how housing regulations may induce slum formation. Households that choose to live in formal housing units must comply with several building constraints, such as minimal lot sizes, height restrictions, and green area ratios, among other building regulations. Households unable to meet these requirements are bound to live in informal settlements. The model thus reveals two reasons why for poor households the (only) feasible option is to live in slums: to avoid binding building regulations and on costs related to formal housing.

Our results show that rural-urban migration and changes in the distribution of income explain much of the variation in slum growth from 1980 to 2000 in Sao Paulo. In particular, the results indicate that the rapid urbanisation process during this period had a considerable effect on slum growth. Our ex-ante policy evaluation analysis shows the following: (i) decreasing barriers to formalisation may largely slow down slum formation; (ii) incomplete slum upgrading interventions can have unintended impacts, as a sole improvement in slum infrastructure can motivate more people to live in such areas; and (iii) the average welfare effect is mainly driven by strong impacts on the subset of the population directly affected by each policy.

Research Output

On the Determinants of Slum Formation, Tiago Cavalcanti and Daniel Da Mata, forthcoming Economic Journal.

Abstract: This article investigates the main determinants of the expansion of slums, home of about one third of the urban population in developing countries. The focus is on how urban poverty, rural-urban migration and land use regulations impact the growth of slum dwellers. We construct a simple model of a city with heterogeneous agents and housing choice to explain the determinants of slums in urban areas. The model supports the main empirical evidence regarding slum formation and is able to quantitatively assess the role of each determinant of slum growth. The model is calibrated and estimated to be consistent with several statistics related to the Brazilian urbanization process from 1980 to 2000. We implement counterfactual experiments and show that income, migration and land use regulation explain much of the variation in slum growth in Brazil. We also perform ex-ante evaluation of the impacts of policy interventions on slums. Our simulations point out that decreasing barriers to formalization has great impact on slums’ reduction. By contrast, slum upgrading interventions may have unintended adverse impacts as upgrading may adversely affect the incentives regarding the decision of whether to live in slums.