Urban slums: Poverty traps or stepping stones?

A- A A+

Millions of people around the world live in urban slums. Relative to the life left behind in rural areas, the net consequences of growing up in these conditions on a person’s education and skills later in life are determined by the country’s stage of development and its policies for access and quality of education. This column uses data from Brazil to explore this relationship, highlighting that while slums in Brazil may have historically been a stepping stone for the country’s human capital formation, by 2010 they had transformed into hindrances. The lessons for developing countries that are still rural as well as for developed countries that are receiving low-skill immigrants are also discussed.

The outbreak of the Industrial Revolution marked the beginning of ‘urbanisation’. This process saw vast swathes of the global population move from rural areas to towns and cities, shifting away from agriculture work and into manufacturing and the services sector.  Urbanisation and industrialisation have long been held as hallmarks of economic development (Lewis, 1954; Kuznets, 1973), and cities are also widely seen as incubators and disseminators of productive ideas and human capital accumulation opportunities (Lucas, 2004; Glaeser, 2011). This reinforces the expectation – and hope – that urbanisation can help push less developed countries towards modern economic growth, allowing them to follow the historical path of developed nations.

While largely completed in developed countries, urbanisation has accelerated in developing countries during the last few decades. For example, in South Korea, the urban population went from only 28% in 1960 to 93% in 2010, while in Brazil, it went from 36% in 1950 to 85% in 2010. These reallocations have motivated extensive research exploring the extent to which moving away from low-productive agriculture may foster economic growth, and whether settling in urban areas can facilitate the dissemination of ideas and the formation of human capital. Broadly speaking, the great shift from village to city has happened, and now researchers are interested in how this has affected people’s lives.

But much of the existing research on economic growth fails to grasp that many rural-to-urban migrants become slum dwellers. Such people are segregated from the formal cities, lack the most basic services, and endure dire living conditions. In fact, data from the United Nations indicate that in 2007, as many as 1.91 billion people lived in urban slums (one in five people in their sample of developing countries). And in countries such as Argentina, Brazil, Peru, Uganda, and Iraq, the urban slum population was higher than the entire population living in rural areas (United Nations, 2024).


What does this new research tell us about urban slums?

A recent paper, Of Cities and Slums, which I co-authored with Pedro Cavalcanti and Luciene Torres, fills these gaps. The paper studies the emergence, persistence, and consequences of urban slums along the structural transformation of countries. In our investigation, we explore how the size and composition of urban slums evolve, giving rise to disparities in housing costs, labour markets, and educational opportunities across a nation’s rural and urban regions. This analysis uses data from Brazil, spanning from 1980 to 2010.

Skills matter. People’s education and professional experience – their ‘human capital’ – plays a pivotal role in the genesis and perpetuation of urban slums. People with modest human capital levels (low skills or limited education) can navigate the exorbitant housing costs of formal cities by residing in slums, albeit at the expense of increased commuting times and other non-financial costs (such as higher crime rates). By broadening the access of lower-skilled households to urban labour markets, slums alter the allocation of a country’s human capital across different sectors, fostering growth in manufacturing and services. Further, urban slums shape the human capital of successive generations. Our findings underscore that, beyond the educational levels of their parents, the area where children grow up significantly influences their educational trajectories and eventual lifetime educational attainment. Growing up in slum conditions has a measurable effect later in life.

Specifically, we reveal that for children born to parents with lower education levels, urban slums can offer vastly improved educational prospects compared with rural areas. Conversely, rural areas present superior prospects for children of parents with higher educational levels compared with slums. So, slums function as stepping stones for some families while acting as barriers for others. This dual nature of urban slums holds profound implications for the current demographic composition across the country’s regions and the prospective distribution of human capital. Where people live, as well as how they live, has a substantial aggregate effect on a country’s economy – and urban slums are part of the story.

In our model, as well as in reality, adults decide on their household’s location based on several factors. These include labour market opportunities, living conditions, and educational prospects for their children. The sorting of households across locations emerges organically from the intricate links between schooling and skills and, subsequently, between skills and occupations. In the economic landscape, higher educational levels correlate with a higher likelihood of households attaining enhanced labour market skills, particularly in non-routine sectors (i.e., working in services). So, there is a positive feedback loop associated with higher human capital and household location.


How do urban slums affect economic outcomes?

The availability of slums as a housing option reshapes urban areas. Slums influence the educational opportunities for all children in the economy, but the implications of their presence can vary significantly across families. For some children, slums serve as stepping stones, propelling them to achieve higher educational outcomes than if they had stayed in rural areas. Conversely, for other children, slums act as impediments, hindering them from attaining higher educational levels that might have been possible if their parents had not opted for slum living. And for yet another group of children, slums serve as shields, by attracting lower-educated households, resulting in a higher average education level in formal cities and reinforcing educational prospects for the children raised there. So, to assess the potential impact of slum reduction programmes, policy-makers should consider all these effects.

What about the poorest in society? For households with limited education, living in a slum presents both an enhancement in adult labour market opportunities and an upswing in educational opportunities for children. A complete eradication of slums in 1980 would have starkly curtailed upward mobility for low-educated families, translating to a potential 13 percentage point decrease in social mobility in 1980.

In contrast, slums appear to be significant impediments for the children of highly educated parents. Eliminating slums could have increased upward mobility, notably reducing downward mobility in the education of children with parents holding 5-11 years of education – potentially by as much as 14 percentage points.

For middle-low and middle-education households, slums represent a trade-off between labour market opportunities for adults and education prospects for children, which might be superior either in rural areas or formal cities. For the highest education groups, slums primarily operate as shields, preserving the high educational outcomes for the children raised in formal cities. Depending on who you ask, urban slums have a very different effect.

For the subsequent generation, these opposing forces tend to balance each other out overall. In the counterfactual scenarios for 1980 (i.e., if slums in Brazil were removed), in our study we observe that, despite variations in the country’s regional and sectoral composition, a complete crackdown on slums would have led to a remarkably similar cross-section education distribution for Brazil in 2010 as in the baseline. Intriguingly, more accessible housing would have less beneficial effects for low-end households.

According to our model, by 2010 the aggregate net consequences of slums are detrimental for the country’s human capital if we then project forward to 2040. The shift in the education distribution between the two points in time elucidates the changing dynamics. From a high concentration of very low-educated households in 1980, where slums acted as stepping stones, the Brazilian population in 2010 transitioned towards more middle and higher levels of education, where slums became barriers. While historically slums may have contributed to the country’s human capital formation, by 2010 slums had transformed into hindrances. Depending on where a nation is on its development journey, urban slums play a very different role.


What might these findings mean for policy-makers?

The seemingly self-contradictory nature of slums offers insights into their emergence and persistence over time. In addition to factors like high agricultural productivity, housing prices, and the demand for personal services by high-income households, the sub-par quality of schools in rural areas also fuels the rise of slums. Ultimately, the lack of educational opportunities in rural areas explains the undereducated population opting to migrate to urban areas for low-skill jobs. Our model also highlights that these households are motivated initially by their children’s relatively better educational prospects in urban slums. But once these families attain middle levels of education, slums transform into barriers, impeding their potential to reach higher education and labour market skills that could further propel them toward formal cities. In short, inadequate rural schools contribute to the emergence of urban slums, while substandard schools in the slums themselves account for their persistence.

Our research delves into the repercussions of enhancing school quality in rural areas or urban slums. We find that improvements in the institutional quality of disadvantaged regions could significantly affect a nation’s human capital (i.e., boost skill and education levels). Intriguingly, the impact of potential reforms in rural areas versus urban slums depends on the country’s developmental stage. In 1980, if rural households had access to the same level of school quality as formal city schools, Brazil might have seen less than half of its population residing in slums. Conversely, if urban slums had access to the same school quality as formal city schools, slums could have been more extensive.

While both scenarios would result in a superior education distribution for adults in 2010 compared with the observed baseline, the two policies would lead to divergent outcomes in terms of slum formation, urbanisation, and structural transformation. For 2010, the responses are similar, but the impact ranking for the subsequent cohort would reverse. In 1980, when Brazil was predominantly agricultural and rural, the more significant impact would have come from improving rural schools. In contrast, in 2010, when Brazil was more urbanised with higher educational attainments, the more considerable impact would have come from improving schools in urban slums.

Having missed the opportunity to improve rural schools and avoid the explosion of slums in the 1980s, our model suggests that policy-makers in Brazil should now target improving the schools within urban slums. These insights provide simple but innovative (and possibly crucial) lessons for numerous countries in Africa, Asia, and Latin America that are in earlier stages of urbanisation compared with Brazil. For developed countries, our analysis of the Brazilian case is also useful. While there might not be urban slums in the form seen in Brazil in these countries, the observed extreme forms of exclusion that the children of low-skill immigrants could end up suffering will have consequences for generations to come, affecting the society as a whole. On a wider level, understanding where a country is in terms of its development journey is critical. Urban slums can be both a symptom and cause of low human capital: knowing which is an important pre-condition of designing an effective policy response.


Author: Alexander Monge-Naranjo