Do urban populations benefit from urban bias?

Conventional wisdom suggests that most poverty is in rural areas. But as there is so little research, how can we know if this is true?

David Satterthwaite's picture
David Satterthwaite is senior fellow in IIED's Human Settlements research group
01 September 2021
Collection
The transition to a predominantly urban world
A series of blogs and interviews designed to share the experiences of community leaders, professionals, researchers and government from the global South
Sewage and water runs down the centre of a makeshift 'drain' surrounded by sandbags and wood

Lacking a proper drainage system, residents of Kalibari community in Bangladesh improvised with wood and sandbags (Photo: SuSanA SecretariatCC BY 2.0)

Do urban dwellers benefit from an urban bias in the policies and funding allocations of government and international agencies?

This was the contention of Michael Lipton in his 1977 book 'Why Poor People Stay Poor: Urban Bias in World Development'. It spawned much discussion and many more publications on this topic, with Google Scholar showing Lipton’s book has been cited more than 4,000 times.

I was asked to review this book in 1977 and never thought to question what it says. But as I began to spend time in cities, working with local research teams on living conditions and health risks, doubts began to grow.

Just looking at the large and often growing informal settlements with their awful living conditions and lack of services made me question what biases actually benefited their residents? Many cities have between 30% and 70% of their population in such settlements, but most donor agencies have given a very low priority to addressing this, with many refusing to work in urban centres. 

This is the latest in a series of blogs and interviews, curated by IIED senior fellow David Satterthwaite, examining different aspects of global urban change. 

The myth of the urban bias?

One definition of ‘bias’ is inclination or prejudice for or against one person or group, especially in a way considered to be unfair. So in regard to development there are many potential biases to consider. 

Surely those who wrote approvingly about urban bias must have seen the very poor living conditions and absence of public services in informal settlements. Yet housing quality, service provision and health outcomes are hardly mentioned or accounted for.

A high proportion of the urban population do not benefit from piped water to their homes or sewer connections. Indeed, most urban centres in sub-Saharan Africa and many in Asia do not have sewerage systems and only a minimal piped water supply.

For those urban centres that do, it is largely middle- and upper-income groups who are served. This is even the case in those large cities where it might be assumed that ‘urban biases’ are most evident.

Urban populations may benefit more than rural populations from publicly-funded schools and (probably) hospitals and clinics, but again there are many city residents in most nations who are ill-served or not served at all. 

Access, not proximity, is what matters

Urban populations may appear to be much better served as they are located closer to services than their rural counterparts. But proximity does not mean a service can be accessed.

Many inhabitants of illegal settlements cannot get places for their children in schools or access other public services because this requires a registered address or official documentation that they do not have. They may live close to mains water supply, schools or hospitals, but have as little chance of using these services as do rural dwellers with no such facilities nearby.

Health outcomes

Even more surprising is that very little of the urban bias literature discusses rural/urban differences in health outcomes. For example, the available statistics for infant, child or maternal mortality rates usually (but not always) show nations with urban populations have some ‘urban advantage’, but the extent of this advantage in many low-income nations is surprisingly small.

Where data on infant and child mortality rates (PDF) are available for urban poor groups in particular cities, these are generally much higher than the rates for rural areas. If a more comprehensive information base was available, it seems likely that it would show a strong ‘urban advantage’ for middle- and upper-income groups in particular cities in almost all nations, but very large differences between nations in the extent of this advantage for low-income urban dwellers.

It could even reach the point where for many nations, there is little or no urban advantage for large sections of their urban population.

Mortality rates for the under-fives (U5MR) reflect the quality of pre- and post-natal services, children’s nutrition and living environment standards (especially provision for water and sanitation). The World Health Organization notes that all the leading causes of U5MR can be prevented or treated with simple, affordable interventions.

U5MR might be expected to be much lower in urban than rural areas given that urban concentrations make it cheaper and easier per person to prevent or treat the main causes. In addition, the concentration of wealthier households in urban areas improves the average rates.

In best practice nations, U5MR are below five per 1,000 live births. Statistics from the latest demographic and health surveys have data for 87 low- and middle- income nations. In 41 countries, U5MR are more than ten times this; and in six nations they are more than 20 times this rate. For 16 nations, urban U5MR are the same or higher than rural rates.

Urban penalty…

In some nations it may be that there is ‘rural advantage’ and an ‘urban penalty’ for low-income groups regarding premature death rates. It is also likely that the extent of the ‘urban advantage’ would vary greatly across the urban population of any nation – for instance between different centres and between districts within cities.

Thirty years ago, it was suggested that a more accurate term for this bias would be ‘middle- and upper-income group, large-city bias’, and this seems unlikely to have changed in most cases.

Another indicator of urban bias would be the high proportion of development assistance going to urban areas – but most aid agencies give very little or nothing to basic service provision in urban areas. 

Poverty lines applied to urban populations appear to show lower levels of poverty than rural areas. But this is in large part because they are set so low and do not take account of the higher living costs faced by low-income households in cities – as discussed in a previous blog.

...or urban advantage

Urban areas have some potential advantages that are not due to urban bias. They include more concentrated demand and agglomeration economies that cheapen the unit cost of providing good quality infrastructure and services. There is a significant urban advantage if these are acted on, as seen in many cities in Latin America.

The extent to which the urban poor benefits from living in urban areas depends heavily on the quality of local government, especially as regards service provision, and attitudes to and relationships with citizens living in informal or illegal settlements and tenement districts. 

But there is a massive urban disadvantage for these people if their settlement lacks good quality, piped water supplies; sewers; storm and surface drainage; household waste collection; and electricity. If these are not provided, government policies are, in effect, anti-poor (and help create poverty). This is also seen in the harassment of the owners of informal enterprises and the eviction of residents in informal settlements.

Now there is, tragically, the devastating impact of COVID-19 to contend with as well. The inhabitants of dense informal settlements face an urban disadvantage from the high transmission risks and impacts on their incomes. But ‘good’ local government working with community organisations is an urban advantage and can help address these challenges as shown in the Philippines, Thailand and India.
 
Data are not available for any comprehensive assessment of the extent and nature of urban bias. What is undeniable is that a high proportion of the world’s population suffering severe deprivation in terms of inadequate nutrition, ill health and premature death, together with inadequate or insecure livelihoods, still live in rural areas.

But it is also undeniable that a large, and probably growing, proportion of such people live in urban areas – and that number has increased very substantially in the last three decades.

Rural specialists rightly remain outraged at the failure of development policies to reduce rural poverty (in all its different aspects). Meanwhile, a growing body of urban specialists rightly remains outraged at the failure of development policies to reduce urban poverty (also with many different aspects), even in nations and cities with successful economies.

Perhaps they should work together more closely to identify the systemic failures of development assistance in addressing both rural and urban poverty, which could offer some hope of alleviating their effects.