52 Urban Livelihoods

Chapter 4

Seeking an Understanding of
Poverty that Recognizes
Rural–Urban Differences and
Rural–Urban Linkages

David Satterthwaite and Cecilia Tacoli

INTRODUCTION
Recent conceptualizations of livelihoods have proposed frameworks that seek to
reflect the diversity and complexity of ways in which different groups make a
living. They also highlight how policies must build on the existing strengths of
people’s livelihood strategies in order to expand their options and choices
(Bebbington, 1999; Carney, 1998; Scoones, 1998). These frameworks have been
developed from a rural perspective, and while they are sufficiently broad to
incorporate non-natural resource-based livelihood strategies – for example,
income diversification and rural–urban linkages (Ellis, 1998; Tacoli, 1998), as
well as some variations in the nature of the vulnerability context – their usefulness
in urban contexts still has to be tested. In particular, it is the frameworks’ ability
to account for the specific characteristics of the livelihood strategies of the poorer
or more vulnerable urban groups and to recognize the non-livelihood related
aspects of deprivation in urban areas that needs to be explored. With this in mind,
this chapter discusses commonalities and differences between rural poverty and
urban poverty, and their implications for policy interventions. The underlying
argument is that, while the often neglected sectoral and spatial linkages and
interdependencies between urban centres and countryside are often critical both
for local economic development and for the livelihood strategies of poor (and
non-poor) groups, there are also crucial differences in the urban and rural
vulnerability contexts which require careful understanding and consideration.
Developing a livelihoods framework for urban areas is also complicated by
diversity in urban contexts, not only between different urban centres but also
between different locations within urban centres (especially larger ones). For
instance, in any major city there are many differences between the various
 Seeking an Understanding of Poverty 53

housing submarkets used by low-income groups in terms of the quality of hous-

ing, the quality and availability of basic infrastructure and services, the level of
insecurity (including the risk of eviction), the nature of social capital and the
accessibility of different income-generation opportunities. So, to understand
poverty and the best means to address it, we need to understand how local
contexts influence poverty and what changes in local contexts can reduce it. The
differences between rural and urban areas is one useful way to emphasize impor-
tant differences in local contexts, but the diversity among different urban areas
(and different rural areas) and the many linkages between rural and urban areas
make the dividing lines between rural and urban contexts imprecise.
The chapter first stresses the importance of agriculture for the economy of
many urban centres, and the extent to which many urban and rural dwellers rely
on access to a variety of urban and rural assets for their livelihoods. It then
highlights the differences in rural and urban contexts of relevance to poverty
reduction, while recognizing the limitations of the distinction. It then relates
differences and commonalities between rural and urban areas to livelihoods
frameworks, with a particular interest in locations that are the interface between
rural and urban areas, and considers how governments and international agen-
cies can respond more effectively to rural–urban differences and rural–urban
linkages.

THE IMPORTANCE OF AGRICULTURE FOR THE ECONOMY

OF MANY URBAN AREAS

Reviews of urban change within nations or subnational regions often highlight

how many of the fastest growing urban centres are within areas with the most
rapid increase in the value of agricultural production (Hardoy and Satterthwaite
(eds) 1986; Blitzer et al, 1988; UNCHS, 1996; Afsar, 2000). Many successful
cities that now have large industrial, commercial and service economies initially
developed on the basis of prosperous agriculture nearby. An analysis of urban
change in Bangladesh during the 1980s found that many of the most rapidly
growing urban centres were serving areas with rapidly growing rural economies
(Afsar, 2000). In the Upper Valley of Rio Negro in Argentina, economic growth
and urbanization were driven by the growing of fruit and high-value vegetables
and the many forward and backward linkages generated by the growing, harvest-
ing, collecting, storing and processing of these (Manzanal and Vapnarsky, 1986).
Important factors influencing the extent to which agriculture supports
prosperous local urban centres include the value per hectare of the crops (the
higher the value, the more local urban development); the potential for local value
added activities (and the scale of forward and backward multiplier linkages); and
the land-owning structure (the greatest stimulus to local urban development
generally being if there are a large number of prosperous, relatively small farms
growing high-value crops).
Other factors also need to be taken into account. These include the market
or institutional arrangements for supplying farmers with inputs (and capital) and
for collecting, processing and marketing their outputs. For example, small
54 Urban Livelihoods

farmers around Paraguay’s capital Asunción are unable to benefit from their
proximity to urban markets as lack of access to credit and low incomes prevents
them from investing in high-value cash crops or intensifying their production
(Zoomers and Kleinpenning, 1996). Commercial crops almost always demand
greater outlays on inputs and even on additional labour than traditional subsist-
ence crops and so are beyond the possibilities of low-income farmers. In many
countries, markets tend to be dominated by large local merchants who control
access to transport and marketplaces and, in many instances, to capital, credit
and information, thereby diminishing the incomes of cultivators and often
steering much of the value of agricultural production out of the locality (Tacoli,
1998).
There is little evidence of governments recognizing the potential for prosper-
ous agriculture to support urban development. Many booming agricultural
towns and cities have been starved of the funds needed to support their economic
expansion and to serve the needs of their rapidly growing populations. In some
instances, agricultural policies have prevented or discouraged rural producers
from diversifying production and trapped them in low-profit crops with few
forward and backward linkages, as in many Asian nations where the national
policy aimed to ensure that rice production could feed urban populations.
Comprehensive rural–urban development frameworks and regional spatial
planning in the 1970s and 1980s generally concentrated on trying to expand
industrial production in smaller urban centres and often failed to identify and
support the potential comparative advantage of each locality. However, this
failure to support prosperous agriculture that could in turn underpin urban
development is also related to political constraints – for instance, inequitable
land-owning structures, limited possibilities for farmers to move to higher value
crops and pricing, and marketing structures that keep down rural incomes. While
many of these reflect constraints at the national level, the growing internationali-
zation of trade and production is an increasingly important dimension which
affects local economies through the rise of international agro-industry and the
resulting marginalization of small farmers (Bryceson et al, 2000).

HOUSEHOLD LIVELIHOOD STRATEGIES ALONG THE

RURAL–URBAN CONTINUUM
Many rural and urban residents rely on a combination of both rural- and urban-
based assets or income sources, and access to these is often essential for the
survival strategies of poorer households as well as for the accumulation strategies
of better-off groups. For example, urban demand (and markets) can be critical
for rural producers, while at the same time many urban enterprises rely on
rural consumers. Small and intermediate urban centres are often linked to the
surrounding rural settlements by complex two-way interactions which include
trade, employment and the provision of services such as hospitals and secondary
education (Kamete, 1998). People may move between rural and urban areas
following employment opportunities, often on a temporary basis – for example,
in Colombia coffee farms provide seasonal work for low-income urban residents
 Seeking an Understanding of Poverty 55

who combine it with informal sector activities in the cities during the rest of
the year (Hataya, 1992). It is now widely acknowledged that access to non-
agricultural employment is increasingly important for rural populations and that
in many cases diversification of income sources is an effective survival strategy
for vulnerable groups with limited access to assets (Ellis, 1998). Some studies
have shown that it is generally farmers with very small holdings who have the
greatest reliance on off-farm income.
For rural populations, migration is an important way to increase or diversify
income and/or to ensure access to assets. In many cases, movement is temporary
and seasonal and complements farm employment. In other instances, one or
several members of the household migrate for longer periods of time but maintain
strong links with relatives in their home areas. These two-way linkages may
include sending remittances from urban to rural areas, but also sending food
from rural to urban areas. In addition, investing in property such as housing, land
or cattle in the home area is often an important element of a migrant’s livelihood
strategy, and relatives and kin are those most likely to take care of these assets in
the migrant’s absence (Afsar, 1999; Krüger, 1998; Smit, 1998). Rural-based
relatives may also perform the crucial role of bringing up the children of migrants
for whom workloads and living conditions in urban centres can make child care
problematic, while urban-based relatives often provide critical support to new
migrants. However, linkages between migrants and non-migrants are not always
strong, especially where migrants have limited or no access to rural assets such
as natural capital, especially land (because of their gender, income, ethnicity, or
religious and/or political affiliation), and as a result have little reason to maintain
links or invest in their home areas. Nevertheless, there is ample evidence to show
that in many circumstances multispatial households are able to secure access to
a range of assets encompassing both rural and urban locations, which in turn can
provide safety-nets or opportunities for cross-sectoral investment.
Strong rural–urban links at household level mean that increased poverty in
rural areas often impacts negatively on urban areas and vice versa. It is assumed
that falling crop prices or declining rural production mean a sharp rise only in
rural poverty, but these also mean a falling demand for the goods and services
provided by many urban enterprises to rural enterprises or households. An
increase in urban poverty also implies that there are fewer job opportunities in
urban areas for rural dwellers, reduced remittance flows from urban to rural
areas, less urban demand for rural products and possibly more urban to rural
migration, which could increase dependency burdens in rural areas.
Policies which affect the viability and effectiveness of livelihoods that straddle
the rural–urban divide can be divided into at least two broad categories. On the
one hand, national and local level policies tend to neglect the importance of
migrants’ remittances and investment in their home areas. For example, non-
residents may not benefit from services, housing loans and relief measures for loss
of property in rural areas, even if they consider these as home and their invest-
ment benefits the whole settlement. On the other hand, macrolevel economic
reform often does not take into account the fact that policies rarely affect only
one sector of the economy. Moreover, global liberalization of trade and produc-
tion is at the root of significant changes in patterns of agricultural production,
industrialization and internal and international labour migration. These changes
56 Urban Livelihoods

bring new sets of constraints but also potential opportunities and are reflected in
the increasing complexity of livelihood strategies which are tending in many
places to include a wider range of spatially separated assets and a growing
diversity in the form, direction and composition of population movements. The
key issue for governance is to ensure that the asset bases of both urban and rural
dwellers are protected, and that they are able to influence the setting of policies
and the allocation of public resources.

RECOGNIZING THE DIFFERENCES BETWEEN RURAL AND

URBAN CONTEXTS
While appreciating the multiple connections between many rural and urban
areas, we also need to recognize the differences between them. One of the most
important, in regard to poverty, is the difference in the level of income needed to
avoid poverty. The methods used by most governments and international agen-
cies to define the income level needed to avoid poverty do not recognize just how
expensive essential non-food items are for many poorer groups in many cities.
Many works on poverty assume that the income level needed to avoid
poverty is the same in rural and urban areas. Many governments explicitly or
implicitly assume this to be the case as they set a single income-based or con-
sumption poverty line to cover both rural and urban households. The World
Development Report 2000/2001 also makes this assumption, as it estimates the
scale of poverty globally based on a US$1 a day poverty line (World Bank, 2000).
National or international income poverty lines are generally based on estimates
of the cost of an ‘adequate’ diet with some minor additional amount added for
non-food expenditures (for instance, a 15–30 per cent upward adjustment from
the cost of a food basket based on what is considered to constitute an adequate
diet).
However, studies of the expenditures of low-income urban households show
that many face particularly high costs for non-food items, typically on water
from vendors, sanitation from pay-as-you-use facilities, healthcare and medicines
(especially where there are no government or non-profit services), housing rent
or the cost of land and self-build, schools (especially where government provision
is poor) and public transport (especially where peripheral/distant locations are
chosen because the land is cheaper and/or households have more chance of
developing their own homes without fear of eviction). Of course, considerable
care is needed in drawing on this evidence because it usually shows not the
income needed to avoid poverty but the high cost of inadequate provision. A low-
income family that is paying 20 per cent of its income to rent a tiny room with
no piped water supply and no sanitation facility and another 10 per cent on water
purchased from a vendor (but at prices that are too high to allow the purchase
of sufficient to meet its household needs), is not avoiding deprivation by spending
30 per cent of its income on these necessities. It might need to spend the equiva-
lent of 60 per cent or more of its income to get adequate quality accommodation
with adequate provision for water and sanitation. The link between the extent of
deprivation faced by low-income households and the quality of government is
 Seeking an Understanding of Poverty 57

obvious, since with efficient infrastructure and service provision the income
needed to avoid poverty is much reduced. Where there is competent, effective
government, poorer urban groups will benefit from better infrastructure and
services because of the economies of scale and proximity that urban areas
provide. But where urban governments are ineffective and unrepresentative,
urban living conditions for poorer groups may be as bad or worse than rural
conditions.
One of the great unknowns is how much the level of income needed by a
household to avoid poverty varies from place to place. Many (or most?) urban
households need a higher cash income than many (or most?) rural households
for:

l Public transport for getting to and from work and essential services; various
studies of urban poor communities show public transport costs representing
a significant part of total household expenditure, especially for poorer groups
living on city peripheries because only here could they find land sites on which
to build housing.
l Schools, where school fees and associated costs, including getting to and from
school, are higher than in rural areas. Even if schools are free, there may be
other costs, such as the cost of uniforms or examination fees, which make it
expensive for poor urban households to keep their children at school (see, for
example, Kanji, 1995).
l Housing for rent or, if living in a self-built house, because access to a land site
for the house and building materials is more expensive. Many tenant house-
holds spend more than a third of their income on rent. Households who rent
rooms or who live in illegal settlements may also be paying particularly high
prices for water and other services.
l Access to water, and in some instances to sanitation and rubbish collection.
For many urban households, the payments made to water vendors represent
a major item of household expenditure – often 10 per cent and sometimes 20
per cent of household income – with particular case studies showing even
higher proportions (see, for instance, Cairncross, 1990). Many urban house-
holds also have to pay for rubbish collection and for access to latrines. There
is a growing literature showing the extent to which large sections of the
population in many cities have no sanitation facility at all in their home, and
public or communal provision is so poor or so expensive that they resort to
defecation outside, or what is termed in the Philippines as ‘wrap and throw’
(this literature is summarized in Hardoy et al, 2001).
l Food, as food is more expensive, especially for urban households who have
no possibility of growing any food and/or raising livestock.
l Healthcare, if this is more expensive in urban areas or no public or NGO
provision is available and private services have to be purchased. A study in a
‘slum’ area in Khulna, Bangladesh, highlighted the very large economic
burden caused by poor health associated with poor quality housing, and how
the economic cost in terms of income lost from days off work and from
medical expenses was greater than the cost of improving the infrastructure to
eliminate the health problems (Pryer, 1993).
58 Urban Livelihoods

l Child care, where all adult household members have to find income-earning
opportunities and child care is needed but there are no low-cost or no-cost
solutions, although often this difficulty is solved through reciprocity at
community level or leaving older siblings in charge.
l Payments to community-based organizations (CBOs), or for bribes to police,
or fines when arrested for illegal street vending.

One reason for the underestimation of the income needed for non-food items
when setting a poverty line may be the inappropriate transfers of experience from
high-income countries. Methodologies for setting income-based poverty lines
which link the income to the cost of food were often developed in countries where
healthcare and education were free and available to all, where virtually all
housing had provision for water, sanitation and drainage, and where there were
separate social programmes to allow people below the poverty line access to
shelter. In such situations, it is more valid to link poverty lines with the cost of
food, since access to adequate quality housing, healthcare and education are
either free or guaranteed through other measures.
There may also be many rural contexts where households face particularly
high costs. Income-based poverty lines may also need to be adjusted regionally
if they are to reflect the income level that is needed to avoid poverty. Much rural
deprivation may also be linked more to the unavailability of services than to the
lack of income. There are also many aspects of deprivation in both rural and
urban areas that are not linked to income levels, including limited or no right to
make demands within the political system or to get a fair response, and discrimin-
ation in (among other things) labour markets and access to services and justice.
Higher incomes do not necessarily guarantee access to basic services, including
good quality education, healthcare, emergency services and protection from
crime and violence.
There are also important differences between most rural and most urban
areas in:

l The mix of assets that best serves poor households in reducing their vulnera-
bility to shocks and stresses – for instance, the economic role of housing as a
production base is important for many low-income urban households (see, for
instance, Kellett and Tipple, 2000). Housing also provides a location within
reach of income-earning opportunities or an income-generating asset in the
form of rooms that can be rented out.
l The constraints on low-income households’ ability to acquire the kinds of
assets that reduce their vulnerability to economic (or health) stresses and
shocks.
l The environmental health risks that low-income households face. Large
populations, highly concentrated in urban areas with a lack of provision for
water, sanitation, drainage and with high risks of accidental fires, produce
some of the world’s most life-threatening settlements (Cairncross et al, 1990;
WHO, 1992).
l The factors that explain the exclusion of low-income households from the
infrastructure and services that are essential for health and development. For
many rural dwellers, the problem is their physical distance from schools,
 Seeking an Understanding of Poverty 59

health centres, emergency services, courts, banks, politicians and institutions

to enforce the rule of law, and the inconvenience and high cost of transport.
For large sections of the urban population, it is not distance but exclusion for
economic, social or political reasons. A squatter household living 200 metres
from a hospital, secondary school or bank, or 50 metres from a water mains
or sewer can be as effectively excluded from these as a rural dweller living 20
miles away from such a facility (although the squatter household may have
the advantage of being able to tap illegally into a water main). Proximity does
not imply access.
l The extent and nature of the influence of ‘good’ or ‘bad’ governance on
poverty. Amis (1999) suggests that one of the most important aspects of a pro-
poor urban policy is to stop urban governments inhibiting or destroying the
livelihoods and homes of poorer groups. One characteristic of urban areas is
the multiplicity of laws, norms, rules and regulations on land use, enterprises,
buildings and products. Many of these deem illegal most of the ways in which
urban poor groups find and build their homes and develop their livelihoods
(Hardoy and Satterthwaite, 1989). As a result, bureaucratic rules and regula-
tions, and formal and informal institutional structures generally have more
influence in urban areas on access to employment, land and basic services.
They also have a larger potential impact on poorer groups if misapplied, for
example, large-scale evictions, harassment of hawkers, access to resources
through exploitative patron–client relationships, and contravention of civil
and political rights or corruption.

These and many other key differences between rural and urban contexts are not
recognized in much of the literature on poverty. For instance, they are hardly
discussed at all in the World Development Report 2000/2001 which focuses on
poverty (World Bank, 2000).

THE IMPRECISION IN RURAL–URBAN DISTINCTIONS

Having stressed how much rural and urban contexts differ, and how much this
affects the nature of poverty and the best means of addressing it, we have to
recognize how fuzzy the lines are between rural and urban, not only because of
the rural–urban flows and interdependencies noted earlier but also because of
imprecise distinctions. Three examples of this are: the imprecision in urban
definitions; the number of urban dwellers who work in agriculture; and the
similarities in poorer groups’ exposure to environmental hazards in rural and
urban settings.
The criteria used by governments to define urban populations are not very
successful at separating settlements with urban characteristics from those with
non-urban characteristics. If the two most important urban characteristics are
taken to be a concentration of non-agricultural production (and, by implication,
an economically active population that makes a living from secondary or tertiary
activities) and a sufficiently large, high-density concentration of people to
warrant different kinds of infrastructure and service provision than more scat-
60 Urban Livelihoods

tered and lower density rural settlement patterns, then urban definitions generally
do not succeed in separating these. Most discussions of rural–urban differences
seem oblivious of the lack of precision in such definitions. Virtually all ignore the
fact that the large differences in the ways that governments define urban areas
compromise the validity of international comparisons of rural–urban gaps. For
instance, it is not comparing like with like if we compare the level of urbanization
(the percentage of population in urban centres) of a nation which defines urban
centres as all settlements with 20,000 or more inhabitants, with another that
defines urban centres as all settlements with more than 1000 inhabitants.
The importance of agriculture for the livelihoods of urban dwellers is difficult
to gauge because of the lack of data. Most data on urban occupational structures
come from censuses and it is often difficult to get census data on particular cities
or these are only published many years after the census was taken. Statistics on
urban occupational structures are also unlikely to include most of those who
engage in urban agriculture, especially those working part time or outside their
work hours, or not registered as working in agriculture. Employment surveys or
census forms often require that each person is registered as working in one
occupation (Smit et al, 1996).
Although there are significant differences between rural and urban areas in
the environmental health context, there are also important commonalities,
perhaps most especially in the environmental hazards faced by many low-income
groups in urban and rural areas. For instance:

l No water supply fully protected from contamination (especially from faecal

matter).
l Difficult access to sufficient water (although for many rural dwellers, this is
because of distance, while for many urban dwellers it is more likely to be
because of long queues at standpipes and irregular supplies).
l Inadequate provision for latrines.
l High levels of overcrowding within homes made of flammable materials and
with high risks of accidental fires linked to the use of open fires, kerosene
stoves or lights, or candles.
l High levels of indoor air pollution linked to the use of smoky fuels and open
fires, or smoky stoves.

DIFFERENCES AND COMMONALITIES IN LIVELIHOODS

FRAMEWORKS
The differences between (most) rural and (most) urban contexts and the extent
to which local contexts influence the scale and nature of poverty (and the most
effective means of reducing it) caution against any attempt to develop a single
‘sustainable livelihoods framework’ of relevance to both rural and urban areas.
It is worth remembering that the sustainable livelihoods framework developed by
Chambers and Conway, on which much of the new thinking about poverty
draws, was developed in large part to emphasize the diversity of local contexts
(see Chambers and Conway, 1992). Chambers’ and Conway’s work explicitly
 Seeking an Understanding of Poverty 61

focused on rural areas to challenge inaccurate assumptions about rural poverty.

It was developed primarily to widen an understanding of the needs of those who
obtain their living from agriculture and livestock, and its stress on sustainability
was in part because of the dependence of most poor rural dwellers on natural
capital, so soils, forests and freshwater systems need to be exploited in a sustain-
able manner (and poor dwellers’ access to them also sustained). This cautions
against assuming that the framework developed for rural livelihoods can be
transferred to urban contexts where a much smaller proportion of poor groups
depend for their livelihoods on access to natural capital. In regard to much urban
poverty, there are also limits on what external agencies can do to improve
livelihoods for the poor, as agencies have limited powers to change the structural
constraints on people’s ability to develop better remunerated livelihoods. How-
ever, in contrast to most rural areas, there are more possibilities of reducing other
aspects of deprivation (including access to justice, law and order, and basic
affordable services). A concentration on ‘livelihoods’, moreover, can miss other
key aspects of deprivation, including those that are particularly relevant for many
urban households.
Another reason for caution is the extent to which attempts to generalize
about sustainable livelihoods frameworks (so they have validity for both rural
and urban areas) fail to draw on urban analysis. For instance, there is a rich and
varied urban literature stretching back several decades which highlights such
sustainable livelihoods issues as: the diversity of livelihood sources for poorer
groups (including the importance of self-production for many); the many kinds
of shocks or stresses from which they are at risk; the importance of assets for
reducing vulnerability (and the role of social relations in this); and the importance
of better governance (and of more accountable and democratic government) for
giving poor groups more protection (for instance, of their civil and political
rights) and more influence on policy and resource allocations. A recent paper that
compared the sustainable livelihoods approaches of different agencies claimed
that one important characteristic of this framework was that it helped to bridge
the gap between macropolicies and microrealities, and to guard against anti-
poverty endeavours being conceived and implemented from the national level
(Carney et al, 2000). It also suggested that little attention had been given to the
manner in which (or where) people live and the resources (assets) used for
pursuing livelihoods. While this commentary may be valid for work in rural
areas, it has less validity for urban areas, as can be seen in the large and well-
established literature on both these themes.
However, the sustainable livelihoods framework has certain key concepts
that are valuable for a better understanding of urban poverty (and how to reduce
it) if considered within an understanding of each urban area’s local context. The
stress on more participatory models of engagement between ‘the poor’ and
external agencies and the stress on supporting capabilities (and improving health
and education) obviously has universal validity. The importance for most poor
households of a stronger asset base both for higher incomes and for reducing
vulnerability to shocks and stresses is valid for both rural and urban areas,
although the assets that are most useful for poorer groups will vary from context
to context. Similarly, the need for an understanding of the causes of vulnerability
and the measures to reduce it has universal validity, although the measures that
62 Urban Livelihoods

most effectively reduce vulnerability will generally differ (both between rural and
urban areas and within different urban areas and different rural areas). Good
governance will be important for rural and urban areas, although differences
exist in the ways in which policies, institutions and governance influence poverty.
As noted earlier, many urban poor groups are particularly vulnerable to ‘bad’
governance.

COMBINING AN UNDERSTANDING OF RURAL–URBAN

CONNECTIONS AND DISTINCTIONS
Thus, we need to evolve an understanding of poverty that:

l encompasses both rural and urban populations and the interconnections

between them;
l acknowledges that where people live and work and other aspects of their local
context influence the scale and nature of deprivation; and
l recognizes that there are typical urban and rural characteristics that cause or
influence poverty, although care is needed in making generalizations because
of great diversity between different urban and rural locations.

Table 4.1 illustrates this. It emphasizes some of the most ‘rural’ characteristics in
the column on the left and some of the most ‘urban’ characteristics in the column
on the right. But these should be regarded as two ends of a continuum, with most
urban and rural areas falling somewhere between these extremes. Earlier sections
highlighted the importance of non-farm income sources for many rural house-
holds (including remittances from family members working in urban areas) and
the importance of agriculture and/or of rural links for many urban households.
For all the other contrasts between rural and urban highlighted in Table 4.1, there
are many exceptions. It is also useful to see in the middle of the continuum
between extreme rural characteristics and extreme urban characteristics a rural–
urban interface in which there are complex mixes of characteristics. For instance,
many of the areas around prosperous cities or on corridors linking cities have a
multiplicity of non-farm enterprises and a considerable proportion of the econ-
omically active population that commutes daily to the city or finds work season-
ally or temporarily in urban areas. Many rural areas also have tourist industries
that have fundamentally changed employment structures and income sources.

THE RURAL–URBAN INTERFACE AROUND CITIES1

Cities generally transform large rural areas around them – for instance, as a result
of growing demand for:

l Land for non-agricultural uses (including housing and industries that seek to
avoid environmental regulations or need large sites).
 Seeking an Understanding of Poverty 63

Table 4.1 The rural–urban continuum

Rural Urban
Livelihoods drawn from crop cultivation, Livelihoods drawn from non-agricultural
livestock, forestry or fishing (that is, the labour markets, making/selling goods or
key for a livelihood is access to natural services
capital)
Access to land for housing and building Access to land for housing very
materials not generally a problem difficult; housing and land markets
highly commercialized
More distant from government Vulnerable to ‘bad’ governance at the
as regulator and provider of services local level because of reliance on
publicly provided services and
restrictive regulation.
Access to infrastructure and services Access to infrastructure and services
limited (largely because of remoteness, difficult for low-income groups because
low population density and limited of high prices, illegal nature of their
capacity to pay) homes (for many) and poor governance
Less opportunities for earning cash, Greater reliance on cash for access to
more for self-provisioning. Greater food, water, sanitation, garbage
reliance on favourable weather disposal, transport to work
conditions
Access to natural capital as the key Greater reliance on the house as an
asset and basis for livelihoods economic resource (space for
production, access to income-earning
opportunities; asset and income-earner
for owners, including de facto owners)
But also
Urban characteristics in rural locations Rural characteristics in urban locations
(for example, prosperous tourist areas, (for example, urban agriculture, ‘village’
mining areas, areas with high-value crops enclaves, access to land for housing
and many local multiplier links, rural areas through non-monetary traditional forms)
with diverse non-agricultural production
and strong links to cities)

l Land for sport, recreation and tourism (which may expand and diversify
employment in many towns and rural areas close to large cities).
l Water for urban uses (often competing with or pre-empting sources previously
used by rural enterprises or households).
l More diverse and often higher value foodstuffs (although many of these come
from distant areas, especially where transport networks are well developed
and there is a demand for foodstuffs for the production of which the local
ecology is not well suited).
l Locations where city wastes can be disposed of cheaply, often giving rise to
a concentration of households and enterprises based on waste recovery, reuse
64 Urban Livelihoods

and recycling, and with serious environmental consequences, especially for

water bodies.
l Building materials (seen in quarries, sites from which aggregate is drawn,
cement plants, timber yards, brick-making units, and so on).

These demands also transform occupational structures. The statistics on employ-

ment or enterprises in rural areas around cities often highlight the importance of
non-farm employment and non-agricultural activities and the coexistence of
agriculture, cottage industry, industrial estates, suburban developments, large
commercial enterprises, sport and recreation and other types of land use (see, for
instance, Jones, 1983; McGee, 1987). There may also be a considerable under-
count of non-farm or off-farm employment, if employment surveys only allow
each person to be registered in one occupation, so the non-agricultural work
undertaken by farmers and agricultural labourers is not registered.
The implications of these economic and land-use changes for poverty are
complex and likely to be very location-specific. For existing rural inhabitants,
they are influenced by the distribution of land ownership (or rights to use land),
the extent to which those with rights to land can benefit from changing demands,
and the responsiveness of political and bureaucratic systems to poorer groups’
rights and demands. The benefits for those who own (or acquire) land can be very
considerable, as the value of land multiplies manyfold as it changes from agri-
cultural to non-agricultural use. Land availability for poor rural dwellers who do
not own land – for instance, tenant farmers, share-croppers and those who draw
resources from open access or common property resources – is likely to diminish.
In many countries, traditional land allocation systems, which had long permitted
local populations to acquire land for agriculture, usually become more com-
mercialized and more oriented to urban demands. However, rural households
with small or no land holdings may also benefit from increased employment
opportunities. In general, those who are most at risk are those who have little or
no land, are dependent on wage or casual labour on other farms for part or all
of their income and are unable to take advantage of alternative economic oppor-
tunities in the urban labour market because households and their members lack
skills, contacts, capital or freedom of movement (Rakodi, 1998).
Many farmers with sufficient land holdings benefit from urban demand or
more accessible markets – for instance, around Kumasi (Sarfo-Mensah and
Adam, 1997) and Yogyakarta (Douglass, 1998). Agricultural production may
respond to external demand (including international demand) as urban expan-
sion improves farmers’ access to roads, ports, airports, credit and information
about market opportunities. It is also common for many low- and middle-income
urban households to find better possibilities for acquiring land and building their
own homes outside the main built-up area (including possibilities for illegal
occupation or purchasing a plot in an illegal or legal subdivision).
Although there is a tendency to refer to the rural–urban interface as ‘peri-
urban’, it is generally more complicated spatially than the term peri-urban implies
– that is, it is not simply a circular zone around the built-up area of an urban
centre in which rural and urban land uses and agricultural and non-agricultural
activities are mixed. For instance, it is common for there to be more intense
urban-type development along major transport corridors or around transport
 Seeking an Understanding of Poverty 65

nodes. Around larger cities, it is also common for there to be a considerable range
of activities and residential communities that are spatially scattered – for instance,
factories, quarries, airports and residential communities from which most of the
workforce commute. Particular settlements spatially separate from the city’s
built-up area but within a peri-urban zone often develop diverse employment
bases with enterprises strongly connected to the main city. Certain natural
features, the preferences of particular landowners or existing road systems can
also keep some peri-urban areas undeveloped, while others that are not con-
tiguous with or close to the built-up area become highly urbanized.
One of the most difficult issues in the rural–urban interface, especially
around or close to prosperous cities, is how to manage the rapid economic and
land-use changes in ways that enhance prosperity while controlling environ-
mental costs, bringing forward sufficient land for housing (so prices are kept
down) and ensuring secure livelihoods for poorer groups. Local government
structures are often particularly weak and ineffective on the edge of large cities’
built-up areas. For instance, they may fall within the boundaries of a large
predominantly rural provincial or district authority (whose headquarters may be
some distance from the large city) or may be part of a newly formed municipality
that remains weak and ineffective. Reviews of growth rates for the districts or
municipalities that make up large cities or metropolitan areas often reveal rapid
population growth rates in peripheral municipalities or districts where local
authorities are particularly weak and household incomes are well below the city
average.
In the absence of an effective land-use plan or other means to control new
developments, cities generally expand haphazardly. Uncontrolled physical
growth impacts most on the immediate hinterland of a city; much of this cannot
be described as urban or suburban and yet much of it is no longer rural. Within
this area, agriculture may disappear or decline as land is bought up by people or
companies in anticipation of its change from agricultural to urban use and the
(often very large) increases in land value that result. There is usually a lack of
effective public control of such changes in land use and no means of capturing a
share of the increased land value, even when it is public investment (for instance,
the expansion of road networks) that creates much of the increment.
Unplanned city expansion produces a patchwork of developments, including
businesses and high-density residential settlements, interspersed with land that
remains undeveloped in anticipation of speculative gain. These include legal
subdivisions for houses or commercial and industrial use that have been approved
without reference to any city-wide plan. Around more prosperous cities, low-
density high-income residential neighbourhoods may also develop, along with
commercial developments and leisure facilities for higher income groups. In many
cities, especially those with high levels of crime and violence, such residential
developments may be enclosed within walls and protected by private security
firms – the gated communities or barrios cerados. There are usually unauthorized
subdivisions as well and where regulation is lax, these may also cater for middle-
and upper-income housing. There are usually illegal squatter communities too,
which originally located here because inaccessibility and lack of infrastructure
gave more chance of not being evicted. In many cities (including Buenos Aires,
Delhi, Santiago, Seoul and Manila), this hinterland also contains settlements
66 Urban Livelihoods

formed when their inhabitants were dumped there after being evicted from their
homes by slum or squatter clearance (Hardoy and Satterthwaite, 1989; Environ-
ment and Urbanization, 1994). The inhabitants of these settlements may find
themselves again under threat of eviction as the physical expansion of the urban
area and its road network increases the value of the land on which they live (see,
for instance, ACHR, 1989).
The uncontrolled and unregulated physical expansion of a city’s built-up area
usually has serious environmental and social consequences, including soil erosion
which contributes to the silting up of drainage channels, and the segregation of
low-income groups in the worst located and often the most dangerous areas. The
haphazard expansion of settlements may bring greatly increased costs for provid-
ing basic infrastructure, public transport and social services, as new developments
spring up far from existing networks. It is also more expensive to provide public
transport and social services. Around cities, one often sees the paradox of
overcrowding, housing shortages and inadequate infrastructure and services in
particular areas, and yet large amounts of land left vacant or only partially deve-
loped. Informal settlements are often concentrated on sites that are subject to
flooding or at risk from landslides or other natural hazards, especially where
these offer the best located sites on which low-income settlers have the best
chance of establishing a home or simply avoiding eviction. But these are also sites
to which it is more difficult and expensive to extend basic infrastructure.
Environmental health problems may become particularly serious for certain
groups within the rural–urban interface, in part because of the increased concen-
tration of population and activities, in part because government controls are less
effective, which is one reason why polluting activities locate there and why land
subdivisions and developments do not conform to official regulations.
Urban land markets can also disrupt agricultural production and the liveli-
hoods of those who depend on it in areas that stretch far beyond the sites devel-
oped for urban use. Conflicts over land-use priorities between urban-based
demands and environmental perspectives include the loss of agricultural land,
forests, wetlands and other undeveloped sites to industrial estates and residential
developments or to golf-courses and country-clubs. These conflicts generally
involve social conflicts too, as the livelihoods are threatened by urban-based
demands (see, for instance, Douglass, 1998). Kelly (1998) analyses such conflicts
in the zone to the south of Manila, describing how national, local and personal
forces ensure that land for residential, industrial or other urban developments is
favoured over the protection and continued use of highly productive farmland.

SOME CONCLUSIONS
1 To understand the particular deprivations that poor people face and the best
means to address them, we need to understand local contexts and the diversity
of livelihoods within these contexts. There is also great diversity in local contexts
both between different urban areas and different rural areas and between rural
and urban areas. Rural–urban differences include sources of income, the range
and nature of environmental hazards and the form and relative importance of
 Seeking an Understanding of Poverty 67

different kinds of deprivation. However, the lines between what is rural and what
is urban are fuzzy, in part because the economic and spatial boundaries between
urban and rural areas are not clear-cut, and in part because of the many rural–
urban interlinkages.

2 Most governments and international agencies still act as if urban and rural
economies and societies are not connected and as if agriculture only affects rural
populations and non-agricultural production only takes place in urban areas.
However, strong rural–urban links at the household level (including livelihoods
that have rural and urban components) mean that increased poverty in rural
areas often impacts negatively on urban areas, and vice versa. As noted earlier,
it is common in the literature to see statements that rural poverty increased more
than urban poverty or (although less often) urban poverty increased more than
rural poverty. But increasing urban poverty will usually mean that there are fewer
job opportunities in urban areas for rural dwellers, reduced remittance flows
from urban to rural areas, less urban demand for rural products and possibly
more urban to rural migration, which could increase dependency burdens in rural
areas. There are tens of thousands of urban centres in low- and middle-income
countries whose economic and employment base is strongly connected to agri-
culture, yet most governments and international agencies operate as if agricul-
tural and urban development are independent of each other or in conflict. In
addition, rural–urban interlinkages are intensifying, partly because of increased
opportunities (for instance, improved transport and communication facilities
which increase access to information on new employment opportunities in
expanding export-oriented industrial sectors) and partly because of stronger
constraints (for example, population pressure on agricultural land but also,
importantly, decreasing agricultural incomes). The nature and scale of rural–
urban linkages are affected by the predominant production base and urbaniza-
tion patterns of specific regions and countries. A country’s position in global
markets is also important in rural–urban linkages.

3 Most national government departments or ministries and international agencies

know little about the local contexts in which their projects and programmes
operate. Local contexts have a strong influence on the scale and nature of poverty
and the most effective means of reducing it. Many studies show the complexity
and diversity of urban livelihoods, rural livelihoods and rural–urban linkages, but
almost everywhere, the projects and programmes of governments and inter-
national agencies are planned with relatively little knowledge of local contexts
and what these imply for the best means to address poverty.

4 Appropriate responses to diversity and complexity within each local context.

The only possible way to operate effectively, given such diversity between locali-
ties and such complexity within each locality, is to ensure that the interventions
of external agencies are influenced by the priorities of those who face deprivation
and build on their knowledge and resources. Also, it is necessary to ensure the
effective functioning of institutions that protect poorer groups’ civil and political
rights, and to ensure that they have access to basic services. This is likely to
require institutional innovation, so that the organizations that allocate resources
68 Urban Livelihoods

and plan and implement interventions can respond to local needs and demands.
This implies the use of more participatory approaches, less to increase the
knowledge of external agencies about local contexts than to allow those suffering
deprivation a greater voice in setting priorities, influencing resource allocation,
preventing measures that threaten their livelihoods and gaining access to justice.
It also means that external agencies must work with and support local organiza-
tions that can implement poverty-reduction measures that respond to local
contexts and work with disadvantaged groups in participatory and accountable
ways. These may include local governments and local NGOs, although with care
taken to ensure that the latter are accountable to low-income groups since many
NGOs work in ways that are non-transparent and non-participatory (see, for
instance, Anzorena et al, 1998). Finally, it implies a better understanding of how
best to support the development of a stronger economic base, while ensuring that
this maximizes livelihood opportunities for those with the least incomes and
assets, despite the difficulty of ensuring pro-poor economic growth when poor
groups have few assets and little political influence (see, for instance, Benjamin,
2000). Perhaps only in these rather general principles are there points that have
the same relevance for all urban and rural areas.

NOTE
1 This section draws primarily on Chapter 5 of Hardoy et al, 2001

REFERENCES
ACHR (Asian Coalition for Housing Rights) (1989) ‘Evictions in Seoul, South Korea’,
Environment and Urbanization, vol 1, no 1, pp 89–94
Afsar, R (1999) ‘Rural–urban dichotomy and convergence: emerging realities in Bangla-
desh’, Environment and Urbanization, vol 11, no 1, pp235–246
Afsar, R (2000) Urbanization in Bangladesh, Institute for Environment and Development,
Working Paper on Urban Change, London
Amis, P (1999) Urban Economic Growth and Poverty Reduction, Urban Governance,
Partnership and Poverty WP 2, International Development Department, School of
Public Policy, University of Birmingham, Birmingham
Anzorena, J, Bolnick, J, Boonyabancha S et al (1998) ‘Reducing urban poverty: some
lessons from experience’, Environment and Urbanization, vol 10, no 1, pp167–186
Bebbington, A (1999) ‘Capitals and capabilities: a framework for analyzing peasant
viability, rural livelihoods and poverty, World Development, vol 27, no 12, pp2021–
2044
Benjamin, S (2000) ‘Governance, economic settings and poverty in Bangalore’, Environ-
ment and Urbanization, vol 12, no 1, pp35–56
Blitzer, S, Davila, J, Hardoy, J E and Satterthwaite, D (1988) Outside the Large Cities:
Annotated Bibliography and Guide to the Literature on Small and Intermediate Urban
Centres in the Third World, Human Settlements Programme, International Institute
for Environment and Development, London
Bryceson, D, Kay, C and Mooij, J (2000) Disappearing Peasantries? Rural Labour in
Africa, Asia and Latin America, IT Publications, London
 Seeking an Understanding of Poverty 69

Cairncross, S (1990) ‘Water supply and the urban poor’, in Hardoy, J E, Cairncross, S and
Satterthwaite, D (eds) The Poor Die Young: Housing and Health in Third World
Cities, Earthscan, London, pp 109–126
Cairncross, S, Hardoy, J E and Satterthwaite, D (1990) ‘The urban context’ in Hardoy,
J E, Cairncross, S and Satterthwaite, D (eds) The Poor Die Young: Housing and Health
in Third World Cities, Earthscan, London, pp1–24
Carney, D (1998) ‘Implementing the sustainable rural livelihoods approach’, in Carney,
D (ed) Sustainable Rural Livelihoods: What Contribution Can We Make? Department
for International Development, London, pp3–23
Carney, D, Drinkwater, M, Rusinow, T, Neefjes, N, Wanmali, S and Singh, N (2000)
Livelihoods Approaches Compared, Department for International Development,
London
Chambers, R and Conway, G (1992) Sustainable Rural Livelihoods: Practical Concepts
for the 21st Century, Discussion Paper 296, Institute of Development Studies, Uni-
versity of Sussex, Brighton
Douglass, M (1998) ‘A regional network strategy for reciprocal rural–urban linkages: an
agenda for policy research with reference to Indonesia’, Third World Planning Review,
vol 20, no 1, pp1–34
Ellis, F (1998) ‘Household strategies and rural livelihood diversification’, Journal of
Development Studies, vol 35, no 1, pp1–38
Environment and Urbanization (1994), Special issue on evictions, vol 6, no 1
Hardoy, J E and Satterthwaite, D (eds) (1986) Small and Intermediate Urban Centres:
Their Role in National and Regional Development in the Third World, Hodder &
Stoughton, London and Westview, Boulder, CO
Hardoy, J E and Satterthwaite, D (1989) Squatter Citizen: Life in the Urban Third World,
Earthscan, London
Hardoy, J E, Mitlin, D and Satterthwaite, D (2001) Environmental Problems in an
Urbanizing World: Finding Solutions for Cities in Africa, Asia and Latin America,
Earthscan, London
Hataya, N (1992) ‘Urban–rural linkage of the labour market in the coffee growing zone
in Colombia’, The Developing Economies, vol XXX, no 1, pp63–83
Jones, G W (1983) Structural Change and Prospects for Urbanization in Asian Countries,
Papers of the East–West Population Institute, no 88, East–West Centre, Honolulu
Kamete, A Y (1998) ‘Interlocking livelihoods: farm and small town in Zimbabwe’,
Environment and Urbanization vol 10, no 1, pp23–34
Kanji, N (1995) ‘Gender, poverty and structural adjustment in Harare, Zimbabwe’,
Environment and Urbanization, vol 7, no 1, pp37–55
Kellett, P and Tipple, A G (2000) ‘The home as workplace: a study of income generating
activities within the domestic setting’, Environment and Urbanization, vol 12, no 1,
pp203–14
Kelly, P F (1998) ‘The politics of urban–rural relationships: land conversion in the
Philippines’, Environment and Urbanization, vol 10, no 1, pp35–54
Krüger, F (1998) ‘Taking advantage of rural assets as a coping strategy for the urban
poor’, Environment and Urbanization, vol 10, no 1, pp119–134
Manzanal, M and Vapnarsky, C (1986) ‘The development of the Upper Valley of Rio
Negro and its periphery within the Conahue Region, Argentina’ in Hardoy, J E and
Satterthwaite, D (eds) Small and Intermediate Urban Centres: Their Role in Regional
and National Development in the Third World, Hodder & Stoughton, London and
Westview, Boulder, CO, pp18–79
McGee, T G (1987) Urbanization or Kotadesasi – the Emergence of New Regions of
Economic Interaction in Asia, Working Paper, East West Center, Honolulu
70 Urban Livelihoods

Pryer, J (1993) ‘The impact of adult ill-health on household income and nutrition in
Khulna, Bangladesh’, Environment and Urbanization, vol 5, no 2, pp35–49
Rakodi, C (1998) Review of the Poverty Relevance of the Peri-Urban Interface Production
System Research, Report for the DFID Natural Resources Systems Research Programme,
London
Sarfo-Mensah, P and Adam, M (1997) Kumasi Natural Resource Management Project:
Highlights of the Findings of the PRA Study on Agriculture, Bureau of Integrated
Rural Development, Kumasi, Ghana and Natural Resources Institute, London
Scoones, I (1998) Sustainable Rural Livelihoods: A Framework for Analysis, WP 72,
Institute for Development Studies, University of Sussex, Brighton
Smit, W (1998) ‘The rural linkages of urban households in Durban, South Africa’,
Environment and Urbanization, vol 10, no 1, pp77–87
Smit, J, Ratta, A and Nasr, J (1996) Urban Agriculture: Food, Jobs and Sustainable Cities,
Publication Series for Habitat II, volume 1, UN Development Programme, New York
Tacoli, C (1998) Bridging the Divide: Rural–Urban Interactions and Livelihood Strate-
gies, Gatekeeper Series no 77, IIED Sustainable Agriculture and Rural Livelihoods
Programme, International Institute for Environment and Development, London
UNCHS (Habitat) (1996) An Urbanizing World: Global Report on Human Settlements,
1996, Oxford University Press, New York
WHO (1992) Our Planet, Our Health, Report of the WHO Commission on Health and
Environment, World Health Organization, Geneva
World Bank (2000) Attacking Poverty. World Development Report 2000/1, Oxford
University Press, New York
Zoomers, A E B and Kleinpenny, J (1996) ‘Livelihood and urban–rural relations in central
Paraguay, Tidschrift Economische voor Geografie, vol 87, no 2, pp161–74