Luxembourg Income Study
Working Paper Series
Luxembourg Income Study (LIS), asbl
Working Paper No. 329
Urban Poverty in Developed Countries
Andrea Brandolini and Piero Cipollone
October 2002
URBAN POVERTY IN DEVELOPED COUNTRIES
by Andrea Brandolini
*
and Piero Cipollone
*
October 2002
Abstract
In this paper we investigate the urban/rural dimension of poverty in developed
countries. We provide original estimates for Italy, we gather published statistics for France
and the United States, and we produce novel cross-country estimates from the LIS database.
We show that the size of urban poverty depends on where the boundaries of metropolitan
districts are drawn and we observe that overlooking geographical differences in the cost of
living is a particularly relevant hypothesis. We find that in France and the United States post-
war economic growth and urbanisation were accompanied by a substantial reduction of the
poverty risk for the rural population, while poverty rates improved less, or even sometimes
deteriorated, for the urban population. The lack of a standard definition of urban/rural area
precludes a rigorous comparative study. Our results indicate, however, that only in few
countries (Denmark, the United Kingdom and the United States) the greatest poverty rates
are found in central cities, while in all other developed countries poor persons are still
relatively more frequent in rural areas. This pattern is stronger in the four non-developed
economies examined here.
JEL classification: I32, R2.
Keywords: poverty, urban/rural areas.
Contents
1. Introduction ..........................................................................................................................1
2. The definition of urban area and the evidence for Italy .......................................................3
2.1 The definition of urban area ..........................................................................................3
2.2 The Survey of Household Income and Wealth and measurement hypotheses..............4
2.3 Urban poverty in Italy ...................................................................................................5
3. Long-run changes in urban poverty rates: France and the United States.............................7
4. International comparisons: evidence from the Luxembourg Income Study......................10
5. Conclusion..........................................................................................................................14
References...............................................................................................................................34
*
Bank of Italy, Economic Research Department.
1. Introduction
1
The urban/rural dimension of poverty has received much attention in social sciences.
In deve loping countries rural poverty rates are still much higher than urban rates, often from
2 to 3 times, although the territorial distribution of poverty is forecasted to change radically
if the ongoing process of urbanisation continues (e.g. World Bank, 2000, Table 4, pp. 280-1;
Haddad, Ruel and Garrett, 1999; Ravallion, 2001; and Eastwood and Lipton, 2000, for a
somewhat different evidence). With regards to developed countries, some researchers,
especially among sociologists, have stressed that poverty may have been increasing more in
cities than in rural areas as a consequence of the downsizing of industry and the growth of
the more heterogeneous service sector, the spreading of contingent work, the retreating of
welfare states, the loosening of family links, the migrations from developing countries. In
North America, the interest for urban poverty has frequently gone hand in hand with the
analysis of the spatial segregation on racial and ethnic bases (e.g. Moynihan, 1968; Peterson,
1991-92; Hajnal, 1995; Jargowsky, 1996; Mills and Sendé Lubuele, 1997; Mayer, 2001).
Nevertheless, the systematic investigation of differences between rural and urban areas in
developed countries is not common both in official sources and the scientific literature.
In the United States, the Census Bureau regularly compares poverty rates in
metropolitan and nonmetropolitan areas (Dalaker, 2001). Recent reports by the National
Council of Welfare (2000) in Canada and by Insee (1997, 1998, 2001) in France examine
how the proportion of low-income families varies with the community or municipality size.
On the other hand, the urban/rural breakdown is not a standard feature of poverty statistics at
the European level (Eurostat, 2000), nor in Italy (Istat, 2001) and the United Kingdom
1
We thank Federico Cingano, Giovanni D’Alessio and Paolo Fabbri for helpful discussions in the writing
of the paper. Microdata of the Luxembourg Income Study for the United Kingdom are taken from the Family
Expenditure Survey for 1995 and are Crown Copyright; they were made available by the United Kingdom
Office for National Statistics through the ESRC Data Archive and their use has been authorised. Neither the
Office for National Statistics, nor the ESRC Data Archive are responsible for the analysis or interpretation of
the data given here. The views expressed herein are those of the authors and do not necessarily reflect those of
the Bank of Italy. Corresponding author: Andrea Brandolini, Banca d’Italia, Economic Research Department,
via Nazionale 91, 00184 Roma, Italy. E-mail: brandolini.andrea@insedia.interbusiness.it.
2
(Department of Social Security, 2000).
2
In the rapidly growing academic literature on the
structure and evolution of economic poverty the attention for the urban dimension appears to
be scarce. Cross-national comparisons are particularly missing. For instance, a recent
valuable collection of essays on poverty in Nordic countries (Gustafsson and Pedersen, ed.,
2000) provides information broken down by several population characteristics but not by
area of residence.
One important obstacle to comparative analysis is the lack of a generally agreed
criterion to distinguish “urban” from “rural” as well as intermediate cases. This deficiency is
recognised in the United Nations’ Demographic Yearbook: “although statistics classified by
urban/rural areas are widely available, no international standard definition appears to be
possible at this time since the meaning differs from one country or area to another” (United
Nations Department of Economic and Social Affairs, 2000, p. 36). The Yearbook therefore
refrains from providing comparable figures on the size of the urban population and reports
statistics computed according to national definitions. In Austria, Norway and Spain, for
example, urban areas are defined on the basis of demographic size only, whereas in France
also the distance between houses concurs to design urban areas. In Italy, the national
statistical office does not provide any classification of urban and rural municipalities,
although demographic size is sometimes used.
3
The main purpose of this study is to review what information is available on the
differential poverty levels in cities and rural districts in some developed countries. We rely
2
At the policy level, the strategy set out by the British Government against poverty and social exclusion
(Secretary of State for Social Security, 1999) devotes considerable attention to “communities” and to targeting
help to the poorest neighbourhoods, recognising the different nature of poverty in urban and rural areas.
However, Shucksmith and Philip (2000, p. 3) remark that: “The Social Exclusion Unit in England has shown a
marked lack of interest in the incidence of social exclusion in rural areas … In Scotland, in contrast, social
inclusion in rural areas has been identified by the Scottish Office as an important issue requiring further
research and policy development”. The report prepared by Harrop, Kenway and Palmer (2000) for the
Countryside Agency contains a number of indicators on poverty and social exclusion, including the proportion
of individuals in low-income households, separately for remote rural, accessible rural and urban areas in
England.
3
The last official classification was based on 13 socio-demographic indicators (e.g. share of the labour
force in agriculture, demographic density, availability of water and toilets in houses, etc.) drawn from the 1981
Census (Istat, 1986). In the frame of regional indicators for the evaluation of development policies recently set
by Istat in collaboration with the Treasury Ministry, a municipality is classified as rural if the population
density is below 100 inhabitants per square kilometre and the agriculture share in total employment is higher
than 12.4 percent, equal to twice the EU average (Istat, 2002).
3
on both national sources for Italy, France and the United States and the international
database assembled at the Luxembourg Income Study (LIS). In section 2 we present our
estimates for Italy in the period 1987-2000, using data drawn from the Bank of Italy’s
Survey of Household Income and Wealth (SHIW). We also exploit the SHIW database to
experiment how alternative definitions of “urban area” impact on poverty rates. In sections 3
we use the information for France and the United States to shed some light on the long-run
evolution of the urban/rural poverty differential. In section 4 we offer comparative evidence
for a number of developed countries around mid-1990s on the basis of the LIS data. We
summarise the lessons to be drawn from our analysis at the end of the paper.
2. The definition of urban area and the evidence for Italy
4
2.1 The definition of urban area
As mentioned above, there is no generally agreed criterion to separate “urban” from
“rural”. This problem overlaps with the choice of the reference territorial unit, which is
typically constrained by the available data. The minimum territorial unit to study urban
poverty in Italy is the municipality, since no information is available on family incomes at
the census tract level. However, this choice may be too restrictive for the largest urban
agglomerations, where residential and business districts may extend over many neighbouring
municipalities. Among larger territorial units, we may select “provinces” or “metropolitan
areas”, if we keep following administrative criteria. Alternatively, we may favour a socio -
economic characterisation and focus on “local labour systems”, i.e. clusters of economically
integrated and adjacent municipalities, whose boundaries are set after analysing daily
journeys to work (Istat, 1997). We test the sensitivity of results for Italy by specifying four
definitions of “large urban area” or “metropolitan area” (henceforth, used interchangeably):
(a) Municipality: the municipal territories of the 6 Italian cities with more than 500,000
inhabitants during the period under examination, namely Palermo, Genoa, Turin, Naples,
Milan and Rome (including Fiumicino, a nearby town independent since 1992);
4
This section draws on Brandolini (2002).
