Neighborhood Effects, Public Housing and Unemployment in France



3 Data and basic evidence

3.1 Data

This paper focuses on Lyon, the third largest city in France. Its agglomeration (defined here by
its urban unit
8) extends over a 958 km2 area and hosts around 1.3 million inhabitants. As shown
in the next subsection, Lyon is characterized by the existence of pockets of unemployment in
the close periphery of its center and thus appears to be an adequate case study to test for the
existence of neighborhood effects.

The empirical analysis conducted in this paper is based on two datasets extracted from
the 1999 French Population Census. The neighborhoods are defined on the basis of Iris, the
finest geographical level available in the French Census (they will be called neighborhoods in
the rest of the paper, for the sake of simplicity). These neighborhoods are either municipalities,
or subdivisions of municipalities if the latter have more than 10,000 inhabitants. They are
created in order to represent homogenous entities in terms of housing and population. They are
generally formed around well identified groups of buildings and respect frontiers such as main
avenues, rivers or railways. Our study area has 540 neighorhoods
9 which have on average 2,428
inhabitants, a figure more or less comparable to the size of American Census tracts used in
previous studies of neighborhood effects in the U.S.

Our first dataset gathers summary statistics at the neighborhood level and includes vari-
ous indicators of the socioeconomic composition and average housing characteristics. This data
is used to define the typology of neighborhoods (see next subsection). The second dataset cor-
responds to a sample of indiduals (1/20th of the total population), for whom detailed personal,
household, and housing characteristics are provided (age, gender, education
10 , employment sta-
tus, household type, housing tenure
11,...) along with the characteristics of the other members of
his/her household. This data allows to link each individual to the neighborhood in which s/he
lives in. It is used to estimate our econometric model. As we already explained, our study deals
8 The urban unit, unit´e urbaine in French, is a set of municipalities, the territory of which is covered by a
built-up area of more than 2,000 inhabitants, and in which buildings are separated by no more than 200 meters.
The urban unit of Lyon consists of 102 municipalities. For practical reasons, we added three municipalities which
are enclosed within the urban unit of Lyon (Quincieux and Poleymieux-au-Mont-D’Or).

9 A few Iris having less than 200 people had to be deleted for confidentiality reasons.

10In the whole paper, the following education levels will be used: No diploma, At most lower secondary school,
Vocational training, High school final diploma, University degree. They correspond to the following French
categories: no reported diploma, CEP or Brevet, CAP or BEP,
Baccalautéat, DEUG or above, respectively.

11Note however that neither housing prices nor incomes are available in the French Census.

10



More intriguing information

1. The name is absent
2. The name is absent
3. Activation of s28-dependent transcription in Escherichia coli by the cyclic AMP receptor protein requires an unusual promoter organization
4. Developmental Robots - A New Paradigm
5. L'organisation en réseau comme forme « indéterminée »
6. The name is absent
7. On Social and Market Sanctions in Deterring non Compliance in Pollution Standards
8. ARE VOLATILITY EXPECTATIONS CHARACTERIZED BY REGIME SHIFTS? EVIDENCE FROM IMPLIED VOLATILITY INDICES
9. Qualifying Recital: Lisa Carol Hardaway, flute
10. The name is absent
11. Regional Intergration and Migration: An Economic Geography Model with Hetergenous Labour Force
12. A Consistent Nonparametric Test for Causality in Quantile
13. The name is absent
14. Midwest prospects and the new economy
15. Factores de alteração da composição da Despesa Pública: o caso norte-americano
16. Determinants of Household Health Expenditure: Case of Urban Orissa
17. DISCUSSION: POLICY CONSIDERATIONS OF EMERGING INFORMATION TECHNOLOGIES
18. The name is absent
19. Credit Markets and the Propagation of Monetary Policy Shocks
20. Dynamic Explanations of Industry Structure and Performance