Inequality, educational choice, and public school quality in income‐mixing communities

AuthorPaolo Melindi‐Ghidi
DOIhttp://doi.org/10.1111/jpet.12336
Date01 December 2018
Published date01 December 2018
Received: 3 August 2017
|
Accepted: 30 August 2018
DOI: 10.1111/jpet.12336
ORIGINAL ARTICLE
Inequality, educational choice, and public school
quality in incomemixing communities
Paolo MelindiGhidi
University of Paris Nanterre Economix
Avenue de la Republique, AixMarseille
University, CNRS, EHESS, Centrale Marseille,
AMSE, Nanterre, France
Correspondence
Paolo MelindiGhidi, University of Paris
Nanterre Economix Avenue de la
Republique 200, AixMarseille University,
CNRS, EHESS, Centrale Marseille, AMSE,
92100 Nanterre, France.
Email: paolo.melindighidi@parisnanterre.fr
Funding information
Agence Nationale de la Recherche, Grant/
Award Number: ANR14FRAL0007
Abstract
Why, in some urban communities, do rich and poor house-
holds cohabit while, in others, we observe sorting by
income? Does income inequality impact residential choices
and community segregation? To answer these questions, I
develop a twocommunity general equilibrium framework of
school quality, residential choice, and tax decision with
probabilistic voting. The model predicts that an equilibrium
with income mixing in which households segregate across
public schools and lowand highincome households choose
to live in the same community might emerge in highly
unequal societies. In this particular equilibrium, income
mixing communities perform lower public school quality
than communities populated by middleincome households.
The effect of inequality on the quality of public schooling in
the incomemixing community is ambiguous and depends on
the relative endowments of private goods, such as housing,
in the two communities.
KEYWORDS
income mixing, inequality, probabilistic voting, randomization,
segregation
JEL CLASSIFICATION
D72, I24, I28, R21
1
|
INTRODUCTION
A striking characteristic of US metropolitan areas is the concentration of poverty close to inner cities. Data from
the US Census show that 17.6% of the population in the inner cities of all US metropolitan areas had an income
below poverty level in 1999, as against only 8.4% of the suburban population.
1
According to the 2012 American
J Public Econ Theory. 2018;20:914943.wileyonlinelibrary.com/journal/jpet914
|
© 2018 Wiley Periodicals, Inc.
Community Survey (ACS), in 2012 the share of the inner city population living in poverty was as high as 19.7%.
2
However, in some urban areas, the city center is also home to highincome households who, not surprisingly, may
coreside in the same area with lowincome households.
The cohabitation of heterogeneous income groups in the same community, in particular in the same school
district, can have important implications for income redistribution, access to highquality education, public policies,
political decisions, and socioeconomic opportunities. A central issue is to understand the consequences of the
emergence of mixedincome communities, since the presence of different income groups within the same
community may create segregation in terms of access to highquality schooling, thereby hindering the upward
mobility of poor households. Two sets of questions arise when mixedincome communities emerge. First, why do
some urban areas show sorting by income at community level while, in others, rich and poor households cohabit?
How does this difference relate to school districtsquality of public schooling and enrollment in private schools?
Are public schools in communities composed mainly of middleincome households qualitatively better than those in
mixedincome districts? Second, does income inequality impact residential choices and community segregation?
In this paper, I develop a general equilibrium model of private/public school choice, political decisions, and
endogenous residential choice to address these questions. I focus on a twocommunity economy in which housing
market and fiscal policies interact with school and residential location choices, and therefore with the quality of
public education. The framework I provide involves an economy composed of two communities with homogeneous
land and fixed boundaries, which can be interpreted as two different school districts. Parents have to decide which
district to live in and which type of school to send their children to, choosing between a taxfinanced public school
or a private school financed by tuition fees. The quality of public education in each district is determined by the
amount of spending per student financed through property taxes on housing value.
3
Moreover, a probabilistic
voting process in each community determines the local tax rate and, therefore, public education spending.
The most recent education financial literature suggests that mechanisms other than majority voting, such as
probabilistic voting, might offer a different description of the aggregation of preferences.
4
Arcalean and Schiopu
(2016), Benzidia, Lubrano, and MelindiGhidi (2016), Bernasconi and Profeta (2012), De La Croix and Doepke
(2009), and Dottori, Estevan, and ILing (2013) introduce probabilistic voting in the context of education and school
choice. In their pioneering work, De La Croix and Doepke (2009) develop a single district economy in which the tax
rate and the quality of public education are determined via probabilistic voting. They show that households
perfectly segregate by education. This segregation pattern is driven by the fact that parents prefer to enroll their
children in private schools when these schools provide higher quality education than the public system.
5
In this study, I construct a hybridapproach able to merge this new strand of the education financial literature
with the standard Tiebout community choice model in which private education options are available. For instance,
Bearse, Glomm, and Ravikumar (2001) and Hanushek, Sarpa, and Yilmaz (2011), and MartinezMora (2006) have
developed general equilibrium models of community choice and school competition. However, all these works
consider majority voting instead of a political mechanism that does not depend solely on the preferences of the
1
Source: US Census Bureau, 2000 Census, Summary File 3. The data also point to declining poverty in inner cities since 1990.
2
In contrast, only 11.2% of the suburban population had an income below the poverty threshold during the same year. See Gabe (2013) for a report on
poverty in the United States, Partridge and Rickman (2006) for a through analysis of poverty trends in America. See also Berube and Frey (2002) for an
analysis of poverty rates in the 102 largest US metropolitan areas based on the 2000 Census, and Berube and Kneebone (2006) for a similar study based
on the 2000 Census and 2005 ACS.
3
While I assume proportional income tax rather than property value tax, the model predicts the same qualitative results.
4
Probabilistic voting models were first introduced by Lindbeck and Jorgen (1993). Glomm, Ravikumar, and Schiopu (2011) note that even though these
models could ease existence problems, whether they provide a better representation of reality remains an open question.
5
More precisely, they examine how the quality of public education is affected by the presence of private schools. The choice between public and private
education has already been studied by many authors: (De La Croix & Doepke, 2004; Stiglitz, 1974), in a single district economy; (Estevan, 2013;
Gutierrez & Tanaka, 2009), in a setting where parents can send children to work instead of educate them; Tournemaine and Tsoukis (2015), in a mixed
regime of education with individuals heterogeneous in their statusmotivation; Lasram and Laussel (2018), in a model where public and private
universities coexist.
MELINDIGHIDI
|
915

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT