Choosing Electoral Rules: Theory and Evidence from US Cities

Abstract

This paper studies the choice of electoral rules, in particular, the question of minority representation. Majorities tend to disenfranchise minorities through strategic manipulation of electoral rules. With the aim of explaining changes in electoral rules adopted by US cities (particularly in the South), we show why majorities tend to adopt "winner-take-all" city-wide rules (at-large elections) in response to an increase in the size of the minority when the minority they are facing is relatively small. In this case, for the majority it is more effective to leverage on its sheer size instead of risking to concede representation to voters from minority-elected districts. However, as the minority becomes larger (closer to a fifty-fifty split), the possibility of losing the whole city induces the majority to prefer minority votes to be confined in minority-packed districts. Single-member district rules serve this purpose. We show empirical results consistent with these implications of the model.

Full text

Page 1

H H I I E E R R

Harvard Institute of Economic Research

Discussion Paper Number 2065

Choosing Electoral Rules: Theory and Evidence
from US Cities

by

Philippe Aghion, Alberto Alesina
and Francesco Trebbi

March 2005

Harvard University
Cambridge, Massachusetts


This paper can be downloaded without charge from:
http://post.economics.harvard.edu/hier/2005papers/2005list.html

The Social Science Research Network Electronic Paper Collection:
http://ssrn.com/abstract=690761

Page 2

Choosing Electoral Rules: Theory and Evidence
from US Cities.
Philippe Aghion∗, Alberto Alesina†, and Francesco Trebbi‡.
March 2005
Abstract
This paper studies the choice of electoral rules, in particular the ques-
tion of minority representation. Majorities tend to disenfranchise minori-
ties through strategic manipulation of electoral rules. With the aim of
explaining changes in electoral rules adopted by US cities (particularly in
the South), we show why majorities tend to adopt ”winner-take-all” city-
wide rules (at-large elections) in response to an increase in the size of the
minority when the minority they are facing is relatively small. In this case,
for the majority it is more effective to leverage on its sheer size instead of
risking to concede representation to voters from minority-elected districts.
However, as the minority becomes larger (closer to a fifty-fifty split), the
possibility of losing the whole city induces the majority to prefer minority
votes to be confined in minority-packed districts. Single-member district
rules serve this purpose. We show empirical results consistent with these
implications of the model.
We thank Matilde Bombardini, Gary Chamberlain, John Friedman, Edward Glaeser,
Richard Holden, Caroline Hoxby, David Lucca, James Robinson, and John Wallis for useful
comments and suggestions. We are grateful to participants of the CIAR meetings in Toronto
and seminars at Harvard University. Dilyan Donchev, Laura Serban, and Radu Tatucu pro-
vided excellent research assistance. Trebbi acknowledges financial support from the Social
Sciences Research Council.
∗Harvard University, Department of Economics, and Canadian Institute for Advanced Re-
search.
†Harvard University, Department of Economics, National Bureau of Economic Research,
and Centre for Economic Policy Research.
‡Harvard University, Department of Economics.
1

Page 3

1Introduction
One of the key questions in political economy is how different electoral laws affect
policy outcomes. In order to provide an answer several authors take electoral
rules as exogenous or predetermined and use them as explanatory variables for
various policies1. However, the rules of government are themselves endogenous
variables chosen at constitutional tables. Ideally, rules should be chosen behind
a ”veil of ignorance”, that is decisions should be taken as if one ignored the
identity of those benefiting from the choices themselves. In reality, however, in
most constitutional tables the veil of ignorance is ”see-through”, in the sense
that there is some knowledge of who would benefit under alternative rules and
of what policy outcomes those rules would produce. Therefore the ”exogeneity”
of constitutions can be called into question.
Our focus is on the question of minority representation, with special refer-
ence to the nature of electoral districts and alternative rules for the choice of
representatives. We have two goals in mind, one more general and one more
specific. The first and more general point is to make progress in modeling in-
stitutional choice as endogenous. On the first point, most of the literature is
normative, i.e. it discusses how electoral laws should be chosen, starting from
the work of Hayek (1960) and Buchanan and Tullock (1962)2. A normative
approach usually characterizes works in Political Science, with some notable ex-
ception such as Riker (1986) and several essays in Colomer (2004).Economists
have only recently began to pay attention to the endogeneity of political insti-
1A classic study is Lijphart (1994). The most recent contributions on the effects of insti-
tutions (taken as exogenous or prestermined) on economic policies are Persson and Tabellini
(2003) for a sample of democratic countries and Baqir (2003) for a cross-section of US cities.
They build upon a vast literature on the effects of alternative forms of government on policies,
a literature that we do not review here. We refer the reader to Persson and Tabellini (2003) for
a survey of the cross-country literature. On US states see in particular Alt and Lowry (1994),
Poterba (1994), and Bohn and Inman (1996) amongst others. Mulligan and Sala I Martin
(2004) offer a dissenting view, namely that policies are determined by lobbying pressure that
are not much affected by institutional forms of government.
2For a survey of the literature on Constitutional Theory, see Voigt (1997).
2

Page 4

tutions from a positive, as opposed to a normative perspective. Alesina and
Glaeser (2004) for instance discuss how the choice of alternative electoral rules,
which are themselves associated with different policy choices over the welfare
state, are indeed the result of strategic constitutional choices. The present pa-
per is a study in ”positive constitutional theory”, with a special emphasis on
the question of how majorities select electoral rules to partially disenfranchise
minorities.
The second and more specific goal of our paper is to analyze the evolution
of minority representation in American cities. We examine empirical evidence
drawn from US municipalities, that adopt two types of electoral rules: either a
single-member district (also called single-district) or at-large rules (or a com-
bination of the two). Councilmen elected by district compete for one or, more
rarely, multiple seats in each geographic subdivision (district or ward) of the
city. Differently, in at-large elections, officials are elected in multi-member plu-
rality districts, voters have as many votes as there are council seats, and the
only multi-member district is identified by the city itself. The basic ”winner-
take-all” logic holds for both rules. For given council size, the difference between
single-district and at-large rules is due to geographic clustering of groups of vot-
ers with homogeneous preferences3. We show why majorities at the constitution
design stage tend to adopt at-large electoral rules in response to an increase in
the size of the minority when the minority they are facing is relatively small.
In this case, as the size of the minority increases, for the majority it becomes
more effective to leverage on its sheer size instead of risking to concede repre-
sentation to voters from minority-elected districts. However, as the minority
becomes larger (closer to a fifty-fifty split), the possibility of losing the whole
city induces the majority to prefer minority votes to be confined in minority-
packed districts. Single-member district electoral rules serve this purpose. This
shift in the preferences of the (constitutional) majority, first towards an at-large
3See Cutler, Glaeser, and Vigdor (1999) for a detailed account of racial segregation patterns
in the United States. The assumption of geographic clustering will be maintained throughout
the paper.
3

Page 5

electoral rule, and then towards a single-district rule as the size of the minority
increases, is precisely what the data show.
When discussing local politics in the US, one is immediately thrown into the
area of race relations.4In fact, it is quite compelling to identify the ”majority”
with the whites and the ”minority” with racial ”minorities” (perfect choice of
words!). The evolution of voting rights, especially in the South, allows us to
test implications of the model regarding the endogenous institutional choice.
Until the mid-sixties, before the civil right movement and the Voting Rights
Act of 1965, racial minorities (mainly blacks) in the South were essentially
disenfranchised by a battery of regulations that, although color-blind on paper,
were in practice directed to severely limit black vote. In this context the choice
of electoral rules and forms of governments had not much to do with a white
majorities’ attempt at controlling black influence on city governments, since
blacks did not vote. After the mid-sixties, due to novel federal Voting Rights
legislation, black influence increased substantially in terms of their ability to
elect representatives. Indeed, after the Voting Rights Act of 1965, we show that
decisions about electoral rules reflected changes in the relative size of the white
and black populations in a way consistent with our model.
Manipulation of electoral rules is not a prerogative exclusive of American
cities. Alexander (2005, p.211) describes in detail the 1947 Gaullist manipula-
tions of electoral rules in France. In the Paris area where the Gaullist alliance
was weak they introduced proportional representation, in rural areas where the
alliance was strong, they introduced plurality rule. Krenzer (2005, p.229) de-
scribes strategic manipulation in Germany. One could go on.
Our empirical results on (within-)US cities variation are quite consistent
with previous findings on the cross-country evidence in Alesina, Aghion, and
Trebbi (2004). In both cases constitutional choices do not seem to occur behind
4For discussion of the importance of race in American local politics, see for instance Hacker
(1992), Huckfeld and Kohfeld (1989) and Wilson (1996) amongst many others. Alesina, Baqir,
and Hoxby (2004) argue that even the design and number of local jurisdictions in the US
depends upon race relations.
4