Do Elections Matter for Economic Performance?

Published date01 February 2015
Date01 February 2015
DOIhttp://doi.org/10.1111/obes.12054
1
©2013 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 77, 1 (2015) 0305–9049
doi: 10.1111/obes.12054
Do Elections Matter for Economic Performance?*
Paul Collier† and Anke Hoeffler
The Blavatnik School of Government, 10 Merton Street, Oxford, OX1 4JJ, UK
(e-mail: paul.collier@economics.ox.ac.uk)
Department of Economics, Manor Road Building, Oxford, OX1 3UQ, UK
(e-mail: anke.hoeffler@economics.ox.ac.uk)
Abstract
In mature democracies, elections discipline leaders to deliver good economic performance.
Since the fall of the Soviet Union, most developing countries also hold elections, but these
are often marred by illicit tactics. Using a new global data set, this article investigates
whether these illicit tactics are merely blemishes or substantially undermine the economic
efficacy of elections. We show that illicit tactics are widespread, and that they reduce
the incentive for governments to deliver good economic performance. Our analysis also
suggests that in societies with regular free and fair elections, leaders do not matter for
economic growth.
I. Introduction
When ruling elites are challenged, theyeither respond by making concessions or by increas-
ing repression. Common concession strategies are the redistribution of income and/or the
extension of voting rights (Acemoglu and Robinson, 2000, 2001, 2006). In our article, we
examine whether elections matter for economic performance. Most developing countries
now hold regular elections, because the fall of the SovietUnion triggered a ‘four th wave of
democratization’.1Autocrats conceded elections because they were faced with pressures
that otherwise threatened their retention on power. The example of the popular revolutions
of Eastern Europe temporarily lowered the cost of popular protest and the political opposi-
tion made elections their obvious goal. Further, with the end of the Cold War, donors were
no longer locked into supporting autocrats in return for their political alignment. Instead,
they could demand democratization as a condition for sustaining public finances. Even if
these pressures were only temporary, once autocrats had conceded elections they had good
reason to maintain them. Cancelling subsequent elections would have signalled a return
*Wewould like to thank the editor, Jon Temple,Jennifer Castle and two anonymous referees for helpful comments
and suggestions. Richard Stanley assisted with the data coding. All remaining errors are our own. Anke Hoeffler
acknowledges funding from the Open Society Institute.
JEL Classification numbers: O11, O43.
1For a discussion see McFaul(2002).
2Bulletin
to less open regimes, which could potentially spark coordinated protest from citizens and
donors.
However, while former autocrats were, in effect, locked in to holding elections, in many
societies the introduction of elections has not been accompanied by institutional checks
and balances preventing illicit electoral tactics. Elections, being events, are much easier to
introduce than effective checks and balances, which are processes that, at best, can only
develop over time. The standard illicit tactics of bribery, intimidation, ballot fraud and
restrictions on candidacy all favour incumbents. An incumbent might well find a stolen
election preferable to either of the alternatives. It is evident that an illicit victory might be
preferable to an honest defeat. But it might also be preferable to the outright cancellation of
elections. Illicit tactics are only ambiguously observable. Individual citizens may well be
able to detect them locally, but cannot be sure that theyhave decisively changed the national
result. Thus, for an incumbent to hold an election which is then won by illicit tactics may
improve the chances of power retention relative to outright cancellation because it dilutes
what would otherwise be a strong signal. In turn, in order to hold contested elections, albeit
marred by illicit tactics, some degree of pluralism and competition must be tolerated, but
such elections do not mark a genuine shift to democracy.As Schedler (2002) argues, they
are designed to generate only the appearance of democratic legitimacy.
In this article, we treat the scope for a leader to use illicit tactics as exogenous: some
leaders inherit norms and institutions which preclude their use whereas others are unre-
strained. This is similar to Besley and Persson (2009), who argue that initial investments
in institutions determine whether leaders are more likely to pursue a strategy of growth
promotion or rent-seeking. This dichotomy of regime types is reinforced through adverse
selection in politics. In Besley’s (2006) model, adverseselection attracts cor rupt individuals
into politics if there are small penalties for embezzlement and illicit tactics. In a companion
article (Collier and Hoeffler, 2009a), we endogenize the emergence of effective restraints.
Where a leader has the option of resorting to illicit tactics, the choice as to whether to hold
an election is of little consequence: the normal discipline of democracy is compromised.
We first present a new global data set which integrates data on changes in incumbency
with data on elections classified according to the quality of their conduct. We use this
to demonstrate that electoral misconduct is both widespread and highly efficacious for
incumbent politicians. In section III, wetur n to whetherthis matters materially for economic
performance. We show that whereas honest elections effectively discipline the economic
choices of political leaders, illicit tactics substantially reduce the incentive for politicians
to deliver good economic performance. In section IV, we investigatea testable implication.
The extent to which economic performance depends on the identity of the leader should
depend upon whether leaders are subject to honest elections.
II. The prevalence and pay-off to electoral misconduct
Elections are a necessary but not a sufficient condition for democracy. Most autocracies
hold sham elections: to be meaningful, an election has to be conducted to certain stan-
dards. Although the quality of elections is assessed by a large number of organizations, for
example, by the Organization for Security and Co-operation in Europe (OSCE),The Carter
©2013 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd

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