Direct Democracy and Local Public Finances under Cooperative Federalism

• Author: Zareh Asatryan,Theocharis Grigoriadis,Thushyanthan Baskaran,Friedrich Heinemann
• Published date: 01 July 2017
• DOI: http://doi.org/10.1111/sjoe.12169
• Date: 01 July 2017

Scand. J. of Economics 119(3), 801–820, 2017

DOI: 10.1111/sjoe.12169

Direct Democracy and Local Public

Finances under Cooperative Federalism∗

Zareh Asatryan

ZEW Mannheim, DE-68161 Mannheim, Germany

asatryan@zew.de

Thushyanthan Baskaran

University of G¨

ottingen, DE-37073 G¨

ottingen, Germany

tbaskar@uni-goettingen.de

Theocharis Grigoriadis

Free University of Berlin, DE-14915 Berlin, Germany

theocharis.grigoriadis@fu-berlin.de

Friedrich Heinemann

ZEW Mannheim, DE-68161 Mannheim, Germany

heinemann@zew.de

Abstract

In this paper, we exploit the introduction of the right of local initiatives in the German state

of Bavaria in 1995 in order to study the fiscal effects of direct democracy. Our identification

strategy combines difference-in-differences and regression discontinuity methods: we com-

pare municipal expenditure and revenue between pre- and post-reform periods at population

thresholds where the signatures needed to launch initiatives and minimum quorum require-

ments decrease discontinuously (difference-in-discontinuities design). The results suggest that

stronger direct democratic institutions lead to an expansion of local government size.

Keywords: Bavaria; fiscal policy; local initiatives; regression discontinuity

JEL classification:D72; D78; H70

I. Introduction

In the last few decades, many countries have complemented their rep-

resentative forms of government, where political decisions are generally

made by elected legislatures and executives, with elements of direct democ-

racy (Matsusaka, 2014). While it is plausible that a direct involvement of

∗We are grateful to Lars P. Feld, Ronny Freier, Benny Geys, Gebhard Kirchg¨

assner, John

Matsusaka, and seminar participants at ZEW Mannheim, University of Mannheim, Walter

Eucken Institute, European Public Choice Society, International Institute of Public Finance,

European Economic Association and Asian Econometric Society for valuable comments.

CThe editors of The Scandinavian Journal of Economics 2015.

802 Direct democracy and local public finances under cooperative federalism

citizens in political decision-making has significant f iscal consequences, it

is ex-ante unclear whether direct democracy leads to larger or smaller gov-

ernments. An intuitive conjecture is that fiscal policies are closer to major-

ity preferences in political jurisdictions that have powerful direct democratic

institutions (Gerber, 1996). Thus, whether more direct democracy causes

higher or lower spending and revenue will depend on the size of govern-

ment preferred by the median voter and on the extent to which elected

politicians systematically deviate from the median voter’s preferences.

On the one hand, most empirical studies at the state level in the US

suggest that direct democracy – the existence of an initiative process –

has reduced spending and (tax) revenue, at least in the second half of

the 20th century (Matsusaka, 1995, 2000; Bails and Tieslau, 2000; Besley

and Case, 2003; Blomberg et al., 2004). On the other hand, studies at the

level of US cities find that direct democracy leads to larger governments

(Matsusaka, 2004; Primo, 2010). Studies from Switzerland reach similar

conclusions: cantons with (more pronounced) direct democratic legislation

have lower spending, but there are also interactions with the local level

(Pommerehne, 1978; Feld and Kirchg¨

assner, 2001; Feld and Matsusaka,

2003; Funk and Gathmann, 2011; Galletta and Jametti, 2012). Beyond the

US and Switzerland, only a few other countries have been studied. One

notable example is Hinnerich and Pettersson-Lidbom (2014) who propose

that direct democracy led to lower public spending at the turn of the 20th

century in Sweden.

We contribute to the literature by studying this issue with a credible

identification strategy and in a novel institutional setting. The setting is the

federal state of Bavaria, where direct democracy was introduced through a

constitutional change in 1995 and where between 1995 and 2011 around

2,500 petitions1were recorded.2That is, we analyze the experience with

direct democracy in 2,056 Bavarian municipalities using a sample that

covers the period from 1983 to 2011.

We make three contributions to the literature. First, the sample of Ger-

man municipalities is an important extension to the existing research pro-

gram, which has been dominated by studies from Switzerland and the US.

A distinctive feature of the German context is the cooperative nature of

Germany’s fiscal federalism, notably the extensive inter-jurisdictional rev-

enue sharing (Baskaran, 2012). Direct democracy can have substantively

1The Bavarian local government code uses different terms to refer to direct democratic leg-

islation. These terms depend on the stage of the process. We adopt the following convention.

We use “petition” to refer to all attempts to initiate direct democratic legislation. Petitions that

receive sufficient signatures and are therefore put to a vote, we call “initiatives”. Obviously,

not all petitions become initiatives. See below for more details.

2The reason to concentrate on the case of Bavaria is that, compared to other German L¨

ander,

local-level direct democratic institutions in Bavaria are by far the strongest.

CThe editors of The Scandinavian Journal of Economics 2015.