Direct Democracy and Local Public Finances under Cooperative Federalism

TLDR: In this paper, the authors exploited the introduction of the right of referenda at the local level in the German state of Bavaria in 1995 to study the fiscal effects of direct democracy.

Abstract: This paper exploits the introduction of the right of referenda at the local level in the German state of Bavaria in 1995 to study the fiscal effects of direct democracy. In the first part of the paper, we establish the relationship between referenda activity and fiscal performance by using a new dataset containing information on all 2500 voter initiatives between 1995 to 2011. This selection on observables approach, however, suffers from obvious endogeneity problems in this application. The main part of the paper exploits population dependent discontinuities in the signature and quorum requirements of referenda to implement a regression discontinuity design (RDD). To safeguard against co-treatments that might affect fiscal outcomes simultaneously at the same thresholds, we validate our results by extending the RDD approach to a difference-in-discontinuity (DiD) design. By studying direct legislation in an archetypical cooperative federation as Germany, our paper extends the literature to a novel institutional setting. The results indicate that in our setting - and in contrast to most of the evidence from Switzerland and the US - direct democracy causes an expansion of local government budgets.

Figures

Figure 3: Annual number of initiatives and referenda in Bavarian towns, 1995-2011

Table 6: RDD: the effect of population thresholds on government expenditure

Figure 4: Histogram of log pre-election-year population around the normalized threshold with Kernel density plots

Table 1: Revenue structure of Bavarian municipalities in 2012

Table 5: Signature and quorum requirements by population categories

Table 10: Robustness of DiD: the effect of direct democracy (1995) reform on fiscal variables at different population threshold separately

Full text

Dis cus si on Paper No. 13-038
Direct Democracy and Local Public
Finances under Cooperative Federalism
Zareh Asatryan, Thushyanthan Baskaran,
Theocharis Grigoriadis, and Friedrich Heinemann

Dis cus si on Paper No. 13-038
Direct Democracy and Local Public
Finances under Cooperative Federalism
Zareh Asatryan, Thushyanthan Baskaran,
Theocharis Grigoriadis, and Friedrich Heinemann
Download this ZEW Discussion Paper from our ftp server:
http://ftp.zew.de/pub/zew-docs/dp/dp13038.pdf
Die Dis cus si on Pape rs die nen einer mög lichst schnel len Ver brei tung von
neue ren For schungs arbei ten des ZEW. Die Bei trä ge lie gen in allei ni ger Ver ant wor tung
der Auto ren und stel len nicht not wen di ger wei se die Mei nung des ZEW dar.
Dis cus si on Papers are inten ded to make results of ZEW research prompt ly avai la ble to other
eco no mists in order to encou ra ge dis cus si on and sug gesti ons for revi si ons. The aut hors are sole ly
respon si ble for the con tents which do not neces sa ri ly repre sent the opi ni on of the ZEW.

Non-technical Summary
A growing literature studies the role of political institutions in shaping fiscal outcomes.
One of the most important differences in political institutions of established democracies
is whether fiscal and economic decisions are made by elected representatives or through
direct voting. Popular influence over policy making brings fiscal policy closer to the
electorate’s preferences. If representatives prefer more spending than voters, then direct
democracy will constrain public budgets. If voters prefer more spending than represen-
tatives, then spending will be higher in political jurisdictions that allow for more direct
democracy.
The public choice tradition argues that elected politicians tend to overspend, overtax
and to overborrow due to the interests and influence of bureaucrats or special interest
groups. Direct democracy may shift power to voters themselves thus alleviating the
principle-agent problems. Direct legislation is one mechanism of the fiscal constitution
that allows voters to constrain the Leviathan government.
In a federal setting, however, local voters may face particular common-pool disincen-
tives. If the funding of municipalities originates from a “common pool” of country-wide
and future taxpayers, voters also welcome overspending on local public goods and ser-
vices. Voter disincentives may even be larger than those of local politicians. The latter
are often constrained by the policies of their countrywide parties. This may result in
voters’ inaction against constraining the Leviathan or even incentivize them to act as a
Leviathan themselves against the taxpayers of other jurisdictions. In this case, direct
democracy increases spending.
The present paper tries to answer empirically which of the two perspectives has va-
lidity for Germany. It is the first analysis of the link between direct democracy and fiscal
outcomes for Germany, an archetypal cooperative federation. Hence, Germany offers an
intriguing case of comparison against the existing Swiss and US evidence.

The starting point of the empirical analysis exploits the variance in the use of voter
initiatives at the municipal level in the German State of Bavaria. We have information
on all initiatives since the introduction of the right of petitioning for referenda in 1995.
Using an selection on observables approach, we find a negative instantaneous correlation
between referenda activity and municipal budgets. This turns positive with increasing
lags.
To address the obvious endogeneity problems of the selection on observables approach,
the main part of the analysis exploits the discontinuous relationship between municipal
population sizes on the one hand and signatures needed to initiate referenda as well as the
minimum quorum requirements for referenda to be approved on the other. A regression-
discontinuity design can be used to identify the causal effect of direct democracy. To
safeguard against other exogenous co-treatments that might affect fiscal institutions si-
multaneously at the same thresholds, we also implement a “difference-in-discontinuity
design” by comparing the discontinuities in the pre- and post-reform periods.
The results indicate that in the German federal setting, direct democracy had a sig-
nificantly positive and robust effect on local government spending and revenue. We
estimate the size of effect to be a 10% average increase in annual per capita municipal
spending and revenue. A plausible explanation for this result diverging from the Swiss
or US experience is that the cooperative form of federalism in Germany results in strong
common-pool disincentives on part of local voters.

Das Wichtigste in K¨urze
Eine wachsende Literatur befasst sich mit dem Einfluss von politischen Institutionen auf
fiskalische Ergebnisse. Einer der wichtigsten Unterschiede zwischen den politischen Insti-
tutionen etablierter Demokratien betrifft das Ausmaß der direkten Demokratie. Der un-
mittelbare W¨ahlereinfluss sollte eine Ann¨aherung der Fiskalpolitik an die W¨ahlerpr¨aferen-
zen erzwingen. Wenn politische Repr¨asentanten ein h¨oheres Ausgabeniveau als W¨ahler
bevorzugen, dann sollte direkte Demokratie folglich die H¨ohe der ¨offentlichen Haushalte
begrenzen. Wenn W¨ahler ein h¨oheres Ausgabeniveau bevorzugen, dann sollten umgekehrt
mit direkter Demokratie gr¨oßere Budgets einhergehen.
Die Public-Choice-Tradition argumentiert, dass gew¨ahlte Politiker ¨uberm¨aßige Aus-
gaben, Steuern und Schulden pr¨aferieren, weil sie von B¨urokratie und von Interessengrup-
pen beeinflusst werden. Direkte Demokratie verlagert dann die Macht zu den W¨ahlern
selber, was Prinzipal-Agenten-Probleme abmildert. Direkte Gesetzgebung w¨are somit ein
Mechanismus in einer Fiskalverfassung, der den W¨ahlern hilft, den Leviathan zu begren-
zen.
In einem f¨oderalen Umfeld sind W¨ahler aber m¨oglicherweise besonderen Common-
Pool-Fehlanreizen ausgesetzt. Wenn die Finanzierung von Kommunen aus einem gemein-
samen Topf (dem “Common Pool”) landesweiter und zuk¨unftiger Steuereinnahmen herr-
¨uhrt, dann werden auch W¨ahler ein ¨uberm¨aßiges Ausgabenniveau zur Bereitstellung
lokaler ¨offentlicher G¨uter und Dienstleistungen bevorzugen. Diese Fehlanreize k¨onnen
auf der Ebene der W¨ahler sogar gr¨oßer als auf der Ebene der lokalen Politiker sein.
Letztere sind oft durch die Politiken landesweiter Parteien restringiert. Das kann dazu
f¨uhren, dass W¨ahler kein Interesse an der Begrenzung des Leviathan haben oder sogar
selber als Leviathan zu Lasten der Steuerzahler anderer Jurisdiktionen auftreten. Dann
w¨urde direkte Demokratie ausgabesteigernd wirken.
Dieses Papier versucht f¨ur Deutschland empirisch zu beantworten, welche dieser bei-
den Sichtweisen stichhaltig ist. Es ist der erste Beitrag zum Einfluss von direkter Demokra-

Frequently asked questions

Q1. What are the contributions in "Direct democracy and local public finances under cooperative federalism" ?

This paper exploits the introduction of the right of referenda at the local level in the German state of Bavaria in 1995 to study the fiscal effects of direct democracy. In the first part of the paper, the authors establish the relationship between referenda activity and fiscal performance by using a new dataset containing information on all 2500 voter initiatives between 1995 to 2011. The main part of the paper exploits population dependent discontinuities in the signature and quorum requirements of referenda to implement a regression discontinuity design ( RDD ). To safeguard against co-treatments that might affect fiscal outcomes simultaneously at the same thresholds, the authors validate their results by extending the RDD approach to a difference-in-discontinuity ( DiD ) design. 

Q2. What have the authors stated for future works in "Direct democracy and local public finances under cooperative federalism" ?

This paper is the first systematic study to extend the discussion on the link between direct democratic institutions and local-level public finances to the case of German municipalities. The authors use the variation stemming from this legal intervention and apply a quasi-experimental design to study the causal effects of the ease of initiating and implementing popular referenda in a large sample of 2099 Bavarian municipalities over the period from 1978 to 2011. Since the sizes of the municipalities are approximately randomized at the exogenously assigned population thresholds, the authors can use a regression discontinuity design to study the effect of the variation of direct democratic legislation on otherwise identical municipal fiscal outcomes. 

Q3. What is the starting point of the empirical analysis?

The starting point of the empirical analysis exploits the variance in the use of voter initiatives at the municipal level in the German State of Bavaria. 

Q4. What is the main part of the analysis used to address?

To address the obvious endogeneity problems of the selection on observables approach, the main part of the analysis exploits the discontinuous relationship between municipal population sizes on the one hand and signatures needed to initiate referenda as well as the minimum quorum requirements for referenda to be approved on the other. 

Q5. What is the purpose of the paper?

To safeguard against other exogenous co-treatments that might affect fiscal institutions simultaneously at the same thresholds, the authors also implement a “difference-in-discontinuity design” by comparing the discontinuities in the pre- and post-reform periods. 

Q6. What is the plausible explanation for this result?

A plausible explanation for this result diverging from the Swiss or US experience is that the cooperative form of federalism in Germany results in strong common-pool disincentives on part of local voters.