Voter preferences, direct democracy and government spending
Introduction
Direct democracy has become increasingly popular in recent decades as a complement for purely representative forms of government. In the United States, for instance, more than two-thirds of the population live in a state or city with popular initiative or referendum (Matsusaka, 2004). Direct democracy has also been popular in Europe, especially in Switzerland, and increasingly so in Eastern Europe. Most recently, several European countries have held referendums on the new European constitution.
An important motivation for granting direct democratic participation rights is to give citizens more weight in the political process. Its attractiveness is rooted in part in the belief that direct voter control would improve the quality of policy-making or could slow down the rapid growth in government spending observed over the past decades. Theoretical models only predict that direct democracy brings policies closer to the preferences of voters (Gerber, 1996, Besley and Coate, 2008). How direct democracy will affect policy outcomes then depends crucially on the policies preferred by the majority of voters.
In this paper, we make use of unique ballot data from Switzerland to provide direct evidence on voter preferences and how they vary across time and space. Since Switzerland allows for referendums and initiatives at the federal level, we can use data from federal ballot propositions to characterize voter preferences in each canton. Our setting has a number of attractive features: Switzerland is the world leader in the use of direct democracy. At the federal level alone, 331 votes have been held between 1950 and 2000. The ballots cover a wide variety of policy proposals from taxes, environmental policy, immigration, to membership in international organizations or subsidies for agriculture. The data on ballot votes therefore provides a rich source to characterize voter preferences expressed at the ballot box over several decades.
Even more importantly, voter preferences are comparable across cantons as all citizens vote on the same federal ballot. We can therefore analyze for the first time how preferences expressed at the ballot box vary across states with alternative political institutions, e.g. different direct democratic institutions. Our results show that voter preferences differ substantially across cantons. Cantons with strong direct democratic institutions have a substantially lower taste for government spending. Further, our measures of voter preferences can better capture heterogeneity in preferences than observable state characteristics (like the strength of left-wing parties, for example).
That preferences vary by political institutions seems to be a more general phenomenon visible also for alternative measures of preferences and other political institutions. Several channels could account for this correlation between institutions and voter preferences. Most plausibly, direct democracy has been adopted by fiscally conservative voters in the past. If political preferences are transmitted across generations (see Dohmen et al., 2012 for evidence), citizens will be more fiscally conservative in cantons with stronger direct democracy today. Selective migration of citizens with preferences for low government spending to cantons with stronger direct democracy would be a second explanation for preference heterogeneity across cantons. Finally, preferences of the electorate might be strengthened by institutions through positive experiences, for instance. We provide suggestive evidence that voter preferences are important determinants of institutional change as shifts in preferences precede reforms in political institutions. We find little support, in contrast, that institutional reforms shift preferences in the years after reforms. Hence, preferences appear to shape institutional choice in our context.
To illustrate the importance of voter preferences for the study of political institutions, we reconsider the relationship between direct democracy and fiscal policy in Switzerland. We hereby focus on the mandatory budget referendum, which gives citizens direct control over cantonal spending. If a canton has a mandatory budget referendum in place, voters need to approve individual projects that exceed a monetary threshold. In other cantons, citizens cannot decide on expensive projects or need to collect signatures to call an optional referendum.1 Budget referendums in Switzerland are similar to tax and expenditure limitations in the United States which require the electorate in some states to approve tax increases or expenditure growth above a certain threshold. They are also similar to budget approvals for local school districts in the United States and other countries. In our analysis, we classify cantons as strongly direct democratic with respect to fiscal policy if their constitutions require a mandatory budget referendum. In contrast, cantons with only an optional or no budget referendum have a weaker or no direct influence on fiscal policy.
We find that a stronger taste for government in a canton is associated with substantially higher cantonal spending and revenues per capita. Many previous studies, in contrast, found no discernible effect of proxies for voter preferences on spending.2 As we argue, the most likely explanation for this pattern is that previously employed proxies for voter preferences are insufficient. Furthermore, our estimates suggest that the impact of the mandatory budget referendum on spending declines by about 20% once we control for voter preferences.
The rest of this article is structured as follows. The next section briefly discusses the related literature. Section 3 provides background information on the structure of direct democracy and fiscal policy in Switzerland. The data are described in Section 4. Estimates of voter preferences are reported in Section 5, while Section 6 shows how institutional rules and voter preferences affect fiscal policy. Finally, Section 7 concludes.
Section snippets
Related literature
Our article contributes to a sizable literature, especially in political science, on the estimation and analysis of voter preferences. Most studies rely on public opinion polls or survey data to study aggregate voter preferences (for example, Ansolabehere et al., 2008, Camobreco, 1998, Erikson et al., 2002, Evans et al., 1996, Gerber, 1996, Lascher et al., 1996).3
Institutional background
Switzerland has a strong federalism where all political responsibilities remain with the canton unless they were granted to the federal government in a national referendum. Cantons thus play a dominant role in the provision of public services, the redistribution of wealth and as a generator of revenues.
In 1998, 34% of all government spending was undertaken by cantons, 39% by the federal and 27% by local governments. Cantons spend 50% of all education, 60% of health and around 27% of social
Data on federal ballots
To characterize voter preferences and the demand for government, we use the fact that Switzerland has direct democratic participation also at the federal level. Citizens can initiate a partial or total revision of the federal constitution, which allows for a wide variety of policy proposals. In addition, any changes to the federal constitution and international treaties need to be approved by voters in a mandatory referendum. Further, all federal laws and executive orders are subject to an
Votes with fiscal consequences
Our first approach starts from the official documents prepared by the federal government, which are distributed to each citizen before the vote. The documents contain the arguments for and against a proposition, an overview of the parliamentary debate (if any) and outside opinions by interest groups. Most importantly for our purposes, they also outline the fiscal consequences for expenditures, taxes or subsidies if the ballot gets approved.14
Voter preferences, direct democracy and fiscal policy
Our results thus far show that voters in cantons with a mandatory budget referendum are more fiscally conservative than voters without such a referendum. We would expect that this preference heterogeneity is reflected in public spending because voters elect more fiscally prudent representatives, or, because electoral competition forces politicians to spend less, for example. We would also conjecture that controlling for voter preferences will reduce the correlation between direct democracy and
Conclusion
This article outlines an empirical strategy to estimate voter preferences from ballot data. Using data on all federal propositions held in Switzerland between 1950 and 2000, we recover aggregate policy preferences in each canton. Our evidence suggests that preferences vary a lot across cantons; these differences are persistent, but also evolve slowly over time. We also find that voter preferences are systematically correlated with political institutions. In particular, citizens in cantons with
Acknowledgments
A previous version of this paper has been circulated as “Estimating the Effect of Direct Democracy on Policy Outcomes: Preferences Matter!” We thank two anonymous referees, Doug Bernheim, Henning Hillmann, Albrecht Ritschl, Barry Weingast, Donald Wittman, participants at the Public Choice Meeting, the Silvaplana Workshop in Political Economy, Humboldt University of Berlin, University of Washington, Warwick, Utrecht and Geneva for many helpful comments. We are grateful to Werner Seitz and
References (31)
- et al.
Direct democracy, political culture, and the outcome of economic policy: a report on the Swiss experience
European Journal of Political Economy
(2000) - et al.
Budget referendums and government spending: evidence from Swiss cantons
Journal of Public Economics
(2003) Supermajority voting requirements for tax increases: evidence from the states
Journal of Public Economics
(2000)- et al.
Auditors and fiscal policy: empirical evidence on a little big institution
Journal of Comparative Economics
(2010) Does right or left matter? Cabinets, credibility and fiscal adjustments
Journal of Public Economics
(2004)- et al.
The strength of issues: using multiple measures to gauge preference stability, ideological constraint, and issue voting
American Political Science Review
(2008) - et al.
Political institutions and policy outcomes: evidence from the United States
Journal of Economic Literature
(2003) - et al.
Issue unbundling via citizens' initiatives
Quarterly Journal of Political Science
(2008) - et al.
Does tax competition tame the leviathan?
- et al.
Bootstrap-based improvements for inference with clustered errors
Review of Economics and Statistics
(2008)
Preferences, fiscal policies, and the initiative process
Journal of Politics
The intergenerational transmission of risk and trust attitudes
Review of Economic Studies
The Macro Polity
Measuring left–right and libertarian–authoritarian values in the British electorate
British Journal of Sociology
The impact of citizen influence on local government expenditures
Public Choice
Cited by (37)
Participatory budgeting and the pattern of local government spending: Evidence from South Korea
2023, European Journal of Political EconomyForm of government and voters’ preferences for public spending
2021, Journal of Economic Behavior and OrganizationCitation Excerpt :This is mainly driven by an increase in administrative and education spending.10 Funk and Gathmann (2013) is the closest papers to this work. The authors analyze the association between direct democratic institutions and Swiss people’s preferences regarding public expenditure using the results of national referendums and initiatives.
The initiative, referendum, and distribution of income: An empirical analysis of Swiss cantons
2021, European Journal of Political EconomyCitation Excerpt :Since all cantons vote on the same issue in a national ballot, the preferences revealed by cantonal voting results are directly comparable. Funk and Gathmann (2013) construct three factors to account for cantonal fiscal preferences over a sequence of 10-year periods. Panel B of Table 5 shows the results when including these factors as additional controls.
Mapping the theory of political representation to the empirics: An investigation for proportional and majoritarian rules
2020, Journal of Comparative EconomicsDoes tax competition tame the Leviathan?
2019, Journal of Public EconomicsFiscal decentralization and government size: The role of democracy
2019, European Journal of Political Economy