European Journal of POLITICAL
,*memmmm. European Journal of Political Economy ECONOMY
ELSEVIER Vol. 16 (2000) 287-306
Direct democracy, political culture, and the outcome of economic policy: a report on the Swiss experience
Lars P. Feld, Gebhard Kirchgässner *
University of St. Gallen, SIAW-HSG, Institutsgebäude, Dufourstr. 48, St. Gallen CH-9000, Switzerland Received 1 April 1999; received in revised form 1 October 1999; accepted 1 November 1999
Abstract
Political culture in Switzerland is, to a large extent, influenced by its direct democracy. Compared to purely representative systems, direct democracy leads to a different type of communication among citizens and also between citizens and representatives. The opportunity of deciding for themselves on political issues provides citizens with incentives to collect more information. Because citizens are better informed, politicians have less leeway to pursue their personal interests. As a consequence, public expenditure and public debt are lower when citizens enjoy direct democratic rights. Citizens also feel more responsible for their community: tax evasion is lower in direct than in representative democratic systems. ©2000 Elsevier Science B.V. All rights reserved.
JEL classification: H10; H30
Keywords: Direct democracy; Public expenditures; Public debt
1. Introduction
Direct democracy has a considerable impact on Swiss political culture. Two political decisions illustrate that: on November 29, 1989, Swiss citizens voted on
* Corresponding author. Tel.: +41-71-224-23-47; fax: +41-71-224-22-98. E-mail address: gebhard.kirchgassner@unisg.ch (G. Kirchgässner).
0176-2680/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0176-2680(00)00003-3
288    L.P. Feld, G. Kirchg'dssner / European Journal of Political Economy 16 (2000) 287-306
an initiative that proposed to abolish the Swiss army. The initiative campaign was accompanied by an intensive and (partly) heated discussion among citizens. Until that time, the Swiss army was supposed to be perceived as a sacred institution not only by politicians, but also by voters. Indeed, the whole Swiss political establishment, all politically relevant parties and interest groups, the federal government and both chambers of the Swiss parliament suggested rejecting this initiative. Only the Social Democratic Party and the trade unions abstained from taking a position. Due to this unified attitude of the political establishment and the traditional support of Swiss citizens for their army, an overwhelming rejection of the initiative was expected. The actual result of the ballot came as a surprise: although the initiative was rejected, accompanied by a relatively high voter turnout of 69.2%, some 35.6% voted for the abolition of the army. This result shook the Swiss self-conception. Soon afterwards, some concessions for opponents of the army, like the introduction of a community service as an alternative to military service, were considered. 1
Already in 1979, Swiss citizens voted on a possible exit from nuclear energy. The proposed initiative 'for the preservation of popular rights and the security for the construction and operation of nuclear power plants' was marginally rejected by the people on February 19, 1979, with a majority of 51.2%.2 However, this initiative led to public discussion about the risk of nuclear energy that was unprecedented in European countries at that time. In Germany, e.g., such discussion was largely prevented because of the unified support for nuclear power of all German parties, employer organisations and trade unions. Environmental policy in Switzerland is, at least in some respects, still supposed to be exemplary today compared to that of its European neighbours. Obviously, the Swiss political system in this case reacted faster to the needs of its citizens than the systems of neighbouring countries with (pure) representative democracies.3
At a first glance, these two cases do not seem to illustrate that citizens' preferences carry through more easily against the position of the political establishment in direct, than in representative democracy, as, e.g., Pommerehne (1978)
1 Thus, the Swiss army received considerable support from Swiss voters, but the relatively strong and unexpected minority vote led to a change in defence policy.
2 See also Aubert (1984, p. 135).
3 Of course, it has to be considered that environmental policy in Germany is partly hindered by the European Union (EU) regulation. It remains open, however, to what extent this reason is a facade in order not to have to take certain decisions. See, e.g., the discussion about introduction of three-way catalytic devices for cars in Germany between 1983 and 1985. In other cases, such as the compromise of the EU member countries on the directive of used cars decided on July 22, 1999, Germany wanted a less restrictive legislation than its partners. The Council decided by qualified majority against Germany that the automobile industry would have to dispose of cars delivered to customers from the year 2006 without imposing additional costs on customers.
L.P. Feld, G. Kirchg'dssner / European Journal of Political Economy 16 (2000) 287-306 289
has shown for Switzerland, or Gerber (1996a, 1999) for the United States.4 Indeed, both initiatives were rejected by voters with a majority following the proposal of the political establishment. It is important, however, to look not only at the decision that was finally adopted, but also to consider the impact of the political discussion among citizens preceding that decision. The political discourse and the accompanying exchange of arguments not only improve general information about a subject, but can lead to changes in the evaluation criteria by which individuals judge policies. Both initiatives mentioned above had a strong impact on future political decisions. The importance of the Swiss army was continually reduced afterwards, and Swiss environmental policy became one of the most progressive in Europe. Both initiatives are also examples of how direct legislation can increase the information of representatives about constituents' preferences.5
The role of discussion has been largely neglected in political economic analyses. Even more forcefully, political discussion is characterised as 'cheap talk'.6 According to this view, there is little opportunity for credible transmission of information in political debates. Unless individuals have sufficiently similar interests, no credible communication occurs. A different view is expressed by Habermas (1981) in his Theory of Communicative Action. He emphasises the positive role of discussion and distinguishes two mechanisms for co-ordinating social actions: strategic interaction, where one person seeks to influence another by means of threatened punishment or promised reward (the typical representation in political economic models), in contrast to communicative interaction, where people seek to convince each other by means of rational arguments. Such communicative interaction is a form of deliberation.
Decisions in direct legislation and in representative government can be expected to be differently influenced by this kind of deliberation. Compared to purely representative systems, direct democracy leads to a different type of communication among citizens, but also between citizens and representatives. The possibility of deciding for themselves on basic political issues provides incentives for citizens to collect more information. This information is not only needed to make a 'correct' decision at the ballots, but also in order to participate in the dialogue with other citizens. Thus, to be well informed about political issues is not only a voluntary contribution to a public good, but has also at least some characteristics of a private good. Moreover, as citizens are better informed in direct, than in purely representative democracies, politicians have less leeway to pursue their personal interests instead of those of the general public. Finally, citizens feel more responsible for their community.
4 See also Romer and Rosenthal (1978, 1979a) or Moser (1996).
5 See Matsusaka (1992) or Matsusaka and McCarty (2000) for this.
6 See, e.g., Riker (1982), Austen-Smith and Riker (1987), or Austen-Smith (1990a, 1990b).
290    L.P. Feld, G. Kirchg'dssner / European Journal of Political Economy 16 (2000) 287-306
In this paper, the interaction of deliberation and direct legislation by the people is considered as a basic feature of Swiss political culture. After a discussion of the importance of deliberation in direct legislation and its impact on mediating information (Section 2), we briefly analyse the role of direct legislation for the principal-agent relationship between representatives and their constituency (Section 3). In Section 4, we survey empirical results on the impact of direct democracy in Switzerland. We show that public expenditure as well as public debt are lower when citizens possess direct democratic rights, so indicating the reduced discretion of politicians. We also show that tax evasion is — ceteris paribus — higher in purely representative systems, indicating a higher responsibility of citizens for their community. We conclude with some final remarks in Section 5.
2. Direct legislation and deliberation
In direct legislation, the discussion preceding the decisions at the ballots is institutionalised at the level of the whole society instead of that of the parliament or an elitist circle that is called by representatives. In this discourse, each individual and particularly each interest group is able to present his/her and its positions, perspectives and arguments to proposed bills to all citizens by using the media.7 Direct legislation can thus be understood as an institutionalised form of deliberation combined with a final voting decision.8
The discourse that takes place in Switzerland preceding a voting decision at the ballots is characterised by four features.9
(i) Drafted bills are considered relative to available alternatives, in most cases to the status quo. The cost of implementing a policy is an important part of the discussion. Both result in an exchange of arguments that does not remain general and non-binding.
(ii) The discourse is the more intense, the more important a proposal is, i.e. the more citizens of respective jurisdictions, communities, cantons (states) or the federal level, feel concerned by a proposal. If some referenda or initiatives are considered to be of minor importance, there will be not very much discussion. If, on the other hand, a referendum or initiative is judged to be highly relevant, such as, e.g., the ones on the partial abandonment of Swiss sovereignty by joining inter-or supranational organisations, a more fundamental and extensive exchange of
7 For a discussion of the role of the new media in direct legislation, see Kirchgassner and Frey (1994) and Budge (1996).
8 It is important to note that this assessment only holds with respect to binding referenda on single issues, and not for advisory referenda, which citizens know representatives can easily neglect without any sanction, or for plebiscites like the ones conducted by the French President de Gaulle, which are only conducted to secure legitimacy for a presidential policy opposed by parliament.
9 See also Frey and Kirchgassner (1993, p. 139).
L.P. Feld, G. Kirchg'dssner / European Journal of Political Economy 16 (2000) 287-306 291
arguments occurs. The discourse preceding a decision is, therefore, not mechanically predetermined, but changes according to the importance of an issue.
(iii) Many different organisations such as political parties or interest groups, but also single individuals, participate in the discourse. Moreover, the discourse takes place and is conducted at different levels of debating culture. It is not restricted to intellectual circles, but also takes place in sporting clubs and in bar rooms. Also, the participation of citizens varies considerably with respect to the perceived salience. At the federal level, the average voter turnout in the 1970s and 1980s was 41.5%.10 If, however, decisions that are perceived to be salient, such as the referenda on joining the UN, on the abolition of the Swiss army, on joining the European Economic Area (EER) and the increase of mineral oil taxes illustrate, voter turnout can become considerably higher.11
(iv) In the course of this discussion, at least in parts of the citizenry, a learning process occurs. Since many citizens are confronted with the arguments of both sides, those opposing and those favouring a certain policy outcome, they are induced to consider each proposal anew. This can lead to a revision of their position. Therefore, the possibility also emerges that citizens examine the extent to which their preferences generalise. This is the deliberative element in this discourse.
Even after the most intense discourse, unanimity in political decisions cannot be expected, because most political decisions lead to a redistribution of income. The outcome is thus rather a compromise, in the sense of Habermas (1973, p. 155), than a consensus.12 Realistically, however, more than that is not obtainable if a policy is to be socially chosen among a set of 'Pareto efficient' alternatives. This does not imply that citizens will decide only on the basis of their particular self-interest. The possibility, that the discourse preceding the decision at the ballots may lead to a revision of preferences over policies, plays a special role. Moreover, citizens may act more altruistically in referenda and initiatives than in economic decisions.13 For example, Pommerehne and Schneider (1985) presented empirical evidence that Swiss people vote for redistributional programmes despite the fact that these result in a loss in their own wealth. In decisions at the ballots, voters, therefore, may follow an ethical position to a larger extent than could be
I See Statistisches Jahrbuch der Schweiz (1993, p. 376).
II The respective figures for voter turnout are: joining the UN (March 16, 1986): 50.7%; abolition of the Swiss army (November 26, 1989): 69.2%; joining the EER (December 6, 1992): 78.3%; increase of mineral oil taxes (March 7, 1993): 54.6%. In general, voter turnout was above 50% in referenda and initiatives that were controversial, such as traffic, immigration, or security policies. See also Eschet-Schwarz (1989).
12 For an assessment of this discourse with respect to discourse ethics, see Frey and Kirchgässner (1993).
13 This can also hold in voting decisions. See Kirchgässner (1992, 1996) or Brennan and Lomasky (1993).
292    L.P. Feld, G. Kirchg'dssner / European Journal of Political Economy 16 (2000) 287-306
expected on the basis of the preferences that guide their behaviour in the market place. This effect can be traced back to the discourse preceding a decision at the ballots. At least a temporary revision of citizens' evaluations of policies is feasible, increasing the probability that common interests will guide the social choice.14
Moreover, the referendum discourse goes beyond a revision of individual positions in the direction of social consensus. Two other arguments are important as well. First, when drafting a bill, government and parliament have to consider the interests of the whole population to a larger extent in a referendum democracy than in pure parliamentary or presidential democracies in which a government, perhaps with a slim majority, mainly follows the interests of its constituents. The possibility in itself that voters may propose a referendum on drafted bills forces the government to adopt a more consensual position.15 Second, the referendum discourse enhances citizens' information on the different arguments for and against certain political decisions. Better information increases the acceptance of policy outcomes by losing minorities, as compared to decisions that are imposed on them 'from above'. Moreover, the positions of minorities are considered to a larger extent in the political administrative process because they are publicly clarified and it becomes obvious that a (considerable) proportion of the population supports the minority position.16
In addition, it can be presumed that citizens' willingness to bear costs of information acquisition is higher in direct legislation than in representative democracies. The reason for this is that it can become privately important for citizens to be well informed about political issues that are going to be decided upon. Such a situation emerges if other citizens belonging to one's social network expect a citizen to be well informed and if the disappointment of this expectation leads to a loss of prestige.17 Nevertheless, information has to be presented in a way that it may be acquired by citizens without high costs in order not to make the citizens, who are willing to bear higher costs, worse off in a direct democracy. Thus, ideologies can play a role as an information-saving device in referenda and initiatives as well, and, like in elections, 'wrong' arguments may occasionally win. It is important, however, that the arguments in the referendum discourse relate to
14 Besides the information theoretic explanation, the evidence that redistributional issues rather than pure procedural and legal decisions are more often decided in initiatives in the US states (see Matsusaka, 1992) also supports the deliberative perspective of direct legislation.
15 See Gerber (1998).
16 Consider again the initiative on the abolition of the Swiss army mentioned in the introduction, which led to a change in policies although it was rejected by the voters. See also again Matsusaka and McCarty (2000).
17 Swank (1999) argues that, when individuals care about prestige, herd behaviour may occur. With herd behaviour, people participate in debates, but these debates provide little information. Up to now there is, however, no indication that herd behaviour impedes the information supply process preceding referenda or initiatives in Switzerland.
L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306 293
questions of specific political issues and not only to people and methods. It is therefore insufficient to motivate citizens by an ideological position exclusively. Such an assessment becomes obvious if supporters of a party vote against the position adopted by their party in the referendum campaign, something that quite often occurs in Switzerland.
Considering that a lot of information is necessary to make 'correct' decisions about political issues, it may be supposed that decisions in a representative democracy are — on average — of higher quality than those under direct legislation. Representatives have stronger incentives to gather information about political issues than citizens.18 However, one should not adopt a Nirvana approach and compare the actual situation in a direct democracy with an ideal representative system. Actually, representatives in parliaments, at least in the parliamentary systems in Europe, are often badly informed about political issues that they do not regard as their expertise. There is division of labour: decisions are predetermined in committees by a small number of specialised and informed legislators. Nearly exclusively, representatives in the parliament follow the positions of their specialists in the committees, if they are present at all. An open discussion rarely takes place. In most cases, an obligation to vote in accordance with party policy prevails: deviations from party positions lead to sanctions that can include the end of a political career. These characteristics of party politics result in sterile debates in parliament. Nobody who is present in parliament has to be convinced of political positions, because the result has been determined previously. Speeches of representatives serve as a means for representation within the party and — particularly if they are broadcast on television — for external representation. But also in the latter case, nobody has to be convinced by a speaker, because voters have no direct influence on single political decisions.
The consequence of these characteristics of parliamentary decisions is that the population hardly discusses important political issues. For example, the question of whether Germany should ratify the Maastricht Treaty was discussed neither before nor after the rejection of this Treaty by the Danish population, although Germany relinquished its sovereignty over monetary policy. Since obvious signs existed that the Treaty would not have a majority in the German population, the
The stronger control of representatives by citizens in a referendum democracy may thus induce policymakers to make less effort to acquire information than in parliamentary democracies because they have to give up discretion to the principal. This argument has been put forward by Aghion and Tirole (1997) in a general model of organisations, and by Feldmann (1998) in the context of direct democracy. It only holds, however, if parliaments are not involved in the final political decision, as is the case in the Californian initiative process. On the contrary, representatives in the Swiss parliament have an incentive to stay informed because they, first, have the possibility of proposing a counterproposal to a popular initiative, and, second, due to roll call votes, voters see how the representatives from their constituency have voted. In the latter case, representatives can more easily be held responsible than in (purely) parliamentary systems such as the United Kingdom or Germany, where members of parliament are expected to strictly follow their party line.
294    L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306
representatives, who supported the Treaty nearly unanimously, had no incentive to start a discussion about it. Moreover, they saw no reason why they should convince the population of the necessity of the Maastricht Treaty. The arrogance with which the German chancellor Helmut Kohl and the French president Fran§ois Mitterand reacted to the rejection of the Treaty by the Danish people on June 2, 1992, illustrates that these representatives were not willing to take the citizens' preferences seriously.
This example shows that not only the demand but also the supply of information is different in direct democracies as compared to purely parliamentary systems. Of course, the supply of information is no less important than the demand: only that amount of information can be used that is supplied. It should be relatively uncontentious in an open society that open access to political information should be provided to citizens, or that competition between suppliers of information is of great importance. Mass media supply information at almost no cost. This holds both for representative and direct democracies. In political systems with direct legislation, an essential additional element occurs: when a referendum takes place, representatives and the interest groups involved have an essential interest in informing voters.19 Representatives, usually belonging to a certain party, are obliged to explain to citizens why they oppose or favour a certain bill. Interest groups, as far as they are affected by a decision, attempt to influence citizens in their favour. Both is possible only on the basis of information on the issue to be decided.
Thus, not only the demand, but also the supply of political information is higher in direct democratic systems than in representative democracies. Of course, interest groups will attempt to influence voters in their favour, which results in biased information supply. But only in the case of an information monopoly is systematic misinformation of citizens to be expected. A political issue usually involves different interest groups with different political positions, and a discussion takes place in the weeks preceding the vote, in which the political parties attempt to refute the arguments of their opponents. The danger of misinformation is therefore limited in referendum campaigns.20
Comparing the constitutional reality in a parliamentary system such as Germany with the Swiss 'half-direct' democracy, parliamentary representatives may thus be expected to be better informed than Swiss citizens on average. The higher average information level in representative democracies is, however, obtained by a strongly asymmetric distribution of expertise among a few representatives in
19 Letterie and Swank (1997) show in a theoretical model that when the implementation of policies needs approval, the policymaker has an incentive to supply information.
20 This process of information provision in direct legislation is analysed in Lupia (1994) and Bowler and Donovan (1998). Although Lupia and McCubbins (1998) generalise these insights beyond direct legislation, they make no attempt to compare representative and direct democratic systems of political decision making.
L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306 295
committees. Ordinary representatives do not have better information on political issues. A crucial advantage of representative democracy as compared to direct legislation is thus lost while representative systems additionally have the disadvantage of a less informed public.21 Swiss citizens are better informed about political issues than, e.g., German voters, because the institutions of direct legislation provide stronger incentives for the government, the parliament and interest groups to supply information, as well as for citizens to gather information about political
22
issues.
The specialists of the different parties should be quite well informed about political issues. They are able to take decisions with more reflection and with better information than ordinary people. In a direct democratic system, these specialists not only provide information to their parliamentary colleagues, but as well to citizens who make the decision at the ballot, because they are usually involved in the public discourse preceding the referendum. Thus, their knowledge is also an input in direct legislation. In pure representative systems, the question additionally emerges whether these specialists have sufficient incentives to consider the (general) interests of citizens, or whether they attempt to follow their own (partial) interests or that of their constituency. They of course often represent the interests of particular groups on whose behalf they occupy a seat in parliament. Modern political economy shows that representatives have incentives to deviate systematically from the interests of the population, in order to capture rents for themselves and their constituencies.23 Representative democracies also give rise to incentives to produce political business cycles.24 These considerations raise at least some doubts as to whether the 'more informed' political decisions in representative democracies made by specialists are 'better' than political decisions in direct legislation, in which proposals of specialists have to get an approval by voters.
3. Direct legislation as a device to control and sanction representatives
Section 2 shows that discussion preceding a decision at the ballots may lead to a revision of citizens' positions from self-interest to more general common interests. Discussion enhances citizens' information about political issues. Knowledge of voters in direct democracy can even be better than that of an ordinary legislator in a parliament. This information problem is one aspect of a fundamental
21 For a comparison of both systems, see also Kirchgässner and Pommerehne (1992).
22 See Nawiasky (1951, 1953).
23 See, e.g., the rent-seeking literature originating with Tullock (1967) and surveyed by Nitzan (1994). See also Kalt and Zupan (1984, 1990) and Gilligan and Matsusaka (1995).
24 For surveys on the political business cycles literature, see Schneider and Frey (1988) or Nordhaus (1989).
296    L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306
political dilemma. On one hand, representative democracy is supposed to have an informational advantage compared to direct legislation. On the other hand, in representative democracy, there is a principal-agent problem between representatives and their constituency. Voters delegate decision making power to representatives in order to save information costs. But representatives then have possibilities to act opportunistically or in the self-interest of narrow groups.
Each political decision can be thought of as a decision about publicly provided goods. Elections in representative democracies or referenda and initiatives can be interpreted as a revelation of citizens' preferences for these goods. If an equilibrium is reached in the political process similar to the market process, then it is possible to derive the demand for publicly provided goods from democratic decisions.26 Following a normative approach, the median voter should be able to enforce his /her preferences in the democratic political process, in both systems with or without additional direct legislation. If competition between political parties and interest groups provides sufficiently strong incentives, then the programmes of political actors will be oriented at the position of the median.27
It has been correctly doubted that this model is applicable to representative democracies.28 Romer and Rosenthal (1978, 1979a) and Ingberman (1985) emphasise the impact of the agenda setter on referendum outcomes. If the government as an agenda setter succeeds in proposing a policy that is closer to the median voter's preferences than the status quo, then the median voter will accept such a proposal although it is not at his ideal position. In contrast to referenda, initiatives shift agenda setting power to citizens. The introduction of both in a representative system would thus enforce citizens' preferences. This assessment changes if there is uncertainty by representatives about citizens' preferences over policies. As Matsusaka and McCarty (2000) show, interest groups may be able to fool representatives, using the threat of an initiative pretending that their position is closer to citizens' preferences than the proposal of representatives, although this moves the policy outcome further away from the ideal point of the median voter. However, as Moser (1996) and Kirchgassner (1997) argue, different majorities in different political decision making bodies induce at least a correction of political outcomes in favour of the preferences of the median voter, even in the case of
We follow Shepsle (1979), in his concept of a 'structure-induced equilibrium', in order to avoid a discussion of cycling problems (see Arrow, 1951). Frey (1994) argues that the discussion and co-ordination process preceding the decision at the ballots reduces the number of available alternatives and increases the homogeneity of preferences. Thus, referenda and initiatives are institutions that decrease the likelihood of cyclical majorities.
26 See, e.g., Pommerehne (1987).
27 The quantity of publicly provided goods demanded by the median voter does not necessarily correspond to the Pareto-optimal quantity. For a discussion of the necessary conditions of an equivalence of the equilibria, see Bergstrom (1979).
28 See, e.g., Pommerehne (1978), Romer and Rosenthal (1979b) or Rowley (1984).
L.P. Feld, G. Kirchgassner / European Journal of Political Economy 16 (2000) 287-306 297
strategic manipulation or agenda setting. On the other hand, possibilities of vote trading29 and the impact of interest groups30 on policy outcomes might induce additional deviations from the median voter's optimal outcome and so render the theoretical concept of the median voter model more or less useless.
Empirical studies have investigated the extent to which political outcomes are closer to citizens' preferences in direct democracy compared to representative democracy. Pommerehne (1978) shows, using aggregate data on Swiss cities in 1970, that the median voter model performs better in jurisdictions in which political decisions are taken by voters directly in referenda and initiatives.31 Gerber (1996a, 1999) presents evidence for the US strongly supporting Pommerehne's result. For two political issues, parental consent laws and capital punishment, she shows that initiatives correct policy outcomes towards the preferences of the median voter. Gerber (1996b), however, provides evidence that the influence of interest groups on representatives leads to a deviation of policy outcomes from citizens' preferences. In Gerber (1999), she also addresses the impact of interest groups in direct legislation using 161 initiatives in eight US states. She finds that interest groups — contrary to citizen movements — have only small prospects of pushing through an initiative in their favour, although they have a good chance of success in rejecting a proposal.32 Overall, these studies indicate that wealthy interest groups do not exert a particularly strong influence on referendum outcomes, and that citizens' preferences are more strongly enforced in representative systems with referenda and initiatives than in pure representative democracies.
These results also imply that competition between political parties is not sufficiently perfect in the sense of Downs (1957). Representatives have leeway to follow their self-interest between elections. Thus, political competition in representative democracies is not sufficient to enforce the median voter's preferences. The restriction of competition can be interpreted as collusion among representatives.33 The possibility of referenda leads to an effective restraint on the ability of government and parliament to collude in order to pursue private interests. The referendum prevents political decisions that result from such collusion. The
2 Breton (1996) argues that direct legislation prevents gains from vote trading. On the other side, Weingast et al. (1981) as well as Weingast and Marshall (1988) emphasise the losses that occur because of vote trading and pork barrel politics.
30 On the one hand, the impact of interest groups is more visible in direct legislation than in representative democracies because their positions are discussed in the referendum campaigns. On the other hand, interest groups influence drafted bills in the early stages in the public administration and by convincing (or bribing) legislators. Finally, since referendum campaigns are costly, particularly well organised and wealthy interest groups have the ability to run successful referendum campaigns.
31 See also Pommerehne and Schneider (1978), as well as the mixed evidence in Megdal (1983), Chicoine et al. (1989) and Santerre (1989, 1993) using a similar approach for the US.
32 See the similar result for Switzerland by Longchamp (1991).
33 See also Breton (1996).
298    L.P. Feld, G. Kirchg'dssner / European Journal of Political Economy 16 (2000) 287-306
initiative creates new policy outcomes and prevents collusion of representatives to suppress a political issue. In presidential and parliamentary systems, the representatives in the government are controlled by representatives in the parliament. Control and sanctions by citizens are only possible via elections. In the Swiss direct democratic system, citizens have the additional possibility to control their representatives directly.
4. Empirical results on the impact of direct legislation
According to our analysis in the previous sections, direct legislation in Switzerland is basically characterised by two features. First, the political discourse preceding a decision at the ballots results in better information of citizens compared to the same citizens and possibly also legislators in representative democracies. Thus, politicians and well informed specialists in the legislature have less flexibility to pursue their personal interests. Second, the referendum discourse leads to reflection regarding policy positions: Do proposed policies benefit narrow self-interest or the public at large? Therefore, citizens should feel more responsible for their community and be more prepared to accept decisions that lead to income or wealth losses for themselves. Both mechanisms are expected to lead to differences in economic policy decisions.
Systematic analyses of the impact of referenda and initiatives on economic policy are available for Switzerland and for the US states.34 The results of the Swiss studies are summarised in Table 1. Schneider and Pommerehne (1983) show for 110 Swiss cities from 1968 to 1972 that expenditure growth in cities with direct democracy was — ceteris paribus — between 1965 and 1975 almost three percentage points lower than in representative democracies. While expenditure growth was, on average, 9.6% in representative democratic cities (according to simulation results), it would have been 6.8% if these cities had been direct democracies.35 In these analyses, cities are considered as directly democratic if their constitutions contain an obligatory or optional referendum or a local assem-
Only the results for Switzerland are reported in more detail here. The US studies are mentioned in footnotes. For a more detailed description of these studies, see Sass (1998), Kirchgassner et al. (1999, Chap. 5) as well as Feld and Kirchgassner (2000).
35 The corresponding results for the US are mixed. In an analysis for the year 1980 for 50 US states and 1305 local communities, Zax (1989) found that communities with initiatives have higher public expenditure. Farnham (1990) finds mixed results for 735 US cities in 1981/1982. Matsusaka (1995) finds a negative impact of initiatives on expenditures per capita for 49 US states in a pooled cross-section times series analysis for 7 years from 1960 to 1990. These results are corroborated by Rueben (2000) in a panel data analysis for the same time period. The difference between the results of Zax and of Matsusaka may be due to the fact that Zax included Alaska, which is a clear spending outlier. Matsusaka (1998) finds higher public spending in initiative states before the Second World War. Finally, the US results may not be robust with respect to other political variables such as ideological positions of the electorate. See Camobreco (1998).
Table 1
Empirical studies on the impact of direct legislation on economic policy in Switzerland
Economic variables
Authors
Sample
Time period
Parameter Estimates
Public expenditures
Public expenditures
Share of revenue from taxes and user charges
Median tax rate
Public debt
Costs of garbage collection
Tax evasion
Gross domestic product
Gross domestic product
Schneider and Pommerehne (1983) 110 Cities
Feld and Kirchgässner (1999) 131 Cities
Feld and Kirchgässner (1999) 131 Cities
Feld and Kirchgässner (1999) 131 Cities
Feld and Kirchgässner (1999) 131 Cities
Pommerehne (1983) 103 Cities
Pommerehne and 25 Cantons Weck-Hannemann (1996)
Feld and Savioz (1997) 26 Cantons
Feld and Savioz (1997) 26 Cantons
1968-1972
1990 1990
1990 1990
1970
1965, 1970, 1978 1990
1984-1993
Expenditure growth
is three percentage points
lower in direct democracy
Expenditure per capita is 14%
lower in the case of a referendum
Share of revenue from taxes and
user charges is 5%higher in
the case of a referendum
Median tax rate is 14%
higher in the case of a referendum
Public debt per capita
is 45% (SFr 10,100)
lower in the case of a referendum
Cost for garbage
collection is 20%
higher in representative democracies
Tax evasion is SFr 1500
lower in direct democracy
GDP 15% higher
in the case of a referendum
GDP 5.4% higher
in the case of a referendum
8
300    L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306
bly to pass fiscal legislation in form of changes in tax rates, the budget draft, or budget deficits.
Direct democratic cities in Switzerland also have lower public debt.36 Feld and Kirchgassner (1999) analyse different aspects of budgetary policy in a simultaneous equations model. In 47 out of the 131 Swiss cities investigated, those where the citizens have no direct rights with respect to budgetary decisions, public debt per capita would — ceteris paribus — have been about SFr 10,000 and thus 45% lower in 1990 than if these cities had had a direct democratic organisation.37 Moreover, Swiss cities with budgetary referenda spent 14% less. They had a revenue share from taxes and user charges, as opposed to transfers and subsidies from other government levels, that was 5% higher, and they levied an income tax on the median income that was 14% higher. This is evidence suggesting that citizens demand fewer public services, but are willing to pay higher and more adequate prices for these services in systems with direct legislation than in parliamentary democracies. This results in a lower public debt per capita under direct democracy. Thus, a referendum democracy appears to fit the Wicksell (1896) idea of a link between the tax price and public services better than a purely parliamentary system.
The question remains, however, whether the lower level of public spending also leads to a more efficient public sector. Lower public spending could, as Breton (1996) argues, be the result of insufficient vote trading in direct legislation, rendering it inefficient. It is very difficult to make profound statements about empirically measured efficiency. Very few studies attempt to investigate this question.38 Pommerehne (1983) analysed costs and prices of local garbage collection in 103 Swiss cities in 1970. He found that average refuse collection costs (per household) were — ceteris paribus — lowest in cities with direct legislation and private garbage collection. These costs were about 10% higher when garbage collection was organised privately instead of publicly. In cities with representative democracy, the costs of private garbage collection were about 20% higher than in direct democracies. Average costs of garbage collection were highest in cities with representative democracy and public organisation (30% higher than in the first case). This provides some evidence that direct legislation enhances the efficiency of the public sector. Together with the results on budgetary policy, this is evidence
3 Similar results can again be obtained for US states. See, e.g., Kiewiet and Szakaly (1996) and McEachern (1978). Bohn and Inman (1996) present evidence that formal fiscal restraints in US states were successful, if they were controlled by constitutional courts of the states. Judges who were directly elected by the citizens rather than being nominated by legislators more thoroughly enforced balanced budget rules.
37 Excluding Zurich, as an outlier with an extremely high debt, the simulated difference was still SFr 4500 or 24%, respectively. Neither formal fiscal restraint, nor a budgetary process with a strong role of the mayor or the secretary of finance, succeeded in putting such a strong constraint on debt issuing.
38 An exception is Noam (1980). The efficiency measure used is, however, not very telling.
L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306 301
in favour of the hypothesis that direct legislation in Switzerland results in less leeway of representatives to pursue their own interests.
Pommerehne and Weck-Hannemann (1996) derive very interesting results with respect to tax morale. They present evidence that tax evasion is lower in those Swiss cantons in which citizens have an impact on budgetary policy in direct legislation. Using data for the years 1965, 1970 and 1978, they show that in those cantons tax evasion is — ceteris paribus — about SFr 1500 lower as compared to the average of the cantons without such direct influence.39 If, however, the willingness to pay taxes is higher the more citizens are satisfied with public services supplied, then these results are evidence for higher satisfaction of citizens and, therefore, for greater efficiency of the provision of public services. Both the results concerning tax evasion as well as the stronger link between tax prices and public services in direct democracy indicate that Swiss citizens feel more responsible for their community. They might be more willing to accept decisions that lead to an income or wealth loss for themselves than citizens of representative democracies. This is indirect evidence for the conjecture that self-interested preferences can at least partially be reversed in the referendum discourse towards the common interest of the community.
These studies lend support for the hypothesis that direct democratic systems are more efficient than representative democratic ones. A more efficient political system should also lead to better economic performance. Feld and Savioz (1997) study the relationship between budgetary referenda and economic performance of Swiss cantons measured by GDP per employee. In a panel with annual data from 1984 to 1993 for the 26 Swiss cantons, they arrive at the conclusion that GDP per employee is — ceteris paribus — by about 5% higher in those cantons with budgetary referenda compared to cantons without those referenda.40 This result is tested in various ways for robustness. In particular, reversed causality is tested under the hypothesis that richer cantons can afford more direct legislation. On the basis of the empirical results, this hypothesis could be rejected. Moreover, the impact of direct democracy is hardly diminished if additional explanatory variables are included in the empirical model.41
There are also theoretical arguments why citizens in direct democracies evade taxes less than those in representative democracies. See Pommerehne et al. (1997).
40 In two recent papers, Frey and Stutzer (1999, 2000) present evidence that people in Switzerland perceive themselves as more satisfied with their life as a whole in direct democratic cantons.
41 Freitag and Vatter (1999) include a variable on the effective use of direct democracy in fiscal referenda as an additional variable influencing cantonal GDP per capita. They conclude that the legal institutions per se do not have an impact on GDP, but that the effective use is the crucial variable. They do not report, however, any correlations between the dummy used by Feld and Savioz (1997) and their variable on the effective use of fiscal referenda. Nevertheless, their results corroborate the hypothesis that GDP per employee is higher in direct democratic cantons. Finally, the results are robust with respect to additional specifications on which cantons use fiscal referenda. See Kirchgässner et al. (1999, Chap. 4).
302    L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306
5. Concluding remarks
In this paper, we have recounted how direct democracy affects economic policy performance as a peculiarity of political culture in Switzerland. Compared to purely representative systems, direct democracy provides two advantages of political decision making. First, the discussion process preceding a decision at the ballots enhances citizens' information about political issues to an extent that their information might even be better than the information of ordinary legislators in parliaments. This places strong restrictions on the ability of politicians or well informed specialists in the legislature to pursue their personal interests. Direct legislation thereby provides for control and sanction of legislators and government, and reduces inefficiency emerging from the principal-agent relationship between voters and their representatives. Second, the open and strongly visible political discourse at different levels of society can lead to a revision of citizens' evaluations of policies, from particular (individual) self-interest to a more general common interest. Citizens might feel more responsibility for their community and be more prepared to accept decisions that lead to an income or wealth loss. Problems characteristic of all democratic systems such as, e.g., the problem of rational ignorance, are better coped with in direct than in purely representative systems.
A review of the empirical evidence on the impact of direct legislation on economic policy in Switzerland reveals strong empirical support for both hypotheses. The Wicksellian link between tax prices and public services is stronger in direct than in representative democracies. Public services are provided more efficiently in direct democracies than in representative democracies. Both empirical results reflect the reduced flexibility of representatives in enforcing their preferences against the interest of the general public. Willingness to pay taxes is also higher in direct than in representative democracies, indicating that citizens are, to a larger extent, satisfied with public services supplied. This points to greater efficiency in the provision of public services in direct democratic systems. Results concerning tax evasion as well as the stronger link between tax prices and public services indicate that citizens in direct democracies feel more responsible for their community. This is consistent with the evidence that economic performance, in general, is also better in direct democratic Swiss cantons than in representative ones.
Political culture in Switzerland is indeed characterised by direct democratic elements, but not exclusively so. Another, perhaps equally important, principle of organisation of the Swiss state that has supposedly a strong impact on Swiss political culture is federalism, in particular fiscal federalism. Although as interesting as direct legislation, an analysis of Swiss federalism is beyond the scope of this paper. The arguments in this paper suggest policy recommendations with regard to the constitutional design of polities such as the EU, or possible constitutional reforms in other countries. Recommendations, which are based on
L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306 303
the Swiss model, have already been made with respect to the fiscal constitution of the EU by Schneider (1993), Kirchgässner (1994), Feld and Kirchgässner (1996) and Feld et al. (2000).
Acknowledgements
This paper was written while the first author was visiting the University of Southern California at Los Angeles and the University of Rennes 1. Financial support from the Swiss National Science Foundation is gratefully acknowledged. We thank Simon Hug, Elisabeth Gerber and John Matsusaka for valuable discussions on the impact of direct legislation on political outcomes, and an anonymous referee for helpful comments and suggestions.
References
Aghion, Ph., Tirole, J., 1997. Formal and real authority in organizations. Journal of Political Economy 105, 1-29.
Arrow, K.J., 1951. Social Choice and Individual Values. 2nd edn. (1963) Wiley, New York. Aubert, J.-F., 1984. So funktioniert die Schweiz: Dargestellt anhand einiger konkreter Beispiele.
Cosmos, Muri/Bern (4th edn., 1984). Austen-Smith, D., 1990a. Credible debate equilibria. Social Choice and Welfare 7, 75-93. Austen-Smith, D., 1990b. Information transmission in debate. American Journal of Political Science
34, 124-152.
Austen-Smith, D., Riker, W.H., 1987. Asymmetric information and the coherence of legislation.
American Journal of Political Science 81, 897-918. Bergstrom, Th.C, 1979. When does majority rule supply public goods efficiently? Scandinavian
Journal of Economics 81, 216-226. Bohn, H., Inman, R.P., 1996. Balanced-budget rules and public deficits: evidence from the US states.
Carnegie-Rochester Conference Series on Public Policy 45, 13-76. Bowler, S., Donovan, T., 1998. Demanding Choices: Opinion, Voting, and Direct Democracy.
University of Michigan Press, Ann Arbor. Brennan, G., Lomasky, L., 1993. Democracy and Decision: The Pure Theory of Electoral Preference.
Cambridge Univ. Press, Cambridge. Breton, A., 1996. Competitive Governments: An Economic Theory of Politics and Public Finance.
Cambridge Univ. Press, Cambridge. Budge, I., 1996. The New Challenge of Direct Democracy. Polity Press, Cambridge. Camobreco, J.F., 1998. Preferences, fiscal policies, and the initiative process. Journal of Politics 60,
819-829.
Chicoine, D.L., Walzer, N., Deller, S.C., 1989. Representative vs. direct democracy and government spending in a median voter model. Public Finance/Finances Publiques 44, 225-236.
Downs, A., 1957. An Economic Theory of Democracy. Harper and Row, New York.
Eschet-Schwarz, A., 1989. La democratic semi directe en Suisse: entre la theorie et la realite: 1879-1987. Canadian Journal of Political Science 22, 739-764.
Farnham, P.G., 1990. The impact of citizen influence on local government expenditure. Public Choice 64, 201-212.
Feld, L.P., Kirchgässner, G., 1996. Omne agens agendo perficitur: the economic meaning of subsidiär-
304    L.P. Feld, G. Kirchgässner / European Journal of Political Economy 16 (2000) 287-306
ity. In: Holzmann, R. (Ed.), Maastricht: Monetary Constitution Without a Fiscal Constitution? Nomos, Baden-Baden, pp. 195-226.
Feld, L.P., Kirchgässner, G., 1999. Public debt and budgetary procedures: top down or bottom up? Some evidence from Swiss municipalities. In: Poterba, J.M., Hagen, J. (Eds.), Fiscal Institutions and Fiscal Performance. Chicago University Press and NBER, Chicago, pp. 151-179.
Feld, L.P., Kirchgässner, G., 2000. Die politische Ökonomie der direkten Demokratie. In: Luthardt, W., Waschkuhn, A. (Eds.), Politische Systeme und direkte Demokratie. Oldenbourg-Verlag, München.
Feld, L.P., Kirchgässner, G., Weck-Hannemann, H., 2000. The introduction of referenda in the EU budgetary procedure. In: Steunenberg, B. (Ed.), Widening the European Union: The Politics of Institutional Change and Reform. Oxford Univ. Press, Oxford.
Feld, L.P., Savioz, M.R., 1997. Direct democracy matters for economic performance: an empirical investigation. Kyklos 50, 507-538.
Feldmann, S., 1998. Bargaining for legislation with voter initiatives. Unpublished manuscript. University of Chicago.
Freitag, M., Vatter, A., 1999. Direkte Demokratie, Konkordanz und Wirtschaftsleistung: Ein Vergleich
der Schweizer Kantone. Unpublished manuscript. University of Bern. Frey, B.S., 1994. Direct democracy: politico-economic lessons from Swiss experience. American
Economic Review, Papers and Proceedings 84, 338-342. Frey, B.S., Kirchgässner, G., 1993. Diskursethik, Politische Ökonomie und Volksabstimmungen.
Analyse und Kritik 15, 129-149. Frey, B.S., Stutzer, A., 1999. Measuring preferences by subjective well-being. Journal of Institutional
and Theoretical Economics 155, 755-778. Frey, B.S., Stutzer, A., 2000. Maximising happiness? German Economic Review 1. Gerber, E.R., 1996a. Legislative response to the threat of initiatives. American Journal of Political
Science 40, 99-128.
Gerber, E.R., 1996b. Legislatures, initiatives, and representation: the effects of state legislative
institutions on policy. Political Research Quarterly 49, 263-286. Gerber, E.R., 1998. Pressuring legislatures through the use of initiatives: two forms of indirect
influence. In: Bowler, S., Donovan, T., Tolbert, C. (Eds.), Citizens as Legislators: Direct Democracy in the United States. Ohio State University Press, Columbus, pp. 191-208. Gerber, E.R., 1999. The Populist Paradox: Interest Group Influence and the Promise of Direct
Legislation. Princeton University Press, Princeton. Gilligan, Th.W.T., Matsusaka, J.G., 1995. Deviations from constituent interests: the role of legislative
structure and political parties in the States. Economic Inquiry 33, 383-401. Habermas, J., 1973. Legitimationsprobleme im Spätkapitalismus. Suhrkamp, Frankfurt. Habermas, J., 1981. Theorie des kommunikativen Handelns. Suhrkamp, Frankfurt, English translation:
Theory of Communicative Action, Vol. 2, Beacon Press, Boston 1984/1988. Ingberman, D.E., 1985. Running against the status quo: institutions for direct democracy, referenda and
allocations over time. Public Choice 46, 19-43. Kalt, J.P., Zupan, M.A., 1984. Capture and ideology in the economic theory of politics. American
Economic Review 74, 279-300. Kalt, J.P., Zupan, M.A., 1990. The apparent ideological behavior of legislators: Testing for principal
agent slack in political institutions. Journal of Law and Economics 33, 103-131. Kiewiet, D.R., Szakaly, K., 1996. Constitutional limitations on borrowing: an analysis of state bonded
indebtedness. Journal of Law, Economics and Organization 12, 62-97. Kirchgässner, G., 1992. Towards a theory of low-cost decisions. European Journal of Political
Economy 8, 305-320.
Kirchgässner, G., 1994. Constitutional economics and its relevance for the evolution of rules. Kyklos 47, 321-339.
Kirchgässner, G., 1996. Bemerkungen zur Minimalmoral. Zeitschrift für Wirtschaftsund Sozialwissenschaften 116, 223-251.
L.P. Feld, G. Kirchgassner / European Journal of Political Economy 16 (2000) 287-306 305
Kirchgassner, G., 1997. Vom Mythos der politischen Führung: Einige Bemerkungen zur gegenwärtigen
Diskussion um die direkte Demokratie. In: Borner, S., Rentsch, H. (Eds.), Wieviel direkte
Demokratie verträgt die Schweiz? Rüegger, Zürich, pp. 247-257, 363. Kirchgassner, G., Feld, L.P., Savioz, M.R., 1999. Die direkte Demokratie: Modern, erfolgreich,
entwicklungsund exportfähig. Helbing and Lichtenhahn/Vahlen, Basel /München. Kirchgassner, G., Frey, B.S., 1994. Volksabstimmung und direkte Demokratie: Ein Beitrag zur
Verfassungsdiskussion. In: Klingemann, H.-D., Kaase, M. (Eds.), Wahlen und Wähler: Analysen
zur Bundestagswahl 1990. Westdeutscher-Verlag, Opladen, pp. 42-69. Kirchgassner, G., Pommerehne, W.W., 1992. Zwischen Parteien-und Bundesstaat: Staatshandeln in der
Schweiz und in der Bundesrepublik Deutschland. In: Abromeit, H., Pommerehne, W.W. (Eds.),
Staatstätigkeit in der Schweiz. Haupt, Bern, pp. 221-245. Letterie, W.A., Swank, O.H., 1997. Learning and signalling by advisor selection. Public Choice 92,
353-367.
Longchamp, C, 1991. Herausgeforderte demokratische Öffentlichkeit: Zu den Möglichkeiten und
Grenzen des politischen Marketings bei Abstimmungen und Wahlen in der Schweiz. Schweizerisches Jahrbuch für Politische Wissenschaft 31, 303-326. Lupia, A., 1994. Shortcuts versus encyclopaedias: information and voting behavior in California
insurance reform elections. American Political Science Review 88, 63-76. Lupia, A., McCubbins, M.D., 1998. The Democratic Dilemma: Can Citizens Learn What They Need
To Know? Cambridge Univ. Press, Cambridge. Matsusaka, J.G., 1992. Economics of direct legislation. Quarterly Journal of Economics 107, 541-571. Matsusaka, J.G., 1995. Fiscal effects of the voter initiative: evidence from the last 30 years. Journal of
Political Economy 103, 587-623. Matsusaka, J.G., 1998. Fiscal effects of the voter initiative in the first half of the twentieth century.
Unpublished manuscript. University of Southern California, Los Angeles. Matsusaka, J.G., McCarty, N.M., 2000. Political resource allocations: the benefits and costs of voter
initiatives. Journal of Law and Economics 43. McEachern, W.A., 1978. Collective decision rules and local debt choice: a test of the median voter
hypothesis. National Tax Journal 31, 129-136. Megdal, Sh.B., 1983. The determination of local public expenditures and the principal and agent
relation: a case study. Public Choice 40, 71-87. Moser, P., 1996. Why is Swiss politics so stable? Schweizerische Zeitschrift für Volkswirtschaft und
Statistik 132, 31-60.
Nawiasky, H., 1951. Die Demokratie der Schweiz. Schriftenreihe der Hochschule für Politische Wissenschaften, No. 12, München.
Nawiasky, H., 1953. Von der unmittelbaren Demokratie: Die Bereitschaft der Schweiz — Die Zurückhaltung in Deutschland. In: Matz, U. (Ed.), Grundprobleme der Demokratie. Wissenschaftliche Buchgesellschaft, Darmstadt, pp. 147-155.
Nitzan, S., 1994. Modelling rent-seeking contests. European Journal of Political Economy 10, 41-60.
Noam, E.M., 1980. The efficiency of direct democracy. Journal of Political Economy 88, 803-810.
Nordhaus, W.D., 1989. Alternative approaches to the political business cycle. Brookings Papers on Economic Activity No. 2, 1-49.
Pommerehne, W.W., 1978. Institutional approaches to public expenditure: empirical evidence from Swiss municipalities. Journal of Public Economics 9, 255-280.
Pommerehne, W.W., 1983. Private versus öffentliche Müllabfuhr — nochmals betrachtet. Finanzarchiv 41, 466-475.
Pommerehne, W.W., 1987. Präferenzen für öffentliche Güter. Mohr (Siebeck), Tübingen.
Pommerehne, W.W., Hart, A., Feld, L.P., 1997. Steuerhinterziehung und ihre Kontrolle in unterschiedlichen politischen Systemen. Homo Oeconomicus 14, 469-487.
Pommerehne, W.W., Schneider, F., 1978. Fiscal illusion, political institutions, and local public spending. Kyklos 31, 381-408.
306    L.P. Feld, G. Kirchgassner / European Journal of Political Economy 16 (2000) 287-306
Pommerehne, W.W., Schneider, F., 1985. Politisch-ökonomische Überprüfung des Kaufkraftinziden-zkonzepts: Eine Analyse der AHV-Abstimmungen von 1972 und 1978. In: Brugger, E.A., Frey, R.L. (Eds.), Sektoralpolitik vs. Regionalpolitik. Rüegger, Diessenhofen, pp. 75-100.
Pommerehne, W.W., Weck-Hannemann, H., 1996. Tax rates, tax administration and income tax evasion in Switzerland. Public Choice 88, 161-170.
Riker, W.H., 1982. Liberalism Against Populism. Freeman, San Francisco, CA.
Romer, Th., Rosenthal, H., 1978. Political resource allocation, controlled agendas, and the status quo. Public Choice 33, 27-43.
Romer, Th., Rosenthal, H., 1979a. Bureaucrats versus voters: on the political economy of resource allocation by direct democracies. Quarterly Journal of Economics 93, 563-587.
Romer, Th., Rosenthal, H., 1979b. The elusive median voter. Journal of Public Economics 12, 143-170.
Rowley, Ch.K., 1984. The relevance of the median voter theorem. Zeitschrift für die gesamte
Staatswissenschaft/Journal of Institutional and Theoretical Economics (JITE) 140, 104-126. Rueben, K.S., 2000. Tax limitations and government growth: the effect of state tax and expenditure
limits on state and local government. Journal of Political Economy. Santerre, R.E., 1989. Representative versus direct democracy: are there any expenditure differences?
Public Choice 60, 145-154. Santerre, R.E., 1993. Representative versus direct democracy: the role of public bureaucrats. Public
Choice 76, 189-198.
Sass, T.R., 1998. The anatomy of political representation: direct democracy, parliamentary democracy and representative democracy. Unpublished manuscript. Florida State University, Tallahassee.
Schneider, F., 1993. The federal and fiscal structures of representative and direct democracies as models for a European federal union: some ideas using the Public Choice approach. European Economy 5, 191-212.
Schneider, F., Frey, B.S., 1988. Politico-economic models of macroeconomic policy: a review of the
empirical evidence. In: Willet, Th.D. (Ed.), Political Business Cycles: The Political Economy of
Money, Inflation, and Unemployment. Duke University Press, Durham, pp. 239-275. Schneider, F., Pommerehne, W.W., 1983. Macroeconomia della crescita in disequilibrio e settore
pubblico in espansione: il peso delle differenze istituzionali. Rivista Internazionale di Scienze
Economiche e Commerciali 33, 306-420. Shepsle, K.A., 1979. Institutional arrangements and equilibrium in multidimensional voting models.
American Journal of Political Science 23, 27-59. Swank, O.H., 1999. Does political competition give political agents incentives to create an efficient
bureaucratic system? Unpublished manuscript. University Rotterdam. Tullock, G., 1967. The welfare costs of tariffs, monopolies, and theft. Western Economic Journal 5,
224-232.
Weingast, B.R., Marshall, W.J., 1988. The industrial organization of congress; or, why legislatures, like firms, are not organized as markets. Journal of Political Economy 96, 132-163.
Weingast, B.R., Shepsle, K.A., Johnsen, Ch., 1981. The political economy of benefits and costs: a neoclassical approach to distributive politics. Journal of Political Economy 89, 642-664.
Wickseil, K., 1896. Finanztheoretische Untersuchungen nebst Darstellung und Kritik des Steuerwesens Schwedens. Gustav Fischer, Jena.
Zax, J.S., 1989. Initiatives and government expenditures. Public Choice 63, 267-277.