Does direct democracy really work ? A review of the empirical evidence from Switzerland

Discussions about direct democracy and its advantages and risks are often superficial, invoke stereotypes and ignore empirical data. This article tests seven common criticisms of direct democracy by referring to the Swiss experience. Evidently, Swiss democracy is not a copy/paste model, but has developed in a specific historical and institutional setting. It is obvious that both conservative as well as left-wing critics overemphasize their case against direct democracy by (wilfully) neglecting the evidence. Direct democracy does not lead to anarchy. The common people can make reasonable decisions. Minorities are not more discriminated against in direct democratic systems than in representative ones. Money plays a role in direct democracy, as it does in representative systems. Direct democracy slows down reforms, but it also makes them steadier and more sustainable. Direct democracy brings contentment to its citizens. Finally, direct democracy is not ideologically predisposed. It is a mechanism to revert policies back to the median voter.


Introduction
F or many years, the alleged crisis of democracy in general and the democratic "deficit" and legitimacy gap of the European Union in particular, have been intensely debated topics (Schmidt, 2002;Thomassen, Schmitt, 2004;Wagschal, 2007;Guastaferro, Moschella, 2012).It is obvious that modern Western democracies are indeed confronted with various endogenous (e.g.demographics, changes in the political culture) and exogenous challenges (e.g.globalization, migration, terrorism, European Union integration) (Helms, 2011).In Europe and elsewhere, many citizens demand more participation and co-decision rights.They feel that representative democracy is "not enough" for them and that they have the capacity to participate more and more intensively.Direct democracy is often presented as a solution to these challenges.Indeed, polls have shown that large majorities of the population in European countries favour more direct democracy (Dalton et al., 2001).
The issue of whether direct democratic instruments can compensate (and at what cost) for the apparent legitimacy gaps in modern democracies is therefore of great relevance.However, discussions about direct democracy and its advantages and risks are often superficial phony debates, nurtured by clichés and prejudices.Critics claim that direct democracy is slow and inefficient.It is said to lead to populism or even anarchy.Citizens are supposedly unable to understand complex political problems and to make reasonable decisions.On the other hand, admirers of direct democracy believe that it is a panacea for all problems which have befallen today's democracies (which it is not, of course).DOI : 10.14746/pp.2014.19.4.6The main problem with these discussions is that they are invariably either hypothetical or based on single cases of controversial direct democratic decisions, mainly taken in Switzerland, as half of the world-wide popular votes take place in that country.These debates also ignore empirical data.They are not about real direct democracy, but about some (evil) imagined one.Even if Switzerland is called a direct democracy, it is a far cry from an Athenian democracy or a utopian rule of the people.Only 7% of all laws proposed by the Swiss parliament are challenged by a referendum, and only half of these referenda are successful (Linder, 2011).
Against this backdrop, this article tests seven common criticisms of direct democracy by referring to the Swiss experience.It is evident that Swiss democracy is not a copy/paste model, but has developed in a specific historical and institutional setting.However, it is also obvious that both conservative as well as left-wing critics overemphasize their case against direct democracy by (wilfully) neglecting the evidence.Serious reform debates about how to narrow legitimacy gaps in "old" democratic countries as well as in "new" democracies in the Eastern and Southern region of Europe in general (Frey, 2003) and in the European Union (EU) in particular may and should look closely at the Swiss model.It can be an inspiration, especially with regard to the sustainable development of local democracy.

Empirical Evidence from Switzerland
From an outsider's perspective, Swiss democracy resembles a labyrinth of political institutions difficult to understand and to disentangle (Lane, 2001).Research about direct democracy in Switzerland and elsewhere however has blossomed in the last few decades (Matsusaka, 2005).Some effects of direct democracy are still popularly debated among scholars (Lupia, Matsusaka, 2004), but a broad, data-based consensus has evolved with regard to several essentials of Swiss direct democracy (Vatter, 2011).
Firstly, direct democracy is an integral element of Swiss political identity (Widmer, 2007).It is supported by a huge majority of citizens independently of social strata, income, gender, or party identification (Geissbühler, 2001, p. 169-173).Secondly, direct democracy has led to remarkable political stability and it has upheld both the legitimacy of the Swiss political system as a whole as well as the trust of citizens in the political process.Historically direct democracy has doubtlessly contributed to the integration of different political, language, religious/confessional and cultural sub-groups of society.Thirdly, direct democratic instruments have better reverted political decisions back to the median voter, and also indirectly as parties and the parliament tend to anticipate referenda and formulate policies and/or laws in such a way as to avoid them: "Political decisions are thereby expected to better approximate the median voters' preferences" (Buetzer, 2011, p. 155).Fourthly, direct democracy does not lead to "worse" policies than representative democracy.Indeed direct democracy has proven to be efficient (Noam, 1980).Fifthly, in a direct democracy different interest groups, some of them very small ones, have an impact on the agenda setting and can launch new and innovative ideas through popular initiatives.Sixthly, direct democracy slows down reforms, but it also makes them steadier and more sustainable.It is an instrument of permanent control by the political elite (Frey et al., 2001;Linder, 1999, p. 238), and it could be one of the factors explaining the still relatively "lean" Swiss (welfare) state (Linder, 2011).Finally, there is empirical evidence that citizens in a direct democracy are more content than their peers in similar representative systems because they have further opportunities to influence the political process and have their voice heard (Stutzer, Frey, 2000;Frey et al., 2001).However, not all these findings have been perceived accordingly by some researchers, by politicians, and by the broader public.Clichés and prejudices are difficult to uproot.They will be addressed below in more detail and confronted with empirical evidence from Switzerland.

Direct Democracy destabilizes the political system
Conservatives in particular have traditionally argued that direct democracy has a destabilizing effect and that it could even lead to anarchy.This prejudice is clearly refuted by the Swiss case.The Swiss example proves "that direct democracy and political stability can both be achieved" (Linder, 2011).In all respective rankings, Switzerland is among the most stable countries in the world (EIU, 2009;World Bank, 2011).Even the end of the so-called magic formula for the composition of the Federal Council (executive) at the end of 2011 will not fundamentally change this stability.Interestingly, this change of the government's composition was not introduced by a popular vote, but by a decision of the parliament.

People are not capable/competent/intelligent enough for direct democracy
Apparently the best argument of many critics of direct democracy is that it leads to "wrong" or "bad" decisions while in "normal" democracies "correct" and "good" decisions are taken.To underline this "argument", these critics base their theories on some individual cases such as the decision of the Swiss people to ban the construction of new minarets.However it could also be argued that nothing is "proven" by individual cases, and in any case it is also easy to find other individual cases to show that direct democracy works.To give just one example: in 2003, the political and economic elites in Sweden vigorously fought for their country to join the Euro zone.The Swedish people said "no" in a referendum.Who was right: the omniscient elites or the "dumb" people?Such arguments were futile (Lupia, Matsusaka, 2004, p. 474).
Nevertheless, it has been argued that direct democracy is simply impossible due to the alleged lack of political understanding and cognitive capacities of the "common" citizen.The "average" citizen is said to be disinterested in politics and not capable of grasping complex political problems and taking reasonable decisions.Even in ancient Sparta, elites thought "normal" people unfit to even speak about politics and policies (Woodruff, 2006, p. 11).However this prejudice falls short of the mark.
Firstly, the "argument" that people are too stupid for direct democracy would mean when taken to its logical conclusion that people are also too stupid for democracy as such: "In any case, the argument that voters are incompetent and uninformed would seem to cut against democracy in general, rather than against direct democracy alone" (Matsusaka, 2005, p. 198).The idea that it is more difficult and challenging to respond "yes" or "no" to PP 4 '14 Does direct democracy really work?A review of the empirical... a concrete political question than to choose the candidate closest to one's preferences among dozens or hundreds of candidates in a proportional parliamentary election is rather odd.Even if there are only two candidates in an election, the choice of the preferred candidate can often be much more challenging cognitively for a "common" citizen than a popular vote.Matsusaka (2005, p. 198) has argued "that, if anything, uninformed voters are more likely to make mistakes when voting on candidates than ballot measures because candidates represent bundles of issues and characteristics, while ballot propositions typically involve only a single issue." Secondly, this prejudice implies that politicians always know better than the "average" citizen, that they are always more competent on all issues and that they always take the "correct" decisions.This suggestion is, of course, far from the truth: "Politicians do not necessarily show expertise and interest" (Budge, 2006, p. 597).Politicians don't know all the issues and are not always well-informed about everything.They depend on "shortcuts" and have to ask other politicians and experts, and trust their knowledge and advice: "Actually, representatives in parliaments […] are often badly informed about political issues that they do not regard as their expertise" (Feld, Kirchgässner, 2000, p. 293).
Thirdly, empirical research has shown that at least the citizens who regularly go to the polls are well informed.They know the issues, frequently discuss politics and policies, and they consult a wide array of sources of information (Linder, 2011: "High information level of a good proportion of citizens").In fact direct democracy can be an incentive for citizens to inform themselves about politics and policies, to discuss political issues, to join interest groups and to participate (Feld, Kirchgässner, 2000;Dalton et al., 2001;Bowler, Donovan, 2002;Tolbert et al., 2003;Benz, Stutzer, 2004;Boehmke, Bowen, 2010).A recent example of the maturity of the Swiss voters was an initiative in March 2012 proposing to introduce six weeks holidays: the initiative was rejected by a huge margin.
Fourthly, we all make hundreds of decisions every day.Most often we do not even realize that we decide.Evidently it is impossible to base all these everyday decisions on thorough research and on the analyses of various sources of information.We base our process on past experience, on "common sense," on simple decision making patterns and on "information shortcuts", in order to make quick decisions.Interestingly, even though we seldom decide as comprehensively informed individuals, our decisions seem to be "good" most of the time (Lupia, Matsusaka, 2004, p. 468).
In popular votes in Switzerland, it is of course not an individual who decides, but all the voters who go to the polls.So even if some voters might be confused and might not understand an issue, the totality of the voters functions as a corrective.Finally, many popular votes do not concern complex issues which are said to be impossible to understand for the "average" citizen and demand huge amounts of information and in-depth knowledge: "[Many] issues are mainly about a community defining its values" (Matsusaka, 2005, p. 193).

Direct democracy discriminates against minorities
Christian Bolliger (2011) discusses this issue in an article recently published in Polish.Therefore I can concentrate here on a few of the key arguments.The thesis that direct de-90 Simon GEISSBÜHLER PP 4 '14 mocracy is a tool in the hands of the majority to discriminate against the minority is an old one.It has been argued that direct democracy leads to or in fact is the tyranny of the majority.Swiss direct democracy -according to Wolfgang Merkel (2011, p. 54) -stimulates patterns of conservative and right-wing discrimination against minorities.However, Merkel's case is rather weak.He bases his theory on two or three popular votes in Switzerland without looking at the broader context.Indeed so far all attempts to reduce or even stop immigration into Switzerland have been rejected by the Swiss people, even though foreigners make up over 22% of the resident population in Switzerland, one of the highest percentages in Europe.Empirical studies have shown that direct democracy as such is neither an instrument of the majority against the minority nor an effective tool of minority protection (Vatter, 2011).
The underlying argument that direct democracy inevitably leads to tyranny is unfounded in the Swiss case.There is no monolithic majority in Switzerland which would force its will upon a similarly monolithic minority through popular votes.That is exactly the point and explains the high legitimacy of direct democracy: majorities and minorities always change depending on the issues to be decided at the polls.In fact one could argue that the tyranny of the majority is much more common in representative democracies: once a parliament is elected and a majority government formed, this government can -at least theoretically -decide whatever it wants.It has in many cases a free hand for four years: the majority remains the majority for four years, and the opposition acts as the minority for four years: "Thus, majoritarian democracies suffer not only from their inbuilt tendency towards a 'tyranny of the majority' […], but also from a tendency to produce conflict resolutions of the zero-sum type, such as in a winner-takes-all game" (Schmidt, 2002, p. 151).
Furthermore, studies have shown that direct democracy gives (social) outsiders and marginal groups a platform to propagate their ideas and to be heard (Höglinger, 2008).Direct democracy creates an additional political arena outside of parliament.Of course discriminations against minorities e.g.foreigners, occur in Switzerland.But they also occur in representative democracies.Democratic systems are never perfect.Only an ideal democracy -whether it is direct, semi-direct or representative -would flawlessly engage "majority and minority elements in a cooperative enterprise" (Woodruff, 2006, p. 218).

Direct democracy is biased towards the higher social strata of society
Wolfgang Merkel (2011) has underlined that the lower the participation in elections and in popular votes, the stronger the bias towards society's higher social strata.In other words the higher social strata participate more actively in direct democracy than the lower ones.Of course this theory is nothing new and it is based on a solid body of empirical evidence.People who earn more and who are better educated are more likely to vote (Harder, Krosnick, 2008).Accordingly Wolf Linder (2011) refers to a middle class bias of direct democracy.
However this argument is both contrary to direct as well as representative democracy.Furthermore direct democracy allows for marginal issues to be put on the agenda PP 4 '14 Does direct democracy really work?A review of the empirical... (Vatter, 2011).This can lead to the political activation of strata of the population that are not interested in "conventional" politics.There is some evidence that popular votes (ballot initiatives) in the US have an educational effect and "increase turnout in mid-term as well as presidential elections" (Tolbert, Smith, 2005;cf. Tolbert et al., 2003;Biggers, 2011).
Voter turnout in Switzerland shows that important popular votes mobilize the electorate.In salient popular votes, voter turnout is much higher than in cantonal or national elections (Bundesamt für Statistik, 2011).Therefore it can be argued that in popular votes -at least in those which are perceived to be crucial -the bias towards the higher social strata of society is less pronounced than in elections.Examples of an above average turnout and also therefore of less social bias would be the popular votes on the abolition of the Swiss army in 1989 (69.2%; initiative rejected) and against the immigration of foreigners in 1974 (70.3%; initiative rejected) and also in 1970 (74.7%; initiative rejected) (IPW/APS 2008).
However Wolfgang Merkel's thesis that direct democracy would accelerate the exclusion of society's lower social strata from the political process is not convincing (Merkel, 2011, p. 55).It is evident that direct democracy is sometimes as exclusive as representative democracy.But quite often it is more inclusive than representative democracy because some ballot issues mobilize the strata of society that do not normally participate politically.

Money buys direct democratic decisions
The fact that money is an important factor in democracies is well known.Wolfgang Merkel (2011, p. 52) writes that money is less decisive in politics in Switzerland than in other countries, but is still a critical element in the political process and in influencing decision-making.Indeed there is no reason to believe that Switzerland would and should be an exception with regard to the influence of money in politics because of its direct democratic mechanisms.The influence of money in politics is not the problem of direct democracy, but of democracy itself.The idea that money buys direct democratic decisions is mistaken (Lupia, Matsusaka, 2004, p. 470): "[Unlike] legislators who may gain directly from lobbyist contributions, citizen voters generally receive no direct benefit by voting one way or another.Instead they are more likely to vote with their conscience on a topic and strive for ethical government" (DuVivier, 2007(DuVivier, , p. 1050)).
Wolfgang Merkel (2011, p. 52) further criticizes that popular initiatives are not really initiated by the people, but by powerful and well-financed interest groups or parties.However the evidence shows a more nuanced picture.While it is true that it is rather cost-intensive to collect signatures for initiatives or referenda, there is much evidence to suggest that "average" citizens and even relatively obscure interest groups can launch their ideas through the direct democratic process.Recent examples are the (successful) popular initiatives against the construction of new minarets (2009) and for the abolition of the statute of limitations for sexual crimes against children (2008).Both initiatives were launched by dedicated individuals and small groups that did not have an abundance of financial and human resources.

Direct democracy leads to an expansion of the (welfare) state and blocks reforms
In Switzerland there is a long tradition of publications, mainly by economists, which argue that direct democracy is responsible for the alleged lack of innovation and for reform blockages (Wittmann, 2011;Brunetti, Straubhaar, 1996;Borner et al., 1994).These authors argue that representative democracies are more efficient and adapt faster to the changing economic and political environment than political systems with direct democratic components.Theoretically this thesis might be convincing, at least at first glance.But in practice, representative or so-called Westminster democracies such as Great Britain run the risk to completely change government policies with each and every change in government ("Jekyll-and-Hyde syndrome").The fact that abrupt policy changes, which can include the short-term discontinuation of policies of the outgoing government and the introduction of diametrically opposed policies by the incoming government, are neither efficient nor effective and does not need to be explained in detail (Abromeit, 1992).Speed alone is certainly not a proof of quality in the legislative process.
It has been underlined that direct democracy can have an innovative effect which compensates at least partially for the policy blockages it creates (Linder, 1999, p. 259-265).Direct democracy reduces the risks of rent-seeking (Vatter, 2011) and the tendency of politicians to spend money and to support their own political clientele.Studies have shown that in Swiss cantons where citizens have more direct democratic rights, public services tend to be better and taxes lower than in cantons with more restricted democratic rights: "Public services are provided more efficiently in direct democracies than in representative democracies" (Feld, Kirchgässner, 2000, p. 302).Furthermore citizens are rather reluctant to expand the (welfare) state.Feld andMatsusaka (2003, p. 2721) have demonstrated that "government spending is lower in Swiss cantons with mandatory referendums".

Direct democracy leads to neoliberal policies
In his latest article about (Swiss) direct democracy, German scholar Wolfgang Merkel (2011, p. 54) contends that the Swiss experience shows that direct democracy leads to the enforcement of a conservative or neoliberal fiscal policy with negative redistributive effects for lower income households.This argument is surprising, as most critics of direct democracy tend to hold precisely the opposite viewpoint.According to these critics, welfare policies are hastily expanded in direct democracy.Voters with lower incomes are said to have a genuine interest in enforcing redistribution through popular votes, thereby "milking" citizens with higher income who are a minority and therefore more likely to lose in a respective popular vote.Direct democratic decisions are seen as tools of welfare state expansion and causing an inefficient allocation of resources (Wittmann, 2011).
It is true that Switzerland is historically a rather "lean" (welfare) state.However, the causalities are not so obvious, and it is rather doubtful that direct democracy is the primary reason for the relatively "small" Swiss (welfare) state.It has been convincingly argued that Switzerland's specific libertarian political culture (Hayek, 1994, p. 235;Abromeit, 1992, p. 161) and the decades' long dominance of political liberalism have contained welfare PP 4 '14 Does direct democracy really work?A review of the empirical... state expansion.But in the last few decades, government activity expanded considerably in Switzerland.While overall government expenditures represented 30.3% of GDP in 1990GDP in , this percentage grew to 35.1% in 2000GDP in and to 36.4% in 2003. .Only after 2003 did government expenditures again marginally decrease (2009: 33.7%) (Armingeon et al., 2011).Clearly direct democracy did not hinder the expansion of government activities.
Overall, there is little evidence to support Wolfgang Merkel's thesis that direct democracy has neoliberal effects.Of course direct democracy has not led to a sprawling welfare state.Clearly it is not an effective tool to rapidly expand the welfare state (Moser, Obinger, 2007, p. 357;Lupia, Matsusaka, 2004).But on the other hand, direct democracy did not lead to radical cuts in social provision either: "Thus, Switzerland with its direct democracy at all governmental levels, can hardly be seen as an example where the welfare state is endangered by the existence of direct popular rights" (Feld et al., 2010).The Swiss welfare state grew in the last few decades (Martin 2002), and the chances of a retrenchment are more limited in a direct democratic system (because of the status quo bias of direct democracy) than in a representative democracy (Moser, Obinger, 2007, p. 344).
But what about direct democracy and redistribution?Again the evidence is rather mixed and unspectacular: "Does direct democracy lead to less redistribution?Considering that it leads to lower welfare expenditure and less tax revenue per capita, one might draw such a conclusion.Such an apprehension is probably behind the strong rejection of introducing additional direct popular rights in Germany by some political scientists: They fear the end of the welfare state.The possible effects might be largely overestimated by such fears; despite the well developed direct popular rights the Swiss welfare state has not yet broken down, and a cutback of public welfare expenditure also occurred in purely representative systems like the German one during recent years.Nevertheless, there might be -ceteris paribus -a negative impact of direct democracy on redistribution.The fact that public welfare expenditure is somewhat lower may, on the other hand, not necessarily compromise redistribution.Because public expenditure might be better tailored to the needs of the electorate in direct democracies, given the amount of public welfare expenditure its redistributive effect might be larger than in purely representative systems" (Feld et al., 2010).

Summary
It can be said that discussions about direct democracy often resemble debates about a phantom."Proxies" are created for the respective argumentative needs and political 94 Simon GEISSBÜHLER PP 4 '14 points of view.These "proxies" or "doubles" (Woodruff, 2006) have nothing to do with real direct democracy.They tell us more about the persons who use them than about direct democracy (Lupia, Matsusaka, 2004, p. 474).This imagined direct democracy looks menacing to some, especially to politicians who fear the permanent control of the people and an additional player with a veto in the political arena (Buetzer, 2011;Linder, 2011;Feld, Kirchgässner, 2000).
It would be essential to discuss direct democracy in a calmer and more evidence-based way.In such a way it would become obvious that direct democracy is neither a panacea for all the problems of modern democracies nor a diabolical, uncontrollable political institution.Direct democracy is neither better nor worse than representative democracy.But it is an interesting institution, both theoretically and practically, and it can be a source of inspiration.With regard to the effects of direct democracy, Lupia and Matsusaka (2004, p. 474) have correctly underlined that it does not make any sense to think that direct democracy is somehow "ideologically predisposed in any particular direction".Direct democracy is simply a political institution.It is a median-reverting institution "that pushes policy back toward the centre of public opinion when legislatures move too far to the right or left".