Confidence in Alternative Dispute Resolution : Experience from Switzerland

Alternative Dispute Resolution plays a crucial role in the justice system of Switzerland. With the unified Swiss Code of Civil Procedure, it is required that each litigation session shall be preceded by an attempt at conciliation before a conciliation authority. However, there has been little research on conciliation authorities and the public’s perception of the authorities. This paper looks at public confidence in conciliation authorities and provides results of a survey conducted with more than 3,400 participants. This study found that public confidence in Swiss conciliation authorities is generally high, exceeds the ratings for confidence in cantonal governments and parliaments, but is lower than confidence in courts. Since the institutional models of the conciliation authorities (meaning the organization of the authorities and the selection of the conciliators) differ widely between the 26 Swiss cantons, the influence of the institutional models on public confidence is analyzed. Contrary to assumptions based on New Institutional-ism approaches, this study reports that the institutional models do not impact public confidence. Also, the relationship between a participation in an election of justices of the peace or conciliators and public confidence in these authorities is found to be at most very limited (and negative). Similar to common findings on courts, the results show that general contacts with conciliation authorities decrease public confidence in these institutions whereas a positive experience with a conciliation authority leads to more confidence. The Study was completed as part of the research project 'Basic Research into Court Management in Switzerland', supported by the Swiss National Science Foundation (SNSF). Christof Schwenkel is a PhD student at the University of Lucerne and a research associate and project manager at Interface Policy Studies. A first version of this article was presented at the 2013 European Group for Public Administration (EGPA) Annual Conference in Edinburgh.


Introduction
In the last decades, many European and non-European states have introduced or increased the use of mechanisms of Alternative Dispute Resolution. 2 This has also been the case for Switzerland, where conciliation has a "long and profound tradition" (Meier 2003: 343) and "always has enjoyed a more prominent role than litigation and adjudication" (Meier 2008: 4).While historical examples for conciliation date back to the Middle Ages, 3 the basis for the introduction of conciliation authorities, according to the model of the French juge de paix, was set in the time of the Helvetic Republic around 1800.By 1883, all Swiss cantons (except Basel-Stadt) had installed justices of the peace to settle disputes and to avoid litigation (Schnyder 1985: 11). 4 In the following decades, the cantons have developed different models for the process of conciliation and included these in their code of civil procedure (cf.Fischer 2008: 25).
When the unified Swiss Code of Civil Procedure (CCP) was enacted in 2011, the historical importance of conciliation was taken into account.Thus, Art.197 CCP requires that each litigation shall be preceded by an attempt at conciliation before a conciliation authority.It is the authorities' responsibility to "reconcile the parties in an informal manner" (Art.201 Abs. 1 CCP).According to Art. 212 CCP the conciliation authorities are also authorized to render a decision in financial disputes with a value in dispute not exceeding 2,000 francs, if the plaintiff so requests.Mandatory conciliation proceedings were introduced to help the courts reduce their workload and to enable the parties with a low-threshold first access to the legal process (Botschaft CCP: 06.062: 7223). 5  In addition, Art.213 CCP provides the possibility to replace conciliation by mediation if all parties request so.The main difference between mediation and conciliation is that in conciliation procedures, the "conciliator plays a relatively direct role in the actual resolution of a dispute and even advises the parties on certain solutions by making proposals for settlement" (Sgubini/Prieditis/Marighetto 2004) -or even decides in small value claims. 6This article focusses on mandatory conciliation, as mediation is still a relatively new development in Switzerland (Meyer 2008: 4) and the choice to enter mediation is still voluntary, there is a limited structure that this study is taking into account.
In the context of the research project "basic research into court management in Switzerland", Lienhard, Kettiger and Winkler (2012: 1) point to "the general dearth of empirical and other theoretical findings on the operations of the justice system and its interaction with society" in Switzerland.This includes the Swiss conciliation authorities, which have been rarely the subject of scientific research. 7  Like the courts on a cantonal level, the organization of the conciliation authorities remains under the jurisdiction of the cantons (Art.3 CCP).While conciliation proceedings are mandatory in every canton, the cantons are free to choose the institutional models of the conciliation authorities and the judicial skills and procedures of the authorities (cf.Fischer 2008: 80).In some cantons, every municipality has its own justice of the peace, selected by popular vote.In other cantons, judges of the cantonal courts are required to conduct the conciliation proceeding (Botschaft CCP 06.062: 7243).Thus, even with the unified CCP, the conciliation authorities vary widely among the 26 cantons.The first part of this paper provides an overview of the different institutional models of conciliation authorities and categorizes the cantons in to three types according to the institutional models of conciliation.
A core factor for successful dispute resolution is illustrated by the public confidence and the legitimation of the conciliation authorities.Similar to the confidence placed in courts (cf.Caldeira 1986), high public confidence in conciliation authorities can influence the parties' willingness to bring conflicts into the system for resolution, to cooperate and to accept the result of a conciliation process.The relevance of public confidence in conciliation authorities is also emphasized by the fact that conciliation authorities are in many civil law cases the only institution within the justice system that people interact with, considering the relatively high rates of success of pre-trial settlement conferences (cf.Staubli 1999: 189ff.).Although several authors (Schnyder 1985: 28ff, Stadler 1998: 2491, Zwickel 2010: 172) indicate that public confidence in Swiss conciliation authorities is high, there is no data available that describes the extent of the population's support or confidence in these authorities.To fill this gap, a comprehensive survey with the inhabitants of all 26 Swiss cantons was conducted in the framework of the above mentioned research project "basic research into court management" and included questions about the conciliation authorities. 8The findings of this survey will be presented in the second part of this paper.
A third part will be dedicated to the analysis of factors influencing public confidence.In addition to the question of whether institutional differences shape public confidence, other factors and their possible effects on public confidence will be discussed.A quantitative analysis will be used to test the working hypothesis of this paper and to provide insight in possible causes for public confidence in the conciliation authorities.Part four sums up the results and provides some conclusive remarks.
The main questions to be addressed by this paper are the following: -What are the institutional models of conciliation authorities in the Swiss cantons?-What level of confidence does the population show towards the conciliation authorities?-What factors affect the public's confidence in the conciliation authorities? 5 The annual report 2012 of the association of justices of the peace in Zurich states that in 70 percent of all cases, a dispute settlement could be achieved.In total, the justices of the peace in Zurich dealt with 3'389 cases in 2012.(Verband der Friedensrichter und Friedensrichterinnen des Bezirks Zürich 2013: 4).The Swiss association of justices of the peace and conciliators also estimates that 50-70 percent of all cases can be resolved on the level of the conciliation authorities, usually within less than two months (Schweizerischer Verband der Friedensrichter und Vermittler 2014).

7
The works of Schnyder (1985), Staubli (1999), Fischer (2008) and Ziegler (2003) provide a good overview on the conciliation authorities and their work (with a focus on the justices of the peace) but are based on the time before the CCP and remain mainly on a descriptive level.

Institutional Models of Conciliation Authorities
Comparative political science frequently makes use of typologies, which provides the opportunity to classify broad empirical diversity into a few categories and thus to reduce complexity (Lauth 2009: 154).This is essential in order to compare a larger number of cases (for example the 26 Swiss cantons).My typology focuses on the conciliation authorities as defined in Art.197 CCP; which defines all responsible authorities for mandatory conciliation procedures in civil law that do not concern matters of labor law or disputes relating to the tenancy and lease of residential and business property or disputes under the Gender Equality Act (Art.200 CCP).For these disputes, cantons have installed special authorities.The exclusion of those authorities can be justified by the attempt to keep the typology simple and to put the focus on the "justices of the peace".Nevertheless, future in-depth projects on conciliation authorities should also include those specialized institutions.
The following three types were defined on the basis of the different institutional models in the 26 cantons: -Type 1: This type includes the cantons where the conciliation authorities do not form a stand-alone-institution, but are incorporated into a court of first instance.This is most often the case in those cantons which have lately introduced mandatory conciliation proceedings as a consequence to the unified CCP.Here, the conciliators are principally professional jurists (judges/clerks).The number of cantons in this group is 5. -Type 2: The cantons that are categorized as type 2 cantons have special conciliation authorities that are institutionally independent from the courts (but often share the same facilities as the courts).These conciliators have been elected by either the cantonal parliaments or by members of the courts and are often, but not always, jurists (but not judges at a court).The conciliation authorities are usually responsible for a larger number of municipalities.The number of cantons in this group is 10.-Type 3: The last type comprises the most decentralized form of conciliation authorities.In type 3 cantons, the justices of the peace or conciliators are directly elected by the population.They are mostly laypersons and often work on a community level, where they have their own facilities (or even conduct the hearings in their homes).The number of cantons in this grouPp is 11.
The following table gives an overview of all 26 cantons.The main factor for grouping the cantons was the cantonal laws on court organisation.To gain further insight, explorative interviews with the following persons have been conducted: Margaretha Reichlin (former justice of the peace, Luzern).Sandra Bättig (justice of the peace, Willisau), Urs Flury (president "Schweizerischer Verband der Friedensrichter und Vermittler"), Urs Wicki (president "Verband der Friedensrichterinnen und Friedensrichter des Kantons Zürich").For a similar, but in some aspects differing categorization, see a recently published article by Meier and Scheiwiller (2014: 170).
Apart from a concentration of French speaking cantons among the type 1 cantons (Geneva, Jura, Neuchâtel), there are no obvious structural factors that would correlate with the allocation of the cantons to either one group.All types would include small cantons (Jura, Obwalden, Appenzell Innerrhoden), cantons with a large number of inhabitants (Geneva, Berne, Zurich) rural (Jura, Uri, Glarus) and rather urban cantons (Basel-Stadt, Vaud, Zurich).

Public Confidence in Conciliation Authorities
Other than the executive or legislative power which can use repressive or financial instruments to enforce their decisions, the judiciary has "no influence over either the sword or the purse" (Hamilton 1788).For this reason, the support of the population is of critical to legitimizing the authority of third power (cf.Caldeira 1986, Hardin 1998: 10), which also incorporates the conciliation authorities.A theoretical basis for my analysis can be found in Easton's (1957) concepts of support.According to Easton, support is a crucial input to keep a political system operating (Easton 1957: 390ff.).Easton distinguishes between specific and diffuse support.Whereas specific support describes the approval of (short-term) outputs and performances of political authorities (cf.Easton 1975), diffuse support focuses on institutions and consists of a "reservoir of favourable attitudes or good will that helps members to accept or tolerate outputs to which they are opposed or the effects of which they see as damaging to their wants" (Easton 1975: 444).This means, that citizens may disagree with what an institution does but continue to concede its authority (Caldeira/Gibson 1995: 357).Empirical research frequently uses public confidence in institutions to measure support in institutions (Benesh 2006; Norris 1999; Wenzel/Bowler/Lanoue 2003; see also Gibson/Caldeira/Spence 2003: 354 andBühlmann/Kunz 2011: 320).
Measuring public confidence in the conciliation authorities can provide information about both, diffuse and specific support -the latter especially in those cases where outputs are known or people took part in a conciliation process.However, considering that the majority of the people are not directly concerned with the outputs of conciliation authorities, I consider public confidence as an appropriate item to measure mostly diffuse support in the conciliation authorities. 10The reflections of Bühlmann and Kunz (2011: 320 f.) are also in favor of this approach.Based on Hardin (2002) I define confidence in the conciliation authorities as the public belief that those authorities have the right intentions towards citizens and are competent to act in specific ways in specific conciliation proceeding.
I put the focus of my analysis on the public confidence in institutions, also described as political trust (Newton 2001) which represents a mostly a vertical relationship between citizens and institutions.Public confidence in institutions or political trust differs from concepts of generalized trust and social capital, which focus on the (horizontal) trust amongst citizens (Freitag 2001, Newton 2001).Social capital theory sees in the fact that citizens trust each other a core factor for the functioning of democracies (Putnam 1993, Almond/Verba 1965, Freitag/Bühlmann 2005, cf. Kaina 2004).Generalized trust and public confidence in institutions are assumed to have a reciprocal relationship in which aspects of generalized trust can be regarded as a precondition for political trust and vice-versa (cf.Kaina 2004, Gabriel/Kunz 2002, Kouvo 2011, Newton 2001).
My data on public evaluations of the cantonal courts and conciliation authorities was collected in an online-survey of 3,484 people which was conducted between March and April 2013 in collaboration with the survey institute LINK. 11The population from which we selected the sample consisted of 120'000 members of the "LINK online-panel" which were actively recruited via telephone.All persons in the online panel were between 18-75 years old, used the internet at least once per week for private purposes and were able to respond in German, French or Italian.To maximize the representativeness of the survey, quotas for canton, gender and age were used when selecting the respondents.
A minimum of 101 respondents from each canton took part in the survey.In the seven most populous cantons, the number of participants amounted to at least 200. 12Participants were asked to rate several institutions according to the confidence they have in these institutions.The respondents were given an 11-point scale ranging from "no confidence" (0) to "complete confidence" (10). 13In addition, the participants were asked whether they had ever been in contact with a conciliation authority and whether they took part in an election of justices of the peace or conciliators.All questions that concerned cantonal institutions focused on the respondents' canton of residence.
The following table shows some of the main results of the survey.In order to make general assumptions for the Swiss population, we used weighted data considering the different population size of each canton.Notes: Confidence is measured based upon the responses to the question: Please tell me on a score of 0-10 how much confidence you have in the following institutions.Zero means no confidence at all, 10 means complete confidence.
Calculations were based on weighted data.
The data shows that confidence in institutions is generally high.This has also been shown in recent other surveys as the Swiss Electoral Study (SELECTS) and an analysis of the ETH centers for security studies (Szvircsev Tresch/Wenger 2013: 101). 14While those studies used a general question on confidence in justice/courts, our survey questions allowed us to distinguish between conciliation authorities and courts at the federal and cantonal level.A remarkable result is that the respondents showed less confidence in conciliation authorities than in courts, but rated their confidence in the conciliation authorities higher than in the canton's government or parliament. 15  Does the confidence in the conciliation authorities differ between the 26 Swiss cantons?The following illustration indicates that there is some, though it is a rather small variance between the subnational states, ranging from a value of 6.2 in the canton of Valais to a value of 7.5 in the canton of Zug.Due to the small N for some cantons (minimum of 101 contacts, with non-response), one should be careful not to overestimate the differences at the cantonal level.
Figure 1: Public confidence in conciliation authorities 6.2 6.3 6.4 6.4 6.5 6.5 6.6 6.6 6.6 6.7 6.7 6.7 6.7 6.7 6.7 6.9 6.9 7.0 7.0 7.0 7.1 7.1 The results of SELECTS 2011 show a mean value (on an score of 0-10) for trust in Justice/Courts of 6.71 and in cantonal authorities of 6.27.(http://www2.unil.ch/selects/).The ETH Centre for Security Studies (Szvircsev Tresch/Wenger 2013: 101) used the same rating scale and shows a value of 7.1 for the item "trust in courts". 15The aim of this paper is to analyze the differences in confidence in conciliation authorities.However, finding explanations for the differences between the levels of confidence in the different institutions (with a focus on the conciliation authorities) would be an interesting task of further research.
Notes: N per canton ranges from 101 to 304.Total N is 3'484.The lower n values used to calculate the mean values per canton are due to non-response in certain questions.
With regard to the three institutional types of the conciliation authorities (see Table 2); the combined averages in terms of public confidence show the following small differences: -Type 1: 6.87 (n=459) -Type 2: 6.82 (n=1'152) -Type 3: 6.79 (n=1'460) The results of the survey also provide information about the interaction of the population with conciliation authorities and courts.The main results are aggregated for Switzerland and illustrated in the following table.Notes: Contacts with institutions is measured using responses (yes, no) to the questions: Did you ever have a contact with a conciliation authority/conciliator/justice of the peace in your canton; did you ever have a contact with a court in your canton?Did you ever have a contact with a federal court?Participation in an election is measured using responses (yes, no) to the questions: Did you ever vote in an election of a justice of the peace/conciliator in your canton?Did you ever vote in an election of judges in your canton?Calculations are based on weighted data.
The findings confirm that conciliation authorities play an important role in the interaction of the Swiss population with the justice system.Twenty-one percent of the Swiss population have once been in contact with a conciliation authority in their canton and a quarter of the population has once taken part in an election of justices of the peace or conciliators. 16The results however differ strongly between the cantons, as shown in the following chart.  1Due to a limited number of questions in the survey, it was only possible to distinguish between positive and negative experiences with those contacts, not between the type/reason for the contact (e.g. as a party in a conciliation, as a person seeking for information, personal contacts with a justice of the peace).The original question in French, German in Italian was formulated as follows: "Vousmême, avez-vous déjà eu affaire une fois…?/HattenSie selbst schon einmal … zu tun?/Lei stesso(a) ha già avuto a che fare con…?".
Notes: The bars show the percentage of positive answers for each canton to the question: "Did you ever have a contact with a conciliation authority/conciliator/justice of the peace in your canton." In terms of direct contacts, the largest interaction can be found in the cantons of Vaud and Ticino with 40 and 37, the fewest contacts are made in the canton of Solothurn, where only 7 percent of the population has dealt with such an authority in some way.
As expected the number of persons that claim to have taken part in an election of justices of the peace or conciliators is the lowest in cantons that do not have popular vote (e.g.0 percent in Jura).The maximum rate of participation in an election can be found in the canton of Valais with 59 percent participation.
If respondents responded that they had been in contact with conciliation authorities, they were asked whether their experience was positive, rather positive, neither positive nor negative, rather negative or negative.The results indicate that fifty-nine percent of the respondents had a positive or rather positive experience with conciliation authorities, 24 percent described their experience as neither positive nor negative and only 17 percent perceived it as negative or rather negative. 17These results are similar to the findings concerning the experience with cantonal courts, where the ratio between positive/rather positive and negative/rather negative is 56 to 19 percent.

Explanations for Public Confidence in Conciliation Authorities
What drives public confidence for the conciliation authorities and what does not?In this chapter I will discuss possible influences, develop my working hypotheses and show what factors may and may not affect public confidence in conciliation authorities by conducting a bivariate and a multivariate analysis.18

Institutional Model
I assumed that the institutional model of the conciliation authorities (meaning the organization of the authorities and the selection of the conciliators) influences the public confidence in these authorities.The theoretical basis for this assumption is found in approaches of New Institutionalism (cf.Hall/Taylor 1996, DiMaggio/Powell 1991). 19Norris (1999: 234) concludes in her empirical analysis that institutional arrangements are significantly related to political support.In the case of the courts, it is emphasized by Benesh (2006: 699) that the public may be sensitive to the potential (or even assumed) effects of institutions and therefore view some institutional models of courts more favorably than others.Other scholars also show how institutional models influence confidence in courts (Bühlmann/Kunz 2011, Kelleher/Wolak 2007).
A key element of an institution is the selection of its members.Thus it is assumed that electoral systems play a major role in linking citizens and the state and affect the support of the population for political institutions (cf.Norris 1999: 225 f.).Kiener (2001: 256) emphasizes that the election of judges by the people in Switzerland provides the third power with direct support by the population and thus fosters the connection of the judiciary with the people.Mürner (2005: 19) also states that judges that have been elected by popular vote are more likely regarded as representatives of the population.Mürner (2005: 19) uses the expression "Vertrauensrichtertum" and assumes that citizen-centered aspects of the judicial institutions in Switzerland (such as oral proceedings, proceedings in local dialects, no statutory requirement to be represented by a lawyer) foster public confidence in the judiciary.Schnyder (1985: 28ff.)refers to the relevance of the popular vote of justices of the peace with respect to their democratic legitimation.The author assumes that the population shows a higher level of confidence in a conciliation authority, whose members have been selected by the population on a local level, than when appointed by a higher level institution.(Schnyder 1985: 30).Stadler (1998: 2481) also considers the democratic legitimation of the justices of the peace in Switzerland with strong roots in the municipalities as a reason for a high acceptance of the conciliation authorities.Stadlers assumption is resumed by Zwickel (2010: 172).Kelleher and Wolak (2007: 710) generally point out that confidence in subnational institutions builds on perceptions that they are more accessible and responsive to the public.Bauer et al. ( 2012) emphasize that people may generally display more trust in decentralized institutions because they are perceived as closer to them and their daily life. 20n this context I assume that public confidence is higher, if the members of the conciliation authorities have been elected by popular vote (type 3), which can also be seen as a proxy for greater decentralization of the conciliation authorities and the importance of lay persons as justices of the peace.In accordance with the findings of Kelleher and Wolak (2007: 718) about US state institutions, I recognized the existence of a negative influence of more professionalised and less citizenlinked institutions on public confidence.Taking into consideration that conciliation authorities should facilitate a "lowthreshold" access to justice, I further assume that this is less easier achieved in cantons where the judges and clerks of cantonal courts act as conciliation authorities (type 1) than in the other cantons.
A possible negative effect of popular vote on confidence is that judges selected by the people may be regarded as less independent (cf.Cann/Yates 2008: 309, Benesh 2006: 702), especially in the case of partisan elections or contested elections where candidates have to prove their accountability and run for re-election.Given the fact that elections of justices of the peace in Switzerland are mostly low key and non-controversial, I consider this to be a negligible factor.However, the independence of conciliation authorities may be jeopardized in those cases where justices of the peace are elected at the small municipality/"local level" and parties may have a personal connection with the person responsible for the conciliation.Justices of the peace or conciliators that work as laypersons could be further regarded as less professional and thus influence public confidence in a negative way.In view of the Swiss system of direct democracy and the importance of common sense in dispute resolution (cf.Ludewig-Kedmi/Angehrn 2008), I think that the positive aspects of popular vote outweigh the negative aspects.This leads me to the following first hypothesis: H1: Confidence in conciliation authorities is enhanced in cantons where the conciliation authorities are a standalone institution and members of the conciliation authorities can be elected by popular vote.Weaver and Rockmann (1993: 446) emphasize that, even where institutional arrangements do contribute to overall differences in specific capabilities, moreover, these effects usually are strongly mediated by other institutional and noninstitutional factors".One important institutional factor on the individual level (which is strongly influenced by the institutional models in each canton) is the participation in an election of justices of the peace or conciliator, especially given the fact that almost a quarter of the Swiss population has at least once taken part in an election of justices of the peace or conciliators.Hence I follow the assumption that the participation in an election strengthens the connection between the conciliation authority and the population (cf Norris 1999: 225ff.).This leads me to the following hypothesis:

Participation in an Election
H2: Confidence in conciliation authorities is enhanced by the participation in an election of justices of the peace or conciliators.

Personal Contacts and Positive Experience
As stated above, our survey data shows that more than 20 percent of the population has had a personal contact with a conciliation authority in their canton.This leads me to the question whether or not those contacts influence the confidence in the conciliation authorities.
Several studies indicate that public support in local or state courts is in fact reduced by interaction with or knowledge about them (Benesh/Howell 2001, Wenzel/Bowler/Lanoue 2003, Benesh 2006, Sarat 1977: 439, cf. Rottman 1998).A reason for this can be found in the "adversary nature of the system, (where) defendants and plaintiffs may have a more negative impression of local courts" than people who did not interact with a court (Wenzel/Bowler/Lanoue 2003: 194).Furthermore, direct contacts with courts could erode favorable stereotypes that people had before experienced the court system themselves (cf.Benesh 2006: 699).Experiencing the formal aspects of a procedure or having difficulties in understanding the technical terminology of judges may, for example, lead to a decrease of confidence in the courts.
When it comes to outcomes, it has be shown that the way people and their problems are managed when they are dealing with a court has more impact upon trust and confidence than the actual outcome of an individual's case (Tyler 2007: 26, van den Bos/Wilek/Lind 1998, Thibaut/Walker 1975, Lind 1995: 11).The evidence for the importance of procedural justice (and the effort to find a way to increase the willingness to accept third-party decisions) is thus regarded by Tyler (2007: 26) as a major motivation for alternative dispute resolution.On the background of the less formal character of alternative dispute resolution, together with the pursuit of the conciliation authorities to acknowledge parties' positions and to create a positive climate, I assume that contacts with conciliation authorities shapes public confidence in these authorities in a positive way (differently than contacts with courts).This leads me to the following working hypothesis:

H3: Confidence in conciliation authorities is enhanced by personal contacts with conciliations authorities.
A second hypothesis puts the evaluation of the contacts people had with the conciliation authorities into consideration and can be connected to the performance of the institutions.I assume that a positive experience with a conciliation authority also shapes confidence in a positive way.
H4: Confidence in conciliation authorities is enhanced by personal contacts with conciliation authorities that have been experienced as positive.
Our data allows us to distinguish between "positive", "rather positive", "neither positive nor negative", "rather negative" and "negative" experiences with conciliation authorities.However, due to the limited number of questions in the questionnaire, no data was collected on the possible outcome of a settlement conference or on the reasons why people were not satisfied with their experience.

Generalized Trust
A core factor of social capital theory is generalized trust or trust between citizens (cf.Freitag 2004: 91).Freitag and Bauer (2013: 26) describe generalized trust as an "attitude towards people in general, encompassing people beyond one's immediate familiarity, including strangers".The theoretical discussion of Kaina (2004: 529) illustrates how generalized trust can be regarded as a crucial precondition for confidence in institutions.Although some empirical research has found that the relationship between generalized trust and institutional confidence to be rather small (Kaase 1999, Newton/Norris 1999), I agree with the assumption that people who express attitudes of trust toward others are more likely to express a higher confidence in authorities (cf.Cann/Yates 2008: 314, Caldeira/Gibson 1992: 647, Zmerli/Newton 2008).This seems to be even more evident in the case of the conciliation authorities as they should provide a low level connection between citizens and the justice system.To measure generalized trust, we used the following, frequently used survey question: "Generally speaking, would you say that most people can be trusted or that you can't be too careful in dealing with people?" (cf.Freitag/Bühlmann 2009). 21This leads to the following hypothesis: H5: Confidence in conciliation authorities is enhanced by a high level of generalized trust.

Individual Level Controls
I will use several control variables that are often found to be associated with public confidence in institutions (see, among others, Cann/Yates 2008, Caldeira /Gibson 1992, Kunz 2009, Wenzel/Bowler/Lanoue 2003, Norris 1999, Hetherington 1998, Gambetta 1988).Based on the findings of Norris (1999: 233), I assume that education is positively related to confidence in institutions.A higher level of education (measured by the highest degree achieved) can enable citizens to better understand processes of the legal system and thus can contribute to more confidence in the conciliation authorities.Furthermore, the knowledge about institutions is probably higher among higher educated citizens.Cann and Yates (2008: 307) discuss the influence of citizens' knowledge about their state courts on public confidence and show that research has found conflicting views on this question.In their analysis about the state courts, they however found a negative effect of court knowledge on diffuse support (Cann/Yates 2008: 314).Since our survey did not include a question on knowledge about the conciliation authorities, I will use the answers to the question "How informed do you consider yourself about courts in your canton" as a proxy to measure information in the justice system in general and also included this variable in my analysis.
Further demographic controls included in my analysis are political views (the respondents were asked to choose their position on a left/right scale), size of the municipality, age and gender.As Switzerland is a multilingual country, I also included language as an important independent variable.Hereby I distinguish between German speaking on the one hand and French/Italian speaking on the other hand (measured by the language in which respondents filled out the questionnaire).

Quantitative Analysis
To estimate the influence of the different independent variables on the confidence in conciliation authorities, I conducted bi-and multivariate analysis. 22The bivariate correlations were calculated for several variables and the (dependent) variable "confidence in the conciliation authorities" The following table shows the results of the correlation analysis: Notes: *significant at a 0.01 level, a value of +1.0 would stand for a perfect linear relationship between two variables.The results show that there is no significant relationship between the institutional type of conciliation authority and the confidence in the institutions.The second independent variable (participation in an election of conciliation authorities) was also not significantly related with public confidence.A small and significant negative relationship was found for contact with the conciliation authorities.This indicates that people show in fact less confidence in the conciliation authorities when they have had a personal interaction with them.
Confidence in conciliation authorities was positively related with positive experience with a conciliation authority (value of .625),which supports my fourth hypothesis.Here, one has to take into account that the number of persons that could answer this questions (due to the majority of people not having contacts with conciliation authorities) is much lower than for the other results of the correlation (n=705).There was also a correlation (value of .497) between positive experience with cantonal courts and confidence in conciliation authorities, which can lead to the assumption that positive experience with the judiciary in general, enhances the confidence in its institutions. 23The highest correlation was found for confidence in cantonal courts (.786) and cantonal governments (.696).This can also be also taken as a strong indicator for conciliation authorities being regarded as one institution among others in the political system.
There is a significant correlation between generalized trust and confidence in the conciliation authorities (with a value of .200).Further calculations show that confidence in other institutions also correlates with generalized trust, but the highest value is in fact reached for the conciliation authorities.
The bivariate analysis shows that confidence in the conciliation authorities is higher among the more highly educated, persons that feel themselves informed about cantonal courts, persons that live in larger municipalities and persons that evaluate their political views as rather left than right.Surprisingly there was no correlation between language and our dependent variable. 24The control variables age and gender also did not correlate with public confidence in the conciliation authorities.
23 Other research has shown a strong relation between the confidence in different institutions in the political system (Norris 1999: 222, Lehne/Reynolds 1978, Fuchs/Gabriel/Völkl 2002).Unlike other studies (Kunz 2009, Benesh 2006), I will however not consider confidence in other institutions as an independent variable to explain variation in my dependent variable and therefore not formulate a hypothesis including confidence in other institutions. 24Concerning questions about confidence in other institutions, there are however (small) differences when considering the language.Thus French and Italian speaking persons show more confidence in all three federal courts.Confidence in cantonal courts and governments is slightly lower among French speaking persons.
In a multivariate OLS-regression, I estimated the influence on my dependent variable, taking into account several independent variables that have seem to be relevant in the bivariate analysis. 25I calculated different models.The first (Model 1) using the independent variables (IV) of my H1, H2 H3, and H5 the second (Model 2.1) using the independent variables of H1, H2, H4 and H5.Since contact with a conciliation authority (IV H3) is a necessary condition for the evaluation of the experience with the authority (IV H4), the two variables are not included in the same model.Due to a correlation between positive experience and information about courts (value of .345),another model (Model 2.2) excludes the latter variable.All models show that the institutional type of conciliation authorities is not related with the population's level of confidence in the conciliation authorities.This leads me to the rejection of my first hypothesis.Despite broad range of institutional models, the confidence in the conciliation authorities is not affected by this.H2 and H3 can also be rejected: neither the contacts with conciliation authorities, nor the participation in an election of these authorities are related with higher confidence.The influence of these independent variables on the dependent variable leads in fact in the opposite direction, and is smaller than for example the evaluation of information about cantonal courts.In Model 1, the independent variables explain 9.3 percent of the variance in the dependent variable.Thus, the overall prediction by the factors included in Model 1 (with generalized trust and information about courts being the best predictors) is rather small.
When considering the experience, people have had with conciliation authorities (Model 2.1), the adjusted R² increases to a value of 43.4 percent.Hence, positive experience explains part of the confidence in the authorities.It is however obvious that in Model 2.1 and 2.2 estimations only include cases where people actually have had an experience with a conciliation authority.The exclusion of the variable "information about cantonal courts" (model 2.1) does not contribute to a relevant difference on the other predictors or the adjusted R². 26 25 Due to a more limited number of responding persons (increase in N) and the absence of an additional contribution to a variance on the dependent variable , I don't show a regression that includes the variable "political view". 26The use of grouped aggregate level data (as type of conciliation authorities according to the canton a person lives in) with individual data (as contacts with conciliation authority) can result in a downward bias in the estimated standard errors of OLS estimates (cf.Moulton 1990).The reason for this is that people within the same canton will often be similar to each other.A common strategy to provide more accurate estimates is the use of multilevel modelling (Freitag/Bühlmann 2005: 586;Kelleher/Wolak 2007).With the relationship between my institutional independent variable (type of conciliation authorities) and the dependent variable being already very limited, I however refrain from conducting a multilevel analysis.

Summary and Conclusions
Alternative dispute resolution plays an important role in the Swiss judicial system.In civil law, it is mandatory for each litigation session to be preceded by an attempt at conciliation (Art. 197 CCP).Each canton has developed its own model for the conciliation authorities, which leads to a broad range in terms of institutional models.Today, more than 600 justices of the peace or special conciliation authorities deal with the conciliation procedures in the 26 Swiss cantons.In 5 cantons the conciliation proceedings are conducted by courts of first instance, 10 cantons have special conciliation authorities whose members are elected by cantonal parliaments and courts and are usually responsible for a larger number of municipalities.In another 11 cantons, justices of the peace or conciliators mostly work as laypersons on the municipal level and have been elected by popular vote.Our survey results show that 21 percent of the population has been in contact with a conciliation authority in their canton.Twenty-four percent of the population indicated that they have taken part in an election of justices of the peace or conciliators.
Taking into account the importance of the institutionalized alternative dispute resolution by the conciliation authorities in Switzerland, my analysis puts a focus on the population's confidence in these authorities.I assume that a high confidence in conciliation authorities will influence the willingness to bring conflicts into the system for resolution and to accept the results of a settlement conference.The results of our survey show that public confidence in the Swiss conciliation authorities is generally high (6.81 on a 0-10 scale).It exceeds the ratings for confidence in the cantonal governments and parliament, but is lower than confidence in the cantonal or federal courts.
The main finding of my analysis is that public confidence in the conciliation authorities is not influenced by the institutional factors taken into account.Despite the differences in terms of institutional models, public confidence is not significantly higher (or lower) among people from cantons with direct vote and decentralized conciliation authorities.The quantitative analysis also shows that participation in an election of justices of the peace or conciliators does not foster public confidence in that institution.
Several studies have shown that experience with courts leads to a decrease of public confidence in the courts or the judiciary (e.g.Benesh/Howell 2001, Wenzel/Bowler/Lanoue 2003, cf.Rottman 1998).Given the "alternative" nature of alternative dispute resolution I hypothesized that personal contact with the conciliation authorities contributed to more public confidence in those institutions.However the results show that an experience with conciliation authorities influences public confidence in a negative way.Hence, my analysis doesn't support the fact that the conciliation authorities are being regarded much different from other institutions within the justice system.This is underlined by the finding that public confidence in courts positively correlates with public confidence in the conciliation authorities.
The results show that public confidence in conciliation authorities is enhanced by a positive experience with these authorities.Taking into consideration that only 17 percent of the respondents with contacts with conciliation authorities deemed their experiences as negative, I assumed that the quality of the settlement conferences and the quality of the involved conciliators contributed highly to the satisfaction with a conciliation.This in turn fosters the confidence in the authorities for the people who have been in contact with a conciliation authority.
Why do the results differ from what I have expected?One explanation can be that if it is not the institutional models of the conciliation authorities, the performance and personal aspects of their incumbents shape public confidence -no matter if they are judges at a court in the canton's capital or laypersons in each municipality of the canton.Further research should therefore focus on qualitative methods and investigate the mechanisms of settlement conferences or character traits of justices of the peace or conciliators.Such an analysis could also provide additional information for institutional factors and put more emphasis on characteristics that may have been neglected in my analysis (for example if elections for justices of the peace are controversial).It has also been shown that the variables "contacts with conciliation authorities" and "positive experience with a conciliation authority" are probably too comprehensive, as they include all types of contacts and experiences.Thus, a more profound examination should include further distinctions, for example about whether respondents were one of the parties in a settlement conference, if they were in favor of the outcome or if they just considered their experience as positive due to a fair procedure.

Table 1 :
Conciliation authorities in all 26 cantons

Table 3 :
Public confidence in institutions

Table 4 :
Contacts and participation in elections

Table 5 :
Bivariate analysis: Correlation with public confidence in conciliation authorities

Table 6 :
Multiple regression: Estimates for public confidence in conciliation authorities Notes: Dependent Variable is confidence in conciliation authorities.Beta values.Numbers in parentheses are t values.*significant at a 0.01 level.