Justice and Court Administrations , Their Workings and Efficiency in Switzerland : Aspects Of Sentencing And Its Outcome In Swiss Cantons

Federal States, their administration of justice and their justice administrations, are characterized by differences in organizational settings, deficiencies in policy implementation, variations of policy outcomes as well as a dearth of information. Defendants of federal States claim that these well-known negative aspects of federalism are – largely – outbalanced by the speedier processes, efficacy and the efficiency of its largely autonomous units. In the field of the administration of justice, especially in penal justice and correctional programs, federalism must however also be considered with regard to the respect of principles of equal treatment, of equality before the law, of the state of law, all presumably protected by the Constitution.


Introduction
Federal States, their administration of justice and their justice administrations, are characterized by differences in organizational settings, deficiencies in policy implementation, variations of policy outcomes as well as a dearth of information.Defendants of federal States claim that these well-known negative aspects of federalism arelargelyoutbalanced by the speedier processes, efficacy and the efficiency of its largely autonomous units.In the field of the administration of justice, especially in penal justice and correctional programs, federalism must however also be considered with regard to the respect of principles of equal treatment, of equality before the law, of the state of law, all presumably protected by the Constitution.
The present study on the administration of justice with regard to the respect of the principle of equal treatment before the law (art.5 Swiss Constitution) is part of a larger project which compares the Swiss federal State composed of 26 cantons 4 and its policies.BADAC -the Swiss cantons and cities database (www.badac.ch)aims at providing data on public administration, government authorities and the policies pursued by cantons and cities.Its objectives are to: (1) follow up reforms at cantonal and city level, including justice and police matters; and (2) to benchmark activities of the cantonal States and to evaluate their impact on the socio-economic environment.The BADAC portal offers a powerful multilevel database and a web-mapping system which make graphic presentations of indicators and geo-localization of data accessible for everybody.Since the 1990s, regular surveys have been conducted through cantonal and city administrations on several subjects which are not collected by other statistical or research organization in Switzerland.In a survey undertaken in 2010, the BADAC collected for the first time, data on cantonal justice and police administrations as well as on the organizations of justice administration at the national and cantonal level.The first analysis of the collected data was published in the Atlas of the State 5 .
In this paper the focus lies on four issues related to the analysis of different applications of penal law in the cantons.First, we want to check out the figures relating to the application of penal law; we work on the total number of convictions as well as on the number of convictions containing criminal code offences broken down by canton.We move then on to the question of the use of pre-trial detention which has to be understood in relation to the number of unsuspended or partially suspended prison sanctions.Further, we describe the use of the prison sanctions and then we move on to question the efficiency of the application of law in terms of recidivism rates.
Initially, we concentrate this analysis on the main issues of the subject and will not address secondary aspects such as the impacts of variables such as gender, age or nationality.As we outline in the course of this exploratory study of important differences in the application of penal law by cantons, we still have to work towards a multidimensional analyses of the application of penal law in the cantons in the future.This means we will have to analyze several characteristics of the geographical location of cantons, the structure of its population, the state of the economy, victimization rates and 1 We want to express our gratitude to anonymous reviewers of our contribution.As non-native English speakers we are also extremely grateful to the proofreader for an in-depth stylistic revision of the text.Remaining errors are all of our own making. 2Dr. Daniel Fink has been leading the Department of crime and criminal justice statistics at the Swiss Federal Statistical Office between 1996 and 2010.He lectures since 2011 in criminology at the University of Lausanne at the Institute of Criminology and Penal law.This is an expanded version of the presentation elaborated by both authors and held by Christophe Koller at the EGPA Conference in Bergen in 2012. 3Dr.Christophe Koller is in charge of the Swiss database of cantons and cities (Banque de données des cantons et des villes suisses BADAC) at the Swiss graduate school of public administration, University of Lausanne. 4Appenzell Ausserrhoden (AR), Bern (BE), Basel-Landschaft (BL), Basel-Stadt (BS), Switzerland (CH), Fribourg (FR), Geneva (GE), Glarus (GL), Graubünden (GR), Jura (JU), Lucerne (LU), Neuchâtel (NE), Nidwalden (NW), Obwalden (OW), St. Gallen (SG), Schaffhausen (SH), Solothurn (SO), Schwyz (SZ), Thurgau (TG), Ticino (TI), Uri (UR), Vaud (VD), Valais (VS), Zug (ZG), Zurich (ZH). 5Koller Ch. et ali., Staatsatlas, Kartographie des Schweizer Föderalismus / Atlas de l'Etat, Cartographie du fédéralisme suisse, Zürich, NZZ-Libro, 2012.The chapter entitled Surveiller et punir (surveilling and punishing; pages 126 till 135) provides an analysis of the organization of the police authorities of the Federal State as well as cantonal and communal police forces and of several asp ects of policing activities.It also contains a study on differences, among cantons, on sentencing regarding different laws, e.g.criminal code, drug and traffic offences.On the website, the pages entitled Justice et Police, contain graphs and maps to many other subjects from the survey conducted in 2010; these expand on the data taken from the Swiss Federal Statistical Office.This additional data, contrary to the latter, the data can automatically be weighted with demographic information at various levels (canton region, among others) and made available in different tabular, graphical and cartographic presentations.police registered crimes.However, we have already advanced the hypothesis that the differences are not only the result of different social, economic and geographical conditions or different crime rates among cantons, but also due to different criminal policies, priority settings, inequality and, possibly discrimination in prosecution and sentencing 6 .

Studies in other Countries
Differences in the administration of justice and equality before the law have been the object of only a few studies in criminology, public policy analysis, history of law or the sociology of legal practice.Most existing research relates to social inequalities in the justice practice and in court proceedings, highlighting the importance of ethnic origin or social status of those prosecuted and sentencedall of which show that socially, economically or otherwise disadvantaged persons historically receive higher sanctions.American, British and Australian studies are particularly prolific on this subject (Bridge 1994, Hagan 1995, Grover 2008, Messmore 2010).Only a few take into account the governmental and public policies which may have a mitigating effect (Braithwaite 1979, Vogel 2007).Belgian, French and Italian researches have been very active on this subject as well; numerous articles were published for instance in the journal "Déviance et société" (see bibliography).The role of the administration in the management of justice and in the prevention of inequality of treatment in prosecution and sentencing has also seldom been the subject of comparative and quantitative research, at the decentralized level, or between federal governments.One has to turn to organizational sociology to find a few studies which consider the role of the federal States in the management of penal justice and inequalities in the treatment of offenders (Heinz 2008, 2010).Regional differences in the organization of justice, its administration and the outcome of its practices have never been studied, whereas in the field of health (Kunst, 1997) or in that of territorial management, these approaches are very common (Terribilini, 2001).

Switzerland
In Swiss studies as in international research, differences in the administration of justice and equality before the law have been the object of studies since the first annual statistical tables on sentencing published by the Swiss Federal Statistical Office (SFSO) in the beginning of the 20 th century 7 . Even at that time, the SFSO noted strong differences which could not only be defined by reference to the crime level in the cantons.In the same way, it was argued that the frequency of prison sanctions was not only related to the severity of crimes committed, but had to be understood with regard to unequal treatment of offenders for equal offences 8 .The subject received no more attention until the beginning 1990s, despite the fact that the publication of the SFSO became from decade to decade more voluminous and more differentiated with regard to cantonal statistics.For the first time in the 1990s, the differentiations of sentencing practices among cantons were used to complete a comparative analysis of the efficiency of sentencing practices.It was supposed that in examining the overall offence rate in the fields of mass delinquency cantons would have the full range of offences, from simple cases to more complex ones.This could be the case with offences such as drunk driving, consumption of drugs, theft, burglary, fraud, threats, among others.By comparing persons sentenced for the first time for such an offence exclusively, one could compare the outcome, e.g. the recidivism rate.The results showed that the recidivism rate was completely independent from the sanction, which means that cantons sanctioning offenders severely and those less severely have the same recidivism rate 9 .Later in the 1990s, differences in sentencing practices were studied with regard to illegal immigrants and the judicial handling of illegal stays by immigrants among cantons 10 .More recently, administrative decision making relating to residence permits for foreigners were the objects of a study of cantonal migration services 11 .Since the year 2000, the SFSO published, through the statistical web portal, a continually more expansive set of statistics on the 6 The first attempts made by the SFSO to establish conviction statistics were in 1906 and in 1909-1911, and then in 1929, extended  analyses were undertaken to evaluate differences in the application of penal law in the cantons.The objectives were to assess the quality of reporting and also evaluate and compare the reported conviction data to the crime level in the cantons and in the country.Since 1946 (using data dating back to 1936), Switzerland disseminates annual data on convictions.Despite the fact that in the study of 1906, there were strong doubts that the differences in cantonal conviction rates expressed differences in crime levels, no validation studies undertaken on such issues until late in the 1990s when differences in drug policies or prosecution of people with refugee or foreigner status were undertaken.The fundamental question of equality before the law has never been studied in Switzerland. 7Schweiz.Kriminalstatistik für das Jahr 1906, Berne, 1908;idem. von 1909idem. von bis 1911idem. von , Berne 1917;;idem 1929idem , Berne, 1931. 8 . 8 As Switzerland had not yet introduced its national criminal code, the differences between cantons could have been linked to differences in the content of the cantonal criminal codes.Nevertheless, it was predicted "that the differences among cantons, as far as they are based on the mentality of the population, the organization of the justice and the practices of the courts, will not disappear as the new Swiss criminal code will have been introduced."in: Schweiz.Kriminalstatistik 1929, Berne 1931, p. 9; translation of the authors. 9See for a summary of these studies of the 1990s: Storz Renate, Strafrechtliche Verurteilung und Rückfallraten, Berne, 1997. 10See Storz Renate, Strafrechtliche Verurteilung von Ausländern, Berne, SFSO 1998. 11Koller Christophe, Services de migration et évolution de la population étrangère dans les cantons suisses, 2002-2008, Cahier  IDHEAP 258, Lausanne, 2010.cantonal level; whereas these data are quite frequently used by cantonal administrations for internal studies, they are also becoming more frequently used for large scale and systematic comparative analysis by universities 12 .In the last five years, the SFSO substantiated former studies undertaken in the field of cantonal sentencing practices and recidivism.The methodologies used to study differences in judicial practice were mostly descriptive statistics; until now no more sophisticated statistical methods have been mobilized in order to find shared judicial policies or to analyze judicial practices.

The Data and Methodology
The data for this explorative study on cantonal differences in judicial practices are all taken from tabulated data sheets disseminated by the SFSO on its website 13 .The Office uses the criminal registry as an initial source.In principal, all convictions for felonies and misdemeanors must be registered; in few cases, even contraventions14 will have to be introduced.The data are aggregated on the basis of convictions and not by person which means that a person who is convicted twice in a year is counted twice once for each conviction.Whereas convictions are counted once, all offences in one conviction will be counted according to the law broken (the law however counted once only) or the offences committed 15 .This means that no major offence rule is applied.The same is not the case for the sanctions: only the major sanction will be counted, the hierarchy being prison sanctions preceding monetary penalties which precede community work orders.The partially suspended prison sanction is counted with the unsuspended prison sanction.
The methods used in this explorative paper are based on comparative analysis of aggregated data at cantonal level.For the time being, the comparisons between the cantons are based on four indicators in sentencing and its outcome, which are: (1) the total number of convictions and the total number of convictions for criminal code offences, (2) the number of pre-trial detentions among all convicted persons, (3) the number of prison sanctions imposed, and (4) the reconviction rate.There are many more aspects of sentencing which will have to be taken into account when comparing sentencing practices including: the number of trials undertaken, the duration of trials, or the amount of fines imposed, just to quote three examples.The comparisons among cantons are undertaken based upon data weighted by resident population.

Sentencing Practices of Cantons
The first point relates to the application of penal law in the cantons.Before starting the analyses of differences, we want to provide some information on the background of the application of penal law: Switzerland has only had a unified criminal code 16 since 1942.Before that date, the penal law was a cantonal matter.Since 1942, some 250 additional federal legal regulations with penal provisions were added, of which the military criminal code, the traffic code, the drugs law and the law on foreigners are the most important and the most often applied ones.The application of the penal laws and the execution of sentences in Switzerland are in the responsibility of the cantons.
In 1985, Swiss courts handed down 46,500 convictions 17 , 22,000 for criminal code offences, 21,000 for traffic offences and 3,500 for offences due to other laws 18 .In 2010, there were 98,000 convictions pronounced (+111%), of which 30,000 were criminal code offences (+36%)over one third concerned theft and other property related offencesand 56,000 traffic law offences.The rest were drug law offences and foreigners sentenced for illegal entry or stay in the country.In terms of the weighted figures, in the year 1985, there were 716 convictions per 100,000 of the population and 335 convictions for offences of the criminal code.In 2010, the weighted number of total convictions increased to 1,200 per 100,000 population 19 whereas the criminal code offences accounting for 360 per 100,000.One further distinction: we speak here of the total work load of the courts and tribunals, because in Switzerland at least one third of the sentenced foreigners every year are not members of the resident population in strictest sense, as they are foreigners in transit, on holiday, without documents (the so-called "sans-papiers") or illegally in the country 20 .
Note: The source for all data is the conviction statistics of the Swiss Federal Statistical Office (SFSO), adult convictions.
The graphs have been designed by the authors.
For a first comparison of conviction policies, two rates are used, on the one side the prevalence of convictions per 100,000 population, keeping in mind that 57% of all convictions relate to traffic offences, on the other side the rate of convictions for criminal code offences per 100,000 population.As the graph shows, there are strong differences among the cantons in their sentencing frequency; the variation is 1 to 6 at the extremes or 1.0 to 1.5 on the average for the bulk of the cases.The extremes are quite easily explained: nearly all cases in the canton of Uri (UR) relate to traffic offences, due to the location of the important transit highway through the Alps in the jurisdiction of this canton; weighted by the population, there is a high sentencing for traffic offences.On the other extreme are also three cantons (AR, GL, AI) which are quite small, located in hilly or even mountainous and more rural areas.For the other cantons, we find quite some opposition in the handling of delinquency by cantons with similar structures, such as the canton of Vaud (VD) convicting many more people than Zurich (ZH), even though Zurich has a much more urban population.But there are other differences: whereas the canton of Vaud has a higher overall rate of sentencing than Geneva (GE), the latter has a much higher rate of criminal code convictions.However, the canton of Geneva conviction rate is overrated with regard to criminal code offences by a smaller much less urbanized canton: Neuchâtel.Generally speaking, we find on the repressive side of the divide the French speaking cantons (FR, JU, GE, NE, VD) with higher conviction rates and on the other the more German speaking ones with lower ones.
Using the differences in conviction rates as a starting point, we want to look at the ways the input is congruous with the sentencing and how the output in terms of sanctions, especially prison sanctions, is in agreement with the sentencing.Due to the fact, that pre-trial detention is one of the most coercive measures prosecutors and judges can take, we first have to examine the use of these measures in the cantons.Two further clarifications on the next graphs: As explained above, the canton of Uri has jurisdiction over the major transit highway through the Alps which results in a high offence rate comprised mainly of traffic offences.The canton has therefore been left out of the following analysis.As detailed analyses have shown, the canton of Basel-City is also quite atypical, because it is a city-canton without any more rural area.We aggregate therefore the statistics for the canton of Basel-City and the canton of Basle-Countryside, as they are more valid for inter-cantonal comparisons.One finds therefore in the following graphs the details for Basel-City (BS) and Basel-Countryside (BL) along with the aggregated ones (BS+BL).
20 Despite of the politicization of the issue, Switzerland will have to improve its reporting system on delinquency of foreigners.The Swiss Federal Statistical Office published a first study for differentiated and weighted population groups in 1998 with the reference year 1994.In 2000 the report of an Interdepartmental working group on "Criminality of foreigners" provided actualized figures on the subject with the reference year 1998.Since then, no study has been undertaken.Currently, it is at the University of Zurich that a research team is getting ready to conduct a new detailed statistical study with judicial data (Prof.Christian Schwarzenegger, University of Zurich).

Application of Pre-Trial Detention Orders
Pre-trial detention is one of the most coercive measures the judiciary may impose on a person presumed innocentand every person is innocent until he or she has been convicted.The Swiss Criminal Procedure Code (CPC) 21 states therefore that "an accused person remains in liberty" (Art.212 CPP).Pre-trial detention has to be used with extreme restraint, as the code regulates that "pre-trial detention may not last longer than the prison sanction which may be expected" (Art.212 para.3 CPP).
In general, as the conviction figures increased regularly over the last 25 years, so did the absolute figures for pre-trial detention which increased from 9,000 cases in 1985 to 14,500 cases in 2010 22 .However, in relative figures, there has been during this period a marked reduction of the frequency of the application of pre-trial detention among all convicted persons; the percentage declined from 19% to 14%.Again, these changes are not equally distributed as one finds 20 cantons with decreased use of pre-trial detention, while a few other cantons, more urban ones, increased their reliance on pretrial detentions (BS, GE, LU, SZ, ZG, ZH).
There are also large differences among the cantons in the application of pre-trial detention in 2010, stretching in terms of percentage differences from 1 to 10.A vast majority of cantons have 10% or less of their convicts who have been placed in pre-trial detention.On the other end of the spectrum is the canton of Geneva, which orders up to 55% of all those convicted to be placed into pre-trial detention, followed by Basel-City, Zurich and Schaffhausen.If we sum with Basel-Countryside, the rate normalizes the figures of the two cantons coming closer to the national average; the figure for the pre-trial detention remains however quite high.
Another way of looking at the use of pre-trial detention is to consider the percentage of those with an unsuspended or partially suspended prison sanction, taken together, with the percentage of those receiving pre-trial detention.Comparing both rates, one would expect, as the criminal procedural code demands, that most of those in pre-trial detention will face an unsuspended prison sanction.This however, is not the case as more than half of those sentenced to an unsuspended prison sanction in Geneva, Basel-City or Zurich, or Schaffhausen will not be sentenced to such a sanction.Nationally, the rate of those being sentenced to an unsuspended prison sanction with a previous pre-trial detention period is 48%.Three cantons show rates which point in the other direction.In the cantons of Bern, Jura, Neuchâtel and Vaud, the number of persons sentenced to an unsuspended prison sanction is higher than the percentage of those having faced a pre-trial detention, showing a higher severity in sentencing.The differences are most important in Neuchâtel, where twice as many persons will get a prison sanction compared with those who were put into pre-trial detention.
These figures provide some initial evidence, on the one side, of a massive use of pre-trial detention in the majority of the cases where the outcome is a conviction without an unsuspended prison term.The cantons where prosecuting offices most frequently impose pre-trial detention -Basle-City, Geneva, Zurichare also among those cantons in which courts most often impose unsuspended prison terms or convictions.On the other side, the four cantons with more unsuspended prison sanctions compared to the use of the pre-trial detention show, to say the least, a problematic severity in punishing. 21Swiss Criminal Procedure Code, RS 312.0, on http://www.admin.ch/ch/e/rs/312_0/index.html 22 Data from SFSO Datasheet 19.3.3.5.2, dated 30.06.2011.Again, included among them are the more severe cantons of the French part of Switzerland (Jura, Neuchâtel, Vaud) and the traditionally severe canton of Berne.
Here as with convictions, we will have to look closer to internal and external causal factors for the frequency of use of pretrial detentionit is however from the outset clear, that it cannot be the crime rate which is solely the cause.The rates just analyzed show no evidence of a strong parallel link between the conviction rate and the pre-trial rate or between the pre-trial part and the prison sanction part, which means there must be extraneous factors involved for the explanation for these diverse relationships.

Unsuspended Prison Sanctions
In 2007, a revised sanction system was enacted.It was explicitly designed to produce a strong reduction of the use of the short prison sanction (up to 6 months of duration) Until 2006, courts handed down annually some 80 to 90,000 convictions, 25% of which contained a fine as a penalty.The other 75% of convictions noted a prison sanction, of which 75% of the cases (2006: 56,000) were suspended, 95% of which were of a duration of 3 months and less 25 .The same holds true for unsuspended prison terms: 85% of those were unsuspended and short, e.g. of 6 months or less.Between 1984 and 1995, there were some 14,000 prison sanctions annually; this figure drops to 12,000 for another 8 years, before elevating to 16'000 in 2004.
The revised penal code of 2007 requires that short prison sanctions of up to 6 months shall only be imposed if no other penalty is suited to punish the offender (art.41 para. 1 CC 26).Moreover, the court has to motivate in a detailed way the choice of the short prison term (art.41 para. 2 CC).This change of the regulation brought the prison sanctions down from 16,000 to 6,500, whereby 9 out of 10 were unsuspended ones, and only 1 was partially suspended.

27
. Since 2007, the trend of unsuspended prison punishments is slowly increasing (2010: 6950).Nevertheless, nearly half of those (45%) have still a duration of 6 months or less, imposed mostly on persons without a permanent residence status in Switzerland, meaning foreigners.The suspended prison sanction fell from 42,000 cases in 2006 to 2500 in 2007 and has been stable since that year.More importantly, the short suspended prison sentence of up to 6 months disappeared completelyit vanished.This information provides a short hand description of the current situation.
In 2010, in terms of differences of sanctioning, we find the following picture.The national average of unsuspended prison sanctions is 80 per 100,000 inhabitants.The cantons of Geneva, Vaud, Basel -City and Neuchâtel impose highly above the national average of unsuspended prison terms.Geneva has a rate of 300 unsuspended prison sanctions for 100'000 of its resident population, Vaud 150, Basel-City 133 and Neuchâtel 100.Geneva and Basel-City are border cities without large rural areas, where as Vaud has a large rural area, as for example do St.Gallen or Aargau, which however have rates quite below average (resp.59 and 50 per 100,000).To indicate that there are differences in sanction practices it must also be suggested that there are 13 out of 26 cantons which impose 10 or more times less prison terms than Geneva.
In the next sentencing graph, the three series of figures are combined.They show the uncorrelated relationship of the frequency of convictions, pre-trial detentions and unsuspended prison sanctions.The cantons with the highest rate of convictions are not those with the upmost number of pre-trial detentions and not those with the largest number of prison sanctions.Geneva, with highest rates of pre-trial detentions has also the highest rate of prison sanctions (possibly to cover the excessive use of pre-trial detentions), but ranks only at 8 th place for the rate of sentenced persons.On the other hand, whereas Zurich places a large number of people into pre-trial detention, in most of the cases, they get away without a prison sanction.
To explain these differences, one will have to pay attention to the input into the process of prosecution and of output in terms of efficacy of the policies pursued.In order to explore these factors, a large number of additional data series will be needed such as geographical, economic, social and cultural information, combined with additional data on personnel, priorities in crime matters and the police crime records.
In order to further complicate these pictures of quite unrelated distributions, one has to pay attention to the question of recidivism which is here used to evaluate the actions and efficiency of the administration of justice.In Switzerland, the use of short prison sanctions has been very strongly criticized on the grounds that it is a massive, life intrusive and a stigmatizing way of sanctioning nonviolent and non-sexual offenders.It has been shown that this way of sanctioning is not more efficient than the other ways of punishing, or said in another way, less severe sanctionsshorter prison terms, community service orders, monetary penaltiesfor less severe offences are substitutable and equally efficient.The SFSO has, in the past years, elaborated several findings on that issue which may here be summarized 28 .

Differences in ways of sanctioning and efficiency of sanctions
There are not many universally valid results in criminology; some regarding the efficiency of sanctions measured on the recidivism rate however are.For instance, it has been demonstrated by dozens of studies 29 , that men have higher recidivism rates than women, young men higher ones than older men, people with previous convictions higher ones than those without previous records.Finally, all over the world, those with prison sanctions have higher reconviction rates than those with any other sanction.Especially with regard to the short prison term, the differences among the cantons are the most valuable statistical raw material to undertake comparative analyses in the way sanctions work.Without going into details on technicalities 30 , let's have a look on people who have been sentenced for drunk driving exclusively, without any previous conviction.The figures are selected according to the most severe to the least severe cantons 31 . 28For an overview of the current analytical principles of recidivism studies and technicalities, see Fink D. and Vaucher S., Recidivism studies in Switzerland 1996-2010, in: Jehle J.-M., Recidivism studies in Europe, forthcoming 2013. 29For an overview on the subject see Kerner H.J., Rückfallstudien, in Kaiser G. et alii.Kleines Kriminologisches Wörterbuch, Munich, 2004. 30For more details consult explanations on the SFSO website about Recidivism analyses. 31The graph is taken from the website of the SFSO www.statistik.admin.ch.
People sentenced for the first time for drunk driving will in general get a suspended prison sanction.There are still however, a great number of cantons which also impose fines in a great number of cases, the unsuspended prison sanction being imposed less.One can group the different styles of sanctioning in three groups, the 10 most severe cantons, the less severe 8 and a last group of the 8 least severe cantons.If one looks however at the specific recidivism rate, there are no differences: in all cantons the recidivism rate lies between 8% and 12%.The two cantons imposing the most severe sanctions, Schaffhausen and Fribourg are quite equal to the cantons of Zug and Obwalden which impose the less severeor more lenient sanctions.And the two neighboring cantons of Vaud on the severe side and Geneva on the side of more tolerant cantons have no significant difference in terms of recidivism of those sentenced.
One may now push forward this examination and look at those who were convicted for a second time: we find that the majority will get an unsuspended prison sanction on their second conviction, whereas the others who were fined will get a suspended prison term.The recidivism rate is more varied, covering a wider range of values, but there is no correlation between cantons sanctioning in a severe or in a less severe way.One finds no argument in favor of the efficiency of prison sanctions.This result has been confirmed for many other offences with similar outcomes, including theft 32 .They show, even confirm, that the recidivism rate is not dependent on the sanction imposed and that sanctions may be exchanged without a loss of efficiency of special or general deterrence, at least in the domain of lesser severe delinquency.Combining the argument with economic considerations, one might consider, in times of major budget constraints, that the imposition of monetary penalties is much more beneficial, in financial and social terms, to society than imposing prison sanctions. 32For Switzerland see : Storz, R., Strafrechtliche Verurteilung und Rückfallraten, Berne, 1997; Vaucher S. et alii., Strassenverkehrsdelinquenz und Rückfall, Neuchâtel 2000; for Germany see: Heinz W., Rückfall-und Wirkungsforschung -Ergebnisse aus Deutschland, see: www.http://www.uni-konstanz.de/rtf/kis/Heinz_Rueckfall-und_Wirkungsforschung_he308.pdf,accessed on 1.10.2010.

Conclusion
The results shown in this research paper are a first illustration of a much more ambitious work currently in progress at the University of Lausanne, based on the data available at the BADAC (www.badac.ch).They confirm that the data allow for strong comparisons among cantons with regard of the application of penal law and measures of crime policy.For the time being, we used only data from conviction statistics.There will be a need to expand these studies in the future using input data such as police crime statistics and output data such as after care of persons released to probation services.
The comparison of the operations of the administration of justice among cantons shows on one side large differences in the three major types of sentencing, in the use of pre-trial detention and the unsuspended prison sanction.When combined, one finds however very weak relationships when considering absolute, percentage or weighted results.On the other side, the outcome of these different policies is much paradoxical as there are no differences when comparing recidivism rates among cantons, despite strong differences in the use of pre-trial detention and the sentencing with prison sanctions.
The paradoxical outcome of crime policies in terms of recidivisme.g. the absence of differences of the outcome based on sanctions in the domain of less severe delinquencysuggests the need for more empirically informed crime policies.The role of justice administrators could be to participate in the dissemination of those findings as well as the dissemination of best practices among cantons with regard to outcomes and the use of resourcesespecially with consideration to the use of the prison sanction as it is the most costly and the most inefficient of all sanctions.Furthermore, the observance of the principle of equality before the law would be most likely be promoted.

23.
Daniel Fink characterizes this process as The Vanishing of the Prison sanction in Switzerland 24 .