(BEING CONTINUED FROM 23/10/13)
3 Switzerland as a Perfect Laboratory for Empirical Analysis
Few countries in the world offer direct-legislative opportunities to their people, particularly
the opportunity to influence the daily political decision-making processes. Basically, only two
nations offer an extensive range of such direct-democratic means: the United States of
America and Switzerland31. The political system of Switzerland is shaped by two important
characteristics. First, in this federal country, direct legislative institutions exist on all three
levels – i.e. the federal level, the cantonal (‘state’) level, and the communal level. Second,
Switzerland shows a very strong fiscal decentralization, which, in contrast to Austria and
Germany, gives each level in the state its own tax sources. Therefore, a direct institutional
link exists between the power to tax and the direct-legislative institutions that provide local
citizens with the political means to influence both sides of the budget equally.
Thus, Switzerland can be viewed as a perfect laboratory for studying the impact of direct
democracy on political outcomes (see also e.g. KIRCHGÄSSNER, 2002, 2000): each canton – or
at the communal level, each commune – can be viewed as one observation with a varying
degree of direct democracy laid down in its constitution. All the cantons and communes,
however, share an identical macroeconomic and political framework at a higher level, so that
some difficulties arising in a cross-national comparison can be avoided. This common
framework is shaped by both policy at the federal level, international politics, and the
economic situation of the rest of the world.
At the cantonal level, there exist two direct-legislative institutions that are of particular
interest because they greatly affect the daily piecemeal works of politics: the fiscal
referendum32 and the statutory initiative (see FREY and STUTZER, 2000; see the review by
KIRCHGÄSSNER 2002 for fiscal aspects). Whereas the fiscal referendum is of a reactive nature,
the statutory initiative provides citizens with an agenda-setting power. Another difference
between these institutions is their focus: the fiscal referendum deals with expenditure projects
of the cantonal government – i.e. it is related more to decrees and by-laws; in contrast, the
statutory initiative directly influences the law-making process – i.e. it is either used for placing new (proposed) laws on the political agenda or for revising or eliminating already
existing laws 33 . Other important institutions of direct legislation are the constitutional
initiative and the statutory referendum34. Chapter II of this dissertation describes in more
detail the form that these four institutions take in the 26 cantons, how they have changed over
recent years, and how they can be used to construct a measure of direct democracy for use in
the subsequent analyses.
4 Structure of the Dissertation
The rest of this dissertation is organized as follows. The second chapter outlines and discusses
in detail the development of important institutions of direct legislation between 1997 and
2003. The goal of this chapter is to provide a basis for an update of the composite index of
direct democracy, which became necessary to carry out the empirical analyses presented in
the subsequent chapters. The third chapter presents an analysis of the impact of direct
democracy on individual well-being using a cross section of 1992 data and Swiss Household
Panel data from 2000 to 2002. Chapter IV analyzes the impact of direct democracy on the
efficiency of redistribution. The fifth chapter is dedicated to the question of how direct
legislation affects levels of property crime, hate crime, and sexual offenses. Chapter VI
presents an investigation into the impact of direct legislation on educational quality as
measured by PISA-like test scores. These empirical chapters are structured like scientific
papers and have already been presented as separate papers at various conferences; for this
reason the reader might find some overlapping in the arguments presented. Finally, chapter
VII concludes with a short summary and evaluation of the findings of this dissertation.
Chapter II:Direct Democracy 1997 – 2003
This chapter describes the changes in direct democratic institutions that occurred in the 26
Swiss cantons between 1997 and 2003. In general, institutions of direct democracy are
initiatives and referenda that are usually regulated in constitutional stipulations. This section
contains a brief introduction to these instruments of direct democracy and a description of the
so-called index of direct democracy that measures the strength of these institutions at the
cantonal level in Switzerland.
On the one hand, initiatives constitute an element of active electoral participation in the
political decision-making process. They allow the electorate to place a proposal for a law or
constitutional change on the political agenda. Accordingly, we speak of a statutory initiative
for altering laws and a constitutional initiative for a constitutional amendment. In
Switzerland, any revision of the cantonal constitution must be approved by the electorate, i.e.
it directly brings about a popular vote, as stipulated in the national constitution (art. 51, l
Swiss Constitution (SC)). In case of an initiative that refers to a cantonal law, a popular vote
must usually take place if the cantonal parliament makes a counterproposal, and only in some
cantons does a popular vote generally follow an initiative (i.e. without a counterproposal).
The requirement for any type of initiative is that a specific number of signatures is to be
collected from among the electorate, and it is not uncommon for the signature requirement to
increase proportionally to the importance of the legislative act to which the initiative pertains
(e.g. law, partial revision of constitution, total revision of constitution).
Referenda, on the other hand, constitute a reactive element of direct legislation because they
can only take the form of a reaction to preceding activities of the legislative body. Referenda
at the cantonal level are, in general, applicable to laws, decrees, international and
intercantonal treaties (concordats), and fiscal issues such as expenditure projects.
Accordingly, we speak of a so-called statutory referendum for laws, decrees and by-laws
(Gesetzesreferendum), an administrative referendum for administrative acts 35(Verwaltungsreferendum), a referendum for treaties (Staatsvertragsreferendum), and the
fiscal referendum (Finanzreferendum). In theory, all referenda could exist in both a
mandatory and optional form: a mandatory referendum triggers a popular vote automatically
following a decision by the representative body, whereas an optional referendum must be held
only if some specific requirements have been met by the electorate– usually a signature
requirement36. In theory, all referenda can also exist in an ordinary or extraordinary form. If a
referendum is ordinary, its (potential) application is directly stipulated in the constitution; it is,
in a sense, part of the daily political process. An extraordinary referendum can be held if a
minority of the representative organ of the canton demands it: in this case, the requirements
for taking this extraordinary optional referendum will be stipulated in the legal act to which it
refers. In general, not all types of referenda exist in one canton, and there is a huge variation
in requirements between the Swiss states.
To ensure a minimum level of direct democracy in each Swiss canton, the federal constitution
rules which institutions of direct legislation must exist at the cantonal level. Article 51, 1 SC
states that ‘each canton shall stipulate a democratic constitution’, where ‘democratic ‘ refers to
the organization of the canton, particularly the division of power, and to the fact that the
cantonal parliament is elected by the cantonal citizenry (EHRENZELLER et al. 2002, p. 624, no.
8). As regards institutions of direct legislation, as already mentioned above, only the
constitutional referendum and the constitutional initiative (both for a partial and total revision
of the constitution) are required in addition by art. 51, 1 SC (EHRENZELLER et al., 2002, p.
624, no. 9) 37. Although at the federal level a statutory referendum is stipulated in its optional
form (art. 141 SC), this institution is not prescribed for the Swiss cantons. Nevertheless, all
Swiss cantons guarantee their citizens more direct democratic rights than the required
minimum. In fact, the statutory referendum of direct legislation exists in all 26 Swiss cantons,
as do most of the other institutions such as the fiscal referendum and the administrative
referendum38. In addition, changes in cantonal territory are subject to approval through a
popular vote of the affected populations and cantons (art. 53, 3 SC), as are secessions or
unifications of cantons (art. 53, 2 SC).
As regards the institutions of direct democracy to be analyzed in this chapter, they are
restricted mainly to those that serve as a basis for constructing the so-called index of direct
democracy as developed by STUTZER (1999). This index is an unweighted average of four
subindices that measure the strength of four specific institutions of direct legislation; in
particular, the initiatives for constitutional and statutory changes, the fiscal referendum
(ordinary optional and mandatory) on expenditure projects, and the referendum for laws (and
decrees)39. To follow this chapter’s development more easily, the reader should recall that
these subindices are based on an evaluation of the requirements for each institution, on the
one hand, the signature requirements necessary for optional referenda and initiatives, and on
the other, the financial threshold for the (optional and mandatory) fiscal referenda. In order to
provide as complete a picture of constitutional changes in institutions of direct democracy as
complete, however, the discussion also lists most of the changes in administrative referenda
and any extraordinary statutory referenda. The administrative referenda are listed in the
section on statutory referenda, because with respect to their political intention, they are
closely related to the statutory referendum.
The original values of the index of direct democracy were presented by STUTZER (1999) and
STUTZER and FREY (2000) for the years 1970, 1992, and 1996 and were then updated for the
missing years between 1980 and 1998 by FELD and SCHALTEGGER (see their various articles
in the list of references). The data used for constructing this index up until 1996 can be found
in TRECHSEL and SERDÜLT (1999)40 (hereafter cited as T/S in this chapter), in which the
authors analyze the institutions of direct democracy and describe the changes from 1970
onwards (T/S 1999, p. 8). The index of direct democracy has since been employed in various
time-series cross-sectional analyses of the impact of direct democracy and is thus essential41.
For the purposes of this dissertation an update for the years 1997 – 2003 became necessary
because the data used to investigate the effect on education, crime, and subjective happiness
encompass only these more recent years42. It should also be noted that neither TRECHSEL and
SERDÜLT (1999) nor STUTZER (1999) analyze the cantonal institutions of the so-called Landsgemeinde cantons, i.e. those cantons that know no form of legislative representation or
delegation but vote on everything in an open meeting43. These Landsgemeinde cantons
excluded from the analysis in T/S (1999) are Appenzell Innerrhoden, Appenzell Ausserrhoden,
Nidwalden, Obwalden, and Glarus because in 1996 they were either still a Landsgemeinde or
had been so until recently (see T/S 1999, p. 7) 44.
The update of this index would not have been possible without the actual wording of the
changes provided by courtesy of various cantonal legal services and administrations.
Therefore, at this point I would like to thank all the various civil servants and honorary
collaborators, most particularly, Mr. Jürg Fehner (ZH), Mr. Urs Rüegg (ZH), Ms. Elisabeth
Vetter (LU), Ms. Kathrin Graber (LU), Ms. Judith Lauber, Mr Georg Zemp (LU), Mr. Heinz
Bachmann (LU), Mr. Peter Huber (UR), Mr. René Zehnder (SZ) and Mr. Peter Gander (SZ),
Mr. Notker Dillier (OW), Mr. Josef Baumgartner (NW), Mr. Hansjörg Dürst (GL), Mr. Bruno
Zimmermann (ZG), Mr André Schoenenweid (FR), Ms. Yolanda Studer (SO), Dr. Denise
Mangold (BS), Mr. René Bolliger (BL), Mr. Christian Ritzmann (SH), Mr. Joe Müggler (AR),
Mr. Köbi Frei (AR), Gabriela Küpfer (SG), Mr. Walter Frizzoni (GR), Mr. Urs Meier (AG),
Mr. Claudio Franscini (TI), Mr. Alex Depraz (VD), Ms. Séverine Despland (NE), and Mr.
Jean-Jacques Tombet (GE). I also thank ALOIS STUTZER, the inventor of this index, for
helpful comments and clarifications.
The rest of this chapter is organized as follows. The next section provides an overview of the
changes in the cantonal constitutions from 1997 to 2003. The subsequent sections separately
examine the institutional changes for the relevant direct democratic institutions and describe
their actual provisions. Addressed by these sections are the constitutional initiative, the
statutory initiative, the mandatory and the optional statutory referendum, and, finally, the
mandatory and the optional fiscal referendum. The outcome of this research is organized in
the form of tables intended to serve as a ‘manual’ for information on the relevant constitutional
requirements and for further consultation. These tables also allow construction of further
controls of direct democratic institutions, such as the signature requirements or the financial
threshold requirements for various time points between 1997 and 2003. Finally, the implications of these changes for the index are described and the constructional changes of
the index are discussed. The chapter concludes with a brief critique of the index.
(TO BE CONTINUED)
Justina AnastasiaValerie Fischer
Approved on the application of Prof. Dr. Gebhard Kirchgässner and Prof. Dr. Simon Hug
of the University of St.Gallen
Graduate School of Business Administration,
Economics, Law and Social Sciences (HSG) to obtain the title of Doctor of Economics
31 In many other countries, however, such as Canada, Austria, Liechtenstein, Italy, Ireland, and France, a popular
vote must only be held on constitutional amendments.
32 An ‘optional’ fiscal referendum can be distinguished from a ‘mandatory’ one. They differ with respect to the
expenditure threshold and the signature requirement, which must be met in the case of the latter. (See also
33 The constitutional initiative aims at changing, eliminating, or adding a new constitutional law. Usually, the
requirements for this initiative are the same as those for the statutory one, so that in econometric analyses the
impact of the two cannot be distinguished. See also chapter II.
34 In some cantons, even more specialized forms of referenda exist, such as those on highway expenditure or
international treaties. See chapter II and TRECHSEL and SERDÜLT (1999) for more information.
35 An administrative act is not necessarily a decision made by an administration; the distinction between an
administrative and a legislative act lies in the scope of applicability. In Switzerland, nomothetic acts are
defined as acts always referring to the regulation of either obligation or rights of persons, organization of the
state or duties and processes of administrations. Furthermore, administrative acts are not directly linked to
financial expenses. The wording in the various cantonal constitutions related to legislative and administrative
acts is not consistent between Swiss cantons. Administrative referenda are very difficult to identify as in many
cantons administrative acts take the form of a legislative act. On the other hand, there is also an almost
unmanageable variety of legal forms of legislative acts (see TRECHSEL and SERDÜLT 1999, pp. 14-15 and p.31).
36 In case of fiscal referenda, both forms require the meeting of a financial threshold in first place.
37 Changes in the cantonal constitutions must be in accordance with the federal law that is ensured by the
approval by the federal assembly (art. 51, 2 SC; art. 172, 2 SC). The federal assembly consists of two
chambers: the federal parliament (Nationalrat) and the representatives of the 26 cantons (Ständerat) (art. 148,
2 SC). In general, any cantonal stipulation should not contradict federal law (federal law breaks cantonal law
(art. 49, 1 SC)).
38 For an overview of the (non)existence of the most important institutions of direct legislation in Swiss cantons,
see LUTZ and STROHMANN (1998).
39 In many cantons, the requirements for the statutory and administrative referendum are identical. However, in
some cantons they differ, and STUTZER (1999) used either their average or either one of them to construct the
subindex of the statutory referendum.
40 Sporadically, the description of the changes reaches 1998, particularly in the second part in TRECHSEL and
SERDÜLT (1999) in which each single cantonal constitution is analyzed.
41 E.g., see the literature cited in KIRCHGÄSSNER (2002, 2001, 2000).
42 Differences between the values constructed by FELD and SCHALTEGGER (for 1997 and 1998) and my own are
mainly due to the extrapolation of old values without taking into account changes in the underlying cantonal
population. See also section 7 in this chapter.
43 In various papers by STUTZER and FREY (e.g. 2000) index values for the Landsgemeinde cantons are also
reported, without, however, a detailed description of their relevant constitutional stipulations.
44 Today (2003), 2 cantons (Glarus and Appenzell Innerrhoden) are still Landsgemeinden; Nidwalden ceased
before 1996, Appenzell Ausserrhoden in 1997, and Obwalden in 1998.