Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=fwep20
Download by: [Masarykova Univerzita v Brne] Date: 27 April 2017, At: 04:49
West European Politics
ISSN: 0140-2382 (Print) 1743-9655 (Online) Journal homepage: http://www.tandfonline.com/loi/fwep20
How does direct democracy matter? The impact of
referendum votes on politics and policy‐making
Yannis Papadopoulos
To cite this article: Yannis Papadopoulos (2001) How does direct democracy matter? The impact
of referendum votes on politics and policy‐making, West European Politics, 24:2, 35-58, DOI:
10.1080/01402380108425432
To link to this article: http://dx.doi.org/10.1080/01402380108425432
Published online: 03 Dec 2007.
Submit your article to this journal
Article views: 1606
View related articles
Citing articles: 19 View citing articles
How Does Direct Democracy Matter?
The Impact of Referendum Votes
on Politics and Policy-Making
YANNIS PAPADOPOULOS
DIRECT DEMOCRACY IN THE FEDERAL POLITICAL SYSTEM
Throughout the world, referendum procedures display a wide variety of
forms. A majority of European countries and more than one-third of UN
members have had some form of referendum in their history. In Switzerland
- the country with the largest number of referendums held at the nation
level1
- this variety is highly codified. The referendum has been mandatory
for any constitutional change since the beginnings of the federal state in
1848. The domain of mandatory referendums has subsequently been
extended to the ratification of major international treaties and for
membership in supranational organisations. In addition, constitutional
referendums have been held increasingly frequently owing to the
constraints of federalism. In federal Switzerland, the growth of central state
activities - a typical feature of most modern welfare societies systematically
requires formal approval by a double majority of voters and
cantons.
The focus of this account is not, however, the consequences of
mandatory referendums. Instead it is concerned with referendum
mechanisms that result from pressure 'from below', which differentiate
Switzerland, Italy at the national level (to a lesser extent), plus several
American states from the many other countries with referendum
institutions.2
As a rule, the countries where referendums are used most
frequently are those where referendums can be initiated from below. In
Switzerland, notwithstanding the increase in mandatory votes (206 since
1848),3
referendums generated by a petition from below are more than half
of the total: 132 optional referendums since the constitutional reform of
1874, and 127 popular initiatives since the reform of 1891.4
These referendums share the property of not being under the control of
the political system.5
By way of petition, a number of citizens (be it an
Yannis Papadopoulos, University of Lausanne
36 THE SWISS LABYRINTH
absolute number or a proportion of the electorate, of voters in a previous
election, or suchlike) can, as a rule at any time, decide that an issue should
be submitted to a vote. The outcome of this vote is not merely consultative
but binding. What matters here is that the initiative for the vote originates in
part of the electorate, not in a political institution - the presidency, the
government, the majority or even a minority in parliament. Hence, as
viewed by those 'above', these referendums cause more uncertainty than
referendums which can be anticipated because of constitutional provisions,
or those decided in a discretionary way by political bodies: '[t]ypes of
popular votes in this category probably raise the most questions and
problems regarding the compatibility and integration of the referendum
phenomenon with constitutional representative government'.6
Significantly,
Swiss citizens were not granted these referendums by the governing elites.
The referendums were introduced into the constitution under pressure from
reform movements in the second half of the nineteenth century, after a
number of cantons had accumulated some experience with them.
There are two types of referendum 'from below' in Switzerland at the
federal level (and others at the cantonal level):7
1. the optional referendum (henceforth 'referendum'): if 50,000 voters sign
a petition opposing a bill 90 days after its passage in the bicameral
Parliament, the bill must then be approved by a majority in a referendum
vote in order for the bill to be enforced;
2. the popular initiative (henceforth 'initiative', equivalent to
'propositions' in the United States): if 100,000 signatures are collected
within 18 months to propose a constitutional amendment,8
then a
referendum must be held. The outcome will be binding, provided a
majority of voters and of cantons supports the proposal.
Thus, it appears that referendums and initiatives serve different
functions in Switzerland.9
The optional referendum allows a group of
citizens to attempt to overrule an existing decision, whereas the right of
initiative enables citizens to put radically new proposals on the agenda.
Optional referendums close the legislative process and thus seek to correct
'sins of commission', whereas initiatives open the process, attempting to
correct the parliamentary majority's 'sins of omission'.10
Referendums
result in 'votes of control', and initiatives result in 'votes of promotion'."
It can be argued that the latter are even less under the control of the system,
for their impulse is not conditioned by any prior decision of parliament.
Hence, initiatives more strongly constrain the margin left to
representatives.
HOW DOES DIRECT DEMOCRACY MATTER? 37
There is little doubt then that '(semi-) direct democracy matters'.12
It
matters when the outcomes of referendum votes are not those desired by the
federal government or by a majority of parliamentarians. It matters even when
these outcomes are congruent with their wishes, for the voice of the people
confers additional legitimacy on policy choices. However, here we should
scrutinise the more subtle indirect effects of direct democracy and their overall
impact on the Swiss political system. On one hand, direct democracy
mechanisms that remain uncontrolled are expected to block the choices of the
ruling elites (referendum) or to upset their priorities (initiative). These
mechanisms can engender governability problems, because they may cause
'irritation' to the political system, as sources of 'noise' impeding its ordinary
operation. On the other hand, as viewed 'from below', they are a useful
resource likely to modify the power balance in favour of outsiders confined to
a marginal position in the official decision-making process.13
Consequently,
established elites and organisations will also, in all likelihood, do their best to
influence the outcome of uncontrolled referendums: by making
recommendations to the rank-and-file and their sympathisers, by campaigning
in the media, and so forth. Yet, the attitudes and behaviour of the populace can
always surprise the rulers, making referendums more than a simple
reproduction of the balance of forces in parliament. For example,
approximately one in four bills rejected by the voters were supported by all
parties in the federal grand coalition as well as most interest groups, and were
opposed solely by very marginal national-populist parties.14
As a result, those
in power often try to erect protective barriers and defensive mechanisms, as a
form of risk management, to cope with the uncertainty caused by referendums
emanating from below. Thus, individual referendum votes are not only
decisive for policy outputs, but also direct democracy as an institution requires
that the political system as a whole adjusts to the pressure caused by it.
Political actors do modify their behaviour in response to the challenge of
direct democracy. This is by no means a simple, deterministic 'iron law' of
direct democracy. There is no mechanistic process, whereby the universal
constraints of direct democracy dictate a single appropriate response by
political actors: political systems are not 'trivial' mechanisms." Strategies
and choices in politics are, more often than not, the outcome of reflection,
calculus, routine, inheritance from the past, and so forth, as stated in the vast
literature on rational choice, bounded rationality or path dependency.
Consequently, similar pressures exerted from below, by virtue of directdemocracy
mechanisms, can be interpreted very differently and can trigger
diverse responses by actors socialised in different settings, as is shown by
comparative research on the issue.16
38 THE SWISS LABYRINTH
Nevertheless, it can be argued that adaptive behaviour to the challenge
of direct democracy has taken three forms in Switzerland. These are the
major strategies pursued by elites in order to lessen risks arising from the
referendum. The first two strategies aim to prevent recourse to direct
democracy, the third to steer the processes it engenders:
1. Widening the executive formula, to encompass all parties likely to make
efficient use of the referendum if not co-opted as partners in the
governing coalition (it should be remembered here that inclusive grand
coalition governments are typical power-sharing devices in
consociational democracies).
2. Anticipating the veto risk by negotiating ex ante with opponents to the
policy reforms that originate in government or parliament. Bills are
amended as early as in a pre-parliamentary phase, to incorporate the
claims of 'referendumsfähig' actors (associations, parties and
movements that enjoy a reputation of winning majorities in referendum
votes). Alternatively, these bills are simply withdrawn if they encounter
too much opposition.
3. Negotiating ex post when recourse to direct democracy cannot be
prevented. This occurs in the case of initiative promoters whose claims
can be partially met in a formal counter-project (a more moderate
constitutional amendment also requiring a double-majority referendum
vote) or, more frequently, can be met in legislative amendments that will
not necessitate a popular vote (unless challenged by an optional
referendum).
One section is devoted to each of these strategies, before an examination of
whether this institutionalist approach thoroughly captures the dynamics of
the Swiss political system, and before conclusions on the validity of the
traditional functions of direct democracy today.
THE GRAND COALITION AS A DISINCENTIVE TO USING DIRECT
DEMOCRACY
In Switzerland, direct democracy contributed to the advent of consensual
practices in spite of, or rather because of, its majoritarian characteristics as
a decision mode. Consensual practices have been established to prevent or
to moderate the use of direct democracy. This well-established thesis on the
impact of the referendum originated with Leonhard Neidhart, in a
pioneering work that was neo-institutionalist avant l'heure." Following this
HOW DOES DIRECT DEMOCRACY MATTER? 39
line of thought, direct democracy is the most important
t
Konkordanzwang':n
a constraint that forces political elites to adopt a
strategy of co-operation rather than confrontation. In his Democracies,
Arend Lijphart may have been too hasty in concluding that direct
democracy is specific neither to majoritarian nor consensual democracies,
though he was right in considering the great variety of its forms that
prevents almost all generalisation. Direct democracy is typically a
majoritarian device. As a result, where direct democracy is well-established,
it may in the long run trigger consensual responses to avoid it, although as
noted earlier - this is by no means necessary.19
The first group to make successful and extensive use of the referendum was
the Catholic-Conservative Party in the second half of the nineteenth century.
Although they managed to keep control over their cantonal strongholds thanks
to federalism,20
the Catholics had been excluded from the central government
since their defeat in the Sonderbund war and the creation of the Swiss
federation in 1848.Hence, the referendum proved to be a useful tool. It enabled
them regularly to challenge important decisions of the incumbent Radical
Party, which concerned the centralisation of competences and secularisation of
society. In 1891, this result, together with the rise of the socialist movement
that forced the bourgeoisie to silence its internal contradictions, led the
parliament to elect the first Catholic-Conservative member of the collegiate
federal executive. Prior experience with coalition government in the cantons
(which are frequently laboratories of institutional innovation) helped allay
fears about what was an important novelty at the time.
Gradually, through a process of learning and imitation, other parties with
a strong referendum power also joined the federal government, leading to
the so-called 'magic formula' of seven members. The 'magic formula',
unchanged since 1959, safeguards the representation of the four major
parties in principle in proportion to their electoral strength: two federal
councillors (ministers) for the Radicals (now a right-wing party, mostly
representing business interests), the Christian Democrats (exConservatives,
mostly based in Catholic cantons), and the Socialists. In
addition, there is one member for the Swiss People's Party (SVP, initially an
agrarian splinter of the Protestant Radical milieu, but increasingly a
nationalist-populist party, and today the strongest party in the country).
Parties able to demonstrate a 'blackmailing power' through direct
democracy thus managed to convert this into 'coalition power'.21
The case of the Socialists, however, shows the limits of this integrative
process, and illustrates its non-mechanistic character: 'paths' are not always
straightforward in path-dependency patterns. Their first member of the
40 THE SWISS LABYRINTH
federal government was elected in the middle of World War II in 1943, in
an atmosphere of 'holy alliance' between the major social and political
forces of the country. Prior to this, the Socialist Party had had to water down
its programme dramatically and had made considerable efforts for over a
decade to be accepted as a partner in the governing coalition. By 1943, its
representation in the Federal Council was similar to that of the People's
Party, which was ideologically much closer to the Radical and Conservative
incumbents, and had been accepted as a partner much earlier, although it
only had half the electoral strength of the Socialists. It was not until 1959
that the Swiss Socialists obtained representation in government proportional
to their electoral support. This suggests that, even if rulers 'learned' from
their opponents' frequent use of direct democracy, this learning process
took considerable time and was also influenced by the ideological profile of
those who were to be co-opted.
Notwithstanding these qualifications, the magic formula in turn seems to
have had some impact on the practice of direct democracy: since the 1960s,
the proportion of bills challenged by referendum has fallen to just seven per
cent. In addition, when referendum votes do take place, voters more
frequently support government policies. Although only 54 per cent of
parliamentary decisions submitted to the vote (mandatory and optional)
between 1848 and 1960 survived the challenge, this proportion increased to
72 per cent between 1961 and 1999. This change is particularly striking in
the case of bills submitted to referendum by petition. It was not only the
proportion of bills challenged that decreased, while the increase in the total
number of optional referendums was simply due to increased legislative
activity by parliament. Prior to 1960, a majority of bills challenged were
rejected by the people (41 out of 65). Subsequently, only a minority of them
was rejected (26 out of 67).
What is more, we observe in Switzerland a differentiation between
direct and representative democracy.22
This created a 'fragmentation of
political risks':23
it is possible that the safety valve of direct democracy has
prevented opposition parties from gaining more influence in the sphere of
electoral competition and of parliamentary politics. This isolating cocoon
prevents contagion from the referendum to the representative scene and
could help explain the low volatility of the federal party system. To give but
one example, the strong support for xenophobic popular initiatives contrasts
with the weak electoral support for parties backing these initiatives.24
The
differentiation of the systems of direct and representative democracy has
been a major ingredient for the overall integrative role of direct democracy
in Swiss politics.
HOW DOES DIRECT DEMOCRACY MATTER? 41
The effects of direct democracy on elite behaviour and on the mechanics
of decision processes are not limited to impacts on the composition of the
federal executive. 'Governmentalising' opponents to public policies is only
the formal aspect of their co-optation, a way to neutralise their veto power
by awarding them some influence. Other less formal aspects - albeit
strongly consolidated now and no doubt equally, if not more, relevant25
accompany
this facet of co-optation, and will be surveyed in the next
section.
EX ANTE NEGOTIATION AS AN ANTICIPATION OF DIRECT
DEMOCRACY
Not only have opposition parties gradually been co-opted into the federal
government, but the views of any group considered to be a credible user of
direct democracy are considered too. The decision process is long and
complex,26
with a high level of institutional redundancy. Several phases
succeed one another, with actors playing 'nested games', seeking in each
phase to anticipate the power balance in the next one, particularly in the
most decisive referendum phase.
There are several moments when, in a direct or in an indirect manner, the
Damoclean sword of the referendum shapes actors' behaviour:
• The decision process often begins with a pre-parliamentary phase, more
frequently when decisions seem important. Sometimes, in addition, an
expert committee will meet regularly to draft a preliminary version of
the bill.27
Despite the name, the members of these preparatory bodies are
primarily representatives of social forces that count. Their activities thus
combine the principles of 'intellectual cogitation' and of 'social
interaction'.28
Furthermore, according to interviews given by their
members, their work takes place in the shadow of the referendum,
whether mandatory (a majority must be obtained) or optional (the
referendum would better be avoided).29
This is not to say that these very
respectable experts, who happen to defend particular interests quite
openly, would overtly make threats. The experts internalise the 'second
face' of power sketched by Peter Bachrach and Morton Baratz in their
Power and Poverty in a much more subtle way. It leads them to selfcensor
with respect to actors having blackmail potential - even those not
represented on the committee.
• In a substantial number of cases, the pre-parliamentary phase continues
beyond the usual informal contacts between the federal bureaucracy and
42 THE SWISS LABYRINTH
various groups. The process continues with the highly institutionalised
'Vernehmlassungsverfahren', whereby major groups are officially
consulted on a first draft of a bill. The outcome of these consultations is
then analysed and interpreted by the administration. Even in the absence
of conclusive empirical findings, it can be reasonably argued that, the
greater an actor's referendum power, the more consideration is given to
its support or criticisms.30
The federal executive must then choose
between simply presenting the bill in parliament, amending it, or
withdrawing it. The federal executive thereby remains an important
filter in the decision process. But perhaps the real power lies in the hands
of high bureaucrats who are better able to assess the life chances of a
bill, or, ultimately, in the hands of actors with veto power, as a result of
the referendum.
• There is another veto point: the federal parliament, where bills are
debated before being submitted to a popular vote. The power of
parliament is subject to some controversy in Switzerland, since the
Federal Assembly makes few amendments to government bills, and only
recently has its input into federal legislation grown.31
But it cannot be
said that the parliament merely ratifies decisions made informally in the
pre-parliamentary phase. The pre-parliamentary phase cannot perfectly
anticipate parliamentary behaviour. Some bills that proved to be very
controversial in parliament had undergone intensive pre-parliamentary
consultations. This suggests that, when conflict is acute at the outset,
consociational procedures do not absorb conflict and that they have
reached their limits.32
More importantly, a simple parliamentary majority
is not enough to avoid defeat in a subsequent referendum.33
The plurality of veto points makes it all the more necessary to form wide
coalitions around reforms. The inevitable consequence is a bias toward
incrementalism in policy-making.34
Stripping Swiss consensual politics of
the sacred aura that surrounds it, Franz Lehner and Benno Homann reduce
it to attempts at securing parliamentary support and popular majorities. Yet
this is not an easy task either. Even when a bill is widely supported in
parliament, many actors - and sometimes governing parties too - may shift
their position in the referendum campaign.35
As a result, the level of
parliamentary consensus has no impact at all on a bill's chances of success
when challenged by the optional referendum.36
In sum, before addressing the systemic impacts of the initiative, we can
conclude that the threat of the referendum was quickly perceived as a source
of stress and uncertainty. The referendum is more frequently used for
HOW DOES DIRECT DEMOCRACY MATTER? 43
'deterrence' rather than 'incapacitation" (Jon Elster): preventing a bill from
being voted on by parliament instead of vetoing a law already passed by
parliament. Integrative devices are crafted to preclude recourse to the
referendum by incorporating opponents into the political system in a
number of ways. In order to limit the referendum risk, policy designers
became very prudent about bold innovations and ambitious reforms.
Preventive strategies, such as the 'Vernehmlassung', which reduce the risk
of a challenge by optional referendum, strengthen the influence of
distributive coalitions that fear losing their rents. Once again, because the
referendum is seldom used, it achieves its impact indirectly rather than
directly. This weapon protects entrenched interests, inhibits the
achievement of redistributive goals and engenders a pro-status quo bias.
This includes not only social measures, but also any policies that entail
some form of resource reallocation. Scholars frequently observe a link
between consensual polities and conservatism, which they explain by the de
facto contractual, and not vertical, nature of policy-making. The most
commonly cited example is federal regimes that require the additional
consent of territorial sub-units in policy-making.37
However, the referendum
imposes similar constraints that make it hard to bring about far-reaching
innovations.
The pursuit of compromise was rewarding, however, as only seven per
cent of parliamentary bills were challenged by referendum. The increased
transaction costs caused by bargaining are thus the price to be paid for
anticipating and neutralising the referendum risk. It must be added,
however, that this form of consociational decision-making was adopted by
the political elites as a reaction to the 'direct-majoritarian'38
institutions that
their predecessors had been forced to establish. Besides, this strategy of
referendum avoidance is not always successful. When referendums cannot
be avoided, the prospects for bills can be grim. It is also a sign that
preventive remedies were not of much use. Alain Valéry Poitry came to the
surprising conclusion that bills were more likely to achieve a popular
majority when they had not been submitted to debate in the preparliamentary
phase!39
This contention underlines the strong element of
unpredictability inherent in the referendum.
EX POST NEGOTIATION AS MANAGEMENT OF DIRECT DEMOCRACY
Avoiding the referendum is clearly a preventative strategy vis-à-vis direct
democracy. It is badly needed because when a referendum on a bill is
requested, the vote takes place with no further possibility for bargaining. Yet
44 THE SWISS LABYRINTH
the political system has some manoeuvring room with respect to popular
initiatives as well. When popular initiatives appear on the agenda, it is proof
positive that the government and parliament were unable to anticipate some
social demands and thus prevent this occurrence. Initiative pressure is,
however, less than is the case with the optional referendum. Experience has
shown that initiatives rarely achieve the required double majority of citizens
and cantons (only 12 of 127 votes since the end of the nineteenth century).
Thus, initiative promoters have less blackmail potential than referendum
promoters do. They are typically outsiders, such as environmental
movements, trade unions or left-/right-wing militant groups.40
Nevertheless,
initiatives also cause indirect effects. They also trigger negotiations. These,
however, occur ex post, after the initiative has been deposited with the
required 100,000 signatures, not ex ante, as is true for attempts to pre-empt
a referendum.
A central institutional technology gives rise to the indirect effects of
initiatives. The parliamentary majority may to some extent endorse an
initiative request,4
' and, in response, the initiative promoters may withdraw
their text. Almost one-third of initiatives have been withdrawn by their
promoters. It is estimated that the parliament responded in some way to
more than half, at least until the end of the 1970s.42
It is also estimated that
one-third of initiatives led to an informal response: a law repealed or
amended, or a vote for new regulations. In these cases, half of the initiatives
were withdrawn.43
The same applies to initiatives that resulted in formal
counter-projects (constitutional amendments). In other words, most
initiative withdrawals can be attributed to a direct or indirect response from
parliament, which was acceptable to initiative promoters.44
Besides, the
voters have approved almost all formal counter-projects once the initiative
was withdrawn.45
The systemic effects of initiatives are not restricted to the
very few that passed the referendum test.
Overall, we note substantial openness and receptivity to popular
initiatives in the political system. There is room for bargaining and giving
consideration to the claims formulated by the proponents. Thus, even when
the use of direct democracy is not anticipated, the federal elites are not
wholly inactive. They have often responded to the initiative pressure, and
have reached agreement with initiative promoters. The consequences of
managing direct democracy are not the same for initiatives and optional
referendums, however. Through their indirect effects, initiatives strengthen
the pluralist dimension of the political system. As a rule, they stimulate
novel options that would otherwise be neglected or, worse yet, be
deliberately ignored.46
Unlike the referendum, the initiative is an
HOW DOES DIRECT DEMOCRACY MATTER? 45
accelerating mechanism. Although we have just noted that this can be the
basis for compromise, the initiative originates with a proposal that has
circumvented the filters (expert committees, 'Vernehmlassung',
bicameralism and the like) of the standard decisional process, which usually
water down innovations. Thus, initiatives may upset the priorities set by
established elites and may create an arena where the power balance is more
favourable to preferences marginally represented elsewhere. This is the
mainstream view on the impact of the initiative, a view which has not gone
unchallenged in the literature.
Indeed, there is a gloomier side to the integrative capacity of the political
system: the initiative deradicalises social movements. Instead of focusing
on the systemic effects of the initiative, Rudolf Epple-Gass47
has adopted the
view from the bottom. He holds that initiatives are a double-edged sword
that should be considered a constraint upon rather than a resource for social
movements. Initiatives fragment claims into single issues, at the expense of
global alternatives. What is more, according to this author, it is not true that
the content of initiatives is unnegotiated. Although this is formally true, in
practice initiative promoters moderate their claims in advance, anticipating
that only modest innovations will be sufficiently acceptable to trigger the
positive responses referred to above.
Besides, direct democracy restricts the action repertoire of social
movements. As a result of organisational bounded rationality and routine,
movements find it harder to opt for less familiar, less institutional and less
conventional strategies. They rule them out even when these are more
profitable to the achievement of their targets.48
To these negative effects we
may add the high costs of signature collection and campaigning, which meet
with varying degrees of success. Between 1979 and 1992, no more than 60
of 98 initiatives achieved the threshold of 100,000 signatures.49
Furthermore, integrative behaviour is also selective in its impact. Rightwing
established parties seldom have to resort to initiatives, but when they
do they are more likely to withdraw them subsequently. This clearly
indicates that insiders have other means of gaining influence and that their
blackmail potential through the initiative is higher too.
Finally, not only do the high costs of initiative campaigns deter minor
groups from using them, but they also seem to strengthen the oligarchic
component of larger organisations. Efficiency requires professionalisation,
resulting in the bureaucratisation and centralisation of organisations.50
Briefly put, the integrative effect of the initiative has its own price in
democratic losses: organisational selectivity and some degree of elitism. If
we add these considerations to the blackmail potential of established groups
46 THE SWISS LABYRINTH
capable of threatening an optional referendum, then we come to a view of
direct democracy that seriously challenges its emancipatory potential.51
But
this is yet another story ...
THE LIMITATIONS OF THE INSTITUTIONALIST THESIS
In previous sections, it appeared that direct democracy - in the form of
referendum or initiative - not only matters directly, but also matters more
indirectly. This form of pressure from below led to considerable
modifications in elites' behaviour, and, in that sense, Leonhard Neidhart's
thesis remains relevant. Nevertheless, the impact of direct democracy on
compromise politics ought to be qualified. There is a degree of
'helvetocentrism' in Neidhart's argument, which does not fully capture all
the dynamics of compromise politics. As noted earlier, similar mechanisms
of direct democracy from below did not result in similar integrative
responses in other systems. In addition, a number of countries, including
Switzerland, are subject to pressures for concertation that have nothing to
do with direct democracy. For example, consociational democracies were
characterised by acute cleavages, mainly religious or linguistic. These
impose a cooperative ethos on leaders of the various subcultures, necessary
to work out agreements to counteract the centrifugal effects of social
heterogeneity in these countries.52
To be sure, several points still remain to be elucidated in the study of
consociational decision-making. Some points require a return to the neoinstitutional
paradigm, such as the still unexplored role of institutional
design in appropriately addressing crucial sources of conflict. Others
require familiarity with the dimensions of deliberative democracy.53
For
example, which arguments had more resonance in debates among political
leaders, and how did the leaders learn to coexist peacefully? The conceptual
lens of consociationalism, however, remains relevant for understanding the
origins of compromise-seeking.
In Switzerland, the first major concession by rulers was a constitutional
compromise on centralisation in 1848. This compromise took place in the
absence of any referendum pressure, because all mechanisms of direct
democracy by petition were introduced into the constitution much later.
After winning a short civil war against the Catholic-Conservative cantons,
the Swiss Radicals (a liberal secular party mainly based in Protestant urban
areas that recruited then from the rising bourgeois class) left much power to
the cantons. This enabled the defeated Catholic Party to remain a dominant
force locally. Although excluded from the federal executive until 1891, the
HOW DOES DIRECT DEMOCRACY MATTER? 47
Catholic Party kept the upper hand in its strongholds: consociationalism
appears then much as a strategy for preventing the repeat of past traumas.
It can be argued that the acceptance by radical elites a few decades later
of reforms that introduced into the federal constitution provisions for direct
democracy 'from below', or the subsequent co-optation of those who
successfully resorted to them would have both been impossible without a
prior experiential learning of consensual politics.54
Hanspeter Kriesi and
Dominique Wisler55
attribute the willingness of the Radicals in power to
make concessions with respect to direct democracy to yet another condition
specific to the Swiss setting: the weakness of coercive resources under
control of the central power, that made harder to defeat opponents.
Switzerland also belongs to the grouping (which partially overlaps with
consociational democracies) of small states with very open economies that
make them vulnerable to the fluctuations of the international environment.56
Here, the aetiology of compromise-seeking differs considerably from the
consociational model. Internal cultural factors no longer matter, but
international market constraints do. Being vulnerable to the international
economy is thought to force politicians, bureaucrats and representatives of
major associations to set up co-operative mechanisms in social-economic
policy. Domestic concertation is thought to result in a decisive comparative
advantage against competitors.
Consociationalism and this version of corporatism are not mutually
exclusive, however, although they are responses to different problems, in
different areas of policy-making. Domestic concertation might be easier in
countries where elites have learned from a consociational tradition. Arend
Lijphart and Michael Crepaz pointed out the parallels between
consociationalism and neo-corporatism. Frans Van Waarden went so far as
to argue that, for the Netherlands (and this also holds true for Switzerland),
there is a causal relation between these two modes of conflict-resolution,
which is more than an isomorphism.57
Thus, a Swiss exceptionalism that is
attributed to the peculiarities of domestic institutions like direct democracy
must then be qualified.
Moreover, a fair number of institutional arrangements are functionally
equivalent to direct democracy by similarly encouraging self-restraint by
decision-makers to avoid vetoes. To give one example, consider systems of
checks and balances, where parliamentary majorities live under the threat
from institutions with a dissuasive blackmail potential, like the presidency
or a constitutional court. Ultimately, we do observe social-economic
limitations on policy-making, with private business interests exerting their
blackmailing power by threatening 'exit', in favour of more propitious
48 THE SWISS LABYRINTH
national settings ('Standortkonkurrenz'). No government will readily ignore
that risk.58
If we take into account these constraints that act as incentives to
pragmatism, moderation and compromise-building, then direct democracy
is just one more constraint. It increases the deterrent power of actors outside
the political system - associations, firms, social movements, and so on actors
who, strictly speaking, would have counted anyway. Switzerland can
thus be depicted as a polity where constraints act cumulatively on decisionmaking,
which enhances the pressure for co-operative behaviour.
To be sure, comparative studies also tend to confirm the validity of
institutional theses à la Neidhart, other things being equal (for example, the
standard explanatory variables used in the 'does politics matter?' debate on
social-economic policies). According to Ellen Immergut's rigorous survey
of health reforms in Switzerland, compared with those in France (during
both 4th and 5th Republic) and Sweden,59
the very liberal and pro-status quo
orientation of Swiss health policy results less from an unfavourable balance
of power between left and right, than from the cumulative effect of veto
points. In this case, the optional referendum was the rescue for opponents
defeated in both Chambers of Parliament. Yet this confirmation highlights
anew the limits of the virtuous role of direct democracy. This proliferation
of veto points can be conducive to policy stalemates, as opponents not
incorporated in the decisional procedures are awarded additional chances to
block reforms and to impair innovation.60
It seems that these limits of the
integrative potential of direct democracy are increasingly prevalent.
A CHANGING ROLE FOR DIRECT DEMOCRACY?
Unlike Italy or some American states, overall the structural coupling between
representative and direct democracy has been harmonious in Switzerland. It
is also thanks to the systemic responses to referendum and, to a lesser degree,
to initiative pressure that this country can be portrayed as 'a paradigmatic case
of political integration' (Karl Deutsch). However, almost all observers
criticise the high price that has been paid: a lack of innovative capacity in the
political system, slow decision-making, adjustment difficulties - European
integration is a case in point61
- or predominantly piecemeal policies, as
demonstrated by health politics studied by Ellen Immergut. The trade-off for
these costs has been the system's superior integrative capacities, which, in
comparative terms, enjoyed very high levels of mass support until very
recently.62
But are things changing now?
Considering first the fate of direct democracy, we can expect a decline
in integrative capacities to result in more bills being challenged by
HOW DOES DIRECT DEMOCRACY MATTER? 49
referendum. Yet the optional referendum remains a seldom-used weapon.
Notwithstanding a slight increase since 1970 compared to the decade
immediately following the advent of the 'magic formula', the proportion of
bills challenged holds steady below a threshold of ten per cent. As shown in
previous sections, it is true that the most efficient referendums are those that
did not take place because government or parliament acceded to the
demands of powerful veto groups. Opponents then achieve their aims by
other means, as they can merely make threats to use the referendum, without
having to carry them out. Although it is hard to say if important legislation
was obstructed by such threats, quantitative data do not support the
argument of non-decision power. The number of bills voted on in
parliament regularly increased, nearly doubling between the 1950s and the
1990s. Stability in the use of the referendum weapon is all the more
puzzling as survey data clearly demonstrate that the Swiss federal system
underwent a decline of confidence in government and parties, and the
emergence of widespread feelings of political alienation.63
This loss of
legitimacy is more severe in Switzerland than in other systems because
popular support had dropped from very high levels.
By contrast, the system's integrative capacities vis-à-vis popular
initiatives seem to have weakened considerably. Since the 1970s, the
number of initiatives has doubled. This is due primarily to the increasing
complexity and differentiation of Swiss society, which leads to an increase
in particularistic demands. Nevertheless, it also illustrates the limits of
systemic adaptiveness: fewer issues are being adequately addressed by
public authorities. What is more, the proportion of initiatives withdrawn
after successful bargaining also decreased: no less than half in the post-war
period, but only slightly more than one-quarter since the 1970s (32 of 122).
Thus we simultaneously observe more initiatives and fewer satisfactory
results (from the promoters' standpoint). A similar polarisation can be
observed in the debates on initiatives in parliament. While in the past
initiatives were usually rejected by an overwhelming majority of MPs
across the political spectrum, they are now accepted by left-wing MPs and
rejected by right-wing MPs.64
The question remains: why does the left mostly support popular
initiatives to promote demands ignored by the right-wing parliamentary
majority, without also making use of the optional referendum to challenge
decisions made by the same majority? The Left-Right conflict in direct
democracy is confined to popular initiatives. Between 1987 and 1999,
nearly two-thirds of referendum votes with a Left-Right divide in party
recommendations were initiatives (23 of a total of 37), which represent less
50 THE SWISS LABYRINTH
that one-third of the total number of referendums (37 of 111). Almost none
of the mandatory referendums (two of a total of 34) were marked by
Left-Right conflict. This is hardly surprising given their nature, for a
popular vote must take place even in the absence of any opposition. What is
more interesting is that less than one-third of optional referendums (12 out
of 39) caused a Left-Right conflict.65
It is hard to explain the Left restricting
its action repertoire in this way. Nevertheless, we know that the optional
referendum is typically a pro-status quo device. So, when the Left parties
disagree with the content of some reforms - finding them too timid, for
example - and decide to challenge them by referendum, their only hope is
a return to the status quo ante, which may not be desirable.
According to Kriesi,66
however, the Left has recently started to make
more use of the optional referendum as well. The most plausible
explanation could be that the late advent of neo-liberal beliefs in
Switzerland, combined with influence from outside,67
unleashed an
unprecedented wave of criticism about Swiss inflexibility on the part of
mainstream economists as well as strong pressures from business interests
and right-wing politicians, in favour of deregulation and liberalisation. As a
result, Switzerland has in the last decade embarked upon large-scale
changes in federal social-economic policies.68
The demand for reform
comes then predominantly from bourgeois forces, with the Left seeking to
defend the former level of welfare measures.69
It is therefore reasonable to
expect that the Left will tend to use the referendum weapon too.
Yet there does not appear to be any confirmation of stronger backing of
optional referendums by the Socialist Party. Instead, there are cyclical
tendencies. Only during the 1975-79 and 1991-95 legislatures were a
significant portion of optional referendums structured around Left-Right
conflict. But this portion is not very important (27.3 per cent in both
periods). During all other legislatures since 1971, including the last one, the
fraction did not exceed 13.6 per cent. What happens is that votes on some
recent reforms have split the Left. Soft-liners - among them the Socialist
Party - seek compromises. Hard-liners - usually minor parties and
associations — are eager to fight reforms through the referendum.
Interestingly, reforms that led to a Left opposition in the referendum (like
the first attempt to amend labour legislation) failed. In contrast, reforms
where compromises had been reached, by combining retrenchment with
improvement measures, avoided or survived the referendum obstacle (for
example, unemployment insurance reform).70
Even though the direction of
policy change has shifted recently, one thing persists: the optional
referendum is likely to split ideological camps into those who are satisfied
HOW DOES DIRECT DEMOCRACY MATTER? 51
with the compromises they negotiated previously, and those that find the
reforms too ambitious or want to stick with the status quo.71
What then is the overall assessment of the systemic role of direct
democracy in Switzerland? As noted earlier, systemic responses to the
optional referendum have been better able to maintain their integrative role
than the responses to popular initiatives. On the whole, however, direct
democracy did not prevent the advent of legitimacy problems. Other
institutional technologies, like the grand coalition formula, did not perform
better in preventing the impressive decline of diffuse support for political
elites and parties among the mass public. In addition, the Socialists'
participation in government is challenged by part of the Right, and the
participation of the increasingly national-populist Schweizerische
Volkspartei is disputed by supporters of European integration. Hence,
Switzerland is facing a wider crisis in the integrative mechanisms at the
centre of its institutional system.
Referendum votes on external relations72
serve to highlight this crisis. To
be sure, they are, in some ways, atypical. Switzerland is increasingly
confronted by decisions negotiated abroad on a multilateral basis, and this
leaves no further manoeuvring room at the domestic level. The only option
is to say 'yes' or 'no' to their ratification. It is no longer possible to search
for compromises that would reflect the domestic power balance. In these
cases, people are more likely to reject parliamentary decisions and the gap
between elite and mass preferences is wider. Opposition to
internationalisation comes predominantly from the underdog strata of
people who perceive themselves to be losers in the modernisation
processes.73
These referendums are, nevertheless, associated with lines of
conflict that seriously undermine internal cohesion, either by weakening
links between linguistic communities (the French-speaking Swiss are much
less sceptical of European integration than the Swiss Germans) or by
undermining prior compromises between economic sectors (namely
domestic and export oriented).74
These votes therefore mirror the decline in
the integrative capacity of conflict resolution strategies which had been
identified as the necessary cement of social integration in small, culturally
divided countries that are open to international economic competition, by
both consociational and neo-corporatist theory.
Hence it comes as no surprise that direct democracy is blamed today for
isolation and lack of adjustment, not only by economists, but also by some
media and politicians too. At the same time, the mass public widely
supports direct democracy. This is also quite logical, for it enhances the
public's opportunity to control political elites that are no longer trusted.
52 THE SWISS LABYRINTH
Direct democracy itself is becoming the object of controversy. People only
support reforms of the referendum system that they expect will grant them
more power. Interestingly, they believe that the increasing importance of
foreign policy issues requires expanding citizens' participation.75
Yet elites increasingly believe that direct democracy threatens
Switzerland's governability and adaptability to international factors. The
government recently tried to increase the number of signatures required for
petitioning initiatives and referendums. This measure was buried by a
majority of parliamentarians, who knew it was doomed from the start.76
To be
sure, the plethora of veto points makes the system unreformable in practice,77
but inflexibility has thus far been accepted as the price to be paid for enhanced
legitimacy. This consensus is being eroded today. Switzerland is confronted
not only by an institutional crisis, but also a 'crisis of crisis management'.78
When adjustment problems are addressed, political elites find it difficult to
overcome a nostalgic popular opposition, and legitimacy deficits thus come to
the fore. Conversely, should the political establishment address the legitimacy
deficit by, for example, giving a stronger voice to traditionalist opponents,
then adjustment problems would no doubt become more acute.
This is indeed a gloomy picture. We note a weakening of the harmonious
coupling between direct and representative democracy in Switzerland, and
the decline of trade-offs between the outcomes of input- and output-oriented
choices.79
Yet this picture is based on a rather short-term view. A major
lesson of Swiss political history is that the long-term virtues of direct
democracy in terms of political integration should be kept distinct from its
short-term vices. What is more, the former appears as an indirect
consequence - a by-product of the learning processes - of the latter. It may
be too soon to assess the present capacities of direct democracy and of the
Swiss political system as a whole. However, there is also no guarantee that
history will repeat itself.
NOTES
The author would like to thank Jan-Erik Lane and Margaret Canovan for their helpful comments.
1. Between 1848, when provisions for referendums were introduced into the federal
constitution, and 1993,of the 799 national referendums held in the whole world, 414 national
referendums had been held in Switzerland. Since the end of World War II, the Swiss share
has increased even more to over two-thirds of the referendums held in the democratic polities
studied in A. Lijphart's Democracies (New Haven: Yale University Press 1984). Data
provided by Trechsel confirm this accelerating trend: approximately half of the votes held in
Switzerland between 1848 and 1997 have taken place since the 1970s (see A. Trechsel,
'Volksabstimmungen', in U. Klöti et al. (eds.) Handbuch der schweizerischen Politik
(Zurich: NZZ Verlag 1999), pp.557-88.
HOW DOES DIRECT DEMOCRACY MATTER? 53
2. Provisions for referendums by petition can be found in the constitutions of some other
countries, but this weapon is much less used elsewhere. See Y. Papadopoulos, Démocratie
directe (Paris: Economica 1998), part I, chap. 2: 'Démocratie directe et systèmes politiques
contemporains: une analyse comparée (Suisse, Italie, Californie)'.
3. This includes 'counter-projects' formulated by the Federal Assembly as a response to
popular initiatives (see below).
4. Unless otherwise noted, data are taken from the web page of the Centre d'études et de
documentation sur la démocratie directe (http://c2d.unige.ch/index.msql) at the University of
Geneva. All data are accurate as of 6 July 1999, including information on all referendums
held before the end of the 1995-99 legislature.
5. Control is one of the dimensions of Gordon Smith's typology of referendums: see his 'The
Functional Properties of the Referendum', European Journal of Political Research 4/1
(March 1976), pp.1-23. Referendums are strongly controlled by the system when public
authorities are able to determine the issues to be decided, the timing of the procedure and
whether the vote is to be binding or not.
6. P.V. Uleri, 'Introduction', in M. Gallagher and P.V. Uleri (eds.), The Referendum Experience
in Europe (London: Macmillan Press 1996), p.6.
7. See the very useful chapter on Switzerland by A. Trechsel and H. Kriesi, 'Switzerland: The
Referendum and Initiative as a Centrepiece of the Political System', in Gallagher and Uleri
(eds.), The Referendum Experience in Europe, pp.185-208. Provisions for direct democracy
may vary considerably across cantons. On the whole, referendum practice is much more
intensive in the German part of the country: see A. Trechsel and U. Serdült, Kaleidoskop
Volksrechte. Die Institutionen der direkten Demokratie in den schweizerischen Kantonen
1970-1996 (Basel: Helbing & Lichtenhahn 1999).
8. There is no right of initiative for ordinary laws at the federal level. This is mainly a formal
limitation, however, which is frequently circumvented by initiative committees whose
proposals for constitutional amendments, if accepted, imply to change the laws that are the
real target of the initiative proponents.
9. This is not always the case. In California, for example, laws can be abrogated by legislative
initiatives and in Italy, the 'referendum abrogativo' indirectly plays a propositional role as well.
10. A. Auer, Le référendum et l'initiative populaire aux Etats-Unis (Basel/Paris: Helbing &
Lichtenhahn - Economica 1989), pp.14 and 35.
11. Uleri, 'Introduction', pp.10-11.
12. There is a substantial body of literature, and considerable controversy, about the factors that
matter in policy-making. I adopt here a perspective that emphasises the impact of
institutions, like direct democracy, which is a centrepiece of the political opportunity
structure in Switzerland.
13. See Y. Papadopoulos, 'Analysis of Functions and Dysfunctions of Direct Democracy: TopDown
and Bottom-Up Perspectives', Politics and Society 23/4 (Dec. 1995), pp.421-48.
14. This depends, in turn, on a number of factors. In a study of all federal referendums held between
1970 and 1996, I found that people regularly follow elites' orientations in economic policy, but
less frequently in the fields of foreign policy and immigration, both of which offer fertile soil to
nationalists. See Y. Papadopoulos, 'Les mécanismes du vote référendaire en Suisse: l'impact de
l'offre politique', Revue française de sociologie 37 (1996), pp.5-35. As a rule, many party
sympathisers are not very familiar with the voting recommendations of 'their' party and tend to
ignore them, not to mention the increasing numbers without any party identification whatsoever.
See H. Kriesi, 'Le défi à la démocratie directe posé par les transformations de l'espace public',
in Y. Papadopoulos (ed.), Présent et avenir de la démocratie directe (Geneva: Georg 1994),
pp.31-72. Furthermore, for a number of reasons, ordinary citizens do not behave in the same
way as political elites, being embedded in a different communication context and facing a
different mix of incentives and threats. See the discussion in part II, chap. 2 ('Autour du débat
contemporain entre élitistes et participationnistes') in Papadopoulos, Démocratiedirecte.
15. I refer here to the sociological theory of N. Luhmann: see in English his The Differentiation
of Society (New York: Columbia University Press 1982).
16. See the chapter of my book on Switzerland, Italy, and California: Papadopoulos, Démocratie
directe.
54 THE SWISS LABYRINTH
17. See his Plebiszit undpluralitäre Demokratie (Bern: Francke 1970), and also Carl J. Friedrich's
law of anticipated reactions in his Constitutional Government and Democracy (Boston: Ginn
1950). Direct democracy also provoked the identification of wider social groups (the petty
bourgeoisie, peasants, workers, etc.) to a common set of democratic values: see A. Tanner,
'Direkte Demokratie und soziopolitische Integration des Mittelstandes, der Arbeiterschaft und
Bauern in der Schweiz 1830-1914', in Eckart Schremmer (ed.), Wirtschaftliche und soziale
Integration in historischer Sicht (Stuttgart: Franz Steiner Verlag 1996), pp.184-212.
18. According to R.E. Germann's concept: see his Staatsreform (Bern: P. Haupt 1994).
19. In his book, Lijphart only counts the number of referendum votes. This quantitative approach
neglects the study of their impact. A more recent article compensates for this shortcoming:
'Changement et continuité dans la théorie consociative', Revue internationale de politique
comparée 4/3 (Dec. 1997: issue on 'Les démocraties consociatives'), pp.679-97. See
especially pp.690-92, where he considers direct democracy in Switzerland as a consociative
device. The link between type of political system and effective use of direct democracy came
to the fore in an intercantonal comparison: Adrian Vatter concluded that referendums and
initiatives are more frequently used as counter-powers by active minorities excluded from
the representative system in the more centralised cantons (with a weak communal autonomy)
and in the cantons where governments approximate the majoritarian model. See his 'Die
Wechselbeziehungen von Konkordanz- und Direkt-demokratie', Politische Vierteljahresschrift
38/4 (1997), pp.743-70. These conclusions were, however, qualified in A. Trechsel's
recent study of cantonal direct democracy: see his Feuerwerk Volksrechte. Die
Volksabstimmungen in den schweizerischen Kantonen 1970-1996 (Basel/Geneva: Helbing &
Lichtenhahn 2000), pp.109-23, 160-63, 182.
20. For the same reason, they were particularly strong in the second Chamber of Parliament, the
Ständerat (Council of States) where all cantons have equal representation - as in the
American Senate - whatever their size.
21. 'Blackmailing' and 'coalition' power are contrasted in G. Sartori's analysis of the role of
parties, theoretically inspired by A. Downs' An Economic Theory of Democracy (New York:
Harper and Row 1957): see his Parties and Party Systems. A Framework for Analysis
(Cambridge: Cambridge University Press 1976). The Swiss case, however, shows that
blackmailing power can be a fungible resource convertible into coalition power. In the
sociological literature, blackmailing power is the property of actors who - by virtue of their
resources in the form of authority, finance, organisation, information, or whatever - control
much uncertainty, and as a result cannot be circumvented. See the seminal work by M.
Crozier and E. Friedberg, L'acteur et le système (Paris: Eds. du Seuil 1981). Finally, for this
power to exist, it has to be acknowledged by decision-makers. In Switzerland, decisionmakers
were quite frequently unable to anticipate the mobilisational capacities of marginal
national-populist movements in direct democracy.
22. Smith, "The Functional Properties of the Referendum', p.16, uses the metaphor of the
'isolating cocoon'.
23. M. Dobry, Sociologie des crises politiques (Paris: Presses de la Fondation nationale des
sciences politiques 1986), p.167.
24. The recent electoral successes of the SVP, however, may mark a radically new phenomenon.
This electoral success was sustained by intensive use of the initiative and of the referendum
to pass national-populist-inspired policies, to veto more generous measures in immigration
policy, and to perpetuate Swiss isolationism.
25. Hanspeter Kriesi, for example, contrasts the formal co-optation of the Left - which lacks
influence on decisions - with the dominance of an informal 'core': a network of right-wing
party members and leaders of economic associations. See his Le système politique suisse
(Paris: Economica 1998, 2nd edn), chap. 9.
26. The most recent accounts of this process are Y. Papadopoulos, Les processus de décision
fédéraux en Suisse (Paris: L'Harmattan 1997), and P. Sciarini's chapter 'La formulation de la
décision', in Klöti et al. (eds.), Handbuch der schweizerischen Politik. See also two major
works on the overall political system, Kriesi, Le système politique suisse, and W. Linder,
Schweizerische Demokratie (Bern: Haupt 1999), which is a revised and extended version of
his Swiss Democracy (London: Macmillan Press 1994).
HOW DOES DIRECT DEMOCRACY MATTER? 55
27. See Kriesi, Le système politique suisse p. 196, who reinterprets data from the 1970s — the only
available so far - collected by A.V. Poitry and presented in his Lafonction d'ordre de l'Etat.
Analyse des mécanismes et des déterminants sélectifs dans leprocessus législatifsuisse (Bern:
Peter Lang 1989). Assessment of decisional importance in advance is of course problematic,
especially in the Swiss system where ordinary citizens have the final say. According to an index
elaborated by E. Gruner and H.P. Hertig in their Der Stimmbürger und die 'neue' Politik (Bern:
Paul Haupt 1983), pp.409-10, people seem to place more value on a referendum making safety
belts compulsory than on one extending the referendum right to international treaties!
28. I refer here to the well-known distinction made by A. Wildavsky in Speaking Truth toPower.
The Art and Craft of Policy Analysis (Boston: Little Brown and Co. 1979).
29. See R.E. Germann et al., Experts et commissions de la Confédération (Lausanne: Presses
polytechniques Romandes 1985), again the only study available, with the same problem of
data obsolescence.
30. It should be mentioned however that a big business union recently complained that in the
'Vernehmlassung', the bureaucracy would not weigh positions in proportion to the
importance of their advocates. I do not share this criticism. Waelti argues that a similar
blackmailing power is awarded to cantons that, in exchange for their cooperation in
implementing federal policies, manifest their claims in the consultation phase too. See S.
Waelti, 'Institutional Reform of Federalism: Changing the Players rather than the Rules of
the Game', Swiss Political Science Review 22/2 (Summer 1996), pp.113-41.
31. For a synthesis of recent — albeit fragmentary - findings, see Sciarini, 'La formulation de la
décision'. For a discussion of Swiss parliamentary power considering institutional rules and
parliamentary resources (for example the so-called 'Milizsystem' of — at least on paper—nonprofessional
parliamentarians), see Papadopoulos, Les processus de décision fédéraux en
Suisse, chap. 6.1 ('Dans quelle mesure le Parlement s'acquitte-t-il de ses fonctions?'). R.
Liithi maintains that the Swiss Federal Assembly is comparatively strong in her chapter on
'Parlament', in Klöti et al. (eds.), Handbuch der schweizerischen Politik, pp.131-57. Annina
Jegher reaches similar conclusions regarding recent parliamentary influence on decisionmaking:
see her Bundesversammlung und Gesetzgebung (Bern: Haupt 1999).
32. See Sciarini, 'La formulation de la décision'.
33. Even though implementation is rather unpredictable too, especially as federal legislation is
usually enforced by the cantons or by non-public bodies, the so-called parastaatliche
Verwaltung: see the research results presented by W. Linder in La décision politique en
Suisse. Genèse et mise en œuvre de la législation (Lausanne: Réalités sociales 1987). It is
widely believed, however, that the referendum confers additional legitimacy to public
policies, reducing thus the risk of implementation conflicts, or that even a negative outcome
of a vote is better than taking the risk to trigger mass mobilisation against a law. On the other
hand, federalism allows opponents who were defeated in a nation-wide referendum to seek
to exploit a more favourable power balance at the cantonal level in the implementation phase.
34. This argument is also used by P. Pierson regarding welfare retrenchment: see his 'Irresistible
Forces, Immovable Objects: Post-Industrial Welfare States Confront Permanent Austerity',
Journal of European Public Policy 5/4 (Dec. 1998), pp.539-60.
35. F. Lehner and B. Homann, 'Consociational Decision-Making and Party Government in
Switzerland', in R.S. Katz (ed.), Party Governments: European and American Experiences
(Berlin: de Gruyter 1987), pp.243-69.
36. P. Sciarini and A. Trechsel, 'Démocratie directe en Suisse: l'élite politique victime des droits
populaires?', Swiss Political Science Review 2/2 (Summer 1996), pp.201-32.
37. See F.W. Scharpf, 'Political Institutions, Decision Styles, and Policy Choices', in R. Czada
and A. Windhoff-Héritier (eds.), Political Choice (Frankfurt/Boulder: Campus/Westview
1991), pp.28-53.
38. J.S. Fishkin, Democracy and Deliberation (New Haven: Yale University Press 1991).
39. See Poitry, Lafonction d'ordre de l'Etat, pp.328-30.
40. It has been calculated that for 87 initiatives registered between 1974 and 1992, 27 were
backed by Green associations, 11 by the Socialist Party and by trade-unions, and eight by
national-populist organisations: see K.W. Kobach, The Referendum: Direct Democracy in
Switzerland (Aldershot: Dartmouth 1993), p.101.
56 THE SWISS LABYRINTH
41. See above for the modalities of this reaction.
42. H. Werder, 'Das Politische System der Schweiz - Eine Skizze seiner Funktionsweise', in W.
Linder et al. (eds.), Planung in der Schweizerischen Demokratie (Bern: P. Haupt 1978),
pp.31-51.
43. Kobach, The Referendum, p.94. Clearly initiative promoters prefer informal reactions that do
not entail a mandatory vote with its inherent uncertainty and the campaigning that precedes
it.
44. B. Hofer, 'Die Volksinitiative als Verhandlungspfand', Schweizerisches Jahrbuch für
politische Wissenschaft 27 (1987), pp.207-35.
45. Kobach, The Referendum, p.107. In contrast to this success, initiative and counter-project
most often failed when simultaneously submitted to the vote. As late as 1987, voters hostile
to the status quo had to choose between the two options, dramatically reducing the chances
of either winning a majority. Thus, formal counter-projects were not only compromiseoriented
but also strategically designed to split the reformist camp. It is interesting that the
propensity to withdraw initiatives has recently declined. Although this is probably due to a
change in the overall political climate (see below), it can be argued that initiative promoters
can now better anticipate their chances and thus have fewer incentives to withdraw their
proposal.
46. See J.-D. Delley, L'initiative populaire en Suisse. Mythes et réalités de la démocratie directe
(Lausanne: L'Age d'Homme 1978), and L. Neidhart, 'Regierbarkeitsfragen in der direkten
Demokratie', Schweizerisches Jahrbuch für politische Wissenschaß 23 (1983), pp.13-43.
47. R. Epple-Gass, Friedensbewegung und direkte Demokratie in der Schweiz (Frankfurt:
Haag+Herchen 1991). It ought to be noted however that the author's conclusions mostly rely
on studies of the pacifist movement.
48. In a comparative study, Hanspeter Kriesi and Dominique Wisler also come to the conclusion
that the repertoire of action of new social movements in Switzerland is more moderate than
in other countries without direct democracy. See their 'Social Movements and Direct
Democracy in Switzerland', European Journal of Political Research 30/1 (July 1996),
pp.19-40.
49. Kobach, The Referendum, p.95.
50. Delley, L'initiative populaire en Suisse, maintains however that direct democracy confers
more weight to the rank-and-file, whose co-operation is necessary for signature collection
and voting campaigns.
51. For a more in-depth discussion of this argument, see Papadopoulos, Démocratie directe, part
III, chap. 2 ('Référendum et citoyenneté: les promesses non tenues de la démocratie directe').
52. See the founding work by A. Lijphart, Democracy in Plural Societies (New Haven: Yale
University Press 1974).
53. There is today a substantial literature on deliberative democracy and policy-making. For a
good discussion on the current state of debate, see J. Elster (ed.) Deliberative Democracy
(Cambridge: Cambridge University Press 1998).
54. This is the argument put forward by the path-dependency school. G. Lehmbruch, for
example, goes back as far as the eighteenth century to find the roots of peaceful resolution
in religious conflicts in Central Europe. See his 'Die korporative Verhandlungsdemokratie in
Westmitteleuropa', Swiss Political Science Review 2/4 (Winter 1996), pp.19-41.
55. H. Kriesi and D. Wisler, 'The Impact of Social Movements on Political Institutions: A
Comparison of the Introduction of Direct Legislation in Switzerland and the U.S.', Cornell
University, Institute for European Studies Working Paper 96/6 (1996).
56. A whole school of thought has now emerged around the seminal book by P.J. Katzenstein,
Small States in WorldMarkets (Ithaca: Cornell University Press 1985).
57. A. Lijphart and M. Crepaz, 'Corporatism and Consensus Democracy in Eighteen Countries:
Conceptual and Empirical Linkages', British Journal of Political Science 21 (1991),
pp.235-56; F. Van Waarden, 'Consociationalism and Economic Performance in the
Netherlands', paper presented to the Conference on 'The Fate of Consociationalism in
Western Europe', 29-31 May 1998, Center for European Studies, Harvard University. The
overlap between the two models is not perfect, however, as a multi-nation comparison
shows: see J.-E. Lane and S. Ersson, 'The Institutions of Konkordanz and Corporatism: How
HOW DOES DIRECT DEMOCRACY MATTER? 57
Closely Are They Connected?' Swiss Political Science Review 3/1 (Spring 1997), pp.5-29.
58. See the thesis on 'the structural dependency' of societies with respect to capitalism: C. Offe,
Contradictionsof the Welfare State (London: Hutchinson 1984).
59. E. Immergut, Health Politics. Interests and Institutions in Western Europe (Cambridge:
Cambridge University Press 1992).
60. To veto points should be added the 'clearance points' previously identified in Pressman's and
Wildavsky's seminal study of policy implementation. Swiss 'co-operative' federalism
frequently leaves the implementation of federal decisions to cantonal administrations, where
either political will or capacity may be missing. On the similarity between 'veto' and
'clearance' points, see B.G. Peters, Institutional Theory in Political Science. The 'New
Institutionalism' (London/New York: Pinter 1999), pp.74-5.
61. The requirement for a double majority (people and cantons) would make EU membership
very hard to attain.
62. High quality of democracy combined with good social-economic performance seem to be
common features of the consensus type of democracies. See A. Lijphart, Patterns of
Democracy. Government Forms and Performance in Thirty-Six Countries (New
Haven/London: Yale University Press 1999).
63. See, inter alia, C. Longchamp et al., Unterstützungvon Bundesrat und Verwaltung (Bern:
GfS-Forschungsinstitut 1994), or any of the above-mentioned reference books on the Swiss
political system.
64. Sciarini and Trechsel, 'Democratie directe en Suisse: l'élite politique victime des droits
populaires?'. Other studies conclude to a more general Left-Right polarisation in parliament,
but Sciarini, 'La formulation de la décision' maintains that the votes taken into consideration
cannot be considered representative.
65. Information from our data base at the University of Lausanne. I wish to thank Jeremias
Blaser for regularly updating the database and providing analyses.
66. Kriesi, Le système politique suisse, pp.103-4.
67. For this aspect of internalisation of globalisation, see the editor's introductory and concluding
chapters in A. Mach (ed.), Globalisation, néo-libéralisme etpolitiquespubliquesdansla Suisse
des années90 (Zurich: Seismo 1999). It should, however, be remembered that some demands
for policy reform in contemporary states arise from domestic structural problems, unrelated to
globalisation (slow growth, ageing populations leading to greater government commitments
that entail more fiscal pressures, etc.): see Pierson, 'Irresistible Forces, Immovable Objects'.
68. For the economists' view, see various essays by Silvio Borner, namely S. Borner et al. Die
Schweiz imAlleingang (Zurich: NZZ Verlag 1994), where Swiss isolationism is attributed to
the entrenched positions occupied by pro-status quo (welfare or protectionist) distributive
coalitions. These coalitions are able to threaten recourse to the referendum in
institutionalised concertation procedures, which are favourable to rent-seeking. For
deregulatory manifestos that served as the basis for the reorientation of some major socialeconomic
policies, see F. Leutwiler et al. Schweizerische Wirtschaftspolitik im
internationalen Wettbewerb. Ein ordnungspolitisches Programm(Zurich: Orell Füssli 1991),
and D. de Pury et al.,Ayons le couraged'un nouveaudépart. Un programmepour larelance
de la politique économiquede la Suisse (Zurich: Orell Füssli 1996).
69. There is also the protectionist-traditionalist coalition of the populist right fighting for the
continued isolationism.
70. G. Bonoli, 'La réforme de l'Etat social suisse: contraintes institutionnelles et opportunités de
changement', Swiss Political Science Review 5/3 (Autumn 1999), pp.57-77. Swiss direct
democracy typically belongs to the category of veto points listed by Pierson, 'Irresistible Forces,
Immovable Objects', which are likely to discourage attempts at dismantling welfare provisions.
71. In the 1970s, it was mostly the Right that split between its government faction willing to
compromise with the Socialists to modernise the political system, and a more conservative
and ultra-federalist 'preventive minority' also belonging to governing parties. See H. Kriesi,
Entscheidungsstrukturen und Entscheidungsprozesse in der Schweizer Politik (Frankfurt:
Campus 1980). Two decades later, the major division within the Right in government is
between those who favour international opening and those of a more protectionist or even
nationalist persuasion.
58 THE SWISS LABYRINTH
72. A generic category encompassing foreign policy issues, among them European integration,
as well as relations with foreigners (immigration and asylum policy). On referendums on
foreign policy issues see P. Sciarini and L. Marquis, 'Opinion publique et politique
extérieure: le cas des votations populaires en Suisse', International Political Science Review
21/2 (2000), pp.149-71.
73. For Switzerland, see S. Kobi, Des citoyens suisses contre l'élite politique. Le cas des
votationsfédérales, 1979-1995 (Paris: L'Harmattan 1999). On new cleavages related to the
international dimension of politics, see H. Kriesi, 'The Transformation of Cleavage Politics',
European Journal of Political Research 33/2 (1998), pp.165-185.
74. For an analysis of the cleavage between linguistic communities, see H. Kriesi et al., Le
clivage linguistique: problèmes de compréhension entre les communautés linguistiques en
Suisse (Bern: Federal Office of statistics 1996). Conflicts between economic sectors have
been emphasised by studies focusing on the reform of Swiss agricultural policy and of cartel
legislation. See P. Sciarini, Le système politique suisseface à la Communauté européenne et
au GATT:le cas-test de la politique agricole (Geneva: Georg 1994), and A. Mach, 'Quelles
réponses politiques face à la globalisation et à la construction europénne? Illustration à partir
de la révision de la loi suisse sur les cartels', Swiss Political Science Review 4/2 (Spring
1998), pp.25-19.
75. See Longchamp et al., Unterstützung von Bundesrat und Verwaltung, and more recent
Univox reports providing information on direct democracy.
76. Both those who have high expectations of reform and those who fear it probably
overestimate its impact. A binary comparison of the cantons which differ most in the
generosity of their direct democracy provisions (Geneva and Aargau) showed that
institutional dissimilarities have only a limited and uneven influence on the practice of direct
democracy and on voters' behaviour. See R. Lachat, 'Réforme des droits populaires: quel
enseignement tirer de l'expérience cantonale?', Department of Political Science of the
University of Geneva, 1998.
77. See the incisive remarks by Germann, Staatsreform.
78. A concept used by Offe, Contradictions of the WelfareState, to describe the strain caused by
conflicting requirements of capitalist prosperity and social policy goals. I use it here in a
different sense: for more details, see Y. Papadopoulos, 'De Charybde en Scylla: le système
politique suisse entre crise de représentation et crise d'adaptation', in C. Honegger et al.
(eds.), Sociétés en construction. Identités, conflits, différences (Zurich: Seismo 1996),
pp.135-48.
79. I refer here to the distinction between institutional orders by Powell and DiMaggio and to the
antinomy between policy choices as depicted by Fritz W. Scharpf. See W.W. Powell and P.W.
DiMaggio (eds.), The New Institutionalism in Organizational Analysis (Chicago: The
University of Chicago Press 1991), pp.1-38, and F.W. Scharpf, Demokratietheorie zwischen
Utopie und Anpassung (Konstanz: Universitätsverlag 1970).