Coping with a crisis: business interest group coalitions in the wake of the financial crisis

The global financial crisis resulted in a re-examination of the business sector as a whole and its lobbying activity. Nevertheless, research has not analysed the impact that the global financial crisis had on the coalitions within the business interest group community. This article describes in detail how coalitions within the business interest group community have emerged in the aftermath of the global financial crisis. I focus my attention on the Swiss case because of the importance of its financial sector as well as its neo-corporatist system. The empirical approach consists of a process tracing on three policy areas: financial regulation, tax policy and corporate governance, which is elaborated through an extensive document analysis and 28 in-depth interviews. The advocacy coalition framework is used to analyse business interest groups’ beliefs. The findings suggest that different beliefs materialise, especially in the field of financial regulation and partially in tax policy.


Introduction
In 2008, the business sector, and especially the financial industry, was affected by a major economic turmoil that caused not only a general loss of confidence in the financial sector but also a reappraisal of the business sector and its lobbying activity as a whole (Culpepper 2011). Concretely, this event may have resulted in an upsurge in negative public opinion towards business interest groups and the consequent threat to their survival, seen as a reputation threat. Yet, considering that financial interest groups have modified their dynamics and strategies to offset this challenging period (Kastner 2018;Bell and Hindmoor 2015), on a broader perspective, what has been its impact on the coalitions within the business community? This paper argues that business interest groups' preferences might also have altered after the crisis, in order to adapt to the hidden contractions, modifying the alliances within the community.
Thus, the study focuses on coalition dynamics. Specifically, following the work of Klüver (2011, p 12), I apply a broader definition of lobbying coalitions: 'a set of actors that share a policy goal'. This definition is wider since it does not take into account the level of coordination but rather the policy aim. Nonetheless, as put forward by some authors, actors' belief congruence and action coordination are typically related (Weible and Sabatier 2005;Henry 2011;Ingold 2011;Matti and Sandström 2011). Furthermore, I am interested in elucidating how business interest groups organised themselves on highly salient policies after the crisis and whether preferences within the community of business interest groups differ from each other. In order to do so, I will draw on the advocacy coalition framework (ACF) since it represents one of the most prominent frameworks when it comes to policy process and change (Henry et al. 2014). By analysing this aspect, I seek to provide a better understanding of both their structure and interest. As a result, I have formulated the following research question: How have coalitions within the business interest groups community changed in the aftermath of the global financial and economic crisis?
This study is centred on Switzerland for two main reasons. First, interest groups have been involved since the end of the nineteenth century in the Swiss decision-making process (Mach 2015). Switzerland is rather closer to the liberal version of neo-corporatism, which is defined by Katzenstein (2003) as being characterised by fragmented labour unions and highly organised business interest groups. Second, its financial sector is characterised by the highest percentage of financialisation in terms of GDP, making it one of the major financial centres worldwide (Tridico 2018). Switzerland is therefore a least-likely case, because if we notice a change in an environment that favours business and financial affairs, then this should also happen elsewhere.
To answer the research question, a qualitative analysis was conducted based on two types of data: 28 semi-structured interviews and a wide range of documents (newspapers, reports, academic research, parliamentary interventions, interest groups' position papers). MPs, directors or presidents of interest groups, specialists, researchers and public authorities were interviewed. Based on this data, the theory-testing process tracing method was applied, which served to elucidate the causal chain linking the crisis to changes in business interest group coalitions building on theories derived from the literature. Concretely, three important economic policies particularly hit by the post-crisis era-financial regulation, tax policy and corporate governance-are examined. As I will explain further on, these policies are also considered least-likely cases. The results suggest that following the global financial crisis, various conflicts emerged creating different coalitions.
In what follows, I will first examine the theory on business interest groups and their coalitions before illustrating the functioning of the ACF. The research design then explains the reasons behind the choice of the qualitative method, how it was implemented and the data collection. The findings are then analysed in the next section and, finally, the main conclusions that can be derived from the study are discussed.

To form or not to form a coalition
Several research projects have focused on the costs and benefits of lobbying together. Mahoney (2007) states that there are two major benefits. First, the unity of interest groups on a specific subject signals to the policy-maker a larger support for a policy position, enabling them to gain the endorsement of decision-makers. Second, the formation of a coalition offers the opportunity to gather more consensus in a less costly manner, thus allowing weaker interest groups to share resources (Junk 2020). Additionally, Hojnacki (1997) claims that interest groups work in broader coalitions only when it can improve their reputation. Hanegraaff and Pritoni (2019) offer another important reason that may induce interest groups to coalesce: interest group survival. In other words, interest groups that perceive themselves to be in a weaker position both financially and politically may be motivated to form coalitions.
Nonetheless, forming a coalition might also engender some costs and uncertainty. The greatest inconvenience is having to align the specific group preferences with those of the main coalition, even if they are not always entirely convergent. Seeking a compromise with the coalition partners might reduce the autonomy of the interest group, threatening the potential benefits of a collaboration (Hojnacki 1997). In the same vein, Hanegraaff and Pritoni (2019) argue that stronger organisations prefer to lobby alone in order to have more room for manoeuvring. Furthermore, groups incorporated in a broad coalition have the tendency to characterise themselves by criticising the activities of other groups participating in the same coalition in order to attract the attention of policy-makers (Browne 1990).
Also of major importance, institutional factors as well as the context might also trigger the construction of a coalition (Klüver et al. 2015). By way of example, corporatist and pluralist systems differ in their ways of incorporating interest groups in their decision-making process, affecting the way interest groups form coalitions. Namely, corporatist systems are more prone to have peak associations compared to the pluralist forms. Moreover, issue salience also seems to play a key role (Beyers and De Bruycker 2018;Mahoney 2007).
Forming a coalition might also depend on the kind of group involved. Indeed, most of the literature asserts that business interest groups are less likely to form coalitions when compared to citizens' interest groups (Klüver 2013;De Bruycker 2017;Hanegraaff and Pritoni 2019). Until now, only one study has found that business interest groups build coalitions more frequently (Hojnacki 1997). Nevertheless, business actors can also engage in a broader coalition in order to signal not only a greater support for a policy, but also their ability to work together (Berry 2015;Hojnacki 1997). It is also important to note that, as explained by Lindblom (1977) in Politics and Markets, business interest groups tend to have a privileged position when compared to other interest groups. This, as mentioned above, could encourage them to lobby alone in order to obtain policy outputs more in line with their preferences. However, their comparative advantage is limited to many context-related elements. For example, recent developments have witnessed the rise of both public and citizen interest groups (Binderkrantz 2012) that are challenging the privileged position of business interest groups. Finally, business interest groups, and interest groups more in general, are mainly looking at two main objectives when taking into account all this calculation: influencing a public policy and defending their own resources. It follows that the focus on beliefs and values attached to interest groups is central to analyse policy process and coalition formation. As a result, in this research I am particularly interested in unravelling business interest group preferences. Yet, as put forward by Baumgartner et al. (2009) and more recently by Baumgartner et al. (2013, p 54), "interest groups that are located on the same side of a policy space on a given issue form a lobbying coalition." This view on coalitions is strongly linked to the structure of the ACF, which in turn focuses on the preferences of the actors forming a coalition. Likewise, this framework asserts that being on the same side and thus sharing the same policy objectives seems to be strongly linked to a form of coordination between actors (Weible and Sabatier 2005;Henry 2011;Ingold 2011;Matti and Sandström 2011). Lastly, the ACF is often used to explain how change occurs during a period of crisis in a given policy area when competing advocacy groups form a coalition. Analysing the dynamics of business interest groups through the lenses of the ACF seems therefore a well-suited technique.

Application of the advocacy coalition framework
The ACF was first formulated by Paul Sabatier and Hank Jenkins-Smith (Sabatier 1986;Jenkins-Smith et al. 1991), who placed special emphasis on the analysis of advocacy coalitions within a policy subsystem. In particular, it posits that actors act in accordance with their beliefs on a policy issue. This framework lies at the interface between the macro and the micro level (Pr et al. 2007). It asserts that not only do the actors incorporated in a policy subsystem influence the policy-making process, but they are, in turn, largely affected by political and socio-economic elements. Sabatier and Weible (2019, p 196) state that the term advocacy coalition involves 'legislators, agency officials, interest group leaders, judges, researchers and intellectuals from multiple levels of government'. According to the ACF, individuals who are a part of a policy subsystem will seek out allies who share their core values. Hence, the ACF is particularly centred on comprehensive advocacy coalitions. Nonetheless, I am interested in disentangling these broader groups, paying particular attention to the role exerted by business interest groups within a given policy process, because they play an important role when it comes to affecting economic policies.
There are different levels of conflict between coalitions, depending on the subsystem. There can be up to three types of subsystems (Weible 2008). In the collaborative subsystem, conflicts are mitigated and cooperation is widely present, whereas the adversarial type is characterised by higher levels of conflicts. Finally, advocacy coalitions can form an unitary entity where various coalitions merge into one (Weible 2008).
It is possible for beliefs to alter in response to either internal subsystem events (policy learning and negotiated agreements) or external events (socio-economic and regime changes, outputs from other subsystems or natural disasters) (Sabatier and Weible 2007;Jenkins-Smith et al. 2014). The latter is depicted as a major explanation and it perfectly suits our case. Furthermore, most of the research has not taken into consideration economic or finance-related policy issues (Pierce et al. 2017), thus it is especially interesting to apply this framework to analyse the beliefs of business interest groups.
According to this framework, actors' beliefs are the main explanatory factors explaining the formation and dissolution of coalitions, and Pr et al. (2007) divided them into three hierarchical layers. The most intrinsic beliefs are labelled 'deep core beliefs' and encompass a range of normative beliefs that are almost impossible to alter. In other words, they are highly abstract and extremely broad in scope. The right-left political dichotomy, for example, fits entirely into this set of beliefs. Then, 'policy core beliefs' depict instead the application of the above-mentioned deep core belief in a policy subsystem. As Sabatier and Weible (2007, p 195) state opposing beliefs on a specific policy can contemplate different aspects of it, such as the role of the state in regulating the policy, the problems underlying it and the action that needs to be pursued. Along with the deep core beliefs, this set of beliefs also tends to remain very stable over time. Lastly, 'secondary beliefs' are the so-called instrumental beliefs because they are conceived as an instrument to achieve a target goal and, consequently, they are narrower in scope. Unlike the others they are more likely to change. However, according to this framework, actors that hold similar beliefsespecially policy and deep core beliefs-tend to act unanimously in order to influence public policies (Weible 2008). In other words, coordination is generally supposed to follow automatically from belief congruence (Weible and Sabatier 2005). Thus, 'the ACF predicts that policy core beliefs will structure an actor's choice of network interactions and these interactions will occur predominantly with other actors of similar policy core beliefs in an advocacy coalition' (Weible and Sabatier 2005, p 183).

Building the argument through process tracing
The main idea is to unravel all the mechanisms that were triggered by the crisis and subsequently had an effect on the coalitions within the business interest group community. To do this, I will make use of the process tracing approach which Collier (2011, p 823) defines as 'the systematic examination of diagnostic evidence selected and analysed in light of research questions and hypothesis posed by the researcher'. Put simply, process tracing traces back the causal mechanism that links two phenomena, allowing a more fine-grained analysis of the black box that connects the independent variable (in this case the financial crisis) and the main dependent variable (business interest group coalitions) through pieces of evidence. In a more systematic fashion, central to the detection of the x⟶ y relation is the presence of one or more elements m representing the causal mechanism: x→ m→ y.
Specifically, I will rely on the theory-testing process tracing, where, as Beach and Pedersen (2019,p 14) explain, 'we know both X and Y and we either have existing conjectures about a plausible mechanism or we are able to use logical reasoning to formulate a causal mechanism from existing theorisation'. This type of process tracing aims to provide evidence for every step of the causal mechanism in order to attest the causal chain. Although it does not allow for the selection of a specific causal mechanism among several rival explanations, it does attempt to render it applicable to other cases (Beach and Pedersen 2019). According to this method, there are three steps to follow. First, the causal mechanism needs to be theorised, and then it needs to be operationalised, describing the expectations that are believed to be found in the empirical part. Finally, the data is collected in order to find evidence in every step of the causal mechanism (Beach and Pedersen 2019, p 16). Thus, in this first part I will theorise the mechanisms and provide the hypothesis.

The theoretical mechanisms
The first mechanism starts from a shock in the financial and economic system, that through increasing public attention and a mobilisation of public interest groups, may led to the intensification of the issue salience related not only to economic policies but also to the actors influencing it. In other words, the appearance of an economic/ financial shock may reveal the pitfalls of the financial and economic system to a wider public. Issue salience is a key concept for business interest groups since during periods of higher issue salience it is more difficult to influence economic policies (Culpepper 2011). Specifically, the financial sector and by extension also transnational companies are held accountable for the severe situation and consequently also for their lobbying activity. It follows that representatives are more prone to listen to their constituency rather than business interest groups due to re-election concerns (Lax and Phillips 2012).
In the second mechanism, there is a window of opportunity that opens up in response to the increase in the issue salience of economic-related policies. According to the ACF, a moment of crisis can indeed lead to policy reforms (Sabatier and Jenkins-Smith 1993;Pr et al. 2007). Specifically, a more difficult environment can induce representatives to be more responsive towards voters (Hayes and Bishin 2012, p 135). Notably, when policies are particularly salient, the behaviour of representatives also becomes more visible. Consequently, citizens and public interest groups seize this moment to call for tighter regulations, while representatives tend to align their preferences with those of their constituents in order to secure re-election. This can eventually lead to a change in economic-related regulations. Figure 1 illustrates the theoretical mechanisms and operationalise the process tracing mechanism based on the table elaborated by Beach and Pedersen (2019, p 15) The financial crisis has severely challenged the world of finance to the point of threatening not only the very existence of some financial institutions, but also their legitimacy in the political arena. Consequently, these threats could have driven financial interest groups and, by extension, business interest groups to perceive a mortality anxiety. As such, adapting their position and changing the way they frame their policy preferences might represent a necessary condition to avoid being eclipsed by public interest groups. Moreover, as depicted by Culpepper (2011), higher issue salience weakens the alliances between business groups and ideologically proximate political parties. Specifically, he explains that political parties allied with managerial organisations have a tendency to 'throw their business allies under the bus' in a high-salience context (Culpepper 2010, p 191). Concretely, an increase in issue salience, due to the crisis, combined with survival fears, would encourage the emergence of opposing preferences and beliefs concerning the most appropriate policy to implement. Yet, traditionally, discrepancies can emerge between domestic and export-oriented firms (Frieden 1988), between technological 'innovators' and 'imitators' (Roemer-Mahler 2013), or between business groups with different levels of resources (Rogowski 1990). That said, the export/domestic-oriented cleavage is the source of one of the main divisions in the business interest group community (Falkner 2012). Some business interest groups place higher emphasis on the domestic market, while others focus predominantly on the international one. In other words, the latter is more in favour of trade liberalisation policies that the former, which tends instead to favour some protectionist actions (Trampusch and Mach 2011, p 75). In recent times, the unity of business was also challenged in Switzerland by Mach et al. (2021) and in the UK by Feldmann and Morgan (2021).
These divisions might have worsened in light of a higher demand for restrictive regulatory policies and as a result of a threat to their survival (understood as a reputational threat). Indeed, conflicts may arise in response to the different effects that new regulatory measures have on different sectors and companies. Therefore, Fig. 1 Operationalisation of the process tracing business interest groups that are domestic-oriented, that is inward-oriented business interest groups, are expected to join forces to reduce regulatory costs. By contrast, outward-oriented business interest groups, which tend to be larger and international are more likely to support international harmonisation and as such might be more willing to compromise when it comes to regulations. As a result, this phenomenon may have further deepened after the onset of the crisis. On top of that, according to the ACF, a change in the socio-economic conditions such as the global financial crisis might engender a transformation in policy and secondary beliefs.
Given these cleavages, it follows that this financial meltdown might have created a difficult environment within the business interest group community, leading to an exacerbation of the divergences. Moreover, peak business associations have a higher chance of encountering discrepancies within and among themselves due to the multitude of associations that they represent and the consequent pluralism of interests. Three hypotheses can thus be formulated: • Conflicts within the business interest group community are likely to form in response to the crisis. • A coalition of outward-oriented business interest groups and a coalition of inward-oriented business interest groups are likely to form. • Conflicts within peak business associations are likely to increase in response to the crisis.

Research design
In the following section, I will provide the fundamental elements that led me to select my case study and the policy issues. This is followed by an explanation of the research design and a description of the data and sample.

The case selection
This study analyses the Swiss case based on tree main reasons: its neo-corporatist regime, its political stability and the influential role of its financial sector. Switzerland is a neo-corporatist country, which means that it tends to integrate organised interest groups into the decision-making process (Christiansen et al. 2018). They benefit from the constitutional right 1 to express their expertise and opinion during the legislation process. Remarkably, they have in fact shaped several fields, such as vocational training, the standardisation of technical norms and the regulation of the banking sector, among others (Trampusch and Mach 2011). In general, in Switzerland there are four umbrella business interest groups (Kriesi and Trechsel 2008, p 101): the 'Swiss Farmer Union' (SFU), which represents the entire farm sector; the 'Swiss Union of Arts and Crafts' (USAM), which uphold the interests of small and medium-sized enterprises (SMEs); the 'Swiss Employers' Union' (SEU), which represents the industry's interest in the negotiations with the unions; and, finally, 'economiesuisse' (previously known as Vorort 2000), which defends the interests of the international business community, especially multinational companies (Trampusch and Mach 2011). The 'Swiss Banking Association' (SBA) is also of great importance, even though it is not a peak association (Kriesi and Trechsel 2008, p 101). Besides, this inclusion of interest groups in the decision-making process, together with Switzerland's consociational system, is one of the factors contributing to the stability of Swiss politics.
Turning to the financial side, Switzerland, thanks to its long banking and financial tradition, is a veritable hub for international financial flows. In 2015, 10% (400,000 full-time equivalents) of all jobs in Switzerland were linked to this sector, and the financial sector accounted for a significant percentage of GDP, although it has declined in recent years (2009: 10.7, 2014: 9.6, 2019: 9.2). Furthermore, in terms of taxes, this sector yielded a tax revenue of 12.7% in 2019 (taking into account the municipal, the cantonal and the federal level) (Peters and Grass 2020). On top of that, two systemically important financial institutions (SIFIs) are Swiss: Credit Swiss and UBS.
Consequently, I argue that Switzerland is a least-likely case, as both interest groups and the financial sector hold a crucial position in a political system characterised by a status quo bias. Indeed, if a more divided environment in the business community is to be found in Switzerland, where the financial sector contributes up to 10% of the GDP and the Swiss policy-making is particularly stable and averse to changes, then it should all the more happen in other countries.
Three policy areas have been particularly affected by the crisis: financial regulation, corporate governance and tax policy. In view of the above-mentioned particularities of the Swiss political system, coupled with the prominent position not only of the financial sector but also of many transnational corporations, I also consider these three policies least-likely cases. First, it is crucial to consider the new set of financial regulations. The bailout of UBS was a watershed moment for the Swiss economic and political landscape, and several financial regulations were discussed in the parliamentary arena in its wake. Second, criticisms of corporate governance also erupted in response to the financial meltdown. A focal moment was marked by the initiative against abusive remuneration, which was voted by Swiss citizens on 3 March 2013. The goal was to keep overly high CEO salaries under control and to give shareholders more decision-making power. The initiative 1:12 is also taken into account, which aimed to limit the difference between the lowest and the highest salary. These initiatives are of utmost importance since the remuneration of senior executives of large institutions is one of the most widely discussed issues in civil society (Guo et al. 2015). Lastly, the post-crisis atmosphere had a significant impact on tax policy. Notably, in an effort to restore the public debt aggravated by the crisis, the international community increased its pressure on tax havens and financially opaque jurisdictions (Longchamp 2010). As a result, Switzerland has been under increasing international pressure to bring banking secrecy and corporate taxation in line with OECD standards.

Data and Sample
The data collection was carried out via a document analysis and semi-structured interviews with open-ended questions. The document analysis included scientific papers, interest groups' position papers, parliamentary interventions, newspaper articles and public records. This type of data made it possible to gain an all-round appreciation of the situation and to detect the official positions of business interest groups. Semi-structured interviews are a common method of data collection in both the ACF and the interest group literature because they enable the collection of beliefs and perceptions related to a policy change (Green and Houlihan 2004). Consequently, the interviews can be compared, which is of utmost importance when determining both similarities and differences between actors.
Based on the positional approach, I selected one or two actors with a leading position in their domain. The sample includes a wide range of actors spanning different domains: MPs, business interest groups, experts, other types of interest groups and public actors. First, both peak business associations and specific business associations were selected. Then, I identified three public institutions which played an important role in the wake of the crisis: the Swiss National Bank (SNB), the Swiss Financial Market Supervisory Authority (FINMA) 2 and the public administration.
In order to collect more information, I also embedded two experts coming from the university. Finally, I decided to favour MPs-members of the Economic Affairs and Taxation Committees (EATC)-that were in office between 2007 and 2011 and, when possible, preference was given to those who had served in the Parliament for many mandates. Importance was also attached to a certain representative equality, which is why all political parties are present in the sample.
In the end, I selected 31 actors to interview (see 'Appendix A'), including 15 members of corporate interest organisations, two experts, three public actors and 11 members of both chambers' Economic Affairs and Taxation committees. The selection of respondents included a variety of actors in order to understand the beliefs of actors who know or have dealt with the crisis and the Swiss economy. Of the selected actors, I managed to interview all of them apart from 3 MPs.
The structure of the questionnaire was basically the same for all participants, although the questions were also geared to the particular field of action of the actors. The interviews lasted on average between 45 and 90 min, and the vast majority of them were conducted in person, although due to the COVID-19 pandemic, 10 interviews were conducted via zoom. They were recorded and transcribed generating a considerable amount of data. The interviews took place over three months between October 2020 and January 2021.
Relying on the method of thematic analysis, all the transcriptions were read in order, not only to gain a general understanding of the situation but also to crystallise the perceptions and the opinions of the actors. Then, in a second phase, I coded the interview manually to identify the recurrent theme, the beliefs regarding policy fields and the relation between each actor. Finally, in order to corroborate the first analysis, I employed the software Nvivo.
Since this research attempts to identify whether a possible variation has occurred after the crisis, it considers the following period of analysis: 2007-2017.

Mechanism 1: issue salience
The crisis reached a tipping point on 15 September 2008, when Lehman Brothers, a systemically significant financial institution (SIFI), collapsed, unleashing a financial tsunami that also hit Switzerland. The failure of this SIFI, along with the failures of others, resulted in a large-scale crash that impacted all global financial markets and created a climate of financial uncertainty.
Switzerland, given the importance of its financial centre, was also affected by the crisis, its two largest banks-UBS and Credit Suisse-severely so. Due to the former bank's huge slump, Switzerland was pressured to step in and bail it out, while Credit Suisse received a private recapitalisation. On October 16, the federal government and the SNB unveiled their 'UBS rescue package': CHF 6 billion from the Confederation to restock the bank funds and USD 54 billion from the central bank to enable UBS to transfer the illiquid asset fund to the 'StabFund'. As stressed by the respondents, it was an unprecedented moment that led the federal government to take an important decision without consulting the parliament.
The bailout of several financial institutions, combined with global financial and economic issues, has eroded public trust in financial institutions, both among investors and the general public. Hence, the distress of the financial world was placed under a critical spotlight, something that all respondents mentioned. This is summarised well by an interest group acting in the technology sector, which, when discussing the conditions of the Swiss financial sector after the crisis, exclaimed: 'the magic is gone!'. Furthermore, Vogler et al. (2016) concluded that not only media coverage of the UBS case but also coverage of the overall financial sector was largely negative. They also stated that the media tended to use more normative standards to evaluate financial meltdown. In the same vein, based on the paper of Bromley-Trujillo and Poe (2020), I used Google Trends to calculate the issue attention 3 of the financial crisis of 2007-2008 in Switzerland. Figure 2 shows the results and adds evidence to the increase in issue salience.
All these elements point to an increase in issue salience. Moreover, smaller financial institutions (cantonal banks and regional banks) blamed the SIFIs for deviating over the years from their basic tasks in order to increase profit: 'the crisis was born abroad and involved mainly system relevant financial institutions and not us' (Cantonal Bank Association).
Problems with reputation did not finish. In 2009, an array of scandals hit the Swiss financial institutions again threatening once more its financial system: 'The perception of citizens on banks has clearly changed: people have noticed that there were some abuses in this field' (economiesuisse).

The output and mechanism 2: increasing regulation
Generally, since the deep core beliefs of business interest groups are basic ontological beliefs based on a liberal form of economy, they are extremely resistant to change (Weible and Sabatier 2005). As such, diversities can be detected mainly in policy core and secondary beliefs. For this reason, in the following section only the policy core and secondary beliefs of business interest groups will be addressed.

Financial regulation
Many members of the EATC explained that they were aware of the window of opportunity subsequent to the crisis: 'Hildebrandt 4 came openly to the commission and made the point that there was a window of opportunity to enforce financial regulations'. Notoriously, the first attempt to regulate the financial sector emerged already at the beginning of the crisis. Important are indeed the so-called too big to fail packages which were elaborated following the problems encountered by UBS. Globally, the respondents asserted that this was a necessary approach. In view of these actions, members of the Christian Democratic Peoples' Party of Switzerland (CVP 5 ), the Socialist Party (SP), two experts and the cantonal bank association stated that the measures taken were the very minimum and that the two largest Swiss institutions are still 'too big to fail'. On top of that, several respondents affirmed that the reckless behaviour of some SIFIs represented the core problem at the root of the crisis and, as such, it was important to regulate them properly. A sentence that summarises this notion came from an MP member of the CVP: 'what the SIFIs did was simply a catastrophe'.
A first division between inward-looking banks (cantonal banks, regional banks and some private banks) and the larger more outward-looking banks was put forward by the vast majority of the respondents. The underling rhetoric advanced by them was that most of the post-financial regulations were designed as 'one size fits all'. This phenomenon seemed to create some division also within the SBA in the field of financial regulation. Remarkably, this trend was openly raised by 13 respondents coming from different fields and political traditions. As a case in point, the USAM stated that conflicts within this association were noticeable and stronger, while an MP of the Swiss People's Party (SPP) added: 'I took part in some SBA meetings and there was UBS and Credit Suisse who did what they wanted, while the cantonal banks were not willing to take any risks'. Another argument that underlines these divisions stems from the willingness of UBS and Credit Suisse clients to shift their bank accounts to other 'less risky' banks (Brown et al. 2020). Specifically, one MP member of the SPP explained that many bank accounts were transferred to more regional banks, such as cantonal banks or Raiffeisen. Lastly, an important element that reveals a jeopardised environment between financial institutions is that several actors, including the Swiss Insurance Company (SIA), asserted that banks were busy protecting themselves rather than finding a common way out.
This trend is to be found also within the business interest group community. A division emerged during the provisions covering the entire banking sector, which came into force in the decade since the crisis. One such example is the incorporation of the Basel III Accord into the Swiss law, which was designed to strengthen both capital and liquidity standards for the entire banking sector. The USAM, the cantonal bank association, Expert Suisse, Swico, SEU and SFU argued that they deemed it unfair that smaller financial institutions with fewer resources should also be subjected to a more stringent regulation. A clear statement that summarises this reticence came from the cantonal bank association: 'although the crisis was engendered manly by big institutions the entire sector had to face the consequences'. Moreover, after the global financial crisis, the USAM began to incorporate financial matters into their political agenda. The association argued that this was conceived to give greater importance to the Swiss market rather than the international one. In particular, they included cantonal banks, some private banks and other regional banks as part of this new cooperation. Also, Swico stressed a major division among the business interest group community: 'I think economiesuisse has not really survived several issues. They are not seen as a positive association by a large section of the population. By contrast, the USAM is much more down to earth compared to economiesuisse, which tend to defend the interest of the companies'. A certain amount of dissatisfaction towards large companies and thus towards the business interest groups defending them was stressed by almost all the respondents.
It therefore transpires that there are two different types of focal points among business interest groups in the field of financial regulation: one group is more outwardly oriented, while the other is more locally focused (Table 1).
Another crucial division is linked to both the SBA's reputational issues and its internal divisions. Respondents contended that prior to the crisis, the bank association held a privileged position within the ranks of economiesuisse: its peak association. In contrast, in the aftermath of the financial turmoil, it was acknowledged that the financial sector had lost a large part of its influence: 'the profits (of the financial sector) remain high, but this sector does not have the same weight any more'. (SwissHolding). By the same token, an MP of the FDP.The Liberal explained that 'banks have lost their reputational power compared to other economic areas'. Consequently, their manoeuvring space may have shrunk. As a member of the CVP stated: 'representatives of the SBA used to tell us what to do. Then, following the crisis, they had to come to the Federal Building to ask for money. That is not easy to face'. Moreover, the respondents noted that the machinery sector had some frictions with the financial centre because they blamed the outward-oriented financial sector for the crisis. It must be acknowledged, as SwissHolding emphasises, that conflicts between these two sectors existed prior to the start of the crisis, albeit in a more covert manner, whereas after 2008, they became more public. However, a consequence of these factors might have been that relatively speaking the industry gained in prominence vis-à-vis the financial sector. In this case, we are rather confronted with a 'industry versus finance' cleavage.

Tax policy
According to an ex-FINMA chief, the gradual loosening of Swiss banking secrecy was a major feature of the post-financial crisis climate. Starting from late 2008, the USA began requesting administrative assistance in order to obtain customer data (SFI 2016). This phenomenon was extensively discussed by the respondents. By way of illustration, two MPs asserted that 'in Switzerland, we always react in response to the pressure exerted by the international arena'. Likewise, a member of the SPP explained: 'We had no choice but to comply with international standards: we couldn't say no. ' In addition to that, several EU countries experienced severe economic distress, prompting them to strive to refinance their public debt. According to the SBA, 'after the crisis, EU countries had a very large public debt. That's why they have pointed the finger at Switzerland to get the people who defrauded them.' Also, economiesuisse supported this argument, noting that 'during the financial crisis countries were looking for every penny to refill their coffers.' Table 1 Policy core beliefs: financial regulation   Outward-looking  Inward-looking   Actors  SBA, large financial institutions,  economiesuisse, export business interest groups Cantonal banks, regional banks and the USAM The interviews reveal that at the very outset of the international pressure on Swiss banking secrecy, the SBA, supported by the centre-right faction, and the federal government tried to preserve it. Similarly, the interest group representing the hedge funds asserted that 'banks were likely to stay in the bunker and were unwilling to confront the truth'. However, as time passed, it became evident that concessions would be required in order to maintain access to the US financial sector. One example is the so-called Lex US, which is an agreement between Bern and Washington on the exchange of information regarding the account of 4450 UBS clients. Both the government and the SBA were in favour of it (Kuntz 2010), while the SPP and the SP were at first against it. An important dynamic that arises from the interviews is that the SBA has increasingly come to terms with the USA for fear of retraction. Another step towards an exchange of information took place at the end of 2009 with the launch of the new Rubik Agreement 6 with Germany. This agreement was conceived by the president of the Foreign Association Bank. However, this project was rejected by the German parliament's lower house (Sainsaulieu and Vinck 2015, p 108). In view of the tense situation, there was a clear shift whereby first the president of UBS declared on October 31 (2012) in the Tagesanzeiger that 'automatic data exchange can be a solution, but only if it applies to everyone worldwide. Otherwise, the current OECD standard applies'. A couple of months later the president of the SBA followed the same line 7 . Additionally, the SBA explained that 'big banks battled less vigorously because they were less bounded by banking secrecy'. By the same token, the actor representing the FINMA noted that: 'the raison d'être of the SBA was the banking secrecy and when that ended, the association encountered major problems'.
This might also be of particular importance given the cleavages within the banking sector discussed in the previous section. Another important consequence of the numerous cases of tax evasion linked with banking secrecy is the damage to banks' reputations. As reported by nine respondents, this might also have contributed to the rise of new economic branches vis-à-vis the financial sector and its relative loss of importance in the Swiss economic arena ( Table 2).
The escalation of the EU debt crisis, along with increased tax competition and widespread tax evasion scandals, resulted in greater cooperation in tax information exchange and corporate tax changes (Fung 2017). A concrete example of this trend is the base erosion and profit shifting (BEPS) project elaborated by the OECD in 2012, whose main objective was to reduce the privilege of several companies residing in tax heavens. This international framework was translated into the federal law through the third revision of the Corporate Tax Reform (CTRIII). 8 Discussions about the new standards began in 2015 and culminated in the federal vote of 2017, without being accepted by the majority of the citizens.
The larger companies supported this measure, while the banking sector, according to SwissHolding, was relatively neutral, recognising the need for revision. Also, the report summarising the consultation phase illustrates that economiesuisse and the SBA were basically in favour of the direction proposed by the steering body (Federal Tax Administration 2019). A wide range of business interest groups argued that they need to comply with the OECD directives in order to avoid any form of retaliation, such as being included in a black list or paying a considerable fine. As reported by Longchamp (2010), holding companies would have been among the main losers from possible economic sanctions. However, MPs asserted that they not only witnessed heated debates in the parliament, but that several elements were introduced to deal with the elimination of holding company privileges. Indeed, respondents noted that cantonal taxes would have needed to be reduced to compensate for the new regulations. This new trend can be directly observed in the canton of Geneva, where the corporate tax rate was reduced from approximately 24.17% in 2019 to 13.99% (for companies without preferential tax regimes). The USAM declared its support for the revision, while emphasising that the tax status of holding companies should not be abolished (Federal Department of Finance 2013). Moreover, it was reported in an article of 'Le Temps' of 13 February 2017 that 'Economiesuisse and the USAM both supported the CTR III, but the strong campaign led by the latter divided minds'. In other words, the USAM's more aggressive attitude in the campaign did not particularly please economiesuisse. As a result, while a compromise between the various economic interests was reached, tensions emerged on the secondary beliefs. It should also be noted that the USAM's intensifying alignment with the SPP coincided with an increase in tensions. This was reported by five respondents and in Le temps (Le Temps 2019). Likewise, according to the SIA, 'there are increasing problems between the USAM and economiesuisse. In terms of international agreements, the USAM is becoming highly conservative and refractory'. As a case in point, in September 2015, the Executive Board of economiesuisse informed the then President of the USAM about the fact that he would no longer be invited to its meetings (TdG 2015).
That being said, the CTR III project was rejected by citizens during the 2017 popular vote. To break through this political impasse, a new part was added to this reform: the old age and survivors' insurance (OASI) revision. In a nutshell, under the impulse of the socialist party, the idea was to increase the OASI funding by CHF 2 billion, primarily through an increase in contributions and VAT. According to swissinfo (2019), this compromise, officially known as the Tax proposal and OASI financing (TRAF), has managed to reunify different opinions. The USAM explained that they were among the first business interest group to support this agreement. They affirmed: 'We were supportive, social compensation had to be found'. Conversely, they explained that at the beginning of the process the SEU and economiesuisse were quite reticent and wanted to provide some changes to this proposal. These opinions were also discussed in the 'Année politique suisse' by Benteli et al. (2021) and in the consultation phase (Federal Tax Administration 2019) ( Table 3).

Corporate governance
As previously mentioned, the various scandals surrounding financial institutions also shed light on the executive compensation of CEOs. These pecuniary incentives were notably central to the risk-taking behaviour applied by several managers before the advent of the crisis (Fahlenbrach and Stulz 2011). Consequently, more specific attention was drawn to the question of remuneration of executives within Swiss financial institutions. 9 Given this issue, revisions were made, such as the Circular 2010/1 on minimum standards for remuneration systems at financial institutions issued by the FINMA on 21 October 2009 (Remuneration Circular). Also of great importance was the initiative against abusive remuneration, which brought this subject into the centre of the public debate. The business interest groups under consideration were particularly opposed to this initiative, 10 while, on the other front, Thomas Wiedmer (member of the SPP), the CVP and the SP supported it. As an illustration, within the business interest group community almost every respondent showed a certain reticence towards this initiative. However, according to the association representing the interests of the textile industry and the SFAMA, it was 'much ado about nothing'. In other words, they argue that this initiative affected people sensitivities, without going to the heart of the problem. By the same token, the insurance association and the SFU stressed that although they understood the problematic behind higher compensations, they did not agree with the measures proposed. Interestingly, five respondents criticised the 'manager denken', stating that, unlike today, in the past managers were sincerely responsible for their workers and committed to improving the Swiss economy. On this topic, an MP of the CVP said 'there is no longer a Swiss culture (within the managers), Table 2 Policy core beliefs: banking secrecy   Outward-looking  Inward-looking   Actors  SBA, large financial institutions,  economiesuisse, export business interest groups Cantonal banks, regional banks, some private banks and the USAM and this has created a widespread dissatisfaction.' This logic might again indicate a certain dissatisfaction among SMEs towards large companies with international managers. Moreover, business interest groups in favour of SMEs like the cantonal bank association and the USAM justified its positioning against higher regulation by arguing that better solutions are to be found within the company.
The USAM also noted that these problems concerned mainly larger companies. It must be emphasised that the countervailing strategies to fight back this initiative were decided by the umbrella association. Specifically, it looks like economiesuisse was especially involved in the campaign against the initiative. Consequently, given their notorious involvement, the defeat was even harder. This also seems to be one of the elements that led both the president and the director of economiesuisse to resign a couple of months later. Another wage initiative was put to the vote shortly after the 1:12 initiative. In this case, the umbrella associations once again led the campaign. Like the previous example, in this case the positioning also seems unanimous. That said, following the defeat of the minder initiative, economiesuisse handed over the reins of the campaign to the USAM and the SEU. As elucidated by the president of the USAM in the newspaper le nouvelliste: 'Managers of SMEs can better defend the 'no' than representatives of large companies' (Le Nouvelliste 2013). This example, coupled with the answers of the respondents, shows once again the reputational problem that large companies faced.

Discussion and conclusion
The evidence collected suggests that the increasingly negative opinion on financial institutions and transnational companies greatly affected the policy issues analysed and thus also the preferences of business interest groups. Indeed, it is a story that is very much centred on the reputational loss and the business interest groups quest for organisational survival-from a reputational perspective. Almost all the actors interviewed stressed that after the crisis, the equilibrium in the business interest group community changed in response to the negative image undermining the financial sector. To summarise the findings, the figure below depicts the primary mechanisms through which the financial crisis increased conflicts within this community and contributed to the rise of two coalitions: inward-oriented and outward-oriented business interest groups (Fig. 3).
It transpires from the analysis that after the global financial crisis a number of conflicts arose within the business interest group community. This shock has thus fuelled the development of several differences regarding both policy beliefs and secondary beliefs. First, frictions between the outward-and inward-oriented banking sectors appear clearly in both financial regulation and banking secrecy. Second, divisions in the banking sector coupled with its general discredit are likely to have harmed the financial sector's important position within the umbrella association economiesuisse. Third, on a more general level, differences in policy beliefs among business interest groups were evident partly in tax policies and fully in financial regulation (see 'Appendix B' for a summary table). Consequently, two types of coalitions can be outlined in these policy issues: outward-oriented and inward-oriented coalitions. By contrast, the initiatives on executive remuneration, the CTRIII and the TRAF had divisions only in the secondary beliefs. With regard to the former, the way the campaign unfolded seems to have divided spirits between the two major umbrella associations (USAM and economiesuisse). As for the TRAF measure, the USAM was positive from the outset, whereas the outward-oriented business interest groups were more reticent. Consequently, even though the crisis engendered mainly some intense cleavages within financial regulation and the banking secrecy regulation, a variation of economic equilibrium was also visible in the other policy issues.
This study tangibly reveals the mechanism underlying the reaction of business interest groups to a financial meltdown, while also providing additional evidence for their survival concerns. As a result, following a moment of high issue salience, reputational concerns and external pressure, business interest groups alliances tend to vary. This in-depth qualitative analysis thus provides a deep understanding of the changes in preferences following the crisis by means of detailed case studies and through the process tracing technique. These methods are especially well suited for investigating business interest groups' preferences and alliances since they allow the researcher to focus on the process and development of coalitions over time.
It should be noted that although this study can clearly serve as an example, its generalisability is limited by a lack of comparisons with other countries. Besides, because the financial crisis occurred a few years ago, there is a possibility of biased and selective reporting. However, now that some time has passed, respondents may feel more at ease talking openly. Ultimately, in order to corroborate the findings, the process tracing framework presented here could be applied to other types of crisis and policy fields.
These results are all the more important considering that the literature suggests that higher levels of business conflicts are directly related to lower levels of policy success (Dür et al. 2015, p 953). However, in order to better grasp this matter future research would be required to assess which kind of coalition is most likely to succeed.

Appendix A
See Table 4. See Table 5.