Effects of Public Corporate Governance Guidelines at the Subnational Level in Switzerland

Public corporate governance guidelines are viewed as a means to increase the transparency and efficiency of fully or partially state-owned organizations. Many OECD countries have thus introduced such guidelines over the past decade. We examine the effects of the introduction of corporate governance guidelines on the governance process and on government entities’ portfolio of holdings. The analysis comprises the procedural effects, measured by the level of control, and the resulting effects, measured by the structure of the portfolio, of corporate governance guidelines. The introduction of guidelines at the subnational level in Switzerland has led to a significant increase in government control. However, the effect on government entities’ portfolios is not obvious, as the introduction of guidelines does not necessarily lead to a reduction in the overall number of holdings of regional governments, an increase in the participation quota, or an increase in the proportion of public-law institutions in the portfolio.


Introduction
In the past decades, governments in the western world have tried to realign their public sectors with the changing global environment (Reith, 1996).Based on the prospect of increased efficiency and effectiveness, state functions have been removed from central administrations or even privatized.Consequently, the political-administrative relationship has been redefined, and the role of the state has shifted from providing public goods and services to ensuring the provision of such goods and services (Verhoest et al., 2012).Through outsourcing, the responsibility for operational provision is delegated to public sector organizations, increasing their autonomy and entrepreneurial accountability.While the responsibility for performance is delegated to the created legal entities, the responsibility for provision remains with the government.
Like many other countries, Switzerland has re-evaluated the role of government in society.The outsourcing of state functions to state-owned legal entities has been common practice for some time (Schedler et al., 2013).Additionally, government entities in Switzerland are holding shares of different organizations for historical or purely financial reasons.Despite the large number of state-owned legal entities, knowledge on the structure and public governance of these legal entities at the subnational level remains limited.Although corporate governance guidelines (regulations and soft laws providing specifications regarding what legal entities shall be owned by the government for what reason and how they shall be steered) and owner strategies were established in Switzerland at the federal level in 2006, at the subnational level, the establishment of regulations or guidelines remains at the discretion of government entities (Federal Council, 2006).As a result, only a few regional governments have a clear framework for the supervision of their holdings (Schedler et al., 2013), and a differentiated landscape of practices for the control of partially or fully state-owned organizations in Switzerland has emerged.
The 26 Swiss cantons offer a basis for an evaluation of the effect of public corporate governance guidelines.Indeed, because Swiss cantons with guidelines can be compared to a reference group of regional governments without such guidelines, the introduction of public corporate governance guidelines and owner strategies in Switzerland provides a compelling opportunity to evaluate the procedural and resulting effects of public corporate governance guidelines.Accordingly, this article aims to analyze the effects of guidelines on the level of control and structure of fully or partially state-owned entities.
Specifically, this article attempts to answer the following question: What are the effects of public corporate governance guidelines on the control of fully or partially state-owned entities and the structure of their portfolio of holdings?
Regarding methodology, a comprehensive survey in all regional governments and municipalities with more than 50,000 residents is conducted.The present study is a replication and extension of the study by Meister (2009), which analyzed the portfolios of fully and partially state-owned entities of regional governments with regulations or guidelines on corporate governance in place based on data from 2007 and 2009.The replication allows for a comparison of the situation five to seven years later and an evaluation of the effects of these regulations.We have sent the questionnaire to the responsible person in the central administration who is in charge of controlling the fully or partially state-owned entities.We focus on comparing the portfolios between the cantons and municipalities that have introduced corporate governance guidelines and those that have not introduced such regulations or guidelines.
Public corporate governance guidelines aim to improve the process of handling public sector organizations and to allow for effective governmental control (procedural effect).We therefore expect these guidelines to affect the structure and composition of fully or partially state-owned entities (resulting effect).

The article is based on two hypotheses:
To evaluate the procedural effect of the introduction of public corporate governance guidelines, we analyze the control activities of the cantonal governments and cities with a population of over 50,000.
Hypothesis 1: An increase in the control activities of a government over fully or partially state-owned entities can be observed after the introduction of public corporate governance guidelines.
We expect to find an increase in the control activities of governments with public corporate governance guidelines because the introduction and standardization of (new) controlling tools should improve the control process.The literature has demonstrated that administrative reforms lead to the introduction of new instruments and mechanisms while the old instruments and mechanisms persist and are not fundamentally revised (see e.g., Gregory, 2006).The introduction of public corporate governance guidelines would therefore increase the control activities of the government by providing them with additional mechanisms of formal control.We measure the use and intensity of control activities by using the gathered data.
To assess the resulting effect of public corporate governance guidelines, we examine the structure of the holdings of the cantonal governments.
Hypothesis 2: The number of fully or partially state-owned legal entities of cantonal governments with public corporate governance guidelines has decreased since the introduction of such guidelines.
We expect the number of state-owned legal entities to decrease in the years after the introduction of corporate governance guidelines because public corporate governance guidelines usually provide guidance on risk management and liability assessment, which in turn affect the evaluation of holdings.To avoid market distortions, government entities should reconsider their ownership in areas with an open market and high risk and should concentrate on their regulatory function (Boston et al., 1996).

Theoretical Reflections
The introduction of New Public Management (NPM), the agencification and privatization trends of the 1980s and 1990s, and the related separation of government and publicly funded services led to the expression of the hollow state (Milward, 1996;Milward and Provan, 1993;Rhodes, 1994).The aim of NPM was to create better performing semi-autonomous public sector organizations through increased managerial autonomy and results-based control (Dunleavy, 1997).In an extreme form, the approach could lead to a hollow state, which is a state that has contracted out all its production capabilities to third parties.However, such extreme cases do not exist in practice, where a wide variety of outsourcing practices can be observed (OECD, 1997).The organizational proliferation caused by NPM has led to the development of a post-NPM doctrine, which attempts to strengthen the role of government (Gregory, 2006, Halligan, 2006).

2.1
The neo-Weberian model and the doctrine of the "whole of government" Political control of fully and partially state-owned entities and the potential democratic deficit associated with the delegation of state functions are major concerns of research in administrative science (e.g., Skelcher, 1998).Control comprises the mechanisms and the instruments that influence the decisions and behavior of the controlled party and that the controlling actor uses to fulfill his objectives.Control relationships can be unilateral or mutually reciprocal, as well as vertical, horizontal, or diagonal (Verschuere et al., 2006).
In the new millennium, the post-NPM doctrine of "whole of government" has gained popularity and has led to reforms in several OECD countries (Gregory, 2006).According to Van Thiel and Leeuw (2002), the lack of justification of agencies has been at least partially repaired in most OECD countries by the increased use of regulators and performance assessment.In addition to marketand network-based forms, these reforms also include hierarchical forms of coordination (Halligan, 2006).With the stronger emphasis on internal control and the implementation of corporate governance guidelines and regulations, new forms of compliance to rules and standards have evolved.However, improving corporate governance in the public sector has been a major challenge in many countries.Since state-owned enterprises constitute a substantial share of the economy, it is important to ensure that governments do not distort competition with the improper application of regulatory or supervisory powers.In 2005, the OECD developed guidelines on the corporate governance of state-owned enterprises (OECD, 2005).These guidelines, which still serve as an international benchmark, suggest the strict separation of the state's ownership and regulatory functions.Although national practices diverge to some extent, a common understanding of good practices has evolved over the past decade.

Public corporate governance
Corporate governance aims at the responsible steering and control of organizations.Public corporate governance applies to the public sector and refers to the responsible steering and control of decentralized public organizations.There are a number of difficulties associated with the steering and control of organizations, most of which are related to conflicts of interest (Conforth 2003).The first and most general difficulty in corporate governance is the problem of delegation.
The state as the provider of resources and (partial) owner delegates resources and goals to public sector organizations.The potential challenge associated with this process is the assurance of goal congruency between the government and the management of organizations.The second problem is related to the first one.
It is the problem of information asymmetries and the opportunistic behavior of board members and management.To foster goal congruency, control and sanction mechanisms have to be imposed.The third problem is the opacity of complex steering systems.These general corporate governance problems are accentuated in the public sector because a multi-layered principal-agent relationship exists in the public sector.The citizen is the principal vis-à-vis the legislative authority, the legislative authority is the principal vis-à-vis the executive authority, and the administration is the principal vis-à-vis the agent of a public organization.
According to Finger (2002), political control must be secured with public corporate governance (Finger 2002).However, the essence of governance is its focus on the governing mechanisms that do not rest solely on the authority and sanctions of government (Stoker, 1998: 17).To minimize conflicts of interest, the functions and tasks of the most important stakeholders must be defined.With regard to the audit function in particular, the functions of management and the board need to be defined.Furthermore, the oversight or control function of the government needs to be clarified in the public sector.Public corporate governance guidelines are believed to help resolve these existing conflicts of interest.

Effects of public corporate governance guidelines
What is the effect of the introduction of corporate governance guidelines?Do they increase transparency, lead to a reduction in risk, and improve the interaction between the governments and the enterprises?
The possible effects of corporate governance guidelines can be divided into two categories.On one hand, there are procedural effects, as corporate governance guidelines are expected to have a noticeable effect on the way governments handle their holdings.On the other hand, there are the resulting effects, as the new processes are expected to influence the structure of governments' portfolios of holdings.

Effects of guidelines on the process of managing holdings
In the literature, control is often mentioned in the context of and compared with the degree of autonomy of public sector organizations.The introduction of NPM and post-NPM principles implies a change in the ways in which agencies are controlled by the government.NPM focuses on results-based control, which means that an agency is judged on whether it has reached its targets (Verhoest et al., 2004).An increase in results-based control mechanisms can therefore be expected.However, some governments are struggling to adopt the appropriate me-chanisms to steer and control the activities of different service providers (Pollitt and Bouckaert, 2011;OECD 2004).The shift to results-based control is especially challenging because governments fear a reduction in service quality and a rise in tariffs caused by the dominance of a for-profit orientation if services are outsourced.Therefore, to ensure the public interest, governments have to develop new control and accountability systems (Grossi and Mussari, 2008).
As implied by governments' attempt to balance the public interest, organizational control is strongly associated with an organization's granted level of autonomy.The two concepts of control and autonomy are therefore related (Verhoest et al., 2004).A high degree of autonomy for agencies is one of the core elements of NPM and post-NPM doctrines.However, studies on the relationship between autonomy and control have not always yielded conclusive results (e.g., Yesilkagit and Van Thiel, 2008;Roness et al., 2008).Indeed, the level of autonomy and control of public sector organizations varies within and between countries, and control by the government can be complemented or substituted by agents from the private and nonprofit sector (Milward and Provan, 2000).A recent study showed that the task of agencies is not as relevant for the degree of autonomy and control as the size of the budget: agencies with larger budgets experience less autonomy and more control than agencies with smaller budgets (van Thiel, 2014).Despite the differences between the different fully or partially state-owned organizations and the different levels of control and autonomy they experience, common controlling measures could help increase transparency.Further, corporate governance guidelines should foster the implementation of results-based control practices, as they formally specify the control instruments and mechanisms.Consistently implemented guidelines should then lead to regular reporting by the public sector organization and provide the government with information on possible problems, thereby improving control (OECD, 2010).
Direct or input control mainly includes the traditional forms of control through direct involvement in the decision-making process of the organizations and the delegation of political representatives on the board of these organizations.Public corporate governance guidelines discourage from this type of control.
Control can be measured by assessing the steering and controlling mechanisms and instruments used by governments.Equally important, however, is the perceived control of the government.Do the responsible government representatives feel that the introduction of corporate governance guidelines changed the government's level of control over its holdings?

Effects of guidelines on governments' portfolio of holdings
The degree of agencification varies considerably between countries (Verhoest et al., 2012), and with an increase in degree of agencification, the number of fully and partly state-owned organizations increases.Previous research has examined the interest of politicians in agencification.As decision makers, politicians have a direct influence on the outsourcing of state functions and therefore the number of fully or partly state-owned organizations.While the most important argument for agencification is usually the expected performance increase, findings on the actual performance of these organizations remain mixed (Pollitt, 2001;Van Thiel 2001).Van Thiel (2004) showed in her study that politicians mainly adhere to their previous code of practice when they decide to outsource additional capabilities and tasks.Thus, governments that outsource many tasks have the tendency to persist in this practice.Whether the introduction of guidelines affects the outsourcing practices remains uncertain.
However, agencification is not the only reason for the growing number of fully and partly state-owned organizations.An additional reason is the adoption of new tasks because of changing public needs.However, with an increase in the number of fully and partly state-owned organizations, the level of risk for the government usually increases.
Governments should consider the potential financial, reputational, and political risks associated with the ownership of fully and partially autonomous organizations.Minority holdings in particular provide the government with only limited influence on the conduct of business (Schedler et al., 2013).With the national and international opening-up of markets, freedom of choice for both customers and regional governments increased.Further, deregulation and the market entry of competitors decrease the possibility for price increases due to long-term investments (Meister, 2009).Public corporate governance guidelines can enhance governments' capacity to understand the risks and liabilities associated with these holdings.With greater transparency and risk awareness, governments will likely try to reduce risk by divesting some of their holdings, and the introduction of corporate governance guidelines will thus lead governments to (re-)evaluate their holdings.The liabilities of a government largely depend on the legal form of its holdings.In the case of a legally independent organization, the liability is limited to the allotted capital.However, enforcing such a liability could be difficult in practice.In many cases, it will be impossible for political reasons for a regional government to allow the bankruptcy of one of its significant holdings.Mismanagement of public sector organizations therefore has a direct effect on the government's finances, and the government would likely reconsider ownership in areas with an open market and high risk.By reducing the number of their holdings, governments could lower their risks and liabilities.In these areas, governments should concentrate on their regulatory function (Boston et al., 1996).

3
Analysis: Corporate Governance in Switzerland at the Subnational Level

Situation and developments
As a federal state, Switzerland is divided into 26 regional governments, the socalled cantons, which are the original states that united in 1848 to form the Confederation.As Switzerland is a multiethnic, multilingual, and multiconfessional country, the cantons differ substantially in not only these aspects but also size and population.In 2012, the smallest canton, Appenzell Innerrhoden, had just over 1% of the population of the most populous canton, Zürich (BFS, 2013).Thus, in absolute numbers, while the canton of Zürich had a population of over 1.4 million in 2012, the canton of Appenzell Innerrhoden had a population of only 16 thousand.Although the governments at the regional level (cantons) in Switzerland are quite heterogeneous, they share many similarities in terms of duties and responsibilities (Linder, 2012).
Based on their political and social mission, cantons traditionally have holdings in the financial industry, in addition to the transport, energy, health, and education industries.However, the largest segment of participation does not belong to one of these categories or traditional public sector industries; participation that does not belong to any of these industries is assigned to the category "other," as shown in Figure 1.The category "other" includes participation in the agricultural sector and the cultural domain as well as an increasing number of holdings in the area of information technology.The aggregated portfolio of cantonal holdings for the year 2007 included 961 cantonal holdings and 608 organizations (Meister, 2009).The total number of legally independent public sector organizations in 2012 was 867 holdings and 633 organizations.The apparent discrepancy between the number of holdings and the number of organizations arises because quite a few of these organizations are intercantonal (Meister, 2009), indicating that two or more cantons are co-owners of these organizations.The greatest decrease in the number of holdings occurred in the category "other."The percentage of holdings in which the canton was the sole owner was 8% in 2007 and 12% in 2012.Parallel to the increase in holdings in which the canton has sole ownership, the proportion of holdings in which the canton is majority owner increased from 8% in 2007 to 19% in 2012.
Since the 1990s and the beginning of the new millennium, the portfolios of cantonal holdings have increased considerably.This increase can be attributed to two developments: first, the cantonal governments have outsourced additional tasks, for example, owing to deregulation; second, they have not reduced their historical holdings.With the increasing number of holdings, the regional governments became increasingly aware that many public services are provided not by the government but by (semi-)autonomous organizations.Service provision by (semi-)autonomous organizations can lead to a perceived democratic deficit because the parliament no longer has direct control over its holdings.This situation leads to the desire for increased transparency and the introduction of certain steering mechanisms.The introduction of corporate governance guidelines is believed to be a way for cantons to improve this situation.The cantons started to introduce public corporate governance guidelines in the beginning of the new century.However, substantial differences have emerged between the different cantons (Schedler et al., 2013, p.1).Indeed, only seven of the 26 cantons had implemented formal laws or guidelines on public corporate governance by 2009 (Meister, 2009).
By 2014, the number of cantons with corporate governance guidelines in place had increased to 14.An interesting case is the canton Jura, which indicated that it had implemented public corporate governance guidelines in 2009 (Meister, 2009) but contradicted this statement in the survey in 2014.Aside from this canton, the cantons that had implemented guidelines in 2009 still had them in place in 2014.Interestingly, some cantons with a large number of holdings have still not introduced corporate governance guidelines.
Based on the OECD guidelines, a corporate governance report at the federal level was published in Switzerland in 2006, which led to a conceptual basis for the steering and organization of outsourced services (Federal Council, 2006).
A question that remains is whether the introduction of public corporate governance guidelines had the expected effect, namely, an improvement in transparency, an increase in control, and a consolidation of the portfolio of holdings in the cantons.

Direct control
Direct control can be attained by the delegation of politicians on the boards of a regional government's holdings.As stated by Meister (2009), many cantons still had politicians on the boards of a substantial number of public sector organizations in 2007.By 2012, the proportion of holdings in which a government representative is present on the board has decreased.

Table 1: Proportion of the cantonal holdings with at least one government representative on the board
As shown in the table, the degree of direct control of regional governments decreased between 2007 and 2012 in all cantons regardless of whether corporate governance guidelines were in place.However, the decrease in direct control is lower for cantons without guidelines and higher for cantons that introduced guidelines between 2009 and 2014.Hence, the introduction of public corporate governance guidelines is likely associated with a substantial decrease in the number of government representatives on boards.Indeed, the decrease is above 10% in cantons with guidelines.A possible explanation for this decrease is a shift from input and process control to results-based control, triggered by the introduction of corporate governance guidelines.To further evaluate the effect of the introduction of guidelines on control, the implementation of steering and controlling instruments is evaluated for the Swiss cantons and the largest cities in the next section.

Implemented steering and controlling instruments
To manage their holdings and to establish a basis for results-based control, governments must introduce steering mechanisms.The implementation of steering and controlling instruments is essential in this context.If strategic goals are formulated, the monitoring of the achievement of these goals is a prerequisite for interventions.Equally important is adequate risk management.Risk management is used not only to identify necessary corrective measures but also to decide whe-ther to end participation in a certain organization.In 2009, risk management was not widely used by the cantonal governments.
To evaluate the use of steering and controlling instruments and the implementation of risk management, an index is developed.The calculation of the index is illustrated in Table 2.In the case of risk management, the existence of a central, superior entity responsible for all holdings was included as an additional element, as an objective evaluation of risks is possible only if this function is centralized.

Table 2: Calculation of the index indicating the existence of steering and controlling instruments and of risk management
Based on this index, the cantons and the cities with a population of over 50,000 are evaluated.Tables 3 and 4 display the results of this evaluation.

Table 3: Use of control instruments by cantonal governments in 2014
For the cantons, the exact Fisher test indicates that a significant relationship exists among the implementation of steering and controlling instruments, the use of risk management, and the existence of public corporate governance guidelines.Based on the data, the null hypothesis can be rejected.The data reveal that the introduction of public corporate governance guidelines and the use of steering instruments are significantly related (N = 26, p < .05,two tails).The introduction of public corporate governance guidelines therefore is associated with the existence of steering and controlling instruments and the use of risk management.All cantons with an index value of three or more have corporate governance guidelines, while only two of the eleven cantons without such guidelines have an index value of more than one.
In addition to the cantons, the cities with a population of more than 50,000 in Switzerland are also evaluated.Only eight cities fall within this category.The results for the largest cities in Switzerland are not equally conclusive, as the following table shows.The exact Fisher test does not reveal a significant relationship between the introduction of public corporate governance guidelines and the implementation of systematic steering and controlling instruments (N = 8, p > .05,two tails).Note that the small number impedes our ability to draw a conclusion; however, the null hypothesis cannot be rejected for the largest cities in Switzerland.Therefore, the introduction of public corporate governance guidelines is not associated with the implementation of systematic steering and controlling instruments at the city level.

Perceived control
We included questions on the government's level of control in the questionnaire to evaluate the level of perceived control of the governments in the cantons and largest cities.A five-point Likert scale with the following format was used for all items: 1. Strongly disagree = -2 points; 2. Disagree = -1 point; 3. Neither agree nor disagree = 0 points; 4. Agree = 1 point; and 5. Strongly agree = 2 points.The scores for the cantons and cities, which depended on their answers in the survey, are aggregated in an unweighted index.Because control is exercised by both the executive and the parliament, both are included in the evaluation.In addition to the perceived control of the executive and the parliament, the development of the decision basis and the achieved reduction in risk are evaluated.The index is constructed according to the following formula: (score executive + score parliament + score risk reduction + score decision basis) With one exception, all the cantons with corporate governance guidelines in place reported an improvement in perceived control.Obwalden, the canton indicating that the perceived control decreased since the introduction of the corporate governance guidelines, mainly attributed this decrease to the insufficient reduction of risk and the deterioration of supervision by the parliament.The cantons with the greatest improvement in control attributed this improvement to the enhanced supervision of the executive authority and the enhanced basis for decision making.
The data show that the introduction of public corporate governance guidelines and the level of perceived control are significantly related (N = 26, p < .05,two tails).The introduction of public corporate governance guidelines is therefore associated with a higher level of perceived control of the cantonal governments.
As with the steering and controlling instruments and the implementation of risk management, the results concerning perceived control differ for the cities with a population above 50,000.Again, the pattern observed in the cantonal governments cannot be confirmed at the city level.The data for the cities with a population over 50,000 show no significant relation between the introduction of public corporate governance guidelines and the level of perceived control.Again, the small number of large cities renders drawing relevant conclusions difficult.The exact Fisher test produced a p-value of 0.72 (N = 8, p > .05,two tails).The introduction of public corporate governance guidelines therefore is not associated with a higher level of perceived control at the city level.
Overall, the introduction of public corporate governance guidelines is associated with a higher level of perceived control and results-based control.Direct control through the delegation of government representatives, however, decreased in the observation period, but only for the cantonal governments, not for the largest cities.

3.5
Changing structure of the holdings The introduction of public corporate governance guidelines is expected to affect the structure of the governments' portfolio of holdings.To evaluate this potential effect, we compared the structure of the holdings of the different cantons between 2007 and 2012.Any effect of the introduction of corporate governance guidelines on the structure of holdings should be reflected in the portfolios of the cantons.
The number of holdings varies between 58 in the canton of Zurich and 15 in the canton of Jura.Between 2007 and 2012, substantial changes can be observed in some cantons, as shown in Figure 2. By contrast, an increase can be observed in only a small number of cantons.The greatest increase occurred in the canton of Aargau across all industries.This result is interesting because Aargau was one of the first cantons to introduce corporate governance guidelines in Switzerland.The increase in the number of holdings for the other cantons can be attributed to the recent outsourcing tendencies in the education industry.To facilitate the comparison, we examine the percentage change in the number of holdings of the regional governments.As the results show, no clear pattern exists.Indeed, the number of holdings did not necessarily decrease after the introduction of corporate governance guidelines, as shown in Table 7. Regarding the median, the cantons without guidelines show the highest decrease in the number of holdings over the past five years, with a decrease of 8%.The decrease in the number of holdings over this five-year period was the highest among cantons that introduced corporate governance guidelines between 2007 and 2012.However, nine of the twelve cantons without guidelines also experienced a decrease in the number of holdings.Based on the data, the hypothesis must be rejected.This result is also confirmed by the exact Fisher test.The data reveal that the introduction of public corporate governance guidelines and the number of cantonal holdings are not significantly related (N = 26, p > .05,two tails).The introduction of public corporate governance guidelines therefore is not associated with the number of the cantonal holdings.When deciding to divest certain holdings, governments have to consider different perspectives: they must consider the effect of the divestment as the guarantor of the provision of a certain good and as the owner or investor.Regional governments must therefore determine which holdings will be necessary in the future to ensure the adequate provision of the service or good and which holdings will be financially attractive in the long term.Competition from the private sector in the traditional publicly owned industries of energy, health, transportation, education, and financial services has increased substantially.Nevertheless, many regional governments are majority owners of these organizations.While the number of holdings is highest for "other" industries, the participation quota is lowest.This category includes holdings in the agricultural, information technology, and culture and leisure sectors.From the perspective of the governments as guarantors of the provision of these goods and services, these holdings are no longer relevant because their minority holdings in the organizations providing these goods and services do not allow the regional governments to influence organizational decisions.Furthermore, the financial importance of these holdings is very low.Regional governments are therefore especially likely to reassess these holdings.We therefore examined the development of the number of holdings in the "other" industries category because the effect of the reassessment of holdings due to the introduction of guidelines should be the most noticeable in this category.However, the expected effect of corporate governance guidelines on the number of holdings in "other" industries is not observed.This result is also confirmed by the exact Fisher test.The data reveal that the introduction of public corporate governance guidelines and the number of cantonal holdings in "other" industries are not significantly related (N = 26, p > .05,two tails).The introduction of public corporate governance guidelines therefore is not associated with a stronger decrease in the number of the cantonal holdings in "other" industries.
Other structural changes that may be connected to the introduction of public corporate governance guidelines are also evaluated.An interesting finding in this context is that in the past five years, the proportion of public-law institutions in cantons with existing public governance guidelines and in cantons that introduced guidelines during this period increased.Such an increase did not occur for all cantons without guidelines.As Table 9 shows, 42% of these cantons experienced a decrease in the share of the public-law institution in their portfolios.However, the data reveal that the introduction of public corporate governance guidelines and the legal form of public sector organizations are not significantly related (N = 26, p > .05,two tails).The introduction of public corporate governance guidelines therefore is not associated with a change in the legal form of the cantonal holdings.The mean participation quota increased in many of the cantons during the observation period.Only five of the 26 cantons experienced a decrease in the participation quota.Thus, the proportion of small holdings relative to large holdings decreased.Although this shift can also be observed in the cantons without corporate governance guidelines, the degree of the increase in the participation quota is higher for the cantons with corporate governance guidelines, especially for those with guidelines in place for a longer time period, as shown in Table 10.However, the exact Fisher test reveals that the introduction of public corporate governance guidelines and the degree of the change in the government's participation quota in these public sector organizations are not significantly related (N = 26, p > .05,two tails).The introduction of public corporate governance guidelines therefore is not associated with a decrease in the participation quota of the cantonal holdings.Based on the data, no significant relationship between the structure of the cantons' portfolios and the introduction of public corporate governance guidelines can be identified.

Conclusion
The introduction of corporate governance guidelines at the subnational level in Switzerland causes a significant effect on government control.However, a distinction needs to be made between control at the cantonal level and control at the city level.The cantonal governments that introduced guidelines show an increase in the level of perceived and results-based control.The reasons for the divergent results at the city level would have to be evaluated systematically.The factors that may have led to this result are the general lower number of holdings for cities and the greater direct involvement of cities in the conduct of business relative to cantons, regardless of the existence of corporate governance guidelines, because of their smaller geographic sphere of influence.Thus, cities have a relatively good overview and control over their holdings even without steering measures.Interestingly, no significant effect of corporate governance guidelines on the structure of the holdings and therefore the portfolios of cantons could be observed.Although changes in the portfolios of cantons could be observed, corporate governance guidelines are not the determining factor for such changes.Changes in the structure must therefore be attributed to other factors.One reason that the number of holdings was not reduced as expected within the observation period may relate to the difficulty of divesting existing holdings (Meister, 2009).Furthermore, regional governments are taking on new tasks.As indicated in the study by van Thiel (2004), politicians are likely to follow old patterns in decision making regarding outsourcing.In Switzerland, where the political system leads to the stable relative power of different parties and active politicians, these patterns are likely to be even more pronounced and persistent (Linder, 2012).
To conclude that corporate governance guidelines are not useful because they do not lead to the expected increase in transparency and decrease in risk would be most likely wrong.Further research is thus necessary to elucidate the effects of corporate governance guidelines.The present research also has several weaknesses.Basing the analysis on the question of whether the guidelines were developed led to several theoretical and practical difficulties.First, the existence of corporate governance guidelines does not allow for a final statement on how well these guidelines were implemented.Second, the content of these guidelines has not been analyzed.
In addition, additional variables that might influence the results, such as potential differences between the different linguistic and cultural areas of Switzerland, could be included in the analysis.Finally, to effectively evaluate the effect of public corporate governance guidelines, it would be interesting to make crossnational comparisons, as the results may differ in countries where the rule of law plays a less prominent role.

Figure 1 :
Figure 1: Number of holdings of all cantons in the different industries for 2007 and 2012.Source: own illustration

Figure 2 :
Figure 2: Comparison of the number of holdings of the cantons in the different industries between 2007 and 2012.Source: own illustration

Figure 1 :Figure 2 :
Figure 1: Number of holdings of all cantons in the different industries for 2007 and 2012 Figure 2: Comparison of the number of holdings of the cantons in the different industries between 2007 and 2012

Table 4 :
Use of control instruments by governments in cities with a population of more than 50,000 in 2014

Table 5 :
Level of perceived control by cantonal government representatives in 2014

Table 6 :
Level of perceived control by the governments in the cities with a population over 50,000 in 2014

Table 7 :
Change in the number of holdings of the cantonal governments from 2009 to 2012

Table 8 :
Change in the number of holdings of the cantonal governments in "other" industries from 2009 to 2012

Table 9 :
Shift in the proportion of portfolios from private-law to public-law institutions from 2009 to 2012

Table 10 :
Shift in the participation quota of the cantons between 2009 and 2012

Table 2 :
Calculation of the index indicating the existence of steering and controlling instruments and of risk management

Table 3 :
Use of control instruments by cantonal governments in 2014

Table 4 :
Use of control instruments by governments in cities with a population of more than 50,000 in 2014

Table 5 :
Level of perceived control by cantonal government representatives in 2014

Table 6 :
Level of perceived control by the governments in the cities with a population over 50,000 in 2014

Table 7 :
Change in the number of holdings of the cantonal governments from 2009 to 2012

Table 8 :
Change in the number of holdings of the cantonal governments in "other" industries from 2009 to 2012

Table 9 :
Shift in the proportions of the portfolios from private-law to public-law institutions from 2009 to 2012

Table 10 :
Shift in the participation quota of the cantons between 2009 and 2012