The good and the bad of family firms: when minority shareholder protection becomes relevant for firm performance Open Access

A substantial share of listed companies worldwide is controlled by families, including firms such as Samsung in South Korea, BMW in Germany, Roche in Switzerland and Walmart in the USA. These so-called “family firms” are often associated with superior firm performance (e.g. Anderson and Reeb, 2003; Duh et al., 2010). Most importantly, controlling families tend to monitor the management actively, thereby mitigating the principal–agent problem between shareholders and managers (Jensen and Meckling, 1976). For example, family shareholders may pressure the board of directors to replace the chief executive officer (CEO) in the event of value-destroying corporate actions. In this role as both shareholders and principals, families can influence corporate decisions, reduce agency costs and enhance firm performance. Such active monitoring may also benefit minority shareholders (Shleifer and Vishny, 1986; Anderson and Reeb, 2003; Miller and Le Breton-Miller, 2006; Barontini and Caprio, 2006; Villalonga and Amit, 2006; Bennedsen et al., 2007).

On the other hand, families can pose a risk to minority shareholders when their interests diverge, potentially leading to a principal–principal conflict. Specifically, families may extract both pecuniary private benefits of control, such as the private use of corporate assets, and nonpecuniary private benefits of control, such as mental satisfaction, social status or the prestige associated with controlling a corporation, all at the expense of minority shareholders (Hansmann, 1988; Shleifer and Vishny, 1997; La Porta et al., 1999; Claessens et al., 2000; Faccio and Lang, 2002; Denis and McConnell, 2003; Dyck and Zingales, 2004; Thomsen et al., 2006; and Anand et al., 2025, for a review).

Minority shareholders may be expropriated through various channels, including the exploitation of information advantages (Anderson and Reeb, 2003), earnings management (Salehi et al., 2020) and the appointment of unqualified family members to management positions or other forms of nepotism (Faccio et al., 2001; Schulze et al., 2001; Chen et al., 2021). Further mechanisms include related-party transactions that are not conducted at arm’s length (also known as “tunneling”) (Khanna and Rivkin, 2001; La Porta et al., 2002; Fan and Yu, 2022), and the pursuit of corporate strategies that primarily serve family interests. Such strategies may involve extraordinary dividends, excessive diversification, political donations to closely aligned parties or the financing of cultural events, arts or sports, patronage and philanthropic activities funded by corporate rather than personal resources (e.g. Miller et al., 2010).

The controlling families’ impact on minority shareholders and firm performance therefore depends on which agency problem predominates (Young et al., 2008). In a principal–principal conflict, corporate governance mechanisms that protect the interests of minority shareholders become essential (Schulze et al., 2001; Chrisman et al., 2018). A well-functioning corporate governance enhances trust in capital markets, facilitates access to capital for financing innovation and ultimately benefits the economy (Claessens and Yurtoglu, 2013).

Complementing this agency-based view, stewardship theory suggests that some controlling families act as long-term stewards, which can strengthen monitoring and support firm performance (Davis et al., 1997). At the same time, the socioemotional wealth perspective implies that families may prioritize the preservation of control, even at the expense of minority shareholders, for instance by adopting control-preserving governance provisions such as dual-class share structures, thereby making minority shareholder protection particularly relevant in listed firms (Gómez-Mejía et al., 2007; Heino et al., 2020; Anand et al., 2025; Bettinelli et al., 2025).

Building on the tension between principal–agent and principal–principal conflicts in family firms, minority shareholder protection becomes economically relevant because it constrains the extraction of private benefits at the expense of minority shareholders. Yet, empirical evidence on which specific minority shareholder protection mechanisms matter in listed family firms remains limited and partly inconclusive. Prior studies examine family control and firm performance (e.g. Anderson and Reeb, 2003; Villalonga and Amit, 2006; Barontini and Caprio, 2006) and also consider control-enhancing mechanisms such as dual-class structures (e.g. Bebchuk et al., 2000; Masulis et al., 2009; Villalonga and Amit, 2009). However, evidence on which concrete minority shareholder protection mechanisms are most relevant within listed family firms remains scarce, especially beyond the broad distinction between family and nonfamily firms. This paper addresses this gap by identifying key minority shareholder protection mechanisms that are relevant in Switzerland and by testing which of them are most strongly associated with firm valuation in family firms.

We investigate how minority shareholder protection mechanisms in listed family firms affect firm performance. This study is motivated by a controversial corporate governance case in Switzerland that challenged the widely held belief that family shareholders invariably act in the best interests of the firm (e.g. PwC, 2014). In 2014, the French firm Saint-Gobain attempted to take over Sika, a chemical industrial company. The case highlighted several corporate governance features that are critical for minority investors. These include the “one share, one vote” principle (i.e. the absence of dual-class shares with differential voting rights or voting restrictions), protection in the event of a change of control (i.e. no opting-out clause) and the presence of corporate directors who are independent of both management and the controlling family (i.e. strong board independence) (e.g. Anderson and Reeb, 2004; Andres, 2008; Villalonga and Amit, 2009; Caprio et al., 2011).

These provisions are analyzed using a sample of 3,229 firm-year observations from Swiss-listed companies. Our results indicate that dual-class structures in family firms are negatively associated with firm performance. However, the other minority shareholder protection mechanisms considered, such as opting-out clauses and board independence, do not exhibit a significant relationship with firm performance. To address potential endogeneity concerns, we use an instrumental variable (IV) approach. In addition, we conduct a series of robustness tests that control for various CEO, board and ownership characteristics, consider different time periods and estimate both random and fixed-effects models.

Our study contributes to corporate governance research in two main ways. First, while prior research shows that family control can have both positive and negative effects, we analyze how minority shareholder protection mechanisms within family firms relate to firm performance. In particular, we examine which of several concrete governance provisions – dual-class share structures, voting rights restrictions, opting-out clauses and board independence – is most strongly associated with firm valuation in the presence of a controlling family. Compared with prior studies that focus on individual governance mechanisms such as dual-class structures or board independence (e.g. Masulis et al., 2009; Gompers et al., 2010; Villalonga and Amit, 2009), we analyze multiple minority shareholder protection mechanisms simultaneously and focus specifically on their interaction with family control. Second, we refine the measurement of board independence by accounting for board members’ relationships with the controlling family and other large shareholders, thereby providing a more accurate assessment of the board’s role in balancing the interests of the controlling family and minority shareholders.

The institutional environment, including laws, norms, values and political structures, is widely considered an important determinant of ownership structures (e.g. La Porta et al., 1999; Black et al., 2012). Switzerland provides an excellent setting for this analysis. Approximately 42% of all Swiss-listed firms can be classified as family firms ^[1]. According to Djankov et al. (2008), Switzerland’s legal environment is “extremely friendly to insiders and hostile to outside shareholders,” which makes the role of minority shareholder protection mechanisms particularly relevant.

Minority shareholder protection in Switzerland is primarily defined in the firms’ articles of incorporation, which allows considerable flexibility in the design of these protections. Provisions that are relevant for minority shareholders include dual-class share structures, voting rights restrictions and opting-out clauses. This setting allows us to examine whether these governance mechanisms are associated with firm valuation in the presence of a controlling family. Removing these protections would require the approval of the controlling shareholders. In contrast to the USA, governance mechanisms such as poison pills, sales of crown jewels, golden parachutes or staggered boards are prohibited under Swiss law (e.g. Daines and Klausner, 2001; Bebchuk et al., 2009). The study is also relevant in light of ongoing debates about dual-class structures in capital markets.

Corporate governance became a highly debated topic in Switzerland following several high-profile corporate scandals in the 2000s, including ABB, Swissair and cases of excessive executive compensation. These events contributed to the approval of a corporate governance-related popular initiative “against rip-off salaries” in 2013, which granted shareholders more legal rights.

Beyond the legal environment, culture has also been recognized as an important determinant of corporate governance practices (e.g. Licht, 2001). Switzerland provides an interesting setting in this regard, as it is a culturally diverse country with multiple languages and religions, which may shape its corporate governance practices (Volonté, 2015).

The Swiss stock market includes firms that differ considerably in terms of size, industry affiliation, age and ownership structure, which makes it a useful setting for our analysis. The paper is organized as follows: Section 2 presents the Sika case and develops the hypothesis. Section 3 describes the data and variables used in the study. Section 4 presents the empirical analysis and Section 5 concludes.

The fundamental purpose of corporate governance is to ensure that financial resources are used effectively to sustain and create corporate value. At the same time, the interests of other shareholders and stakeholders must be carefully considered. This issue is particularly salient in family firms, given the prevalence of family control and its economic importance in the corporate sector worldwide (Faccio and Lang, 2002; Claessens et al., 2000). In contrast to firms with dispersed ownership, where small shareholders face high monitoring and coordination costs that limit their voting behavior and their ability to act collectively (Olson, 1971), family shareholders are typically large, long-term owners with strong incentives to actively monitor management and influence corporate governance (e.g. Burkart et al., 2003).

However, restrictions on share sales, for example, through a family trust, may limit exit options and negatively impact firm performance (Fan and Leung, 2020). In addition, controlling families may extract private benefits of control at the expense of minority shareholders. Given the power of families, minority shareholder protection is crucial to mitigating agency costs that arise from potential conflicts of interest between controlling families and minority shareholders. Therefore, while family control may alleviate the principal–agent problem, it may increase the principal–principal problem.

Thomsen et al. (2006), as well as Claessens et al. (2002), demonstrate that the mismatch between powerful controlling shareholders on the one hand and weak minority shareholder rights on the other is prevalent in Continental Europe and Asia, respectively. Bebchuk et al. (2000) and Masulis et al. (2009) further show that agency costs increase as the divergence between voting rights and cash flow rights widens, such as in situations where the “one share, one vote” principle is violated. Moreover, various empirical studies provide evidence that stronger shareholder rights are positively associated with firm performance (e.g. Gompers et al., 2003; Bebchuk et al., 2009; Cremers and Ferrell, 2014).

In addition, independent directors can help mitigate agency costs (e.g. Nguyen and Nielsen, 2010). However, the active role of controlling families in governance often becomes apparent through the appointment of family representatives to the board of directors, which may reduce the board’s independence from controlling shareholders. Prior evidence suggests that family board involvement can be beneficial when it is proportional to the family’s ownership stake (Jaskiewicz et al., 2024), underscoring the importance of aligning control and incentives. Nevertheless, the value of independent directors may be particularly valuable in family firms (Anderson and Reeb, 2004). Dahya et al. (2008) show that independent directors help reduce the value losses associated with the dominant shareholder’s potential to extract private benefits and harm firm wealth. Hence, independent directors may play a crucial role in family firms by balancing the interests of controlling families and minority shareholders.

The distinctive features of corporate governance in Switzerland are well illustrated by a prominent and recent case. Many commentators have cited this case as a prime example of shortcomings in corporate governance and the weak protection of minority shareholder rights (Finanz und Wirtschaft, 2016).

Sika, a chemical manufacturing firm, was founded in 1910 by Kaspar Winkler. Since 1930, the Burkard family has exercised indirect control over the company through Schenker–Winkler holding (SWH). Sika went public in 1968. On the evening of Friday, December 5, 2014, the Burkard family decided (without informing the company’s board of directors) to sell SWH, and thereby their controlling stake of Sika, to the French competitor Saint-Gobain for CHF 2.75bn. The corporate governance structure at the time was characterized by the following key features:

• The firm operated under a dual-class share structure, consisting of unlisted registered shares with one voting right per CHF 1.5 of nominal value and listed bearer shares with one voting right per CHF 9 of nominal value. This structure allowed the controlling family to hold 53% of the voting rights (through the unlisted shares in SWH, which they intended to sell) while owning only 16% of the cash-flow rights. Under the terms of the transaction, the family was set to receive a premium of around 80%, while minority shareholders were not entitled to any compensation upon the change of control (see point 3). Within four days of the announcement, the market value of the listed shares declined by 28%, resulting in substantial losses for minority shareholders. Several additional corporate governance issues further complicated the deal.

• The articles of incorporation also included voting rights restrictions (“Vinkulierung”), which allowed the board of directors to cap the voting rights exercised by any shareholder at 5%. The restriction was introduced to limit the influence of outside shareholders. At one point, when the family’s shares were publicly listed, the family held only 42.6% of the voting rights. Although the family did not sell Sika shares directly to Saint-Gobain in the 2014 transaction, the courts later classified the sale of control via SWH as a circumvention of these rules and upheld the limitation of voting rights to 5%.

• Since January 1, 1998, Swiss law has required an acquirer to make a public tender offer upon obtaining control of a company. However, Sika’s articles of incorporation included an opting-out clause that exempted the acquirer from making an offer to all shareholders, as otherwise mandated by Swiss takeover law ^[2]. As a result, Saint-Gobain was able to obtain control without extending an offer to all shareholders at a higher price. Furthermore, because no mandatory bid was required, the controlling family was able to receive a control premium, a practice that has been prohibited in Switzerland since 2013 in the context of public tender offers.

• The board consisted of six independent directors and three family representatives. Because independent directors held a majority, the board was able to restrict the family’s voting rights (according to the articles of incorporation) to prevent changes to the board’s composition. Despite this, the family then attempted unsuccessfully to appoint new board members at the subsequent general meetings.

The Sika case highlights that minority shareholder protection in family-controlled firms depends on specific governance provisions embedded in the articles of incorporation, and on the effective independence of the board of directors. Paradoxically, by introducing voting rights restrictions at an earlier stage, the controlling family ultimately undermined its own interests. More broadly, this case demonstrates how corporate insiders can install governance provisions that weaken minority shareholder protection and hinder hostile takeovers. The interaction of these various elements of Swiss corporate governance in the Sika case has therefore been described as an «explosive mixture» (by the proxy advisor Ethos; SonntagsZeitung, 2014) or «[The] Burkard-Schenker-Code» (Tages Anzeiger, 2015).

On October 28, 2016, the Cantonal Court of Zug, in its first-instance ruling, confirmed the lawfulness of the voting rights restriction imposed by Sika’s board of directors on the Burkard family. As a result, the family would have been unable to change the composition of the board of directors through a shareholders’ resolution at the general meeting. However, on May 11, 2018, Sika, Saint-Gobain and the Burkard Family announced a settlement agreement. Under this agreement, the Burkard family agreed to sell control to Saint-Gobain, while Sika committed to repurchasing part of Saint-Gobain’s stake to preserve its independence. As a result, Saint-Gobain became a shareholder holding 10.75% of the voting rights. This agreement precluded a potential final ruling by the Swiss Federal Supreme Court (Bundesgericht).

The Sika case comprehensively demonstrates the relevance and the interaction of key minority shareholder protection mechanisms in Switzerland. Building on this case, we focus on the following mechanisms: ^[3]

• Dual-class share structures. Firms may issue different classes of shares, such as super-voting or nonvoting shares. Super-voting shares typically grant one vote per share but have a lower nominal value, thereby conferring greater voting power relative to their cash-flow rights. Under Swiss law, the maximum permissible ratio is 1:10. Furthermore, companies can issue nonvoting shares, such as participation certificates, which provide full economic rights but no voting rights. In both cases, voting rights are separated from cash-flow rights. A prominent example is Richemont, a luxury goods holding company with two share classes: listed “A” and unlisted “B” shares. Because the par value of “B” shares is ten times lower than that of “A” shares, “Compagnie Financière Rupert,” which holds all “B” shares, controls 50% of the voting rights while owning only 9.1% of the cash-flow rights. As a result, minority shareholders possess disproportionately less voting power relative to their economic stake. US companies such as Google, LinkedIn, Groupon and Facebook have adopted similar dual-class share structures. Gompers et al. (2010) characterize such arrangements as the most “extreme example of antitakeover protection.”

• Voting rights restrictions ^[4]. Many firms’ articles of incorporation limit shareholders’ voting rights to a certain threshold, often around 3%. As a result, even a shareholder holding a 20% ownership stake may exercise only 3% of the voting rights at the general meeting, making it nearly impossible to initiate corporate changes. These restrictions (referred to as “Vinkulierung”) have traditionally been used to protect companies from foreign takeovers. Initially, no justification was required to restrict a shareholder’s voting rights. However, following the 1991 revision of Swiss company law, only percentage-based voting rights restrictions are permitted. For example, the British hedge fund Laxey was forced to abandon its takeover attempt of Implenia, a Swiss construction firm, after the company’s board registered only 4.8% of Laxey’s voting rights despite the fund holding up to 38% of the shares. Interestingly, even minority shareholders opposed Laxey’s attempt to lift the voting restriction at an extraordinary general meeting. Similarly, voting rights limitations at Georg Fischer, an industrial group, prevented the private investor Giorgio Behr from increasing his voting stake.

• Opting-out clauses. In 1998, the Swiss Stock Exchange Act (SESTA) introduced a mandatory public tender offer requirement in the event of a change of control, triggered when a shareholding exceeds 33% of the voting rights ^[5]. However, the law also allows companies to opt out of this requirement or to opt up the threshold to 49%, thereby preventing minority shareholders from tendering their shares. In addition, the 2013 revision of SESTA abolished control premiums in public tender offers by requiring the acquiring party to extend an offer to all shareholders under identical conditions. The reform was largely motivated by the Quadrant takeover case. Importantly, when a company has adopted an opting-out clause, takeover regulations no longer apply, rendering the prohibition of control premiums ineffective.

• Board independence. Board independence is among the most extensively studied aspects of corporate governance (e.g. Dalton et al., 1998; Nguyen and Nielsen, 2010; Johnson et al., 2013; Knyazeva et al., 2013; Masulis and Zhang, 2019). The Swiss Code of Best Practice defines directors as independent if they are neither current nor recent former executives (within three years of leaving office) and have no significant business relationships with the company. By this definition, all of Sika’s directors would have qualified as independent. However, in practice, three directors maintained close ties to the controlling family, highlighting the importance of assessing directors’ relationships with major shareholders rather than relying solely on formal independence criteria (Ansari et al., 2014).

Family firms are often seen as a solution to the principal–agent problem, as controlling shareholders have strong incentives to actively monitor management and align managerial actions with the company’s long-term interests (e.g. Anderson and Reeb, 2003; Duh et al., 2010; Combs et al., 2023). Conversely, conflicts between controlling families and minority shareholders may give rise to a principal–principal problem. The risk that controlling families extract private benefits or expropriate minority shareholders is mitigated by stronger minority shareholder protections (Hansmann, 1988; Shleifer and Vishny, 1997; La Porta et al., 1999; Claessens et al., 2000; Faccio and Lang, 2002; Denis and McConnell, 2003; Dyck and Zingales, 2004; Thomsen et al., 2006). In environments with weak minority shareholder protection, family firms may engage in practices such as nepotism, tunneling and strategic decision-making. They may also resist governance reforms, resulting in organizational inefficiencies. Key corporate governance mechanisms that influence these risks include dual-class share structures, which grant controlling families disproportionate voting power (Masulis et al., 2009; Villalonga and Amit, 2009), voting rights restrictions that limit the influence of external shareholders (Claessens et al., 2000) and the degree of board independence (Anderson and Reeb, 2004; Villalonga and Amit, 2009). Moreover, as demonstrated in the Sika case, Switzerland’s opting-out clause further undermines the protection of minority shareholders.

In family firms, dual-class share structures are expected to weaken minority shareholder protection because they separate voting rights from cash-flow rights and allow controlling families to preserve control with limited economic exposure. Voting rights restrictions may further reduce the ability of outside shareholders to influence corporate decisions. Opting-out clauses weaken protection in takeover situations by allowing controlling shareholders to transfer control without extending equivalent terms to minority shareholders. Finally, effective board independence may mitigate conflicts between controlling families and minority shareholders by strengthening oversight and balancing interests. Stronger corporate governance – such as eliminating dual-class share structures, enforcing “one share, one vote” principle, removing voting rights restrictions and opting-out clauses and ensuring effective board independence – can better align decision-making in family firms with shareholder interests and, ultimately, enhance firm value. As a result, minority investor protection plays a crucial role. Based on this reasoning, we propose the following hypothesis:

We collect data on listed companies in Switzerland (excluding real estate and investment firms) from 1998 to 2023. After excluding observations with missing data, our final sample consists of 3,229 firm-year observations for the 2005–2023 period. The complete data set for the full 1998–2023 period (without board independence data) comprises 4,444 firm-year observations. Ownership data for 13,500 shareholders holding at least 3% of the voting rights were manually collected from the SIX Exchange database and firms’ annual reports. Corporate governance data from 1998 to 2015 were hand-collected from annual reports, while data from 2016 to 2023 were obtained from Inrate Ltd., a Swiss sustainability rating agency. Financial data were sourced from Thomson Reuters Datastream.

We define Family firms as firms controlled by families or individuals holding at least 20% of the voting rights. This threshold is motivated by two considerations. First, a 20% ownership stake is generally considered sufficient to exert effective control. Second, it is often challenging to distinguish between family and individual ownership in practice. Furthermore, this 20% threshold has been widely used in prior studies to identify family-controlled firms (La Porta et al., 1999; Faccio and Lang, 2002; Villalonga and Amit, 2006; Isakov and Weisskopf, 2014).

The variable Dual class is assigned a value of 1 if a firm has issued two or more classes of equity and 0 otherwise. Regardless of the ownership stake, the presence of dual-class share structures reduces the voting rights of certain shareholders (e.g. La Porta et al., 1999; Bebchuk et al., 2000; Cronqvist and Nilsson, 2003). The variable Voting rights restrictions equals 1 if the board of directors is authorized to restrict the voting rights of shareholders once their ownership exceeds a certain threshold, typically 3%. The variable Opting-out is assigned a value of 1 if the firm has opted out of, or opted up (to a threshold of 49%), the public tender offer requirement under Swiss takeover law, which otherwise mandates that shareholders exceeding 33% of the voting rights must make a public offer to all remaining shareholders.

As board independence is a key corporate governance mechanism, we define Board independence as the proportion of board members who are neither current nor former executives of the company and who have no material business relationships with the firm. Moreover, independent directors must not be shareholders themselves nor representatives of shareholders; specifically, they must not be affiliated with a family or individual shareholder holding more than 3% of the voting rights. Crespí-Cladera and Pascual-Fuster (2014) emphasize the importance of not relying solely on firms’ self-declarations regarding board independence. Please note that, for the period starting in 2016, we rely on the independence classifications provided by Inrate Ltd. This change in data source may affect the measured level of board independence. However, as the empirical specifications include year fixed effects, any systematic level shifts are absorbed by time effects.

We use Tobin’s Q as a proxy for firm performance. Tobin’s Q is calculated as the sum of total assets and the market value of equity, minus book equity, divided by total assets. This measure serves as an approximation of the firm’s replacement value (e.g. Agrawal and Knoeber, 1996).

Furthermore, both corporate governance and firm performance are influenced by firm-specific characteristics. To mitigate potential omitted variable bias, we include several control variables commonly used in the literature (e.g. Bebchuk et al., 2009; Knyazeva et al., 2013). Size is defined as the natural logarithm of total assets and serves as our measure of firm size. Diversification equals 1 if the company reports more than one significant business segment and 0 otherwise. Sales growth is calculated as the median annual sales growth over four periods. Firm age is measured as the natural logarithm of the number of years since the firm’s founding (Hülsbeck et al., 2019; Chirico and Kellermanns, 2024).

Profitability is defined as the ratio of earnings before interest, tax, depreciation, and amortization to total assets. Liquidity is calculated as the ratio of cash holdings to total assets. Investments capture capital expenditures divided by total assets. Tangibility is defined as the ratio of property, plant and equipment to total assets. R&D is measured as research and development expenditures relative to total assets. Leverage is calculated as total debt divided by total assets. Furthermore, we include industry-fixed effects to control for time-invariant industry characteristics such as regulation, competition or growth opportunities, and time-fixed effects to account for time trends, such as recessions and economic expansions.

The definitions of variables are summarized in Table 1, and the descriptive statistics are presented in Table 2. Tobin’s Q, sales growth, profitability and leverage are winsorized at the 5% level.

As Table 2 shows 42% of all listed firms in Switzerland are classified as family-controlled. Dual-class share structures are present in 15% of companies, 26% have adopted an opting-out clause and 27% impose voting rights restrictions. On average, independent directors account for 50% of board membership. The table further reveals significant differences in corporate governance between family firms and nonfamily firms. Dual-class structures and opting-out clauses are more prevalent among family firms, while voting rights restrictions and board independence tend to be lower on average. Voting rights restrictions are less relevant in the presence of a dominant shareholder, and board independence reflects a lower proportion of independent directors in family-controlled firms.

Figure 1 illustrates that family firms remain a stable component of the Swiss corporate landscape, with their proportion even increasing in recent years. Regarding minority shareholder protection, the proportion of firms using multiple share classes has decreased since 1998. The increase in reported director independence reflects the use of new data from Inrate Ltd. While these independence classifications are comparable to earlier measures, they are not identical and may be subject to periodic redefinitions. However, as we incorporate year fixed effects in the regressions, we believe that these deviations are adequately controlled for. Overall, the elements related to minority shareholder protection demonstrated in the Sika case appear to be widespread among listed firms in Switzerland.

The correlation matrix presented in Table 3 indicates that multicollinearity is not a concern. No pairwise correlation exceeds 0.7, a conclusion further supported by the variance inflation factor values, all of which are below 2.

We evaluate the effect of minority shareholder protection on firm performance in the presence of controlling family shareholders.

Our baseline model to test these effects is specified as a linear regression model:

where the indices i and t correspond to firm and year.

Consistent with prior studies, we use pooled ordinary least squares (OLS) methodology, including year and industry-fixed effects and cluster-robust standard errors. Because our main variables of interest exhibit limited within-firm variation, firm fixed-effects regression models would yield meaningless results (e.g. Knyazeva et al., 2013; Huang et al., 2015). To address endogeneity concerns, we use an IV approach and conduct a series of robustness tests to validate the results.

The results reported in Table 4 show that family firms are negatively associated with firm performance (Columns II, IV and V). However, this negative relationship is primarily driven by family firms with dual-class share structures (Columns III, VI, VII and VIII). These findings are based on the broader sample of 4,444 firm-year observations, which excludes board independence. The results suggest that capital markets penalize family control when it is achieved by separating voting rights from cash-flow rights. By contrast, voting rights restrictions and opting-out clauses, which exempt controlling shareholders from the obligation to make a mandatory takeover offer, do not show a significant relationship with firm performance.

In addition to the preemptive provisions embedded in firms’ articles of incorporation, we next account for board independence using a sample of 3,229 firm-year observations from the 2005–2023 period. The results reported in Table 5 indicate that board independence, as a stand-alone variable, is positively related to Tobin’s Q (Column I). However, we find no evidence that board independence moderates firm performance within family firms (Column II). In contrast, the negative relationship between dual-class structures in family firms and firm performance remains robust. The effect is also economically meaningful. Based on the pooled OLS estimates in Table 5, the interaction coefficient implies a valuation discount of roughly 30% relative to the sample mean of Tobin’s Q. The absence of a robust association for voting rights restrictions, opting-out clauses and board independence should not necessarily be interpreted as evidence that these mechanisms are irrelevant. Rather, these findings may indicate that their effect on firm value depends on the context and may become particularly relevant in specific control contests or takeover situations rather than in average firm-year observations. In the case of board independence, the null result may also reflect measurement issues, as formal independence may overstate actual independence when directors are ultimately selected and elected by the controlling family. Another explanation is that board structure may reflect an equilibrium outcome rather than an exogenous source of variation in firm value (Hermalin and Weisbach, 2003). More generally, the null results may also reflect institutional specificities of the Swiss setting or offsetting effects across firms. Overall, the evidence is consistent with our hypothesis. Among the minority shareholder protection provisions considered, adherence to the “one share, one vote” principle – that is, the absence of dual-class share structures that separate voting rights from cash-flow rights – emerges as the key mechanism associated with higher firm valuation in family firms. By contrast, the other mechanisms examined do not show a robust relationship with firm performance.

To address potential endogeneity concerns, we estimate simultaneous equations for Tobin’s Q, family firm status, dual-class share structures and their interaction using three-stage least squares (3SLS) regressions (Coles et al., 2008).

The Family firm variable is instrumented using the 1990 Divorce rate across Swiss regions (cantons), as reported by Robert-Nicoud (2014). We expect a higher divorce rate to be associated with a lower prevalence of family firms, as lower divorce rates may indicate stronger family cohesion, which we consider as an essential factor for the existence of family firms.

The Dual class variable is instrumented using the Political voting participation rate, measured by voter turnout in the 1983 National Council election, the earliest year for which complete canton-level data are available from the Federal Statistical Office. We expect that cantons with higher electoral participation exhibit stronger norms of shareholder democracy and, consequently, a lower prevalence of firms with dual-class share structures.

Finally, the interaction term Family firm × Dual class is instrumented using the proportion of People without religious affiliation per canton in 2010, the earliest year for which such data are available from the Federal Statistical Office. We interpret religious affiliation with one of the official churches in Switzerland as a proxy for cultural conservatism, which may foster a preference for stability – as it is also achieved by combining family ownership with dual-class share structures. This assumption aligns with Xie and Yuan (2025), who point out that regions with stronger family-oriented cultures tend to exhibit a higher prevalence of family firms.

These canton-level variables are historical and largely predetermined relative to current firm valuations. We therefore use them as proxies for social, cultural and political environments that are associated with family control, dual-class structures and their interaction. While this historical structure helps mitigate reverse-causality concerns, the exclusion restriction remains an identifying assumption. To assess instrument relevance, we report the excluded-instrument first-stage F-statistics in the text accompanying Table 6: F  =  21.6 (p < 0.001) for the divorce rate, F  =  57.4 (p < 0.001) for political voting participation and F  =  8.2 (p  =  0.004) for people without religious affiliation. Because the system is exactly identified, with three instruments for three endogenous variables, overidentification tests such as the Sargan–Hansen test are not available. We therefore interpret the 3SLS estimates as complementary evidence rather than as the sole basis for inference. The estimates are consistent with our main finding that family firms with dual-class share structures are associated with lower firm valuation.

Dual-class equity structures have experienced a resurgence among technology companies in the USA, alongside the development of new corporate governance practices and regulations over the past 25 years. Furthermore, Crespí and Martín‐Oliver (2015) suggest that family firms may provide greater stability in times of crisis, while Hansen et al. (2020) find that this effect varies across countries depending on institutional factors and governance systems.

We therefore examine the effects across five distinct time periods. The 1998–2002 period predates the introduction of the SIX Swiss Exchange Corporate Governance Directive and the Swiss Code of Best Practice for Corporate Governance, as well as major corporate scandals in Switzerland and abroad. The period between 2008 and 2012 is characterized by the global financial crisis and the subsequent awareness of corporate governance. Finally, the years from 2020 to 2023 were marked by the COVID-19 pandemic.

The results reported in Table 7 show that financial investors’ perceptions of dual-class share structures in family firms have changed significantly over the 25-year period under review. In particular, dual-class family firms have been increasingly perceived negatively over time, with the magnitude of the interaction effect nearly doubling from the 1998–2002 period to 2020–2023. These findings suggest that investors’ concerns about dual-class structures in family-controlled companies have intensified over time.

As mentioned earlier, due to the quasi-time-invariant characteristics of our main explanatory variables, our primary analysis relies on pooled OLS regressions. However, for completeness, we also estimate both random-effects models and fixed-effects models. The results reported in Table 8 show that the negative relationship between dual-class family firms and Tobin’s Q remains robust across these model specifications.

We conduct several additional robustness checks, which we do not report here, none of which alter our main finding regarding the negative effect of dual-class family firms on firm performance. First, we account for family members in management positions, including representation on the board of directors and in the CEO role (Isakov and Weisskopf, 2014; Åberg et al., 2024). Second, we include additional controls for board characteristics, such as gender diversity. We also estimate models using Tobin’s Q without winsorization and with winsorization at the 10% level as dependent variables, and we redefine family firms using varying voting rights thresholds of 10%, 25% and 50%.

The results remain qualitatively consistent. However, the magnitude and statistical significance of the negative association of dual-class family firms and firm performance diminish as the voting-rights threshold increases. In other words, dual-class structures are perceived more negatively when a family holds 10% of the voting rights than when it holds 50% of the voting rights, with the effects for 20% and 25% falling in between.

Finally, we analyze the stock market reactions of family firms with different minority shareholder protection provisions to the announcement of improved shareholder rights. The study period provides the opportunity to investigate this effect. In March 2013, Swiss voters approved the popular initiative “against rip-off salaries,” which substantially strengthened shareholder rights. For example, the initiative introduced binding say-on-pay votes and the annual individual election of members to the board of directors.

To assess the impact of the new regulatory framework, we follow Wagner and Wenk (2019), who identify four key events to analyze stock market reactions. The first event is the announcement that enough voters have signed the initiative to trigger a constitutional referendum (Event 1, February 26, 2008). The second event is the acceptance of the constitutional amendment by Swiss voters (Event 2, March 03, 2013). The third event is the publication of the draft legislation (Event 3, June 14, 2013) and the fourth event is the publication of the final Ordinance against Excessive Compensation (Event 4, November 20, 2013).

To calculate cumulative abnormal returns (CARs), we follow Wagner and Wenk (2019) and use the event-study methodology from Kothari and Warner (2007) and MacKinlay (1997). We pool the four events and calculate CARs as the sum of abnormal returns (ARs) over a three-day event window. ARs are estimated using the market model, with the Swiss Performance Index as the reference. The estimation window spans 250 trading days and ends two days prior to the event date.

Given the improvement of shareholder rights resulting from the initiative, we expect a positive stock market reaction for firms with weaker minority shareholder protection, such as dual-class family firms. Even though minority shareholders may not always have a decisive influence on voting outcomes, signals of dissatisfaction at general meetings (such as the percentage of “No” votes by minority shareholders) and the public scrutiny of poor corporate governance could put pressure on family-controlled firms to adopt better governance practices. However, this process of change may be slowed down or turned impossible by restrictions on voting rights.

As reported in Table 9, the stock prices of dual-class family firms react significantly positively to the announcements of improved shareholder rights. The expansion of minority shareholders’ rights at general meetings may increase the visibility of their opinions, thereby putting pressure on poor corporate governance practices and potentially leading to gradual improvements. By contrast, companies with voting rights restrictions experience negative stock price reactions. A possible explanation is that voting ceilings (such as limits capping voting rights to, e.g. 3%) restrict the voting power of minority shareholders. This dispersed ownership makes it more difficult to drive improvements in corporate governance at general meetings.

Family shareholders are often viewed as having a potentially positive influence on firm performance. On the one hand, as large shareholders who have invested a substantial share of their financial wealth in a single company, they have strong incentives to monitor management closely, thereby mitigating the principal–agent problem. On the other hand, controlling families may also be in a position to extract private benefits of control, resulting in poor corporate decisions and weaker performance, thereby creating a principal–principal problem that harms minority shareholders. In addition, family shareholders may perceive themselves as only weakly accountable to minority shareholders of “their company.” These adverse effects are likely to be particularly pronounced in firms with weak minority shareholder protection.

Our study shows that dual-class family firms, where minority shareholders hold disproportionately fewer voting rights relative to their cash-flow rights, are negatively associated with firm performance. Moreover, the magnitude of this effect has increased over the 25-year study period. Other minority shareholder protection mechanisms or the absence thereof (such as voting rights restrictions, opting-out clauses and board independence), do not exhibit a consistently significant relationship with firm performance. These findings suggest that a misalignment between control rights and economic ownership distorts incentives and ultimately leads to lower firm performance.

Understanding how different corporate governance arrangements in family firms affect firm performance is essential for both listed and unlisted companies. Family-controlled firms represent one of the most common business structures worldwide. Effective governance fosters trust among capital market participants, thereby facilitating access to capital that can be used to drive innovation, create employment and support economic growth (Claessens and Yurtoglu, 2013). Conversely, poor corporate governance structures introduce risks that may lead to resource misallocation, value destruction and economic disruptions. To mitigate these risks, family firms should avoid provisions that disadvantage minority shareholders, while investors must carefully assess corporate governance structures prior to investing, because deficiencies may become evident only after conflicts arise, as illustrated by the Sika case.

Given the strong and robust evidence of the negative impact of dual-class share structures, our results indicate that the “one share, one vote” principle remains a crucial issue in corporate governance in our setting. For controlling families and boards of directors, these findings highlight the importance of avoiding dual-class share structures. For investors, the presence of dual-class shares represents a critical governance feature that should be scrutinized when investing in family firms. In the Swiss context, the findings are also consistent with governance arrangements that strengthen shareholder democracy and limit the separation of voting rights from cash-flow rights.

Nevertheless, several limitations should be acknowledged. As with any empirical study, the results always depend on the sample used and may vary across specific industries, company characteristics (e.g. firm size), time periods or jurisdictions. Because this study focuses on Switzerland, the findings may not be fully generalizable to countries with different institutional frameworks.

Future studies could explore how different legal systems influence the effects of minority shareholder protection on family firm performance. In addition, as suggested by Dyer (2018), future research could benefit from collecting more granular data on firm-specific characteristics.