2. How Do Family and Control Structures Affect Group Affiliated Firms’
2.2 Theory And Hypotheses
國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
10
The results confirm our conjecture that excess control rights and controlling shareholdings are for the ambivalent needs of control and growth and then improve group affiliated firms’ performance. In addition, family firms with such control structures will perform better in non-electronic industry while worse in electronic industry. Lastly, active control, namely family members control management and chairman at the same time, will perform better in electronic industry while worse in non-electronic industry.
2.2 T
HEORYA
NDH
YPOTHESES2.2.1 Control structures for growth
Until 2000, Taiwan’s manufacturing sectors has experienced with a consequence/seires of industrial reform. During this period, business groups (qiye jituan) favored by deregulation and varieties of subsidies from government emerge (Chu & Amsden, 2003).
Initially (Since 1950s), Taiwan’s government proposed import substitution strategy and imposed protective tariffs on imports to promote local upstream production (Chu, 1994).
However, the domestic market is too small to develop the industries where the economies of scale are optimal for large firms. In this phase, most industries promoted are capital intensive industries (e.g. petrochemical industry). Therefore, for survival, the enterprises in these industries started to use vertical integration or diversification through building affiliated firms for reducing production costs. Later, after finding the limitation of import substitution strategy, Taiwan’s policy maker implemented export-led growth strategy as a supplementary strategy for import substitution strategy. In addition, Taiwan government also decreased the exchange rate on purpose to increase domestic firms’ comparative
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
11
advantages in international trade. Thereby the entrepreneurs persist for scaling up via forming business groups captured the growth opportunities created by the government.
The second wave of economic growth in Taiwan, which was driven by electronic industry, took off from 1970s. During this phase, although the strategy of “from import substitution to export-led growth” remained, Taiwan government used different policy instruments to promote this industry; by means of systematic import and assimilation of foreign advanced technologies, Industrial technology Research Institute of Taiwan (ITRI), a government subordinate research institute, facilitated the development of electronic industry in Taiwan. ITRI not only incubated and establish spin-off companies, like TSMC, but also developed and transferred advanced technological to local enterprises. Therefore, Taiwan’s enterprise which took advantage of technological knowledge of new products provided by ITRI, involved in related research and development, and invested in scaling up capture the growth opportunities arose from developed countries’ outsourcing policies.
Since 2000, the increasing demand for “one-stop shopping” from buyers of these countries impelled Taiwan’s contractors/OEMs to scale up further (Chu & Amsden, 2003).
During economic transformation, Taiwan’s entrepreneurs establish their own markets internally within business groups not only for the purposes of substituting for external markets, but also for control and growth. Business groups are sets of legally independent firms bound together by varying degrees of legal and social connection. These firms are mostly controlled by an agent (a family, a person, or a group) through a chain of ownership stakes. In emerging markets, the most common structure of business groups is
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
12
pyramid, in which the ultimate owners are located at the apex and often hold only a fractional ownership. Since such ownership structure allows the ultimate owners to raise funds without sacrificing their control rights, this gives them the incentives to scale up and capture the growth opportunities created by the government.
In Taiwan, the largest firms exhibit the most separation of ownership and control (Claesssens et al., JFE, 2000). In addition, Chung and Mahmood also found that affiliate ownership rose significantly while family ownership dropped sequentially from 1988 to 1998. These researches demonstrate that Taiwan’s entrepreneurs achieve the needs for growth and control simultaneously through the ownership structures which divorce ownership and control.
Therefore, we propose that when firms scale up with separation of ownership and control, they will perform better:
Hypothesis 1: the higher the level of divergence between ownership and control rights, the better the firms’ performance.
Another common way for ultimate controllers to increase their control rights is cross-shareholding. Although the pyramidal control structure can lead to the divergence between ownership and control in firms located at the lower layer of the pyramid, cross-shareholdings among group-affiliated firms within a business group can increase the level of separation of ownership and control in firms located at the top. Both kinds of control enhancing structures allows the ultimate owners handle “the ambivalent needs of growth and control” (Chung and Pasha ;Chang and Hong 2000; Chung 2004).
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
13
In terms of control, Cubbin and Leech (1983) defined “the critical control level” which is necessary to accumulate the majority of votes for control as the line between ownership control and management control. In diffusely held firms, management with “an absolutely large but relatively small bloc of shares” exercises its control rights. However, in business groups, ultimate owners with relatively large amounts of shares comparing to other shareholders control the firms. The critical control level is the threshold between these two kinds of control structure. Since the ultimate owners can just hold a relatively large amount of shares to be in a position to dominate a corporation, we regard the parts of the controlling ownership which exceed the critical control level as the parts for growth. Hence, we propose that:
Hypothesis 2: the more the ownership exceed the critical control level, the better the firms’ performance.
2.2.2 Agency Problems in Business groups
Although excess control rights or controlling shareholdings may be beneficial in some specific situation, they are also regarded as the sources of agency problems. From the agency theory perspective, giving ultimate controllers excess control rights or controlling shareholdings may cause serious conflicts of interest between controlling shareholders and non-controlling shareholders. Through dual-class share structure, pyramidal ownership structure or cross-shareholding, an individual or a family can financially or administratively control a set of legally independent firms with only a small fraction of equity (Bebchuk, Kraakman, & Triantis, 2000; Claessens et al., 2000). Since these
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
14
control-enhancing mechanisms allow the controlling shareholders to exercise control rights without commensurate shareholdings, they have the incentive and ability to pursue for private benefits at the costs of outside shareholders. And because of this so-called
“other people’s money” problems (Morck & Yeung, 2003), group-affiliated firms with high degree of separation between voting rights and cash flow rights may be subject to the expropriation of minority shareholders (Shleifer & Vishny, 1997).
Another agency problem in business groups is entrenched management problems.
Although the increase of ownership will raise large shareholders’ incentive to monitor CEOs’ behavior and mitigate agency costs (Agrawal & Knoeber, 1996; Shleifer & Vishny, 1986), the advantages comes from increasing ownership will be overwhelmed by the disadvantages result from excess control rights when inside shareholders’ controlling stakes exceeds over a threshold (Morck, Shleifer, & Vishny, 1988). In business groups, through a chain of ownership, ultimate controllers are entrenched in all group-affiliated firms. Moreover, when ultimate controllers also have sufficient controlling shareholdings to dominate a firm, faithless insiders are motivated to make special dividend policy or use specific trading relationship for their own benefits (Shleifer & Vishny, 1997), then, the problems of entrenched management occur (Morck et al., 1988; Morck & Yeung, 2003).
Bebchuk and Cohen (2005) also demonstrate that entrenched boards are associated with lower firm value.
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
15
2.2.3 The contingent effect of environmental uncertainty on the relationship between family control and control structures in Business groups
To reconcile the contradictory points of view discussed above, the contingent effects of environmental uncertainty are introduced to shed light on financial logics of different owner identities. A firm’s investment behavior which is strongly influenced by its owners’
financial logics will affect its performance and survival. Villalonga and Amit (2010) argue that families are more likely to preserve their control rights of a firm when the efficient scale is small. Gallo et al. (2000) also find that families are risk-averse and fear for losing control of their firms, which impede them to seize the opportunity for growth and development. However, the control structures of business groups allow family owners to handle the need for ambivalent needs of growth and control” (Chung and Pasha ;Chang and Hong 2000; Chung 2004). Therefore, family owners’ risk preference becomes the determinant of how they employ control structures through business groups.
Since families are risk-averse, we predict that when environmental uncertainty is low and families can preserve their control rights, they will utilize control structures to raise funds and scale up for growth. Thus we propose that:
Hypothesis 3: for group affiliated firms, family firms with higher excess control rights and excess voting rights will perform better when environmental uncertainty is low.
On the other hand, since families have the incentive and ability to pursue family interest at the costs of outside shareholders when the market for corporate control is absent or
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
16
weak (Schulze et al., 2001), environmental uncertainty will exacerbate this problem.
Grossman and Hart (1983) argue that the more inside controllers possess information which outside shareholder does not, the more difficult outside shareholders can check inside controllers’ decision. Since environmental uncertainty increase the information asymmetries between family controllers and outside investors, monitoring schemes turn into invalid. Furthermore, when environmental uncertainty is higher, family controllers become more risk averse and will leave investment opportunities out of consideration.
Thus we propose that environmental uncertainty will worsen the problems of underinvestment and expropriation of outside shareholders:
Hypothesis 4: for group affiliated firms, excess control rights and controlling shareholdings will increase family owners’ incentive and ability to exploit outside owners and then worsen firms’ performance when environmental uncertainty is high.
2.2.4 Family management and environmental uncertainty
The other way to director control a firm is to take over the position of management.
According to previous researches, the effect of family management comes with pros and cons simultaneously. Therefore, it is hard to shed light on the still obscure mechanism of family management on firm performance without further analysis (e.g. Villalonga & Amit, 2006). However, for family firms, family management also implies owner-manager, thus some implications from the upper-echelons theory are helpful to ascertain the contingencies which can disentangle the relationship between family management and
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
17
firm performance (Boyd, 1990; Boyd, 1995; Brickley, Coles, & Jarrell, 1997; Dalton, Hitt, Certo, & Dalton, 2007; Faleye, 2007; Finkelstein & Hambrick, 1990; Hambrick &
Finkelstein, 1987).
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
18
Environmental uncertainty is an essential contingency to understanding the benefits and disadvantages of unitary leadership structure (Brickley et al., 1997), which is the case when family members hold the positions of both CEO and board chairperson. Appointing an outside board chairperson may cause the agency costs of controlling the behavior of the board chairperson (Brickley et al., 1997); also, hiring outside CEO can increase the agency costs of controlling the CEO’s behavior. Furthermore, environmental uncertainty will increase the extent of information asymmetries between insider and outsider, thereby exacerbate agency problems. Unitary leadership structure in family firms not only gets rid of such kind of information asymmetries and agency problems but also has the
advantages of effective leadership and responsiveness which are critical to successful adaptation in uncertain environments (Bourgeois & Eisenhardt, 1988; Finkelstein &
Hambrick, 1990). In addition, since involvement-oriented management will lead to long-term effectiveness, empowering has significant advantages than control in unstable, uncertain environments (Davis et al., 1997). Miller, Le Breton-Miller, and Scholnick (2008) find that family involvement will promote long-term effectiveness and involvement-oriented management. Thus we speculate that family firms with unitary leadership structure will positively affect firm performance in high uncertainty environment.
Hypothesis 5: for group affiliated firms, family members holding the positions of both CEO and board chairperson will be positively related to firm
performance in high uncertainty environments.
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
19
By contrast, family management may impair firm performance in a stable environment.
From the perspective of upper-echelons theory, since the discretion of top managers is endowed by environmental uncertainty, top managers have lesser need for power in a stable environment (Finkelstein & Hambrick, 1990; Hambrick & Finkelstein, 1987).
Finkelstein and Hambrick (1990) exhibit that top-management-team tenure, which signals managerial entrenchment (Berger, Ofek, & Yermack, 1997), is associated with firm performance that deviated from industry averages in low-technology industries (e.g.
chemical industry and natural-gas distribution industry). In a related vein, giving CEOs more discretion in a stable environment may be harmful for firm performance.
Gomez-Mejia et al. (2001)demonstrate family firms are more vulnerable to managerial entrenchment problem by showing that family CEOs face lower risk of dismissal than their non-family counterparts not only in uncertain environments but also when firm performance is negative; meanwhile, powerful CEOs who can get rid of accountability will bring about higher agency costs in a stable environment (Davis, et. al., 1997).
Accordingly, we posit that family members holding the positions of both CEO and board chairperson may cause poorer firm performance when environmental uncertainty is low.
Hypothesis 6: for group affiliated firms, family members holding the positions of both CEO and board chairperson will be negatively related to firm
performance in low uncertainty environments.
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
20