The emergence of portfolio restructuring in Japan.
* We examine the role of stakeholders as an influence in the practice of portfolio restructuring. We contend that with the presence and creation of new legal arrangements and regulations, portfolio restructuring achieved widespread usage, but only when the initiatives were consistent with the interests of the most powerful social actors in a firm.
* Building on a stakeholder power approach to corporate governance, we examine whether the interests of relational banks, managers, and business groups were consistent with the practices of portfolio restructuring we observed in Japanese firms in the 1998 to 2005 period.
* Regressions using data on 174 Japanese machinery firms lend support to predictions that portfolio restructuring increases in frequency with the extent to which business groups' strategies are facilitated and decreases with the degree to which banks' interests are protected. The association between the frequency of portfolio restructuring and managerial interests depends on the level of managerial ownership.
Keywords: Portfolio restructuring * Divesture * Asset expansion * Corporate governance * Stakeholder influence * Japan
Corporate restructuring is an action that alters the structure, composition or orientation of a firm in response to changing external environments or altered internal organizational conditions (Bowman/Singh 1993, Kang/Shivdasani 1997). By realigning a firm's business and resources with its environment, corporate restructuring aims to improve returns from the deployment of the firm's assets.
A firm's managers can implement a variety of restructuring activities including portfolio, financial, and organizational restructuring (Gibbs 1993). In this study, we focus on an important subset of portfolio restructuring activities: namely, asset divesture and asset expansion restructuring. Different forms of restructuring interact with each other. For instance, portfolio restructuring can bring in organizational and financial restructuring, such as executive turnover, layoff, costly financial charges, and negative public perception (Hoskisson/Hitt 1994).
Even though restructuring activities can be prominent ones in a firm, a number of scholars have identified that weaknesses in ownership and governance systems can be key obstacles to restructuring (Bethel/Liebeskind 1993, Hanson/Song 2003). The governance literature emphasizes that with properly defined agency relationships, managers would implement a strategy that is in the best interest of shareholders. However, in economies where the agency relationships are not clearly defined, for example, Japan and Germany, the coordination of stakeholders becomes a major obstacle to the implementation of shareholder wealth maximizing strategies (Buck/Filatochev/Wright 1998). When the influences of stakeholders over corporate decisions become stronger relative to that of the shareholders, the initiation and implementation of corporate restructuring requires a complex cooperation among stakeholders. Principal-agent models hence become of less use to explain such behavior. Furthermore, the mechanisms of coordination are particularly important when firms are faced with radical environmental uncertainties or operational risks. If there is substantial uncertainty about the process and consequence of a strategic action, policymakers and managers can easily become hesitant to push for such a change in action.
Existing studies of portfolio restructuring have focused on the diversification behaviour of the firm, as pioneered by Wernerfelt and Montgomery (1988). Comment and Jarrell (1995) argue that a focusing strategy is consistent with the maximization of shareholder value, and Daley, Mehrotra, and Sivakumar (1997) find that a firm can increase its value through spinning-off unrelated lines of businesses. These studies focus on firms in the US, but literature on portfolio restructuring in Japan after the banking crisis in 1997 is scarce. Our purpose is not to complement the previously cited literature in evaluating the diversification strategy of Japanese firms, but instead to focus on portfolio restructuring in Japan from the perspective of both divesture activities and related-asset expansion activities.
Accordingly, we seek to explain the differing levels of portfolio restructuring activity across firms. We adopt a stakeholder power perspective on corporate governance and consider differences among stakeholders in their power to defend their interests relative to those of other stakeholders as a primary influence on the type and intensity of restructuring undertaken in these firms. We build on earlier work by providing a systematic justification for the observed choices of the most relevant stakeholders on portfolio restructuring and by offering a specific conceptual and empirical analysis of their interests and power. Specifically, we argue that with the presence and creation of an appropriate regulatory and legal infrastructure, as in the case of Japan since late 1990s, portfolio restructuring could not achieve widespread usage unless it was consistent with the interests of a firm's most relevant stakeholders. In terms of stakeholders, we examine whether the interests of relational banks, business groups and top managers were consistent with the practice of portfolio restructuring.
To address these questions, we utilized an empirical context in which there were changing institutional and regulatory conditions, and one in which shareholders were not clearly dominant in influencing the strategic direction of a firm. The Japanese economy after the collapse of the bubble economy in early 1990s and the banking crisis worsened in 1997 yielded this setting. Following the collapse of the economy, the increased awareness in the public and private sectors in Japan on restoring industrial profitability led to substantial changes in the regulatory and legal environment. At the same time, Japanese firms' stakeholders had developed more influence than owners as shareholders. Up until the onset of the 2000s, Japanese firms had been embedded in firm-bank relationships, and banks acted as important stakeholders to industrial firms. We capitalized on these characteristics of this setting, and utilized a sample of 174 Japanese machinery firms over the period 1998-2005 to test our hypotheses.
Background anal Hypothesis Development
When considering the action of portfolio restructuring, a firm's managers can choose to refocus on the core business of the firm via the divestment of unrelated businesses (downscoping), or choose to integrate operations through the selective acquisition or establishment of businesses related to the core business (asset expansion).
Portfolio restructuring may lead to a change in a society's impressions of the viability and sustainability of an organization (Davis/Diekmann/Tinsley 1994). Moreover, portfolio restructuring can be difficult: divestitures were described as "failed acquisition attempts" (Montgomery/Wilson 1986). This leads to the question concerning why we observe portfolio restructuring? Previous studies have found that the gains of portfolio restructuring come from the elimination of negative synergies (Hite/Owers/Rogers 1987). For example, assets unrelated to core operations may prevent a firm from focusing on developing and exploiting its core competencies. Portfolio restructuring can also reduce market inefficiencies or agency costs. For instance, managers typically are reluctant to raise capital through asset sales unless asymmetric information, debt overhang, or managerial discretion makes an equity offering too expensive (Lang/Poulsen/Stulz 1995). However, if the proceeds from asset sales are used to repay debt, be distributed as dividends, or buy back shares, agency costs are reduced (Hanson/Song 2003).
New institutional theorists contend that corporate restructuring emerged as a dominant strategy in the last three decades as a response to major changes in the business environment, such as the relaxation in the enforcement of antitrust legislation, changes in tax laws, innovations in external markets and changes in competition (Shleifer/Vishny 1990, Bowman/Singh 1993, Mitchell/Mulherin 1996). For example, changing regulatory environments and shifts in power and resources can transform corporations that were the embodiment of entrenched notions of the firm and its appropriate organizational form, as evidenced by the wholesale breakup of US business conglomerates through takeovers, leveraged buyouts, and investor pressure (Davis/Diekmann/Tinsley 1994).
If a portfolio restructuring increases the level of discrepancy between the actual organizational structure and the institutionally prescribed ideal, a firm can be confronted with significant risks, including the loss of legitimacy and reputation, which might also be met by a declining performance (Heugens/Schenk 2004). To avoid such an outcome, institutional theorists make the point that when a firm is subject to institutional pressures, it will adopt the predefined roles and routines that reflect the prescriptions conveyed by the wider, previously rationalized and legitimated institutional environment. Top managers anticipating these risks can judge the relative magnitude of the perils of restructuring and its uncertain benefits, while other corporate stakeholders who structure the operating environments of the firms may exercise their power to influence corporate decisions on portfolio restructuring (DiMaggio/Powell 1983, Palmer/Barber 2001).
While institutional pressures operate at the institutional field level (DiMaggio/Powell 1983) or societal level (Meyer/Rowan 1977), stakeholder relations principally operate at the firm level. Once initiated, portfolio restructuring is expected to eventually uproot the status quo and shift the distribution of burdens and rewards in favor of some stakeholders and to the detriment of others (Heugens/Schenk 2004).
Relevant stakeholders for a firm in its restructuring initiatives include suppliers, customers, political parties and the surrounding communities (Donaldson/Preston 1995). Following the arguments of Aguilera and Jackson (2003), we expect that stakeholders with significant firm-specific investments will enjoy the most substantial level of influence in decisions in a portfolio restructuring strategy. Therefore, we focus our analysis on such stakeholders that include business groups, managers, and financial institutions, especially banks, because these stakeholders have a comparatively direct claim on the allocation of the company's resources and its performance. These stakeholders can also potentially contribute to the spread of portfolio restructuring in different countries (e.g., Hoshi/Kashyap/Scharfstein 1990, Bethel/Liebeskind 1993, Gibbs 1993, Schneper/Guillen 2004).
Financial institutions, business groups and managers make different types of claims on a business corporation. For example, banks have a primary interest in obtaining commercial businesses from companies and in financing their activities. The power to influence the management of the firm to set a strategy that can meet their interests extends from the ownership or other influential positions a stakeholder has in a firm. For example, a stakeholder with sufficient power may discourage portfolio restructuring proposals from being attempted, or encourage alternative measures to achieve a focused strategy or cost reduction.
In this study, we focus on an empirical analysis of portfolio restructuring attempts. Because past research suggests that even unsuccessful restructuring attempts can have a profound effect on a firm and its constituents (Chatterjee/Harrison/Bergh 2003), excluding attempts that are announced but not completed would seriously bias and underestimate any assessment of the role that stakeholder power may play in deterring or encouraging portfolio restructuring initiatives. Following the descriptive view of the stakeholder approach (Mitchell/Agle/Wood 1997), in our hypotheses, we analyze which stakeholders matter and how they influence the frequency of portfolio restructuring.
In countries where an efficient market for corporate control is absent, and there is no clear separation between the management and the board of directors, for example, Japan, large shareholders rely more on internal monitoring mechanisms to exert disciplinary pressure over managers (Mikkelson/Partch 1997). Differences in block shareholders' monitoring abilities affect the implementation of corporate restructuring strategies.
Banks are different from other shareholders in Japan, who tend to monitor managers by relying on passive threats of exit and entry in the capital market. As in Germany, banks in Japan can be important stakeholders as they often make significant firm-specific investments (Aguilera/Jackson 2003). In Japan and Germany banks fulfill a central and powerful role in the corporate governance not only by providing loans but also by acting as equity holders and prominent stock market intermediaries (Schneper/Guillen 2004). Influential banks can help firms recover quickly without initiating risky and costly restructuring activities.
A bank shareholder is usually concerned with protecting the quality of its loans. The lending relationship provides a bank with the capability to influence where and when managers invest corporate resources. Firms with close bank ties face less pressure to make abrupt cuts in employees or engage in divesture restructuring because they are assured of funding, even given a poor performance (Hoshi/Kashyap 2001). When a firm is facing liquidity problems or it is in default of a secured loan, the typical remedy by the bank is to nominate a bank officer to the firm's board of directors in the hope of effecting a turnaround strategy (Gerlach 1992, Kaplan/Minton 1994).
Banks can prevent portfolio restructuring attempts from occurring for two main reasons (Hoshi/Kashyap/Scharfstein 1990). First, banks that directly control large blocks of shares in business corporations generally prefer stable systems of corporate control. When faced with a firm's poor performance, they often eschew portfolio restructuring, which may involve a high level of uncertainty and may result in a high level of indebtedness of the firm. Second, banks that are insiders to a company can dispatch financial expects to the corporate board as stewards of the banks' commercial interests. Banks, especially those acting as references, have an exclusive access to lucrative commercial transactions and the costs of protecting the commercial relationships can ultimately be borne by other shareholders (Gerlach 1992). Therefore, the interests and strategic objectives represented by directors dispatched by these banks and by other stakeholders are often incompatible, and these bank directors are more likely to prevent the portfolio restructuring proposals from being approved. Thus, we propose the following hypothesis:
Hypothesis 1. The level of bank power is negatively related to the frequency of portfolio restructuring activities.
Japanese firms typically form networks of affiliated companies which we can refer to as business groups (Lincoln/Gerlach 2004). Business groups are composed of legally independent firms that are bound together by multiple ties, including cross-ownership, inter-firm transactions and social relations through which they coordinate to take actions (Khanna/Rivkin 2002). One feature of business groups is that firms affiliated with business groups prefer separating businesses into their subsidiaries and affiliates to managing them in-house. For example, Japanese firms affiliated with business groups have a comparatively lower adoption ratio of the multi-divisional form than American firms (Itoh 2003). Business groups provide affiliated firms with access to substantial tangible and intangible resources that are typically unavailable to independent firms (Chang/Hong 2000). Although yet to be the focus of extensive empirical examination, restructuring by firms affiliated with business groups is one of the most important challenges firms and governments face in many countries (Khanna/Palepu 1999, Hoskisson et al. 2004). For example, firms affiliated with business groups may transfer one of their business segments to another affiliate, or divest it to an independent firm. Thus, the viewpoint of business groups is important for the analysis of portfolio restructuring, as business group affiliated firms have restructuring options that might not be available to independent firms.
Business group affiliation can facilitate a firm's portfolio restructuring initiatives in several ways. First, business group affiliates can continuously realign their assets and financial resources, as they can obtain superior information and resources from the business groups. Second, business group affiliates have a better financial resource pool to acquire other firms, by relying on their group's internal capital market. Third, when facing financial distress, business group affiliates can expect group-wide rescue, which is often led by a bank in the group (Peek/Rosengren 2003). In contrast, independent firms do not have access to group resources to act as a buffer against restructuring pressures, and they may have to consider corporate downsizing as the strategic response to cost reduction pressures. Fourth, business groups have a higher degree of diversity (Khanna/Palepu 1999, Chang/Hong 2000), such as network ties which provide group affiliates with the information needed to identify acquisition targets and other business partners. Accordingly, we have our second hypothesis:
Hypothesis 2. The frequency of portfolio restructuring activity is higher in firms affiliated with business groups than in other firms.
Top managers are important stakeholders who invest their human capital in the firms. A shareholder's wealth is tightly tied to the market value of a firm, but top managers consider other aspects of a firm's performance including corporate size and the risk of bankruptcy, because of the added power, prestige and compensation benefits they derive from managing a large corporation (Jensen/Murphy 1990). Consequently, while shareholder activism or the threat from external environment may place pressure on top managers to engage in portfolio restructuring, top managers often have a limited incentive to do so (Shleifer/Vishny 1986). As top managers in a firm cannot diversify their human capital invested in the firm, across other firms, these executives prefer risk-reducing corporate strategies (Wright/Ferris 1997).
To help counter the incentives to avoid risky corporate strategies, which might arise as the impediments to portfolio divestiture attempts, agency theory suggests aligning the interests of top managers and shareholders with ownership incentives (Jensen/Murphy 1990). However, such an approach fails to capture the complexities of the portfolio restructuring phenomenon. At the very least it is necessary to take into account the differences in the ownership level of top managers which in turn are likely to influence their objectives and behaviour.
When top managers hold a low equity stake in the firm, they may have little economic incentive to divest assets that generate negative synergies. Instead, they may have incentives to retain these assets if they justify a higher level of compensation (Shleifer/Vishny 1990) or reduce the manager's risk (Amihud/Lev 1981). Moreover, as Boot (1992) suggests, top managers could be reluctant to divest because the sale would highlight past mistakes or negatively affect their reputation. Key stakeholders may lose confidence in the top managers perceived to be responsible for the past mistakes highlighted in the restructuring (Davidson/Worrell/Dutia 1993). Similarly, when top managers hold a low equity stake in the firm, they tend not to engage in asset expansion portfolio restructuring because of the high risks involved. Instead, they tend to use their positional power, as well as any ownership-derived power to pursue risk-reducing expansion alternatives, such as internal expansion and growth which may better serve their personal, non-pecuniary, interests (Agrawal/Mandelker 1987). Taken together, in situations where the top managers hold a low equity stake, higher managerial ownership is associated with a lower frequency of portfolio restructuring because the increased ownership rights only supplement the power of managers to avoid risky, but potentially profitable strategies.
When the top managers hold a high equity stake in the firm, portfolio restructuring can actually be stimulated by further increases in the levels of managerial ownership. This positive association emerges from two features in the corporate decision-making process: First, the central position in the dominant coalition that top managers enjoy implies that they are often placed under intense scrutiny by other stakeholder groups (Boot 1992). For example, pressures from non-profit-oriented stakeholders may generate reputation costs to the top managers, relating to the negative public opinion on asset divesture attempts. When they hold a high equity stake in the firm, top managers will have the incentives to engage in portfolio restructuring before the negative synergies become too large, and they will consider the potential financial gains associated with their ownership stake to become too large to ignore, thus, creating a strong incentive to engage in asset divesture restructuring. Second, when making corporate decisions, shareholders not only care about the divergent interests of the top managers, but also how often they have to rely on the top managers' potentially divergent project proposals and expertise. Asset expansion activities can exacerbate the information gap between the external market and the firm, and increase the level of managerial complexity faced by the top managers. When the top managers hold a high equity stake in the firm, shareholders are more likely to delegate the authority to these managers. These top managers can then use their expertise to gather information on potential projects, and use the improved managerial incentives to compensate for information asymmetry, thus, facilitating the initiation of asset expansion restructuring.
In this sense, as ownership incentives rise, the financial interests of top managers and shareholders will begin to converge (Bethel/Liebeskind 1993, Palmer/Wiseman 1999). However, such incentives may not consistently provide senior managers with the motivation to lessen the agency costs associated with portfolio restructuring. Instead, we contend that the relationship between managerial ownership and the frequency of portfolio restructuring may initially decrease and then subsequently increase with rising managerial ownership. Accordingly, we propose the following hypothesis:
Hypothesis 3. The level of managerial ownership has a U-shaped relationship with the frequency of portfolio restructuring activities. At low levels of managerial ownership, increases in managerial ownership are associated with a lower incidence of portfolio restructuring. But at high levels of managerial ownership, increases in managerial ownership are associated with a greater incidence of portfolio restructuring.
Data and Methodology
Our choice of Japan as the institutional setting complements previous studies on corporate restructuring that were conducted in the context of Anglo-American economies. This study also complements previous work on restructuring activities of Japanese firms (e.g., Hoshi/Kashyap/Scharfstein 1990, Kang/Shivdasani 1997). Historically, Japanese firms dealt with declining performance through wage adjustments, reductions in overtime, and dismissal of contract laborers (Kang/Shivdasani 1997). According to Ministry of Labor statistics (2000), the percentage of downsizing due to management decisions, as opposed to retirement and other such related activities, increased from 2.3 percent in 1980 to 9.3 percent in 1998. Having suffered stagnation and a macroeconomic crisis since the bursting of the asset bubble in the late 1980s, a broad consensus emerged in Japan in the late 1990s that consolidating a firm's operations was not sufficient to attain a meaningful transformation of the dysfunctional corporate system (Peek/Rosengren 2003).
To cope with the prolonged economic slump and the banking crisis that was worsened in 1997, the Japanese government began to take a series of policy measures to promote portfolio restructuring. In December 1997, the Commercial Codes rationalized the procedures for corporate mergers and acquisitions. In 1999, the revision of the Anti-Monopoly Act liberalized the previous prohibition on genuine holding companies, and the revision of Commercial Codes introduced a share swap system that allowed a parent firm to buy a subsidiary through an exchange of shares. In 2000, the revision of Commercial Codes introduced procedures for splitting-up companies to facilitate restructuring through spin-offs and divestures, and the enactment of Civil Rehabilitation Law simplified bankruptcy reorganization procedures. The 2002 revision of Commercial Codes introduced the adoption of an American-style corporate governance system that utilized external directors. The implementation of the Corporate Restructuring Law in April 2003 eased restructuring provisions and allowed some flexibility in the restructuring measures in line with those of the Civil Rehabilitation Law. In addition, the Japanese government established public entities to dispose of non-performing loans and promote corporate revivals (Ministry of Finance, OECD). Following the implementation of these new policies and regulations, the pace of portfolio restructuring picked up quickly.
The Japanese government also enacted policies to dissolve cross-shareholding, including the implementation of the new regulations on banks' Tier 1 capital shareholding and the enactment of Banks Shareholding Restriction Law in 2001. Furthermore, the revision of the Commercial Codes in the same year abolished restrictions on share buy-backs and treasury stock, allowing firms to hold onto their shares after acquiring them. After the implementation of these policies, companies started to buy back shares, and banks sold off large portions of their holdings of corporate shares (Hoshi/Kashyap 2001). Using data collected from PACAP Database, we found that the average level of ownership held by financial institutions decreased from 31.6 percent in 1997 to 23.9 percent in 2004.
Corporate governance also changed in other respects. Japanese boards are usually dominated by insiders; that is, senior officials within the company. The president is usually the most powerful member of the board, except in rare cases when the chairman, who is often a former president, has higher authority than the president (Kaplan/Minton 1994). In recent years, outside directors are encouraged through the new company-withcommittees system. A growing number of managers received stock options as an element of executive compensation, although these schemes remained quite modest compared to those in the US and the United Kingdom (Jackson 2007).
We started with an industry-specific sample that included all 230 machinery firms that were listed in the period 1998 through 2005. The machinery industry is a coherent industry, in terms of not having as substantial amount of within industry heterogeneity as perhaps the electronics industry. Further, the Japanese machinery industry was one of the worst hit by economic stagnation and the banking crisis, making it a good setting in which to study portfolio restructuring, as there were substantial incentives for firms to engage in restructuring in this industry (Japan Development Bank 1996).
From this sample, we removed 23 companies because of liquidation (10) and takeovers (13). After developing this sample, we obtained market capitalization and accounting data from the PACAP Database. We supplemented this information with data drawn from the NEEDS tape and the Japan Company Handbook. After elimination of cases with missing data, as described above, our final sample included 174 firms with 1,129 observations over the period 1998 through 2005.
There is no archival source of data on portfolio restructuring. Existing databases and reporting services do not capture such complex non-financial information. We therefore adopted the method used by Kang and Shivdasani (1997) to study restructuring in Japan, by using articles published in the business press. Our search centered on each firm in our sample and was conducted on the Dow Jones Interactive Database. After eliminating repeated articles, we found 130 unique reports related to asset divesture, 56 reports on downsizing and 577 unique reports related to asset expansion activities.
This method of data collection assumes that media sources comprehensively report business activities. The limitations of this approach center on two aspects. First, it is possible that only the most significant restructuring activities were reported. Second, newspaper reports typically devote more attention to large and high technology industries. To address these concerns, we focused on firms in the same industry and we included corporate size and R&D intensity as control variables in our analysis.
We identified 9 types of portfolio restructuring activities through our content analysis of reports, and classified them into two broad categories. The first category is asset divesture, which is the count of sell-off, spin-off, plant closure (liquidation) and business withdrawal activities. The second category is asset expansion, which is the count of activities that include mergers and acquisitions, business establishment, business expansion, investment (including buying equity) and joint venture. For comparison, we also collected three types of corporate downsizing activities, which are early retirement schemes, wage cuts, and lay-off announcements. Downsizing may or may not change the composition of businesses in a company's portfolio.
Using Toshiba Machine Co as an example, we identified reports that relate respectively to asset divesture, asset expansion and downsizing.
Toshiba Machine Co Ltd in Numazu, central Japan, plans to spin off its core shaping machine business into a separate company in the spring of 2004 (Source: Japan Industrial Journal, 12 May 2003: Asset divesture: Spin-off) Japanese industrial machine maker Toshiba Machine Co Ltd plans to open a 500 mln Japanese yen ($4.23 mln/3.7 mln euro) factory in Shanghai, China, in 2004, which will be the company's second factory in Shanghai. (Source: Japan Industrial Journal, 8 July 2003: Asset expansion: Business establishment) Toshiba Machine Co Ltd Wednesday said it will reduce 900 workers, or about 20 percent, of its group workers to 4,000 by March 31, 2002, as part of its restructuring plan to ride out the severe business environment. (Source: Dow Jones International News, 28 April 1999: Downsizing: Lay-off)
We used Tobin's q to measure firm performance. We defined Tobin's q as the ratio of market value of equity plus the book value of liabilities to the book value of assets. We used the closing stock prices of the last trading day in a year to calculate the market value of equity. Tobin's q has advantages over accounting measures as performance reflected in accounting measures can have serious lagged effects, which is a potential problem for our study. Furthermore, in the post-crisis period, Tobin's q was an important measure for firm performance because during our sample period government bureaucrats were particularly attuned to the performance of the stock market to monitor the health of the slow-to-revive Japanese economy. Market measures of performance, such as Tobin's q, were also highly influenced by the activism of foreign investors and bank stakeholders, which accordingly acted as a gauge that could stimulate reform on a macro-economic level, such as the revisions of the Commercial Codes (Tiberghien 2002).
We measured bank power using two items: The percentage of shares held by financial institutions, and the number of banks which were not only one of the largest ten blockholders, but also acted as reference banks for the client firm. We used the level of ownership by financial institutions instead of that only by banks because of the extensive level of cross-shareholding between financial institutions in Japan. We standardized the value of these two items and calculated bank power as the aggregate of these standardized values. Our second independent variable is managerial ownership, which we computed as the percentage of shares held by the president of the firm and Chairman. Our third independent variable is business group affiliation, which we coded as an indicator variable that takes the value of one if a firm was affiliated with a horizontal keiretsu such as the Mitsubishi Group, or affiliated with a vertically structured keiretsu, such as found in the typical buyer-supplier relationships in the Japanese automobile industry.
We used return on assets (ROA) as the measure of firm profitability as perceived by the top managers. Managers tend to use internal accounting information rather than external market metrics such as Tobin's q to measure performance. Operating performance can influence restructuring decisions as managers receive more pressure to initiate portfolio restructuring if a firm's performance is in decline (Kang/Shivdasani 1997).
We determined a firm's level of product diversification using the standard formula: PRODIV = 1 - [[summation][s.sup.2]j], where [s.sub.j] is the proportion of a firm's sales in each industry j. Resource availability and bankruptcy risks may influence the extent of restructuring initiatives. We accordingly measured a firm's debt ratio as long-term debt over the sum of long-term debt and the market value of equity. For similar reasons, we measured liquidity as the ratio of current assets to current liabilities. We also included the ratio of research and development expenses to total sales (R&D intensity) to account for the newspaper reporting bias mentioned previously.
We also controlled for two aspects of the corporate governance system: The presence of stock options and the level of ownership by foreign investors. Finally, we included a measure of firm size (log), sales growth rate (the ratio of sales in the current period to that in the previous period), alongside a set of indicator variables to mark the year of an observation. We note here that all independent variables and control variables were lagged by one year relative to the dependent variables.
Model Specification and Estimation
After developing our data on restructuring activities, we aggregated the observations to annual firm-level counts. Our dependent variables have certain important characteristics: (1) they are non-negative; (2) they are integer-valued, denoting counts of asset divesture and asset expansion activities; (3) they exhibit overdispersion, with 93 percent of the values on asset divesture and 68 percent of the values on asset expansion equaling zero; and (4) they are longitudinal. When the outcome variable is non-negative and integervalued, Poisson regression is more appropriate than ordinary least squares analysis. To adjust for the overdispersion, we used the zero-inflated negative binomial (ZINB) model, a generalization of the Poisson model in which the assumption of equal mean and variance is relaxed (Hausman/Hall/Griliches 1984).
We pooled our data to take advantage of the greater degrees of freedom offered by pooling and to capture both the dynamic information of time series and the variation due to cross-sections. As we do this, we also need to address two potential sources of bias that arise in this longitudinal panel data: A lack of independence in observations and unobserved heterogeneity. A lack of independence arises because our sample contains repeated observations across years, and unobserved heterogeneity arises because our models cannot control for all the remaining effects that influence the frequency of portfolio restructuring activities. These two issues may lead to an underestimation of the true standard error, which inflates the significance tests that are associated with the parameter estimates.
To address these issues, we employed a random-effects version of the ZINB and report the robust standard errors that are derived from the robust variance estimator. The robust variance estimator produces consistent standard errors irrespective of the correctness of the correlation structure that is assumed by the regression model, and yields asymptotically consistent estimates even when errors are heteroscedastic. Using robust standard errors also allows us to relax the assumption that observations across years are independent, which helps us to obtain better estimates of the parameters. In addition, the Vuong statistic indicates that the Z1NB model is more appropriate than the general negative binomial model (Vuong 1989).
Table 1 presents the sample correlations and descriptive statistics, including the decomposition of each variable's standard deviation into its within- and between-firm elements. Most of the variables show substantially more variation between firms, than across observations within the same firm. These standard deviation statistics reinforce the importance of accounting for the longitudinal nature of the data and the lack of independence across records, as we do in our models.
Table 2 presents our primary expansion on the description of the sample. It lists 9 types of detailed portfolio restructuring activities that finns undertook, and 3 types of downsizing activities for comparison. We examined these trends in a series of bi-variate comparisons. The first t-test was for changes in restructuring frequency in the period 2000-2002 as compared to the period 1998-2000, as the Civil Rehabilitation Law was implemented in 2000. The second t-test was for changes in restructuring frequency in the period 2003-2005 as compared to the 2002-2003 period, as the Corporate Restructuring Law was implemented in 2003. The results show that Japanese firms engaged in significantly fewer asset divesture restructuring and more asset expansion restructuring only after the implementation of the Corporate Restructuring Law. Japanese finns undertook fewer downsizing activities after the implementation of both laws.
Downsizing is a response to diseconomies of scale resulting from excessive growth while asset divesture or downscoping is a response to diseconomies of scope brought about by over-diversification (Markides 1992). These results imply that Japanese finns engaged in restructuring activities selectively, and institutional changes had significant influences on the types of corporate restructuring implemented. The fact that asset expansion was actively pursued reveals the importance of policies aimed at financing and information provision for middle and small sized firms that helped to transform their business configurations effectively.
Table 3 shows the ZINB regression results with levels of significance reported for two-tailed tests. Models 1 through 3 all lend consistent support for the hypotheses that the stronger the power of banks, the lower the frequency of portfolio restructuring activities (H1), the stronger the association with business groups, the greater the frequency of portfolio restructuring activities (H2), the frequency of asset expansion restructuring will initially decrease and then subsequently increase, alongside a greater level of managerial ownership (H3). However, H3 is only partially supported. The U-shaped relationship between the level of managerial ownership and the frequency of asset divesture activities was not supported by the results shown in Model 4.
The observed effects of stakeholder power are not only significant but important in magnitude. Using the coefficient estimates from Models 1 and 2 in Table 3, we can determine that the frequency of asset expansion restructuring decreases with the level of managerial ownership when managerial ownership is lower than 20 percent, but it then increases with the level of managerial ownership, when it exceeds 20 percent. A finn affiliated with a business group has on average 45 percent more asset expansion activities, while a firm with one additional point on the bank influence scale has on average 11 percent fewer asset expansion activities.
Using the coefficient estimates from Model 3 in Table 3, we can determine that a firm affiliated with a business group has on average 1.15 times more asset divesture activities, while a firm with one additional point on the bank influence scale has on average 20 percent fewer asset divesture activities. The U-shaped relationship proposed between managerial ownership and the frequency of asset divesture was not supported. The coefficient on the squared term of managerial ownership, as shown in Model 4, was insignificant. Instead, based on the results in Model 3, we found that a firm with one additional percent of ownership held by the key managers witnesses on average 16 percent more asset divesture activities than an otherwise comparable firm.
Several of the control variables turned out to be significant and in the expected direction, including the presence of stock options, growth rate of sales, liquidity and firm size. The two-period lagged R&D intensity, which corrects for newspaper reporting bias and industry effects, was insignificant. Debt ratio and the level of foreign ownership were generally not significant. In some models, the diversification index exhibited a sign opposite to the expected one, but this result was not consistently observed. This result casts some doubt on the premise that portfolio restructuring is simply a response to over-diversification.
Our finding that foreign ownership had no significant impact on determining the occurrence of portfolio restructuring supports the home bias hypothesis, which predicts that foreign investors tend to purchase large and well-established stocks (Kang/Stultz 1997). As a supplementary analysis, we used the level of foreign ownership in the next period as the dependent variable, and found that foreign ownership tended to increase in firms with a better market performance (Tobin's q), a larger asset size, a higher intensity of R&D investment, and a lower level of debt and diversification. Therefore, it might be that portfolio restructuring risks determine the level of foreign ownership, not vice-versa. However, there is no doubt that foreign investors play a significant monitoring role in Japanese firms by partly substituting for the banks. The amendment to the Company Law in 2007 made it possible for foreign investors to buy Japanese firms through exchanges of stock, which opened the door wider for mergers and acquisitions activities.
To inspect the robustness of these results, we conducted three supplementary analyses. First, we performed regressions that included an interaction between bank influence and growth rate of sales, and an interaction between bank influence and ROA in the previous period. The rationale for this test was that relational banks may monitor through a contingent governance mechanism (Kang/Shivdasani 1997). When the demand for external bank financing is low, management can retain considerable autonomy; however, when performance or growth rates decline below a certain threshold, relational banks can intervene and become active in corporate rescues. The interaction terms were not significant; while other results remained similar (results are available upon request). We suspected that these non-significant results can be attributed to the weak performance of banks during the study period. The poor performance of the banks themselves diminished their monitoring incentives and capabilities over their client firms. Peek and Rosengren (2003) argue that Japanese banks tend to bail out nearly bankrupt firms by evergreening old loans and keeping their ties with unprofitable firms. Our results suggested that the evergreen policy taken by banks might not be so prevalent among Japanese firms. Their strong influences provided banks with incentives not to push restructuring measures onto the client firm. Thus, we argue that in this period relational banks were undergoing a transition of their roles. On the one hand, they continued to help firms avoid inefficient corporate restructuring, to maintain firm-specific skills. On the other hand, banks delayed the necessary restructuring for firms facing a performance decline or a substantial insolvency risk.
Second, we performed a test using the count of downsizing activities as the dependent variable. As shown in Model 5 of Table 3, two of the stakeholder variables, bank influence and managerial ownership were significant, while business group affiliation was insignificant. We next examined the subsample for the period 1998-2001 as most downsizing occurred during this period. The results in Model 6 showed that the impact of bank influence and business group affiliation on downsizing was similar to that on portfolio restructuring; however, the impact of managerial ownership on downsizing was negative and significant. These findings imply that long-term employment and increasing opportunities for employees were still a primary objective for top managers. As such, managers consider the reputation costs associated with negative public opinions on downsizing as outweighing the potential financial gains from this strategic move.
Third, we used Tobin's q and ROA to predict the performance consequences of portfolio restructuring. We accounted for the longitudinally clustered nature of the data using a random-effects generalized estimating equation (GEE) approach. The GEE algorithm accounts for correlation between records within the same cluster, thus providing improved standard error estimates (Liang/Zeger 1986). As shown in Table 4, unfortunately, neither asset expansion nor asset divesture was significant in this test. Two stakeholder variables, managerial ownership and business group affiliation, were significant and in the expected direction. Overall, the GEE results suggest that decision-makers were reluctant to engage in portfolio restructuring because of the uncertain future performance associated with a restructuring.
Discussion and Conclusion
We argued and established empirically that with the presence of an appropriate set of legal arrangements, the practice of portfolio restructuring, as measured by both asset divesture and asset expansion activities, achieved widespread implementation only when it was consistent with the interests of the most powerful stakeholders in a given context.
Using a sample of 174 Japanese machinery firms in the 1998 to 2005 period, we found strong and robust support for our predictions that portfolio restructuring increases in frequency with the extent to which business groups' strategies are facilitated and decreases with the degree to which banks' interests are protected. The association between the frequency of portfolio restructuring and managerial interests depends on the level of managerial incentive ownership. Thus, our stakeholder power approach to the corporate governance issues we discuss in this study yields insights that help explain why portfolio restructuring is more prevalent in certain firms and periods of time, than in others.
While there is no agreement as to whether portfolio restructuring is "good" or "bad" with respect to a firm's performance, our analysis deepens our understanding of the drivers of this important phenomenon. The different types and frequencies of portfolio restructuring that we observed across firms and time periods show that firms themselves have substantial variance in terms of what levels of stakeholder power they experience. An important point that we develop in this study is that discussing the virtues of portfolio restructuring is but one way of looking at this issue. We contend that examining the extent to which portfolio restructuring is consistent with the structure of stakeholder power is another important avenue of investigation that yields further insight into this phenomenon.
In line with these statements, the arguments we developed and the analyses we conducted speak to the different ways in which banks influence corporate strategy. First, the way that bank stakeholders influence the incidence of portfolio restructuring serves as an approach to examining the relational contingent governance in Japanese firms (Hoshi et al. 1990, Kang/Shivdasani 1997). Our findings support the contention that a close bank relationship can benefit a firm through avoiding inefficient early liquidations and job layoffs, or via the bank's exertion of discipline on client firms' business expansions. Second, our results imply that as relational banks continued to help firms avoid portfolio restructuring, they might also delay necessary restructurings for firms facing performance decline or insolvency risks. Furthermore, as the Japanese economy faced a prolonged period of stagnation and banks faithfully held shares in their client firms for long periods, relational banks potentially fostered a moral hazard in their client firms' incumbent managers. As the management became entrenched, they were less likely to push for necessary portfolio restructuring. It is clear that the restructuring of the banking sector has been an urgent and necessary undertaking in the Japanese economy since the late 1990s.
Our empirical results also shed some light on the debate on whether there is convergence in corporate governance practices across countries as a result of globalization and of the liberalization of capital flows. With globalization, some of the changes in corporate practices in Japan, such as the implementation of stock option plans and increasing levels of foreign ownership, moved the Japanese corporate governance system toward a shareholder-oriented system. Yet, such moves may lead to conflicts with the existing set of stakeholders. As an example, foreign institutional investors may demand a business portfolio focused on core competence development and short-term profit maximization that could lead to conflicts with business groups over growth strategies and the decision to divest or liquidate non-core units. Further, equity-oriented performance targets create conflicts between institutional investors and bank stakeholders over performance criteria, time horizons, and how to discipline poorly performing units. Finally, the introduction of managerial stock options may provoke controversy over perceived short-termism in a firm's strategy. In sum, all these factors create pressure to either restructure the firm to make it consistent with the structure of stakeholder power, or to improve the firm performance to match the expectations of shareholders. For the moment, corporate governance reform in Japan has not led to radical shifts: The market for corporate control has grown in importance but hostile takeovers remain relatively rare. Thus, it may be possible for firms to engage in portfolio restructuring even in an economy that is transitioning towards a more market-oriented corporate governance environment, to the extent that the strong linkage between stakeholders and corporate strategy remains intact.
Acknowledgements: The authors would like to express their sincere thanks to both editors of the journal as well as to the anonymous reviewers for their constructive comments.
Received: 12.07.2006 / Revised: 13.09.2008 / Accepted: 22.01.2009
Agrawal, A./Mandelker, G., Managerial Incentives and Corporate Investment and Financial Decisions, Journal of Finance, 42, 4, 1987, pp. 823-837.
Aguilera, R./Jackson, G., The Cross-national Diversity of Corporate Governance Dimensions and Determinants, Academy of Management Review, 28, 3, 2003, pp. 447-465.
Berglof, E./Perotti, E., The Governance Structure of the Japanese Financial Keiretsu, Journal of Financial Economics, 36, 2, 1994, pp. 250-284.
Bethel, J./Liebeskind, J., The Effects of Ownership Structure on Corporate Restructuring, Strategic Management Journal, 14, 1993, Summer Special Issue, pp. 15-32.
Boot, A. W., Why Hand on Losers? Divestitures and Takeovers, Journal of Finance, 47, 4, 1992, pp. 1401-1423.
Bowman, E. H./Singh, H., Corporate Restructuring: Reconfiguring the Finn, Strategic Management-Journal, 14, 4, 1993, pp. 5-14.
Buck, T./Filatochev, I./Wright, M., Agents, Stakeholders and Corporate Governance in Russian Firms, Journal of Management Studies, 35, 1, 1998, pp. 81-104.
Chang, S. J./Hong, J., Economic Performance of Group-affiliated Companies in Korea: Intragroup Resource Sharing and Internal Business Transactions, Academy of Management Journal, 43, 3, 2000, pp. 429-448.
Chatterjee, S./Harrison. J./Bergh, D., Failed Hostile Takeover Attempts, Corporate Governance and Refocusing, Strategic Management Journal, 24, 1, 2003, pp. 87-96.
Comment, R./Jarrell, G. A., Corporate Focus and Stock Returns, Journal of Financial Economics, 37, 1, 1995, pp. 67-87.
Daley, L./Mehrotra, V./Sivakumar, R., Corporate Focus and Value Creation: Evidence from Spinoffs, Journal of Financial Economics, 45, 2, 1997, pp. 257-281.
Davis, G. F./Diekmann, K. A./Tinsley, C. H., The Decline and Fall of the Conglomerate Form in the 1980s: The Deinstitutionalization of an Organizational Form, American Sociological Review, 59, 4, 1994, pp. 547-570.
DiMaggio, P./Powell, W., The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields, American Sociological Review, 48, 1, 1983, pp. 147-160.
Donaldson, T./Preston, L. E., The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications, Academy of Management Review, 20, 1, 1995, pp. 65-91.
Gerlach, M. L., Alliance Capitalism: The Social Organization of Japanese Business, Berkeley: University of California Press 1992.
Gibbs, P. A., Determinants of Corporate Restructuring: The Relative Importance of Corporate Governance, Takeover Threat, and Free Cash Flow, Strategic Management Journal, 14, special issue, 1993, pp. 51-68.
Hanson, R. C./Song, M. H., Long-term Performance of Divesting Firms and the Effect of Managerial Ownership, Journal of Economics and Finance, 27, 3, 2003, pp. 321-336.
Hausman, J. A./Hall, B. H./Grilliches, Z., Econometric Models for Count Data with an Application to the Patents-R&D Relationship, Econometrics, 52, 4, 1984, pp. 909-938.
Heugens, P. A./Schenk, H., Rethinking Corporate Restructuring, Journal of Public Affairs, 4, 1, 2004, pp. 87-101.
Hite, G. L./Owers, J. E./Rogers, R. C., The Market for Inter-firm Asset Sales: Partial Sell-offs and Total Liquidation, Journal of Financial Economics, 19, 2, 1987, pp. 229-252.
Hoskisson, R. E./Hitt, M. A., Downscoping: How to Tame the Diversified Firm, New York: Oxford University Press 1994.
Hoskisson, R. E. et al., Asset Restructuring and Business Group Affiliation in French Civil Law Countries, Strategic Management Journal, 25, 6, 2004, pp. 525-539.
Hoshi, T./Kashyap, A./Scharfstein, D., The Role of Banks in Reducing the Costs of Financial Distress in Japan, Journal of Financial Economics, 27, 1, 1990, pp. 67-88.
Hoshi, T./Kashyap, A., Corporate Financing and Governance in Japan: The Road to the Future, Cambridge, Massachusetts: MIT Press 2001.
Itoh, H., Corporate Restructuring in Japan Part I: Can M-Form Organization Manage Diverse Businesses? Japanese Economic Review, 54, 4, 2003, pp. 49-73.
Jackson, G., Employment Participation, Adjustment and Distributional Conflict in Japanese Firms, in Aoki, M./Jackson, G./Miyajima, H., Corporate Governance in Japan: Institutional Change and Organizational Diversity, Oxford: Oxford University Press 2007, pp. 282-309.
Japan Development Bank, Recent Research and Development Trends in Japanese Enterprises: Technological Fusion, Research Report No. 91, Tokyo 1996.
Jensen, M.C./Murphy, K.J., CEO Incentives: It's Not How Much You Pay, But How, Harvard-Business Review, 68, 3, May/June, 1990, pp. 138-153.
Kang, J. K./Shivdasani, A. Corporate Restructuring during Performance Declines in Japan, Journal of Financial Economics, 46, 1, 1997, pp. 29-65.
Kaplan, S. N./Minton, B. A., Appointments of Outsiders to Japanese Boards: Determinants and Implications for Managers, Journal of Financial Economics, 36, 2, 1994, pp. 225-258.
Khanna, T./Palepu, K., The Right Way to Restructure Conglomerates in Emerging Markets, Harvard-Business Review, 77, 4, 1999, pp. 125-134.
Khanna, T./Rivkin, J.W., Estimating the Performance Effects of Business Groups in Emerging Markets, Strategic Management Journal, 22, 1, 2002, pp. 45-74.
Lang, L./Poulsen, A./Stulz, R., Asset Sales, Finn Performance, and the Agency Costs of Managerial Discretion, Journal of Financial Economics, 37, 1, 1995, pp. 3-37.
Liang, K-Y./Zeger, S. L., Longitudinal Analysis Using Generalized Linear Models, Biometrika, 73, 1, 1986, pp. 13-22.
Lincoln, J.R./Gerlach, M.L., Japan's Network Economy: Structure, Persistence, and Change. New York: Cambridge University Press 2004.
Markides, C. C., Consequences of Corporate Refocusing: Ex Ante Evidence, Academy of Management Journal, 35, 2, 1992, pp. 119-136.
Meyer, J./Rowan, B., Institutional Organizations: Formal Structure as Myth and Ceremony, American Journal of Sociology, 83, 2, 1977, pp. 340-363.
Mikkelson, W. H./Partch, M. M., The Decline of Takeovers and Disciplinary Managerial Turnover, Journal of Financial Economics, 44, 2, 1997, pp. 205-228.
Mitchell, M. L./Mulherin, J. H., The Impact of Industry Shocks on Takeover and Restructuring Activity, Journal of Financial Economics, 41, 2, 1996, pp. 193-229.
Mitchell, R. K./Agle, B. R./Wood, D. J., Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts, Academy of Management Review, 22, 4, 1997, pp. 853-886.
Montgomery, C. A./Wilson, V., Mergers that Last: A Predictable Pattern? Strategic Management Journal, 7, 1, 1986, pp. 91-96.
Palmer, D./Barber, B. M., Challenges, Elites, and Owning Families: A Social Class Theory of Corporate Acquisitions in the 1960s, Administrative Science Quarterly, 46, 2002, pp. 87-120.
Palmer, T. B./Wiseman, R. M., Decoupling Risk Taking from Income Stream Uncertainty: A Holistic Model of Risk, Strategic Management Journal, 20, 11, 1999, pp. 1037-1062.
Peek, J./Rosengren, E., Unnatural Selection: Perverse Incentives and the Misallocation of Credit in Japan, NBER Working Paper No 9643, 2003.
Schneper, W. D./Guillen, M. F., Stakeholder Rights and Corporate Governance: A Cross-National Study of Hostile Takeovers, Administrative Science Quarterly, 49, 2, 2004, pp. 263-295.
Sheleifer, A./Vishny, R. W., Large Shareholders and Corporate Control, Journal of Political Economy, 94, 3, 1986, pp. 461--488.
Sheleifer, A./Visimy, R. W., Equilibrium Short Horizons of Investors and Finns, American Economic Review Papers and Proceedings 80, 2, 1990, pp. 148-153.
Tiberghien, Y., State Mediation of Global Financial Forces: Different Paths of Structural Reforms in Japan and South Korea, Journal of EastAsian Studies, 2, 2, 2002, pp. 103-141.
Vuong, Q. H., Likelihood Ratio Tests for Model Selection and Non-nested Hypotheses, Econometrica, 57, 2, 1989, pp. 307-333.
Wernerfelt, B./Montgomery, C. A., Tobin's q and the Importance of Focus in Firm Performance, American Economic Review, 78, 1, 1998, pp. 246--250.
Wright, P./Ferris, S., Agency Conflict and Corporate Strategy: the Effect of Divestment on Corporate Value, Strategic Management Journal, 18, 1, 1997, pp. 77-83.
PhD C and. Z. Wu ([mail])--Assoc. Prof. A. Delios
Department of Business Policy, National University of Singapore, Singapore
e-mail: email@example.com; firstname.lastname@example.org
Table 1: Descriptive Statistics and Correlations (N=1,129 firm-years, 174 firms, 1998-2005) Within Mean Over- Between firm Variable all SD SD SD 1. Tobin's [q.sub.t] 1.07 0.60 0.46 0.39 2. [ROA.sub.t] 0.01 0.06 0.04 0.05 3. Divesture 0.08 0.36 0.20 0.28 4. Downsize 0.03 0.20 0.09 0.17 5. Asset expansion 0.46 0.89 0.66 0.62 6. Bank power -0.03 1.68 1.52 0.75 7. Managerial 2.13 5.54 4.98 2.78 ownership 8. Foreign 0.07 0.09 0.08 0.03 ownership 9. Business group 0.14 0.35 0.33 0.03 10. Debt ratio 0.21 0.22 0.19 0.08 11. Liquidity 2.02 1.57 1.36 0.73 12. Finn size 4.73 0.56 0.55 0.05 13. Growth of sales 0.03 0.22 0.10 0.20 14. Diversification 0.50 0.19 0.17 0.09 15. R&D [intensity.sub.t-2] 0.02 0.02 0.01 0.01 Variable 1 2 3 4 5 1. Tobin's [q.sub.t] 2. [ROA.sub.t] 0.19 3. Divesture 0.00 -0.06 4. Downsize -0.05 -0.10 0.27 5. Asset expansion 0.02 -0.02 0.31 0.23 6. Bank power -0.08 0.00 0.04 0.03 0.04 7. Managerial 0.25 0.14 -0.02 -0.07 -0.11 ownership 8. Foreign 0.19 0.05 0.08 0.04 0.23 ownership 9. Business group 0.02 0.02 0.14 0.10 0.20 10. Debt ratio -0.15 -0.15 0.11 0.08 0.08 11. Liquidity 0.04 0.13 -0.09 -0.02 -0.07 12. Finn size 0.01 0.07 0.23 0.18 0.40 13. Growth of sales 0.14 0.28 -0.05 -0.05 -0.01 14. Diversification -0.09 -0.09 0.05 0.03 0.04 15. R&D [intensity.sub.t-2] 0.12 0.07 0.01 -0.01 0.06 Variable 6 7 8 9 10 1. Tobin's [q.sub.t] 2. [ROA.sub.t] 3. Divesture 4. Downsize 5. Asset expansion 6. Bank power 7. Managerial -0.21 ownership 8. Foreign 0.14 -0.08 ownership 9. Business group 0.18 -0.12 0.16 10. Debt ratio 0.09 -0.04 -0.17 0.13 11. Liquidity 0.01 0.10 0.21 -0.09 -0.35 12. Finn size 0.37 -0.15 0.45 0.28 0.18 13. Growth of sales -0.10 0.06 0.08 -0.03 -0.08 14. Diversification -0.06 0.01 -0.07 0.03 0.06 15. R&D [intensity.sub.t-2] 0.08 0.01 0.20 0.04 -0.14 Variable 11 12 13 14 1. Tobin's [q.sub.t] 2. [ROA.sub.t] 3. Divesture 4. Downsize 5. Asset expansion 6. Bank power 7. Managerial ownership 8. Foreign ownership 9. Business group 10. Debt ratio 11. Liquidity 12. Finn size 0.00 13. Growth of sales -0.04 0.04 14. Diversification -0.08 0.04 -0.07 15. R&D [intensity.sub.t-2] 0.24 0.24 0.22 -0.04 Note: Correlations greater than |0.055|, p < 0.05; correlations greater than |0.076|, p < 0.01; two-tailed tests Table 2: Frequencies of Types of Restructuring (1) 1998-1999 (2) 2000-2002 Restructuring Actions Mean Median Mean Median Asset divesture 0.118 4 0.125 4 - Business sell-off 0.037 2 0.037 2 - Spin off 0.015 2 0.015 1 - Liquidation 0.022 1 0.022 2 - Withdrawal 0.011 1 0.013 1 Asset expansion 0.354 5 0.438 6 - Merger and acquisition 0.059 2 0.057 2 - Business establishment 0.055 2 0.129 2 - Business expansion 0.033 2 0.035 1 - Investment (buy equity) 0.018 1 0.015 1 - Joint venture 0.052 2 0.090 2 Down-sizing 0.085 3 0.048 Z - Early retirement scheme 0.022 1 0.004 1 - Cut wages 0.004 1 0.011 1 - Cut jobs 0.041 1 0.022 1 [H.sub.0] : (3) 2003-2005 (2)-(1) = 0 Restructuring Actions Mean Median t-statistic Asset divesture 0.074 2 0.20 - Business sell-off 0.033 1 0.02 - Spin off 0.010 1 0.05 - Liquidation 0.021 2 -0.02 - Withdrawal 0.010 1 0.24 Asset expansion 0.531 5 1.30 - Merger and acquisition 0.083 3 -0.11 - Business establishment 0.200 3 2.84 ** - Business expansion 0.115 3 0.12 - Investment (buy equity) 0.056 1 -0.32 - Joint venture 0.076 2 1.66 Down-sizing 0.014 1 -1.72 * - Early retirement scheme 0.002 1 -2.23 ** - Cut wages 0.000 0 1.05 - Cut jobs 0.012 1 -1.46 [H.sub.0] : [H.sub.0] : (3)-(2) = 0 (3)-(1) = 0 Restructuring Actions t-statistic t-statistic Asset divesture -1.98 -0.16 - Business sell-off -0.32 -0.27 - Spin off -0.68 -0.49 - Liquidation -0.12 -0.13 - Withdrawal -0.40 -0.09 Asset expansion 1.62 * 2.73 *** - Merger and acquisition 1.42 1.10 - Business establishment 2.55 *** 4.76 *** - Business expansion 4.32 *** 3.51 *** - Investment (buy equity) 3.34 *** 2.45 *** - Joint venture -0.67 1.18 Down-sizing -2.90 *** -4.03 *** - Early retirement scheme -0.63 -2.77 *** - Cut wages -2.31 ** -1.34 - Cut jobs -1.12 -2.51 *** Note: * p < 0.05; ** p <0.01; *** p < p.001, two-tailed tests Table 3: Zero-inflated Negative Binomial Regression on Restructuring Determinants Model 1 Model 2 Expansion Expansion 1998-2005 1998-2005 Bank power -0.10 -0.10 (0.04) ** (0.04) ** Managerial ownership -0.03 -0.08 (0.02) * (0.03) ** Managerial ownership 0.002 squared (0.001) * Foreign ownership 0.68 0.35 (0.71) (0.72) Business group 0.36 0.37 (0.17) * (0.16) * Stock option 0.35 0.34 (0.11) ** (0.11) ** ROA -3.40 -3.31 (1.21) ** (1.25) ** Growth of sales -0.34 -0.36 (0.29) (0.30) Debt ratio -0.39 -0.34 (0.25) (0.26) Liquidity -0.09 -0.08 (0.05) * (0.05) ([dagger]) Size 0.87 0.93 (0.14) *** (0.15) *** Diversification 0.23 -0.02 (0.34) (0.31) R&D [intensity.sub.t-2] -0.83 0.33 (2.77) (2.74) Constant -4.76 -4.96 (0.62) *** (0.66) *** Fixed effects for year Included Included Log pseudo likelihood -922.34 -921.05 N (firm-years) 1,129 1,129 Number of firms 174 174 Wald Chi-square 249.16 *** 273.85 *** Vuong statistics 3.00 *** 3.79 *** Model 3 Model 4 Divesture Divesture 1998-2005 1998-2005 Bank power -0.18 -0.22 (0.09) * (0.11) * Managerial ownership 0.15 0.10 (0.02) *** (0.07) Managerial ownership 0.001 squared (0.002) Foreign ownership 1.58 1.43 (1.59) (1.42) Business group 0.79 0.77 (0.36) * (0.35) * Stock option -0.99 -1.21 (0.61) ([dagger]) (0.59) * ROA 0.09 2.29 (2.27) (2.02) Growth of sales -3.35 -2.32 (1.74) * (0.98) * Debt ratio 0.99 1.21 (0.91) (0.85) Liquidity -1.68 -1.68 (0.24) *** (0.28) *** Size 1.56 1.38 (0.35) *** (0.32) *** Diversification -5.24 -5.58 (1.10) *** (0.98) *** R&D [intensity.sub.t-2] -0.07 -1.24 (7.52) (7.99) Constant -4.57 -3.37 (1.96) * (1.51) * Fixed effects for year Included Included Log pseudo likelihood -260.77 -256.65 N (firm-years) 1,129 1,129 Number of firms 174 174 Wald Chi-square 114.45 *** 100.05 *** Vuong statistics 11.97 *** 22.76 *** Model 5 Model 6 Down-size Down-size 1998-2005 1998-2001 Bank power -0.28 -0.40 (0.12) * (0.18) * Managerial ownership -0.64 -11.07 (0.24) ** (0.95) *** Managerial ownership squared Foreign ownership -1.33 -3.11 (2.40) (3.47) Business group 0.28 0.94 (0.47) (0.50) * Stock option -0.47 -0.13 (1.51) (0.75) ROA -10.19 -16.98 (5.09) * (7.56) * Growth of sales 0.06 0.94 (1.06) (1.60) Debt ratio 0.34 1.41 (0.77) (1.09) Liquidity -0.32 -0.30 (0.21) ([dagger]) (0.29) Size 0.79 1.04 (0.82) (0.61) ([dagger]) Diversification 0.38 -1.09 (2.02) (1.65) R&D [intensity.sub.t-2] 3.78 57.23 (31.17) (24.16) * Constant -6.68 -6.79 (3.43) * (3.43) * Fixed effects for year Included Included Log pseudo likelihood -137.05 -79.66 N (firm-years) 1,129 505 Number of firms 174 154 Wald Chi-square 73.86 *** 485.12 *** Vuong statistics 21.67 *** 13.35 *** Note: ([dagger]) p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001, two-tailed tests Standard errors are shown in parentheses beneath regression coefficients. Table 4: Random-effects GEE Regression on Performance Consequences Dependent Variable = Tobin's q Model 1 Model 2 Asset divesture 0.01 (0.02) Asset expansion 0.02 (0.02) Bank power -0.03 -0.02 (0.02) (0.02) Managerial 0.02 0.02 ownership (0.01) ([dagger]) (0.01) ([dagger]) Foreign 1.08 1.01 ownership (0.33) *** (0.34) ** Business group 0.11 0.10 (0.05) * (0.05) * ROA 0.34 0.31 (0.47) (0.48) Debt ratio -0.18 -0.17 (0.10) ([dagger]) (0.10) ([dagger]) Size -0.02 -0.03 (0.06) (0.06) Diversification -0.30 -0.31 (0.16) ([dagger]) (0.16) * R&D [intensity.sub.t-2] 1.98 2.21 (1.31) (1.32) ([dagger]) Fixed effects for Included Included year Constant 1.23 1.32 (0.26) *** (0.28) *** N (firm-years) 1,054 1,054 Number of firms 158 158 Wald Chi2 258 *** 253 ** Dependent Variable = ROA Model 3 Model 4 Asset divesture -0.01 (0.01) Asset expansion -0.002 (0.003) Bank power -0.01 -0.01 (0.01) (0.01) Managerial 0.001 0.001 ownership (.0003) ** (.0003) ** Foreign -0.04 -0.03 ownership (0.06) (0.06) Business group 0.01 0.01 (0.003) * (0.003) ([dagger]) ROA 0.31 0.33 (0.09) *** (0.09) *** Debt ratio -0.02 -0.02 (0.01) * (0.01) * Size 0.01 0.01 (0.01) ([dagger]) (0.01) Diversification -0.02 -0.02 (0.01) * (0.01) * R&D [intensity.sub.t-2] 0.12 0.09 (0.12) (0.12) Fixed effects for Included Included year Constant -0.02 -0.01 (0.02) (0.02) N (firm-years) 1,054 1,054 Number of firms 158 158 Wald Chi2 270 *** 351 *** Note: ([dagger]) p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001. Standard errors are shown in parentheses.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||RESEARCH ARTICLE|
|Author:||Wu, Zhonghua; Delios, Andrew|
|Publication:||Management International Review|
|Date:||May 1, 2009|
|Previous Article:||Institutional environment and subsidiary survival.|
|Next Article:||Understanding environmental risk for IJVs in China.|