Abstract
This study considers whether the strategic decision to enter voluntary administration (VA) rather than to trade the company’s business for a protracted period of declining performance is systematically related to the effective monitoring of management decision-making. Analysis that tests the association between strategic entry into VA and the likelihood that a company will reorganize in VA is also presented. We find about half of the companies in our sample entered VA as a strategic choice. The likelihood of strategic entry to VA increased with the proportion of independent board directors, the existence of an audit committee and a dual CEO/chair board structure. Subsequent analysis of reorganization outcomes suggests that strategic entry into VA improves prospects for a successful reorganization.
1. Introduction
When a company experiences financial distress, effective management is critical because a diverse range of stakeholders are likely to be affected by the substantial losses that can follow poor decision-making (Anderson and Davis, 2009; Brown and Ma, 2011). A timely and appropriate response to the company's problems is likely to limit the costs of ongoing decline and failure, and is therefore desirable (Moulton and Thomas, 1993). When financial distress is sufficiently severe, managers might decide that the appropriate strategy is for the company to enter a formal insolvency administration – commonly liquidation. The difficulty with liquidation is that it is terminal, marking the end of the company's business, and accordingly brings to an end the company's productive economic relationship with its stakeholders, including employees. In this context, business recovery options are typically contained within relevant corporations or bankruptcy codes, such as the United States of America's (US) Chapter 11. In Australia, the Corporations Act (2001) (Cth) (the Act) provides for companies to enter into voluntary administration (VA).
VA insolvency provides an opportunity for the company to reorganize and continue its business as usual, or to make an arrangement with its creditors with a view to full recovery at a later time. Both possibilities represent an alternative to liquidation. If reorganization is not possible, VA allows for disposal of the company's business and assets in a way that seeks to preserve company value. VA is therefore an attractive strategy for dealing with financial distress because of the flexibility afforded companies, including the range of options open to distressed companies electing VA. However, for VA to be used to its best effect, companies must enter administration in a timely manner as delay is likely to result in continuing decline of already ailing companies. The possibility of successful reorganization decreases as available resources are diminished and the company's operating environment becomes more difficult to manage (Dawley et al., 2002; Meeks and Meeks, 2009; Moulton and Thomas, 1993).
Company management are best able to determine if VA is an appropriate strategy, and the optimal time to make that decision. However, it is questionable whether managers have sufficient incentives to make an early decision for VA. From an agency theory perspective, self-serving managers will tend to make decisions that cause them the least harm (Fama and Jensen, 1983; Jensen and Meckling, 1976). If managers perceive that VA is not in their interests, they are likely to behave opportunistically and delay the entry decision. One of the disincentives for managers to make a timely decision is the requirement for their replacement by an independent trustee administrator upon commencement of VA. Hahn (2004: 139) explains that the prospect of termination will lead managers to delay action and to undertake ‘risky activity with the hope of rescuing the corporation from its difficulties’ (Hahn, 2004: 139). The Corporations and Market Advisory Committee (CAMAC) addressed the management replacement policy in its report into the operation of VA, and questioned whether it is appropriate to alter the VA regime to more closely mirror the US Chapter 11 approach that allows managers to retain control of the firm. The Committee report noted that submissions suggested that ‘a debtor in possession reorganization system may encourage boards to take early remedial action, given that they remain in control’ (CAMAC, 2004: 13).
In the current study we examine a sample of listed companies utilizing VA and consider three critical questions related to the timeliness of the management decision to enter VA. First, we assess whether companies enter VA as a strategic choice. For the purpose of the study, entry into VA is considered a strategic choice if it does not follow a protracted period of negative industry-relative performance. We examine each company’s industry-adjusted return on assets over a period of 3 years prior to VA and identify strategic choice as having occurred when: (1) the first period that the company reported a negative industry-relative return on assets is immediately prior to VA, or where (2) the company does not report a negative industry-relative return on assets in any of the observed periods. Second, we examine whether strategic choice to enter VA is systematically related to the effectiveness of mechanisms that provide monitoring of management decision-making. The tension between managers’ self-interest and the company’s interests in making a decision to enter VA suggests that effective monitoring is likely to mitigate delaying behaviour. We expect that more effective monitoring by formal and informal governance mechanisms such as the board of directors and its committees, debt providers, and shareholders will mitigate opportunistic behaviour and result in improved clarity around the decision to enter VA. Third, we examine whether strategic entry into VA has benefits in terms of VA outcomes. To this end, we test the association between strategic entry into VA and the likelihood of a company being able to reorganize and emerge from VA relatively intact.
The structure of insolvency legislation varies between jurisdictions, and the differences have been shown to have an effect on distressed firm behaviour and performance (Franks et al., 1996; Gutierrez et al., 2011). Thus, the advancement of insolvency law policy is enhanced by consideration of the variation in legislative approaches taken across jurisdictions. There is little prior empirical analysis of the operation of the VA legislative scheme. Most of the related prior research examines companies that entered the US Chapter 11 scheme. This study provides a useful point of comparison with these prior studies because, unlike VA, Chapter 11 allows for continued control of the company by incumbent managers and in the US companies that are not in dire financial circumstances may elect for Chapter 11 to apply to them. The study seeks to assist policy makers concerned with the development of optimal insolvency regulation that facilitates key decisions made by managers of distressed companies. The study is also of practical value for those required to make decisions regarding the fate of distressed companies, since it identifies characteristics that are systematically associated with improved decision-making by distressed companies. This is important given the level of information asymmetry that is associated with distressed companies and the resulting difficulty that stakeholders experience with ascertaining the true position of the firm (Charitou et al., 2011; Frino et al., 2007).
We find that a little more than half the sample companies enter VA prior to a protracted period of decline, making this a strategic choice according to our criteria. This is an encouraging result considering the possible incentives for delay by managers facing removal in accordance with the legislature. We find evidence that some of the tested monitoring mechanisms improve decision-making. Of the monitoring mechanisms tested, we find companies are more likely to enter VA as a strategic choice if they have: greater board independence; a dual CEO/chair board structure; and an audit committee. The results are consistent with prior studies that find the monitoring role of independent directors and board committees are important in the context of financial distress. While prior research indicates that a dual CEO/board chair structure reduces board capacity for independent oversight, it appears that the same structure improves the quality of decision-making about entry into VA. We suggest that this is because the CEO has valuable inside knowledge of company operations. Finally, our additional analysis provides evidence that entry into VA as a strategic choice increases the likelihood of a company exiting VA with an agreement with its creditors – the equivalent of a fresh start.
The paper proceeds as follows: next, we provide a brief summary of the salient features of VA, review relevant literature and develop hypotheses that are tested by our empirical analysis; then the research design including sample selection and measurement of variables; we then report and discuss the results of the analysis; and finally present the research results, conclusions and identify limitations of the study.
2. Institutional setting
The distressed companies examined in this study entered the Australian VA reorganization as set out in Part 5.3A of the Act. VA commenced operation in June 1993 as an option to enhance the opportunity for corporate rescue of distressed companies as distinct from liquidation. VA is loosely based on the existing reorganization in the United Kingdom and its implementation resulted from the recommendations of the Australian Law Reform Commission Report Number 45 (1988), which closely followed the English Cork Committee Report. It is similar in nature, although not in legislative detail, to the US Chapter 11 provisions.
VA is required to be completed within about 3 months except when there has been a court-approved extension. The short time frame is aimed to reduce costs and be attractive to stakeholders as an alternative to a lengthy administration process (Bebchuk, 2000; LoPucki and Doherty, 2002) and reduced returns. Commencement of VA requires the appointment of an independent administrator to take control of the company. Importantly, entry to VA triggers a moratorium period that prevents creditors from proceeding with legal claims against the company. The administrator investigates the company's affairs, and makes a recommendation that the company attempt reorganization or, alternatively, enter into liquidation. A meeting of the company's creditors makes the final decision regarding the fate of the company. If a reorganization plan is adopted, the company enters into a statutory ‘deed of company arrangement’ that is binding on all creditors.
The automatic removal of existing management is a key aspect of the institutional setting for this study. Analysis by Lee et al. (2007) and Hahn (2004) suggests that managers are reluctant to enter insolvency administration if it will result in them being automatically removed. Moreover, once the firm has entered VA, the decision regarding the firm's fate and any future involvement of incumbent managers rests with creditors – the likely residual claimants in liquidation. From an agency perspective, this aspect of the operation of VA creates conflicts for managers as they decide on an appropriate strategy to deal with distress. Self-serving managers are motivated to pursue risky strategies that may not be as effective as entering VA (Hahn, 2004). Accordingly, in this setting, the importance of monitoring management by formal and informal governance mechanisms is evident and the development of our hypotheses proceeds from that standpoint.
3. Literature and hypotheses
We expect that effective monitoring by formal and informal governance mechanisms mitigates opportunistic behaviour and results in increased likelihood that entry to VA is a strategic choice. Prior literature suggests that both formal and informal governance mechanisms are relevant to the quality of decision-making for distressed firms. Such mechanisms include the board of directors and its committees, the external audit process, debt providers and shareholders. Hypotheses regarding the association between each of these mechanisms and the likelihood of entry into VA as a strategic choice are discussed below.
From an agency perspective, the board of directors has a duty to protect stakeholders’ interests by monitoring management and limiting their opportunistic behaviour. Independent directors are considered to be better able to exercise their judgment in a manner that constrains opportunistic managerial decision-making (Fama and Jensen, 1983). Greater independence of the board of directors is associated with more effective monitoring and control (Beasley, 1996; Dechow et al., 1996). The positive effect of board independence on management of distressed companies is consistently demonstrated in prior research. Studies show that boards with fewer independent directors are associated with an increased likelihood of bankruptcy (Daily and Dalton, 1994a, 1994b; Elloumi and Gueyie, 2001; Lai and Sudarsanem, 1997). Lai and Sudarsanem (1997) show that board independence is associated with the choice of effective recovery strategies for declining companies. Further, Daily (1995, 1996) finds a positive association between successful US Chapter 11 bankruptcy reorganization and board independence. This suggests board independence is likely to improve reorganization prospects to the extent that it enhances monitoring and limits or prevents decisions that reflect a high-risk survival strategy prior to VA. Therefore, we expect to observe a positive relation between board independence and entry into VA as a strategic choice. This leads to our first hypothesis:
Hypothesis 1: Companies with greater board independence are more likely to enter VA as a strategic choice.
Another important characteristic associated with board effectiveness is the existence of independent leadership. Separation of the roles of CEO and board chairperson promotes independent leadership. For this reason, the ASX Corporate Governance Guidelines (ASX, 2007) recommend against the dual role. Centralized authority related to the dual CEO/board chair structure increases the probability of a dysfunctional and rigid response to the threat of financial distress (Daily and Dalton, 1994b; McKinley, 1993; Staw et al., 1981). As Daily and Dalton (1994b: 645) point out, the adoption of the CEO/board chair structure will ‘exacerbate any tendencies which limit the organization's adaptive abilities’. Prior distress studies report that the dual CEO/board chair structure is associated with a greater probability of bankruptcy and a reduced likelihood of reorganization (Daily, 1994; Elloumi and Gueyie, 2001; Lai and Sudarsanem, 1997). In the context of VA, we expect that the disincentive to enter VA because of manager termination to be compounded by problems associated with concentrated authority, reduced monitoring and the prospect of censure or reputational damage. Therefore, we conclude that the dual CEO/board chair structure will increase the probability that management will favour high-risk survival strategies over entry into VA.
Hypothesis 2: Companies with a dual CEO/board chair are less likely to enter VA as a strategic choice.
From an agency perspective, larger (in number) boards may be better able to maintain independent oversight of decision-making. A larger board of directors is less readily dominated by the CEO due to the dilution of influence and increased effort required to gain consensus (Muth and Donaldson, 1998). Empirical evidence suggests larger boards are associated with improved monitoring of management due to their ability to commit more resources, time and skill to the task (Anderson et al., 2004; Chiang, 2005; Klein, 2002; Williams et al., 2005). Some of the research evidence related to board size suggests that problems with organizing and coordinating large boards has a negative impact on the board's ability to engage in long-term strategic planning and firm value (Beasley, 1996; Lipton and Lorsch, 1992; Yermack, 1996). In the context of financial distress, the immediacy of the firm's problems minimizes the need for long-term strategic planning. Nonetheless, the monitoring role of the board will continue to be important as to the quality of decision-making. This leads to our third hypothesis:
Hypothesis 3: Companies with a larger board of directors are more likely to enter VA as a strategic choice.
Board activity is associated with effective board operation in the context of financial distress. Vafeas (1999: 140) assessed the relation between board activity and performance of distressed companies, finding that increased board activity follows poor performance as ‘boards respond to tough years of operation’. More frequent meetings are associated with improved performance, suggesting a ‘day-to-day’ effect of board monitoring on performance. Therefore, greater activity in the period before a company enters administration indicates an appropriate response by the board in circumstances of financial distress.
Hypothesis 4: Companies with an active board of directors are more likely to enter VA as a strategic choice.
From an agency perspective, it is suggested that shareholding by directors leads to their having a greater incentive to monitor management. Consistent with this view, prior studies report a negative association between director shareholding and the probability of bankruptcy (Abdullah, 2006; Lee and Yeh, 2004; Parker et al., 2002).
Hypothesis 5: Companies with a higher proportion of director share ownership are more likely to enter VA as a strategic choice.
Prior studies show the external audit function and the existence of an audit committee play an important monitoring role. Larger audit firms provide better quality audit services and a higher level of monitoring (DeAngelo, 1981; Francis et al., 1999; Kim et al., 2003). Audit firm size measure is generally based on whether a firm is one of the recognized top-tier audit firms (the ‘Big 4’). The audit committee is the monitoring mechanism most likely to maintain the quality of a company's financial information (Davidson et al., 2005; Koh et al., 2007; Rainsbury et al., 2008). The quality of financial information is important to ensure properly informed decision-making in circumstances of financial distress. Our next two hypotheses focus on the role of the external auditor and the audit committee:
Hypothesis 6: Companies that engage one of the ‘Big 4’ audit firms are more likely to enter VA as a strategic choice.
Hypothesis 7: Companies that form an audit committee are more likely to enter VA as a strategic choice.
As financial distress worsens, debt providers have a greater interest in the company's operating decisions and compliance with debt agreements. The increase in monitoring of management decisions serves to reduce the opportunity for managers to behave opportunistically. As the ratio of assets to debt increases, there is greater propensity for monitoring by debt providers. Our hypothesis for the monitoring role of debt providers is:
Hypothesis 8: Companies with higher leverage are more likely to enter VA as a strategic choice.
4. Data and method
4.1. Data
Data collection commenced by searching the Aspect DatAnalysis database for listed company announcements related to the appointment of a voluntary administrator. The search of announcements was made for the 10-year period from 1998 to 2008. Of the companies identified, we selected those for which 3 years of data was available prior to the company entering VA. The final sample constitutes 62 companies for analysis. Governance and shareholding data was hand-collected from each company's first full annual report prior to entry into VA. Collection of financial data proceeded by first downloading relevant information from the Aspect FinAnalysis database. For companies with missing data, relevant information was hand-collected from annual financial reports. For some companies, there was a substantial lag between the last financial report and entry into VA. For these companies, both annual and half-yearly financial reports were used to obtain data for calculation of the industry-adjusted return on assets used for the dependent variable. This enabled us to determine the industry-adjusted return on assets closer to VA.
4.2. Dependent variable
Our objective in defining the dependent variable is to identify companies in the sample that entered VA as a strategic decision in response to the onset of financial distress. While the decision to enter VA is publicly announced and directly observable, the decision by managers to delay entry to VA and adopt high-risk survival strategies is not. Therefore, we rely on measures from financial statements to deduce management decision-making in the period prior to the company entering VA. Our use of financial statement measures to deduce management decisions is consistent with the approach taken in a substantial number of prior financial distress studies (Peat, 2007). In this case, management decision-making is deduced from the reporting of protracted period of poor industry-adjusted accounting performance. To determine whether the company enters VA as a strategic choice we examine the industry-relative return on assets over the three reporting periods immediately before the company enters VA. 1 Our rationale is that two periods of negative industry-adjusted operating performance prior to entering VA suggests protracted decline, and is inconsistent with entry into VA as a strategic choice.
An alternative to our static model design is to develop a hazard model to determine how variables of interest are related to entry into VA. This approach has been applied in prior studies for prediction of bankruptcy (Fich and Slezak, 2008; Hillegeist et al., 2004; Shumway, 2001) and the duration of bankruptcy administration (Partington et al., 2001; Wong et al., 2007). This disadvantage of our analysis compared with the hazard model approach is that it only considers the effect of the observed variables in the year immediately prior to VA. Thus, it does not incorporate time-related changes in the variables of interest, which can result in biased coefficient estimates (Shumway, 2001). There is evidence, however, that corporate governance characteristics tend to be consistent over time (Brown et al., 2011), which is likely to mitigate potential econometric problems. Moreover, problems with availability of more extensive and reliable data for some of the earlier VA companies suggest our static model approach is appropriate.
Using industry-adjusted operating performance to measure financial health is consistent with a number of prior company reorganization studies (Denis and Kruse, 2000; John et al., 1992; Kang and Shivdasani, 1997). Return on assets is calculated as the ratio of net profit after tax to total assets, and the industry-relative measure is determined by reference to companies in the same Global Industry Classification Standard (GICS) category. We identify those companies where: (1) the first period that the company reports a negative industry-relative return on assets immediately prior to VA, or (2) the company does not report a negative industry-relative return on assets over the analysis period. The dependent variable is a dichotomous indicator that is coded 1 (one) for entry into VA as a strategic choice, and 0 (zero) otherwise.
4.3. Independent variables
We assume board independence by measuring the proportion of non-executive directors against the total number of board directors. A more refined measure of board independence is desirable; however, the annual reports for companies that entered VA prior to 2003 do not reflect the ASX-issued corporate governance recommendations (ASX, 2003). For earlier annual reports it is difficult to determine the independent status of directors. We therefore use the proportion of non-executive directors, as it can be consistently measured for the entire sample. Independent board leadership is indicated by a dichotomous variable that is determined by whether the roles of the CEO and board chair are combined (coded one (1) for dual CEO/board chair and zero (0) otherwise). The number of board meetings per year measures board activity. Board size is a function of the number of appointed directors. Director shareholding is measured as the percentage of total issued ordinary shares held by directors. Dichotomous indicator variables are included for the existence of an audit committee (coded one (1) for the existence of a committee, and zero (0) otherwise), and the engagement of a ‘Big 4’ audit firm (coded one (1) for engagement of a ‘Big 4’ firm, and (0) otherwise). Financial leverage is included in the analysis as a proxy for the extent of monitoring by debt providers, and is measured by the ratio of total assets to total liabilities.
4.4. Control variables
Firm size is included as a control for several reasons. Moulton and Thomas (1993) note that larger firms with more varied assets are better able to survive protracted periods of poor performance than small firms. Moreover, larger firms are more likely to have at least a part of their business operations that is profitable. Larger firms also tend to have in place more substantive formal corporate governance mechanisms, due to the complexity of their operations (Boone et al., 2007; Dedman, 2000). Accordingly, we include the natural log of total assets in the analysis to control for any effect associated with company size.
Substantial shareholders are motivated to monitor management and, via board representation, influence decision-making. Concentrated shareholding has been found to be an effective monitoring substitute for formal governance structures (Bédard et al., 2005; Birt et al., 2006; Dechow et al., 1996; DeFond and Jiambalvo, 1991; Rainsbury et al., 2008). In the context of financial distress, prior US studies show that substantial shareholders play a role in monitoring and disciplining managers (Charitou et al., 2007; Daily, 1995; Donoher, 2004; Parker et al., 2002). However, substantial shareholders also use their position to expropriate corporate resources (Dennis and McConnell, 2003). Hahn (2004: 133) suggests that a close association is likely to develop between strong shareholders and managers, and that, in the context of insolvency, shareholders may direct management to ‘engage in overly risky projects and gamble for a yield with creditors’ money.’ Analysis by Gong (2004) suggests that shareholder preference for risk-seeking rather than risk-avoidance is likely when levels of company debt are high. Shareholders have no formal decision-making role in VA, and are not in a position to act in coalition with incumbent managers due to the removal of the latter upon VA commencement. Therefore, we suggest that substantial shareholders are averse to the company entering VA. To the extent that substantial shareholders are able to influence management, we expect the existence of concentrated share ownership increases the probability that the firm will adopt risky survival strategies rather than enter VA. Therefore, we include in the analysis a measure of shareholder concentration as a control variable. Concentrated shareholding is measured as the percentage of issued ordinary shares held by parties that are independent of the directors with a 5% or greater shareholding.
A further control variable is included for the lag between the date of the reporting period that the company return on assets is calculated and commencement of the VA. As mentioned (in the ‘data’ discussion previously), there is a lag between the last financial report and entry into VA. To control for any possible effects associated with this lag, we include a variable that measures the number of months between the reporting date and the date of VA.
4.5. Statistical analysis
Our hypotheses are tested by logistic regression analysis. Logistic regression is a suitable analysis technique given the dichotomous independent variable and because it is robust to violations of the assumptions for independent variables (Chi and Tang, 2006; Hosmer and Lemeshow, 1989). Equation (1) below shows the logit used in our analysis:
Strategic Choice = Variable coded 1 for entry to VA as a strategic choice, and 0 otherwise.
Proportion of Non-Executive Directors = Number of non-executive directors divided by number of directors.
Dual CEO/Board Chair = Dummy, coded 1 if CEO is also chair of the board of directors; 0 otherwise.
Board Size = Number of directors.
Board Meetings = Number of board meetings in the year.
Directors’ Shareholding = Percentage of ordinary shares held by directors.
‘Big 4’ Auditor = Dummy, coded 1 if company engaged a ‘big 4’ audit firm; 0 otherwise.
Audit Committee = Dummy, coded 1 if company has an audit committee; 0 otherwise.
Leverage = Total assets divided by total liabilities.
Size = Natural log of total assets.
Concentrated Shareholding = Percentage of issued ordinary shares held by independent parties with a five percent or greater shareholding.
Lag= number of months between the date of the reporting period from which return on assets for t-1 was calculated and the date of VA.
5. Analysis and results
5.1. Descriptive statistics
Panels A and B of Table 1 provide descriptive statistics for all companies in the sample. The mean of industry-adjusted return on assets is negative in both t−2 and t−1 and declining over the two periods. This poor average industry-adjusted performance is evidence of the financial distress experienced by the sample companies. For the governance variables, the descriptive statistics indicate that boards are structured to promote independence. The mean proportion of independent directors is 62%, and only 11% of companies have a joint CEO/board chair arrangement. The number of board meetings per year ranges from a minimum of zero to a maximum of 31, with a median value of 12. The number of directors ranges between 2 and 10, with a median of 4.
Pooled descriptive statistics (n=62)
Notes: All variables are measured at t−1 unless noted.
Industry-Adjusted ROA = Ratio of net profit after tax to total assets, less ratio of net profit after tax to total assets for companies in the same Global Industry Classification Standard (GICS) category.
Proportion of Non-Executive Directors = Number of non-executive directors divided by number of directors.
Board Size = Number of directors.
Board Meetings = Number of board meetings in the year.
Directors’ Shareholding = Percentage of ordinary shares held by directors.
Leverage = Total assets divided by total liabilities, winsorized at 10%.
Size = Natural log of total assets.
Concentrated Shareholding = Percentage of issued ordinary shares held by independent parties with a 5% or greater shareholding.
Lag = number of months between the date of the reporting period from which return on assets for t−1 was calculated and the date of VA.
Strategic Choice = Variable coded 1 for entry to VA as a strategic choice, 0 otherwise.
Dual CEO/Board Chair = Dummy, coded 1 if CEO is also chair of the board of directors; 0 otherwise.
‘Big 4’ Auditor = Dummy, coded 1 if company engaged a ‘big 4’ audit firm; 0 otherwise.
Audit Committee = Dummy, coded 1 if company has an audit committee; 0 otherwise.
Of the control variables, the mean value of leverage is 1.92. The mean percentage of ordinary shares held by independent parties with a 5% or greater shareholding (concentrated shareholding) is 29% and directors’ shareholding is, on average, 21%. Just over half of the companies engage one of the ‘Big 4’ audit firms, and 84% have an audit committee.
Table 2 reports disaggregated descriptive statistics for the sample companies based on their classification according to whether their entry to VA is determined to be a strategic choice or otherwise. The difference in mean values of the industry-adjusted return on assets lends support to the classification regime applied to identify companies that enter VA as a strategic choice. The companies with a protracted period of negative performance (dependent variable coded as 0) show substantially lower industry-adjusted return on assets over t−1 to t−3. A test of the performance difference between the groups of companies is significant in the periods t−2 and t−3. Of the governance variables, the mean proportion of non-executive directors is significantly greater for the strategic choice companies (at p<0.05). This is consistent with the expectations of hypothesis one. We also find that 94% of the group of companies that enter VA as a strategic choice formed an audit committee. This is significantly greater than the rate of formation of 71% for the group of companies that did not enter VA as a strategic choice (at p<0.05), and is consistent with hypothesis seven. Of the control variables, company size (measured by total assets) is significantly different between the groups with larger firms more likely to enter VA as a strategic choice.
Disaggregated descriptive statistics (n=62)
Notes: Descriptive statistics after sample companies are grouped according to whether their entry to VA was a strategic choice or otherwise. Results of an independent samples t−test are reported for continuous variables and a chi square test for dummy variables; * p<0.1, ** p<0.05, *** p<0.01.
All variables are measured at t−1 unless noted.
Industry−Adjusted ROA = Ratio of net profit after tax to total assets, less ratio of net profit after tax to total assets for companies in the same Global Industry Classification Standard (GICS) category.
Proportion of Non−Executive Directors = Number of non−executive directors divided by number of directors.
Board Size = Number of directors.
Board Meetings = Number of board meetings in the year.
Directors’ Shareholding = Percentage of ordinary shares held by directors.
Leverage = Total assets divided by total liabilities, winsorized at 10%.
Size = Natural log of total assets.
Concentrated Shareholding = Percentage of issued ordinary shares held by independent parties with a 5% or greater shareholding.
Lag = number of months between the date of the reporting period from which return on assets for t−1 was calculated and the date of VA.
Strategic Choice = Variable coded 1 for entry to VA as a strategic choice, and 0 otherwise.
Dual CEO/Board Chair = Dummy, coded 1 if CEO is also chair of the board of directors; 0 otherwise.
‘Big 4’ Auditor = Dummy, coded 1 if company engaged a ‘big 4’ audit firm; 0 otherwise.
Audit Committee = Dummy, coded 1 if company has an audit committee; 0 otherwise.
Table 3 reports correlation measures for the independent variables included in the regression analyses. Some correlations are of a magnitude that raises concerns about multicollinearity for the logistic regression analyses conducted (Tabachnick and Fidell, 1996). To determine whether these correlations affect the results, logistic regression analyses were run with and without each of the more highly correlated control variables. The tenor of the results is unchanged, indicating the reported results are robust to the effects of correlations between independent variables.
Correlation matrix (n=62)
Notes: ** p<0.05, *** p<0.01
All variables are measured at t−1 unless noted.
Proportion of Non-Executive Directors = Number of non-executive directors divided by number of directors.
Dual CEO/Board Chair = Dummy, coded 1 if CEO is also chair of the board of directors; 0 otherwise.
Board Size = Number of directors.
Board Meetings = Number of board meetings in the year.
Concentrated Shareholding = Percentage of issued ordinary shares held by independent parties with a 5% or greater shareholding.
Directors’ Shareholding = Percentage of ordinary shares held by directors.
‘Big 4’ Auditor = Dummy, coded 1 if company engaged a ‘big 4’ audit firm; 0 otherwise.
Audit Committee = Dummy, coded 1 if company has an audit committee; 0 otherwise.
Leverage = Total assets divided by total liabilities, winsorized at 10%.
Size = Natural log of total assets.
Lag= number of months between the date of the reporting period from which return on assets for t−1 was calculated and the date of VA.
5.2. Regression results
Table 4 reports logistic regression results for the main analysis. We find that a greater proportion of non-executive directors is associated with a significantly increased likelihood of entry into VA as a strategic choice (p<0.05). We also find that the audit committee indicator variable is significant at p<0.05. The coefficient for this variable is positive. This means the existence of an audit committee increases the likelihood of VA as a strategic choice. The results provide support for hypotheses one and seven. The existence of a dual CEO/board chair is also significant at p<0.05. However, the existence of a dual CEO structure is associated with an increased likelihood of entry into VA as a strategic choice, contrary to hypothesis two.
Logistic regression: governance and strategic entry to VA (n=62)
Notes: Significance test is based on change in chi square statistic when variable is included in the model; ** p<0.05, *** p<0.01.
All variables are measured at t−1 unless noted.
Proportion of Non-Executive Directors = Number of non-executive directors divided by number of directors.
Dual CEO/Board Chair = Dummy, coded 1 if CEO is also chair of the board of directors; 0 otherwise.
Board Size = Number of directors.
Board Meetings = Number of board meetings in the year.
Directors’ Shareholding = Percentage of ordinary shares held by directors.
‘Big 4’ Auditor = Dummy, coded 1 if company engaged a ‘Big 4’ audit firm; 0 otherwise.
Audit Committee = Dummy, coded 1 if company has an audit committee; 0 otherwise.
Leverage = Total assets divided by total liabilities, winsorized at 10%.
Concentrated Shareholding = Percentage of issued ordinary shares held by independent parties with a 5% or greater shareholding.
Size = Natural log of total assets.
Lag= number of months between the date of the reporting period from which return on assets for t−1 was calculated and the date of VA.
Of the control variables, firm size is significant (at p<0.01) and the coefficient shows larger firms more likely to strategically enter VA. We also find the reporting lag variable is marginally significant (at p<0.10), suggesting that the timing of the reports in relation to the administration influences the results. To test the extent of this problem, we repeated the analysis by removing the companies in the top ten percentile of the lag measure. The lag variable was not significant in this analysis, and the other results remained unchanged.
The results show that board governance characteristics are associated with entry to VA as a strategic choice. In particular, the results support prior findings that board independence contributes to improved decision-making in the context of company financial distress. Our finding that the existence of a dual CEO/board chair enhances strategic choice is unexpected. We suggest that weakness in governance resulting from the dual role may be offset by the benefits that accrue from the dual CEO/board chair's inside knowledge of the firm's operations.
5.3. Additional analysis
One question that follows from the results of our main analysis is whether strategic entry into VA is associated with better VA outcomes. Additional logistic regression analysis is presented in this section to address this question. We anticipate that strategic choice will be associated with better outcomes for several reasons. First, timely entry into VA halts the downward spiral of decline (Hambrick and D’Aveni, 1988) and preserves available resources. Second, the longer that the company stays outside of VA, the more difficult the management of the company's operations becomes. As decline continues, the strategic options available to managers are reduced as a result of declining resources, which can be compounded by constraints imposed by a hostile external environment (Dawley et al., 2002; Hambrick and D’Aveni, 1988; Meeks and Meeks, 2009; Moulton and Thomas, 1993; Partington et al., 2001; Wong et al., 2007). Third, the positive signal sent by timely action is likely to be received favourably by creditors and enhance the possibility of agreement being reached in VA.
The dependent variable for the analysis conducted is a dichotomous indicator that takes a value of 1 if the company was able to reorganize its affairs and emerge intact from administration, and 0 otherwise. The independent variables in the analysis include leverage (ratio of total assets to total liabilities) and liquidity (ratio of current assets to current liabilities). Both of these variables indicate the stock of financial resources available to the company at the time of entry into VA. Available resources have previously been shown to improve ability to implement effective strategies for dealing with financial distress (Hambrick and D’Aveni, 1988; Smith and Graves, 2005). The strategic choice indicator variable is included, along with interaction variables for leverage by strategic choice and liquidity by strategic choice. Including interaction variables allows for observation of the association between financial characteristics and VA outcome, conditional on strategic entry into VA. We also include the change in industry return on assets as a proxy for health of the external environment of the distressed companies, and company size measured by total assets. Results of the additional analysis are presented in Table 5.
Logistic regression: VA outcome (n=62)
Notes: Significance test is based on change in chi square statistic when variable is included in the model; *p<0.1, ** p<0.05, *** p<0.01.
All variables are measured at t−1 unless noted.
Liquidity = Current assets divided by current liabilities, winsorized at 10%.
Leverage = Total assets divided by total liabilities, winsorized at 10%.
Strategic Choice = Variable coded 1 for entry to VA as a strategic choice, and 0 otherwise.
Size = Natural log of total assets.
Change in industry ROA = Change from t−2 to t−1 for ratio of net profit after tax to total assets for companies in the same Global Industry Classification Standard (GICS) category.
The dichotomous strategic choice variable is not significant in the logistic regression. This shows that timely entry into VA of itself does not increase a company's prospects for reorganization in VA. However, the leverage by strategic choice interaction variable is significant (at p<0.05) and the coefficient is positive. This shows that for a given level of leverage (measured by total assets to total liabilities) strategic entry to VA increases the likelihood of VA ending with reorganization. Larger companies and companies operating in an industry with increasing return on assets are also more likely to reorganize. The results for the liquidity variable were not significant. Overall, the analysis provides some evidence that strategic entry into VA improves the prospects reorganization in VA.
6. Summary and conclusions
This study considers questions related to the timeliness of the decision to enter VA. We assess whether companies enter VA as an optimal strategic choice. We then examine whether strategic choice to enter VA is systematically related to the effectiveness of mechanisms that provide monitoring of management decision-making. Analysis that tests the association between strategic entry into VA and whether a company is able to reorganize is also presented.
Of the sample companies, a little over half make a strategic choice to enter VA, as indicated by the company not experiencing a protracted period of negative industry-relative decline before VA. Descriptive statistics show that companies using VA as a strategic choice have better industry-adjusted performance, more independent boards, are larger and are more likely to have formed an audit committee. Multivariate analysis shows the likelihood of strategic entry to VA increases with the proportion of independent board directors and the existence of an audit committee. This demonstrates the important monitoring role of the board and its committees in the context of financial distress and confirms the benefits of having an independent board. Contrary to expectations, the existence of a dual CEO/board chair structure is associated with a strategic VA, which indicates that benefits from the dual CEO/board chair structure outweigh the problems arising from centralized leadership. We suggest that the benefit of the dual CEO/board chair structure is that the board has access through the CEO to internal information that assists with making well-informed critical decisions (Baysinger and Hoskisson, 1990). Our analysis of the association between strategic entry into VA and a reorganization outcome suggests that companies are better able to take advantage of their financial resources when they strategically enter VA. We suggest that timely entry into VA results in better management of financial distress and an increased likelihood of VA ending with reorganization. Larger companies and companies operating in more profitable industries are found to be more likely to reorganize. Taken together, these findings suggest that effective monitoring enhances the resolution of financial distress in VA because it improves decision-making.
The findings have practical implications for stakeholders in distressed companies. The results show that the prospects of a distressed company are likely to be associated with the monitoring capacity of the board and the strength of its leadership. Information regarding these governance characteristics is readily observable, and can be used by creditors or other stakeholders to inform their decision-making. This is important because of the level of information asymmetry that is likely to exist for distressed companies (Frino et al., 2007). Creditors might be well advised to exercise particular caution when it is apparent that monitoring mechanisms are weak. For managers of distressed companies, the results suggest that delaying entry into VA is not an optimal strategy.
The results demonstrated suggest useful implications for legislative policy around structure of the Australian insolvency administration provisions. The findings regarding the dual CEO/board chair suggests that knowledge of the firm's operations may be useful as a company grapples with the problems of financial distress. Therefore it is posited that formal insolvency administration is enhanced if incumbent managers continue to play a significant role. In the VA procedure, existing managers are sidelined once the administration starts. While removal of management has some appeal in that those responsible for a company's failure are seen as being somewhat removed from the actions required to stabilize and improve the company's business welfare, such practice may be costly. As Mumford (2003: 57) notes: it is ‘impractical to dispense altogether with the services of existing managers. In particular, their specialist knowledge is likely to be valuable in maximizing the value of the company where it is worth more as a going concern than it is in liquidation’. Hahn (2004) suggests that an alternative to the removal of incumbent managers is the incorporation of a trustee into the existing management structure upon entry into VA. The empirical evidence presented in this paper lends support to this policy approach.
There are several limitations to this study. The sample size is relatively small and includes only listed public companies. This is due to a lack of readily available data for smaller companies that do not have public reporting requirements. These sampling issues limit our ability to generalize the findings to smaller private company business operations. Our measure of performance is limited to considering only financial performance based on industry-adjusted return on assets, and there are prior studies that have identified earnings manipulation by managers of distressed firms (Charitou et al., 2007; DeAngelo et al., 1994; DeFond and Jiambalvo, 1994; Rosner, 2003). This represents a limitation if accounting data has been manipulated to the extent that it does not represent the true underlying economic performance of the company.
The analysis is also subject to the effectiveness of the measures used to operationalize the chosen characteristics of corporate governance. Further, it is difficult if not impossible to determine ex ante the optimal time for a company to enter insolvency administration. At best, this is an ex post determination that benefits from hindsight, and therefore the interpretation of the patterns of performance to proxy for strategic choice is limited.
Footnotes
Funding
Thanks to the Accounting and Finance Association of Australia and New Zealand for funding the data collection for this project.