Corporate board vigilance and insolvency risk: a mediated moderation model of debt maturity and fixed collaterals

Board vigilance has two sub-constructs, namely board independence and CEO non-duality. There is an abundance of literature focusing on the relationship these two share with insolvency risk. However, most researches have used CEO duality instead of non-duality; yet, this doesn’t create any difficulty when interpreting the past literature. Three primary theoretical justifications explain the relationship between board independence and insolvency risk. According to the reputation hypothesis, the managers are advised by the autonomous boards to invest in less risky projects which reduce the chances of failure and save a firm reputation [Pathan, 2009]. The monitoring hypothesis also supports the argument that firms with more non-executive members onboard control the high risk-taking behavior of managers through continuous monitoring. Finally, the agency theory also advises in limiting the conflict between agents and principals by safeguarding the interests of shareholders [Brick and Chidambaran, 2008]. Undue risk-taking can put firms under distress and hence more independent boards prevent managers from indulging in actions which cause the firm to undergo considerable risk-taking. Karkowska and Acedański [2019] obtained data from 239 firms from 40 different countries to ascertain the role of board independence and risk and found a negative association between them. Chong et al. [2018] investigated the data of 290 public firms listed in Malaysia from 2010 to 2014 and they reported an inverse relationship between board independence and insolvency risk as well. Mathew et al. [2016] and Akbar et al. [2017] have also supported a negative association between risk and board independence among UK based firms. Berger et al. [2016] studied this relationship in US commercial banks and also confirmed the existence of a negative association between board independence and chances of firm failure. Rossi [2016] analyzed the impact of a number of independent members in a firm's board on profitability and risk based on data of 82 Italian firms. His reports of positive impact on profitability and adverse effects of risk were also consistent with the results of other studies.

Figure 1

Most of the past literature points toward the availability of a positive linkage between CEO duality and insolvency risk, which enables us to interpret that there should be a negative association between CEO non-duality and risk. Dong et al. [2017] reported a positive impact of CEO duality on risk and also captured a negative effect on firm efficiency in their analysis. Carmona et al. [2016] also discovered similar evidence by studying the link between CEO duality and risk disclosure. Daily and Dalton [1994] made a comparison between failed businesses and businesses in operation. These findings revealed that a higher percentage of businesses fail because the same individual holds the position of both CEO and chairman. There is also empirical evidence that establishes a negative association between CEO duality and insolvency risk; theoretically, this is justified through the stewardship theory, which postulates that a goal-oriented and enthusiastic person can bring more good results. Such a person is spared by procedural delays and unnecessary objections in making better decisions for organizational well-being. Recently, Felício et al. [2018] reported a negative association between CEO duality and firm risk. Armeanu et al. [2017] also supported these results as they found a negative relationship between CEO duality and failure risk, but their results were insignificant. Similarly, Boussaada and Labaronne [2015] analyzed the insignificant role of the duality while regressed on credit risk. Based on the literature discussed in this section, we propose the following hypothesis:

H₁=Vigilant boards reduce corporate insolvency risk.

There is a plethora of researchers who have established the relationship between board vigilance and financial leverage [Kieschnick and Moussawi, 2018; Aktas et al., 2019]. However, evidence on debt maturity in comparison to overall debt is scarce and needs to be studied more frequently. There is proven diversity among the results generated over the years related to debt usage. However, there are specific characteristics of boards which result in particular outcomes. A more independent board has consistently been ascertained to be capable of taking unbiased decisions on various problems as well as on deciding the debt maturity. The agency theory pushes the firm to opt for the non-duality of the CEO to avoid aggressive adventures of individuals who are vested with overly vested power. On the other hand, stewardship theory and resource dependency theory favor CEO duality. In this scenario, a determined CEO can work with complete freedom and without any interruption. Moreover, he can deploy organizational resources with ease to get results. However, the literature related to this theory disagrees with the concept of separation of duties at the top strategic level. Past research suggests that conservative boards accumulate more long-term debt in the corporate debt structure as it is less risky as compared to the short-term debt. However, more vigilant and informationally efficient boards have shown progress while using short-term debt. A corporate board is considered more useful if it has good negotiating power and better market relations that enable the firm to re-acquire short-term debt regularly and at reasonable terms [Harford et al., 2008].

The literature on one hand portrays the short-term debt as risky, but on the other hand considers it is a tool to discipline the managers. Superior skills are needed to handle the short-term debt to get excellent results because it enhances the chances of bankruptcy generally. Long-term debt, on the other hand, delays the probability of financial default but cannot be as rewarding as short-term debts. Recently, Tosun and Senbet [2020] confirmed the positive effects of an internal board monitoring mechanism. They observed a positive relationship between board independence and long-term debt and this evidence is found to be consistent in firms with CEO duality and with other robustness tests. Sheikh and Wang [2012] also made the same proposition regarding independent directors. Alves et al. [2015] stated that boards with more independence prefer short term leverage on retained profits, long-term debt over short-term debt, and equity financing over long-term debt. Huang et al. [2016] have stated that overconfident CEOs prefer more short-term debt. Shyu and Lee [2009] took data of 611 Taiwanese listed firms from 2002 to 2006 and observed a negative association between CEO duality and short-term debt by employing GMM regression. So, the resultant hypothesis can be constructed as:

H₂= Vigilant boards attract more long-term debt in leverage structure.

The available literature supports the notion that debt with longer maturities is less risky when compared to short-term debt. Cathcart et al. [2020] observed the impact of leverage on the default risk of large, medium-sized, and small enterprises. They observed a significant effect of leverage on default risk but it was more severe in SMEs due to their exposure to short-term financing. The difference between defaulted and survived firms lies in the maturity of their consumed debt. Della Seta et al. [2020] also opposed the view that short-term debt can be a value enhancer. They suggested that the rollover risk and other operational losses enhance the default risk due to the presence of high financial friction and fair debt pricing. Wang et al. [2020] confirmed the rollover risk hypothesis in US-based firms. They explained that an increase in short-term debt causes an increase in the cost of debt and credit risk eventually. However, this phenomenon was not severe in high growth firms. The asset substitution hypothesis is in contradiction to the associated rollover risk. Its proponents claim that short-term debt is closely monitored and negotiated. However, the majority of the recent researches are in agreement with the hypothesis concerning rollover effects. Chen and Duchin [2019] believed that unpaid high levels of debt with shorter maturities increase asset risk leading to distress as such firms invest more in financial securities.

Short-term debt puts the firms on run risk and long-term debt generates risk-shifting tendencies for the firms [Cheng and Milbradt, 2012]. He and Xiong [2012] supported the association of rollover risk and short-term debt, and the hypothesis that the probability of default intensifies since it causes conflict between shareholders and debtholders. Wang and Chiu [2019] confirmed the claims that short-term debt increases the chances of default. They analyzed the data of five Pacific Basin nations during 2002–2016. They also evidenced that it is highly undesirable for firms under financial difficulty to accumulate liquid reserves to cope with this situation. Adachi-Sato and Vithessonthi [2019] showed that an increase in debt maturity reduces the volatility in firm performance in the future.

H₃= Debt with long term maturity reduces corporate insolvency risk.

Past studies prove the importance of governance in corporations, especially in today's dynamic business environment [Wan and Ong, 2005; Kula and Tatoglu, 2006]. The research on the board's decision-making in performing their duties is still insufficient [Hasnah and Hasnah, 2009] especially in the context of risk avoidance. Past literature has mainly focused on the study of a direct influence of board attributes, structure, and composition, on corporate performance, financial leverage, and risk [Rohana et al., 2009; Noriza, 2010]. Research on board processes and decision making is necessary since the literature focusing only on board characteristics is insufficient to ensure adequate quality. The board process refers to the tactics adopted by board members to perform their duties and the reflection of company decisions [Macus, 2008]. The Board of directors has a monitoring role to play in governing the financial affairs of the company. So, they are supposed to observe intensely the decision making and performance, keeping in mind all relevant risk tolerance approaches [Jensen and Meckling, 1976]. There is recent evidence on the mediating role of financial leverage [Detthamronga et al., 2017; Ramli et al., 2018; Naseem et al., 2020], but there is not enough empirical evidence that supports the mediating role of debt maturity except financial leverage [Hussain et al., 2020b]. By focusing on the mediating role of debt maturity on growth and insolvency risk relationship, some studies have established the relationship between board independence and debt maturity [Sheikh and Wang, 2012; Alves et al., 2015] and between CEO duality and debt maturity [Ataullah et al., 2018; Tosun and Senbet, 2020]. There is also plenty of evidence that supports the relationship between debt maturity and insolvency risk [Javadi and Mollagholamali, 2018; Brancati and Macchiavelli, 2020]. Therefore, the current study proposes the following hypothesis to evaluate the mediating role of debt maturity on the relationship between board vigilance and insolvency risk.

H₄= Relationship between board vigilance and corporate insolvency risk is mediated by debt maturity.

There are several theoretical explanations for the moderating role of fixed assets that can act as potential collaterals in pursuit of financial leverage. Resource dependency theory explains that business assets are possible resources which can be used to perform business operations and generate profitability. Fixed assets do not act only as operational wheels but also can be used as a source of bringing liquidity to the business through loan acquisition [Lowe et al., 1994]. This scenario establishes a positive link between the presence of fixed assets and financial leverage [Alipour et al., 2015]. Plenty of research proves a negative impact of fixed collaterals on insolvency risk [Barton et al., 1989]. Daskalakis and Psillaki [2008] reported that firms that invest more in fixed assets feel less distressed. Recently, Alfaro et al. [2019] determined the importance of total assets in defining the association between financial leverage and financial fragility in emerging markets. They further insisted that large firms are more fragile and equally crucial for economic growth. Lee et al. [2011] studied the relationship between leverage and financial distress in the restaurant industry in the USA. This study found out the positive moderating role of the capital intensity measured through fixed assets to total assets ratio on the relationship between financial leverage and distress. So, the positive association between financial leverage and financial difficulty may be moderated by the volume of tangible assets possessed by firms, other things remaining constant. Joshi [2018] reported that firms with better risk management systems issue more debt and acquire more tangible assets. Increased debt causes more risk, but it is neutralized through better risk management practices. Firms having a higher debt component as compared with equity also exhibit better cash flows with stability in sales and profits. Recently, Xuezhou et al. [2020] analyzed the interaction of asset tangibility in Pakistan-based non-financial firms that are directly or indirectly associated with agriculture. They reported that tangible assets negatively moderate the relationship between leverage maturity and insolvency risk.

H₅= Fixed collaterals moderate the relationship between debt maturity and corporate insolvency risk.

A total of 369 non-financial firms were listed in the Pakistan Stock Exchange (PSX) at the end of the year 2017, which was the last year of the study period presently adopted. The detailed study sample comprising of 284 listed and non-financial firms of Pakistan is presented in Appendix A. These firms are divided into 14 different sectors and a fair representation of firms is given to each sector. The highest number of firms are from the textile sector i.e., 101 firms out of 136 firms. The distribution of firms in other sectors is comparatively small but within sectors, even the smallest contribution of firms is 50% of what belongs to other services activities, as can be seen in the percentage column of Appendix A. Further, the firms are included under sectors “Cement Sector and Motor Vehicles, Trailers & Autoparts” in the final sample as represented by 100% in the percentage column. The data for this study was extracted from multiple sources and made part of the sample based on several criteria. The five-year financial data (2013–2017) is obtained from the financial statement analysis (FSA) published by the State Bank of Pakistan (SBP) for the years 2012–2017. The data regarding board vigilance was manually extracted from the audited financial statements of individual firms in annual frequency available at the official website of the Pakistan Stock Exchange (PSX) and also the respective websites of the different firms. The criteria for exclusion of firms was as follows: (a) those firms with missing values were dropped, (b) firms not in operation during the study period were excluded, and (c) and firms with no return on assets (ROA) data for five years before the study period were also excluded because it was needed to calculate the volatility of the profits.

The primary purpose of this study is to capture the mediated moderation effect of potential fixed collaterals with debt maturity as mediators. We followed the footsteps of Cheng et al. [2019] and used the hierarchical multiple regression approach to detect the mediated and moderated role of debt maturity and fixed collaterals. This approach involves multiple regressions conducted stepwise on eight different models to prove our hypothesis. The following equations (2) and (3) will explain the final regressions in this eight-step process that incorporate all the study variables, but deal with the two independent variables separately.

The notation IR represents insolvency risk measured through the emerging market z-score. We used NDU in equation (2) that stands for our first independent variable non-duality of the CEO. In equation (3) we used the second independent variable known as board independence and represented by BI. The debt maturity ratio is shown as DMR and fixed collaterals stand as FC. The interaction term between debt maturity ratio and fixed collaterals is represented by DMR*FC. This study used six control variables namely size, taxes, firm growth, profitability, risk, and liquidity posed by SIZE, TAX, FG, ROA, σROA, and LIQ, respectively. We also controlled the time and industry effects in our regression models using year and industry dummies represented by DummyYE and DummyIE, respectively. The notation α is intercept or constant term and βi12 \beta _i^{12} represents our regression coefficients.

The study sample comprises 1420 observations for each variable, making it a balanced pool of panel data as reported in Table 1. This table reports the mean values of each variable with their standard deviations. Further, it also reports the correlations among study variables with their reported level of significance.

The mean value for insolvency risk is 9.96, which shows that on average, the sampled firms are in a safe zone revealing the low level of insolvency risk and shows a variation of 76.86 from this mean value. The mean value for board independence is 2.64 and its standard deviation is 0.24. CEO non-duality is a binary number and is not explained in descriptive terms. The data regarding debt maturity reveals that there is 27% long-term debt in the debt structure of selected firms and it can vary up to 21% across firms. However, it is found that the proportion of fixed assets that can be used as collaterals for loan acquisition is 55% of the total assets and it can vary up to 22%. This shows the disparity between the debt maturity and asset maturity structure of Pakistani firms. The reported correlation between debt maturity and fixed collaterals also shows a significant negative association between these two variables. This evidence is against the maturity matching principle according to which loan repayment period should match the asset maturity period. Instead, it depicts that firms in Pakistan are extensively using debt with short term maturity. Moreover, the correlations among all study variables are smaller which means there is no problem of multicollinearity among the data series used.

Table 2 presents the OLS regression results with robust standard errors by including CEO non-duality as the independent variable. We employed hierarchical regression which includes a set of variables under eight different models. The first two models consider debt maturity as the dependent variable and the next six models assume insolvency risk as a dependent.

Among control variables, size and taxes have an insignificant impact in all of the eight models, but they have positive coefficients for debt maturity as dependent variables and negative coefficients for insolvency risk as dependent variables. The firm growth has a significant positive impact on debt maturity, but its impact is negative and insignificant for all the remaining models which consider insolvency risk as a dependent. However, size, taxes, and growth have a negative but insignificant impact in the final model that has incorporated all the independent variables having coefficient values of −1.10897, −0.53800, and −0.28125, respectively. The remaining control variables have a significant impact on insolvency risk. Profitability reduces insolvency risk, as it has a positive impact (β = 0.30508**). The liquidity also reduces insolvency risk (β = 1.13018***), and risk impact insolvency risk adversely (β = −0.43472**).

CEO non-duality has a significant negative impact on the debt maturity ratio (β = −0.02763*), but it consistently had a positive but insignificant impact on insolvency risk (β = 3.77374). Debt maturity has a significant positive impact on insolvency risk which means that firms with more long-term financing reduce insolvency risk to listed firms in Pakistan. The significant impact of non-duality on debt maturity and significant impact of debt maturity on insolvency risk completes the indirect path of CEO non-duality to insolvency risk through debt maturity. Further, it confirms the presence of full mediation between CEO non-duality and insolvency risk using hierarchical regression.

The interaction term DMR*FC is introduced to capture the moderating effects of fixed collaterals on the relationship between debt maturity ratio and insolvency risk. The OLS regression results with robust standard errors prove a negative moderating effect of fixed collaterals (β = −140.6002***). As suggested by previous literature, the regression results provide new evidence that in an emerging economy a business with more fixed assets which can be used as collaterals should have low insolvency risk. This new evidence suggests that potential fixed collaterals increase insolvency risk in the listed firms of Pakistan. It confirms the results generated by Xuezhou et al. [2020], which suggest that Pakistan firms do not use tangible assets to their advantage and they accumulate non-productive assets that have more maintenance and associated fixed costs. This scenario makes fixed collaterals a risk enhancing factor when compared to the other countries where they caused a reduction in insolvency risk [Lee et al., 2011].

Table 3 reports board independence as an independent variable while other things remain the same. The impact of size, taxes, profitability, and liquidity on debt maturity is insignificant, but it is negatively correlated to profitability and liquidity. Growth is a significant positive predictor of debt maturity whereas risk is a substantial negative predictor of debt maturity in the listed firms in Pakistan. Size, taxes, and growth proved to have an insignificant negative impact on insolvency risk. Results for profitability, risk, and liquidity were significant and only volatility of returns had a negative influence on insolvency risk among these three. Board size, our second independent variable, turns out to be the significant positive (β = 0.07101***) influencer of the debt maturity as reported under model 2.

Further, debt maturity had a significant favorable influence on emerging market z-score (β = 135.3171***) showing a reduction in insolvency risk. Further, it also completes the indirect impact of board independence on insolvency risk via debt maturity and once again confirms the mediating role of debt maturity in this relationship. Fixed collaterals once again negatively moderated the impact of leverage maturity structure on insolvency risk (β = −140.8334***). This evidence once again confirms that non-productive fixed assets with high associated costs increase insolvency risk in Pakistan firms.

We also employed panel corrected standard errors (PCSE) regression as an alternative, to make our analysis more rigorous and robust. The PCSE regression calculates superior standard error estimates efficiently and hence generates correct standard errors of coefficients [Reed and Webb, 2010]. It creates robust standard errors that not only controls for heteroscedasticity, but also for serial correlations [Bailey and Katz, 2011]. According to Marques et al. [2016], the PCSE estimator is also highly suitable when the number of cross-sections is greater than the time intervals i.e., (N > T).

Table 4 reports the hierarchical PCSE regression results by considering CEO non-duality as the independent variable. The firm size, taxes, and firm growth had a positive, but insignificant impact on the debt maturity ratio. The profitability, volatility of returns, and liquidity showed a negative effect on the debt maturity ratio, but coefficient values for the volatility of returns were significant. For insolvency risk, the firm size and volatility of returns showed significant negative coefficients, but profitability and liquidity displayed significant positive beta coefficients. The influence of taxes and firm growth on insolvency risk was insignificant at a 5% level of confidence.

The CEO non-duality had a significant negative impact on the debt maturity ratio, having a beta value of −0.02763 and a p-value of less than 0.01. The effect of debt maturity ratio on insolvency risk is also significant but positive (β = 134.6859***) confirming the indirect path coefficients through debt maturity. Interestingly, the direct impact of CEO non-duality on insolvency risk also tested to be positive and significant (β = 3.77374**), hence proving the partial mediating role of debt maturity ratio using PCSE estimates. The moderating role of fixed collaterals between leverage maturity and insolvency risk once again turned out to be significantly negative proving our previous observation of association of non-productive assets with high costs (β = −140.6002***).

Table 5 is also based on PCSE regression estimates, but it uses board independence as an independent variable. Similar results as compared to Table 4 are evident for the control variable in this case also i.e., the significant negative role of the volatility of returns in defining debt maturity ratio at a 1% level of significance. Similarly significant negative impact of size and volatility of returns on insolvency risk are observed, but there is significant positive impact of profitability and liquidity on insolvency risk and insignificant negative impact of taxes and growth on insolvency risk in this case also. The board's independence had a significant positive impact on the debt maturity ratio confirming the predictions OLS regression and (β = 0.07101***). The debt maturity also had a significant positive impact on insolvency risk (β = 135.3171***). The direct impact of board independence on insolvency risk is insignificant, referring to the full mediation in this case. The results for the interacting effect of fixed collaterals consistently proved to be significantly negative (β = −140.8334***) in this scenario also.

The board's independence had a significant positive impact on the debt maturity ratio, confirming the predictions of OLS regression and (β = 0.07101***). The debt maturity also had a significant positive impact on insolvency risk (β = 135.3171***). The direct effect of board independence on insolvency risk is insignificant, referring to the full mediation in this case. The results for the interacting effect of fixed collaterals consistently proved to be significantly negative (β = −140.8334***) in this scenario also.

Table 6 presents the summarized mediated moderation results using both OLS with robust standard errors and PCSE estimators. It is easy to compare the results in this table to arrive at the inference that the fixed collaterals have negatively moderated the mediating role of the debt maturity ratio in all of the four independent regression scenarios. Full mediation is detected by the debt maturity ratio on the relationship between board vigilance and insolvency risk even if it is negatively moderated by fixed collaterals as reported in the OLS regression results. However, partial mediation is detected for CEO non-duality and insolvency risk relationships, but the results are unchanged as far as the moderating role of fixed collaterals is concerned. There is no difference in results when we assess the intervention of debt maturity ratio on board independence and insolvency risk relationship; further, the negative moderating effect of fixed collaterals confirm the previous results.