MULTIPLE DIRECTORSHIPS AND RELATED PARTIES TRANSACTIONS: THE WEAKNESS OF NUMBERS

We examine whether the presence of outside directors with multiple boards seats influence firms related-parties transactions. These non-executive directors with vast skill, experience, knowledge, prestige and shared networks are part of major Boards Committees responsible for key corporate policies, strategy and management. Subsequently, it remains an untested assumption whether the outcome of related parties’ transactions are influenced by the presence of this class of ‘busy directors’. We obtain data from 142 companies across five sectors between 2009 and 2014 and conduct analysis using a two-stage multiple regression. The results reveal that the existence of multiple directorships on boards failed to evolve as predictors for related parties’ transactions. We thus conclude that the presence of these ‘busy directors’ on boards does not alter a firm’s related parties’ transactions significantly. Overall, this may suggest that the influence of executive directors and other outside directors and key management personnel do play an important role in explaining organizationally complex strategic decisions in this regard.


INTRODUCTION
Prior literature on corporate boards highlights a number of the most important features of boards of directors, such as board composition and board size. A substantial body of evidence detailing the potential advisory benefits associated with experienced directors who serve on multiple boards. There is also extensive research on the limit of multiple directorships allowable by publicly traded firms (see, for example, Holmstrom and Milgrom, 1991;Fich and Shivdasani, 2007). Thus far, however, corporate governance research has not explored why and how the presence of multiple directorships on boards, which has attracted a great deal of attention, affects industryspecific strategies. In this study, we focus on this interesting class of directors whose unique characteristics can either enhance or weaken strategic boards' decision making. Specifically, we examine multiple directorships (MuDs) in Nigerian corporations. We present an empirical evidence on the relationship between MuDs and related parties transactions (RpTs). The literature on multiple directorship in corporate governance has dramatically strengthened in recent years due to a number of propositions. First, the "Reputation Hypothesis" (Fama and Jensen, 1983) which signal director reputation suggests that directors with more outside board seats contribute a valuable set of resources and bring experience to the firm, thus, have greater influence on corporate decision. These outside directors comprise affiliated and grey directors constituting about 10 percent of board membership (Hermalin and Weisbach, 2003). They often have potential or existing business ties with the firms, either as former manager, solicitors, business consultants, commercial or investment bankers and insurance executives. By virtue of being more networked, they can influence the direction of related parties transactions to third party companies (in which they also hold seats) because their holding of multiple board seats is necessitated by being good as directors (Jiraporn et al., 2008). If this is the case, an increase in their proportion may have implications for the structure of corporate third parties related transactions portfolio. As such we predict that multiple directorships on corporate boards could evolve as a good predictor for third parties related transactions. The "Busyness Hypothesis", however, holds that directors with too many outside board appointments have limited attention, capacities, and time constraints and are less likely to monitor managers effectively (Ferris et al., 2003). In Nigeria, more of these directors are less likely to engage in key operational decision. As such, we contend that their presence is unlikely to have significant impact on relatedparties transactions, frequently classified in annual reports as "related parties".
There are mixed empirical evidence on the theoretical opinion of the "busyness hypothesis". These literature suggests different directions of causality in the relation between multiple directorships and firm performance. One line of literature argues that multiple directorships can increase the chances of corporate fraud (Beasley, 1996), lead to excessive CEOs remuneration (Core et al., 1999), and ultimately affect firm's performance adversely (Fich and Shivdasani, 2006). A variant of this argument is that multiple directorships commit to more transparent maximization of shareholders wealth through the offering of large premiums in tender offers (Cotter et al., 1997) and superior returns from acquisitions (Brown and Maloney, 1999). These suggestions however remain inconclusive as others find no relations between multiple directorships and firm performance (see, for example, Kiel and Nicholson, 2006). Arguably, one of the most important roles of the board is the decision to retain or replace top managers, which may be influenced by the desire to secure existing business ties. Accordingly, directors with multiple outside boards' appointments might favor internal management successors through whom they can sustain current business arrangements, even when such candidates are less qualified replacements (Borokhovich et al., 2014). To this end, we analyze how and to what extent multiple boards' directorships and multiple committee memberships on boards influence related parties transactions.
The current study offers a thorough examination of the nature of multiple directorships influence in the light of multiplicity of codes of corporate governance in Nigeria. Specifically, we emphasize that the role of outside directors is to evaluate rather than facilitate related parties transactions. To the knowledge of the authors, it is the foremost empirical study documenting the incentives of multiple directorships alongside related parties transactions across quoted companies in Nigeria. A better understanding of the motives of these connected board membership is important because the section on "related parties transactions" is increasingly becoming a part of annual report and accounts, given the recent direction and attention of codes of corporate governance in general, and in the light of the call for increased disclosure by listed entities in particular. Second, we provide evidence to suggest that board members with multiple director appointments should be classified separately from other outside directors in the annual reports. This is particularly essential, given the continous emphasis on directors' independence.
The remainder of the current study proceeds as follows. Section 2 discusses multiple directors incentives and related literature. Section 3 describes empirical method and data. Section 4 presents the results while the fifth section concludes the study.

RELATED LITERATURE AND HYPOTHESIS
The concept of multiple directorships in corporate governance is far from a new issue. There is a well established stream of research devoted to the analysis of multiple directorships. It has long been a matter of interest to stakeholders in academics and practice. Proof of this is the intense literature developed in this area and we refer below.

Prior evidence on multiple directorships
Multiple directorships is a crucial issue in corporate governance. We connect two separate literature that examines the relations between multiple directorships and firm performance and the relations with influence on corporate strategy. Early contributions point to a picture with value-enhancing capabilitties of multiple directorships. Outside directors on multiple boards established through networks and business relations (Mace, 1986) have exposure to learning different management styles and strategies that influence their decision making skills (Booth and Deli, 1996). Serving on multiple boards also signal director competence (Kaplan and Reischus, 1990) and quality (Brickley et al., 1994). Interestingly and contrary to a good number of findings, directors who serve on multiple boards also present firms with lower probability of receiving qualified audit opinion and are better negotiators for CEOs and external auditors remuneration (Méndez et al., 2015).
In contrast to the above findings some studies however notes the negative effect of overcommitment when holding multiple board seats (Core et al., 1999;Fich and Shivdasani, 2006;Falato et al., 2014). For example, Fich and Shivdasani (2006) report that representation on multiple boards result in weak governance, lowers profitability, reduces market-to-book ratios and lower sensitivity of CEO turnover to firm performance. Contrary to Sarkar and Sarkar (2009), such directors are often absent from board meetings (Jiraporn et al., 2009). In addition, their "busyness" encourages them to recommend excessive CEO remuneration as monitoring decreases, serving as an incentive for earnings management in addition to increased agency costs (Faleye et al., 2011). The main finding of Falato et al. (2014) is that multiple board seats weaken their monitoring quality and erodes shareholder value. In the seminal paper, the authors report that the adverse effect of attention shocks on firm value persists over time leading to reduced board monitoring, i.e., higher CEO rent extraction and reduced earnings quality. In a related empirical analysis presented by Ahn et al. (2010), the authors note that firms with busy directors allow valuedestroying acquisitions, although this adverse effect does not, nonetheless extend across the entire range of multiple directorships examined.

Corporate governance regulation in Nigeria
In addition to individual industry guidelines, listed firms regulation in Nigeria are, of course primarily imposed by the Securities and Exchange Commission (SEC) and the Nigerian Stock Exchange (NSE). In response to the global corporate governance crisis in the early 2000s, many stock exchanges and securities regulators around the world issued coporate governance codes. Nigeria was no exception.
The "Code of Corporate Governance for Listed Companies" was issues in 2003, by SEC. The codes share similarity with most others around the world, serving as mere guiding principles, rather than explicit regulations. Further, subsequent reforms into corporate governance in Nigeria saw the emergence of other codes, often times conflicting and encouraging noncompliance (Adegbite, 2014

Corporate governance and regulation of multiple directorships
The Nigerian Securities and Exchange Commission (SEC) code of corporate governance for public companie makes no specific disclosure on the limit of concurrent directorships a company may hold (Securities and Exchange Commission, 2008a). According to the revised guidelines (Securities and Exchange Commission, 2008b, p. 8): "There shall be no limit on the number of concurrent directorships a director of a company may hold. However, concurrent service on too many boards may interfere with an individual's ability to discharge his responsibilities. The Board and the shareholders should therefore give careful consideration to other obligations and commitments of nominees in assessing their suitabilityfor appointment into the board." In developed markets however, a large number of publicly listed firms have consistently limited the number of multiple directorships held by board members (Falato et al., 2014). These measures may be associated with reasons highlighted in earlier section. Banks are vital institutions in any society as they contribute significantly to economic development through facilitation of business (Moi, 2014). Given its role, prestige and greater visibility, bank directors are highly likely to obtain additional directorships than others. Consistent with this perspective, directors are less likely to resign from high performing firms with high prestige and greater rewards (Fahlenbrach et al., 2010), even when they already hold other outside board appointments.
There is a long history of economic liberalization strategy in Nigeria giving rise to four main groupings of corporate enterprising (Ahunwan, 2002). These corporations, including publicly listed entities are characterized by weak and unimpressive corporate governance. Adegbite (2014) summarizes the rationale behind the poor state, hinged on weak institutional and regulatory climate: weak board governance; weak executive monitoring and accountability; corporate (private) corruption; and publicprivate corruption. Although "there are not that many highly experienced executives, such that you have to appoint the same people on different boards" (Adegbite, 2014), the author argue that placing a limit on multiple directorships may hinder board performance. While this may sound plausible, the behaviour of directors with existing or potential business affiliations with the banks they govern becomes increasingly "interest-centered" as they hold other boards seat. Moreso, as those other firms may belong to other industries and essentially active bank clients who continously seek credit facilities for operational and strategic reasons.

Related parties transactions
International Accounting Standards (IAS) 24: Related Party Disclosures states that "a related party transaction is a transfer of resources, services or obligations between a reporting entity and a related party, regardless of whether a price is charged". This standard also describe a related party as any person(s) and/or entity(ies) that is related to the reporting entity. This definition include any such person(s) or group of persons who may be expected to influence, or be influenced by, that person in their dealings with the entity. The membership of the board of directors expressly and impliedly qualify as related parties within this context. In particular, the standard notes that: "… if an entity has had related party transactions during the periods covered by the financial statements, it shall disclose the nature of the related party relationship as well as information about those transactions and outstanding balances, including commitments, necessary for users to understand the potential effect of the relationship on the financial statements.
In particular, such entity shall disclose at a minimum, the amount of the transactions; the amount of outstanding balances, including commitments (their terms and conditions, including whether they are secured, and thenature of theconsideration to be provided in settlement; and details of any guarantess given or received); provisions for doubtful debts related to the amount of oytstanding balances; and the expense recognized during the period in respect of bad or doubtful debts due from related parties." (International Accounting Standards Board, 2012) In Nigeria, disclosure on the consideration of related parties include the parent company, directors, their close family members and any employee who is able to exert a significant influence on the operating policies of the company. Hence, we expect directors with multiple boards seats to fall within this category.

Sample selection and data
In order to test the hypothesis proposed, our sample selection starts with data from the Nigerian listed companies excluding service companies. Service companies were dropped, due to the differences in regulatory standards, financial reporting standards and compliance requirements (Manzaneque and Chamisa, 2008;Muller and Whiteman, 2008). The study period is between 2009 to 2014, a period which coincide with the revision of the code of corporate governance for listed companies, Second, we found more companies being experiencing more regulatory sanctions. We exclude completely, foreign and state-owned banks for two reasons. First, the motivation for third parties transaction in government banks is complex. Second, related parties transactions in foreign banks is likely to be motivated by other factors, regardless of the existence of multiple boards representation. In specific instances, we also exclude some companies based on the initial data obtained, when it was discovered that in some cases such companies do not make specific disclosure of the class of directors. Consequently, we draw from information made available through official information disclosed on individual websites and press releases. The final sample is made of 142 companies across five sectors (namely consumer goods, healthcare, industrial goods, oiland gas, and natural resources), representing 86 per cent of non service firms listed in the NSE daily summary (Hermalin and Weisbach, 2003).

Dependent variables
In line with extant studies (see, for example, Bikker and Metzemakers, 2005;Fonseca and Gonzalez, 2008;Kanagaretnam et al., 2009), we made use of factor-proportions. Accordingly, we introduce three elements as indepedndent variables: related parties transactions (RpTs), the ratio of RpTs to assets and the ratio of RpTs to profits. We note that RpTs include exposure to all categories, i.e., key management personnel having authority and responsibility for planning, directing and controlling the activities of the enties, directly oridirectly (whether executive or otherwise). We use the dummy variable RpTs to measure the existence of this class of transaction during each accounting period. If it does exist, RpTs is coded one, otherwise zero. The ratio of RpTs is defined as the amount the company reports as transactions in which these group ofpersons has direct or indirect interest in each financial year. While controlling for directors tenure (subject to reelection, exit and replacement), we use two ratios of director-related transactions measures, where the ratio of RpTs to assets equals the log-transformed (ratio of naira value of related parties transactions to total assets × 100 + 1) and ratio of RpTs to profit equals the logtransformed (ratio of naira value of related parties transactions to net profits + 1).

Independent variable
We use the ratio of directors with multiple board seats as the measure of MuDs. This is consistent with prior studies in the use of magnitudes (see, for example, Adams and Ferreira, 2009;Campbell and Mínguez-Vera, 2013).

Control variables
We include a number of control variables in the analysis. The ratio of Director RpTs to net current assets is measured to control for corporate exposure to directors. Director RpTs is specificed in the model as a control variable because it is found to have direct relationship with related parties transactions due to its size relative to other class of RpTs in individual sectors. We also include relationship type to control for director-class effects. If a director RpTs is not related to outside directors, the type is coded one, otherwise zero. Logarithm of (ratio of naira value of RpTs to total assets × 100 + 1). ***

Ratio of RpTs to profits
Logarithm of (ratio of naira value of RpTs to net profit + 1). ***

Ratio of MuDs
The proportion of multiple directorships on Boards of directors. Brickley et al., 1994 Endogeneity Board experience Dummy variable that equals 1 if the number of years as a director prior to the current firm is in the top quartile of the sample distribution of director industry tenure [available only for a sub-sample of hand-matched directors whose complete employment histories were disclosed on Boards profiles).

Endogeneity correction
In the current study, we anticipate that MuDs on boards have a negative association with companies' RpTs. However, it is probable that regulatory requirements for board positions in terms of experience, knowledge, skill and expertise influence the pattern of MuDs. If a director has more Board experience, chances exists that he gets multiple appointments. Further more, if a director has a good relationship networking with executives in the industry or elsewhere, then he will have a chance of gaining multiple board seats. To control for potential endogeneity, we introduce outside directors cummulative experience and networking as instrumental variables. This is included to lend credence to the requirement of efficient instrumental variables. Tab. 1 presents definitions and sources of data. Our analysis is conducted in two stages. First, we model the influence of instrumental variables on the ratio of MuDs. To test the hypothesis, we employ ordinary regressions. To model the probability that a board is expected to comprise of directors with multiple seats, the endogeneity correction variables are included and a binomial logistic regression is used. Further, Tobit regression models is used to analyze ratios of specific values. Given that a large number of the dependent variable of interest, RpTs is only observable under certain conditions, i.e. not all RpTs is traceable to outside directors, the censorship regression is appropriate.

EMPIRICAL ANALYSIS
The descriptive statistics for variables used in the empirical analysis are presented in Tab. 2. It can be noted from the descriptive statistics in Tab. 2 that variables from the firms included in the sample are very heterogenous. Tab. 3 reports the regression results indicating the effectiveness of the instrumental variables. It can be noted from the table that all of the instrumental variables are significantly related to the ratio of MuDs. As described earlier, we control for endogeneity using a two-stage regression. We create a variable to measure the residual of the ratio of MuDs on the board, and include this in the following regression models.
We performed a two-step regression model and the results for RpTs revealed in Tab. 4. The significant coefficients of the residual of the ratio of MuDs confirm the endogeneity problem for MuDs, hence the basis for the twostage regression method. The second-stage regression includes model 1, 2 and 3. The control and independent variables are includedin the dependent variables RpTs, the ratio of RpTs to assets and the ratio of RpTs to profits. Thus, when model 1 predicts RpTs probability, the coefficient for the ratio of MuDs returns significant (β = −8.217, p < 0.05). Further, when models 2 and 3 predicts the RpTs ratio, the coefficients are both significant at the ratio of RpTs to assets (β = −29.561, p < 0.10) and the ratio of RpTs to profits (β = −27.384, p < 0.10) respectively for both model. These results support our Hypothesis.

DISCUSSION
We examine the implications of multiple directorships on related parties transactions across 142 companies in the real sector. We use a 1,216 firm year observations with specific disclosures on various related parties transactions with companies whose directors are also directors in the respective firms, in particular, at values and terms comparable to other related parties trsnasctions across all the sectors. Previous studies has not studied the relationship between these variables or considered the ways in which individual firm characteristics moderate the existence of outside directors with multiple board seats. Using a data sets covering the period 2009 to 2014, the current study provides convincing evidence that RpTs is negatively related to the ratio of MuDs. We can conclude that the existence of multiple directorships on boards do not influence corporate related parties transactions.
The study makes several contributions to the literature on corporate governance. First it extends the corporate governance literature that multiple boards directorships affect related parties transactions. We promote an increased understanding of the impact of director-class, director-related transactions, social networking, and experience on related parties transactions. Second, our study depicts an understanding of multiple directorships function as evaluators of the agency costs, which challenges the notion that the precense of outside directors with multiple board seats increases the adverse effect of attention shocks on firm value and persists over time leading to reduced board monitoring, i.e., higher CEO rent extraction and reduced earnings quality.
The results are also nuanced by our findings that the relation between multiple board seats holding and director related transactions is influenced by whether the firm: (i) has high director related transactions; (ii) has outside directors with fairly long cummulative directorship experience; and (iii) has gray directors on board with expanded cross-industry clouts network. Notes: a Logarithm; * p < 0.1 for tw-tailed tests; *** p < 0.01 for two-tailed tests.

Source: SPSS Output
Our study also supports the presumed, but generally untested assumption that is common in corporate risk management literature that directors with multiple boards representation can and do influence related parties transactions.
Overall these results provide an empirical estimate of the extent of positive contributions that outside directors bear for corporate risk management. These results are useful for outside directors to assess their extent of exposure, given that academic researchers continually follow the trend of corporate strategies and departure from disclosure, transparency and accountability. While the proportion of multiple directorships is small compared with the overall population of directors, individual outside directors who occupy multiple seats can weigh their risks differently. Finally, our results serves as a policy directive on the influence played of multiple boards in corporate strategy.
Our research has some noteworthy implications for firm leaders and policmakers. The latter should be aware that outside directors on supervisory roles with multiple boards seats may not have a significant influence on a firm's related parties transactions. The incumbent management, other outside directors as well as other executive directors might therefore Notes: a Logarithm; * p < 0.1 for two-tailed tests; ** p < 0.05 for two-tailed tests; *** p < 0.01 for two-tailed tests.
benefit from an intensified exchange of business transactions with their former employers and other firms with which they are networked. Policy makers should consider that corporate networks via board connections might provide an effective mechanism to allocate scare specific related parties transactions in favour of executive directors than do outside directors with multiple boards seats.
A justification for the regulation of simultaneous board memberships in the new Nigerian corporate governance framework might been seen in potential positive effects on public firms. At the sametime these outside directors seem to provde an efffective mechanism to transfer external knowledge to other supervisory board members.