The Impact of CEO Tenure on Cooperative Governance

MANAGERIAL AND DECISION ECONOMICS

Manage. Decis. Econ. 34: 218–229 (2013)

Published online 2 January 2013 in Wiley Online Library

The Impact of CEO Tenure on CooperativeGovernance

Michael L. Cook* and Molly J. Burress

University of Missouri, Columbia, MO, USA

(wileyonlinelibrary.com) DOI: 10.1002/mde.2585

*CorrespondenE-mail: Cook

Copyright ©

This paper investigates whether long-tenured cooperative chief executive officers (CEOs)are successful in negotiating less monitoring, resulting in the cooperative being agent driven.Utilizing a sample of the 1000 largest US agricultural cooperatives, we examine whetherboards of long-tenured CEOs exhibit differences in composition, formal committees, orprocedures that may indicate these boards are more lenient monitors. We find long-tenuredCEOs experience less board monitoring. This result is primarily due to a difference inprocedural mechanisms, rather than board composition. However, it is unclear whethermonitoring leniency is an indication of the CEO’s ability to negotiate less monitoring. Itremains a possibility that CEOs with shorter tenures also influence the board; but theirrecommendations may be heavily influenced by non-compulsory conformance with strictercorporate governance regulations. Copyright © 2013 John Wiley & Sons, Ltd.

INTRODUCTION

Principals, as represented by boards of directors, playan important role in monitoring agents in the corporategovernance literature. Agency arguments suggest thedirection of control optimally flows from board tochief executive officer (CEO) agent. Simultaneously,there is a growing literature that suggests the flow runsin the opposite direction (Bebchuk and Fried, 2003;Adams et al., 2010; Dey and Liu, 2011), thus support-ing the work of Mace (1971). Mace rejected the ideathat boards can be an effective governing force incorporate control problems and suggested directorsare co-opted and dominated by the CEO. In this paper,we probe this potential paradox as it relates to cooper-ative governance.

We begin by reviewing corporate governanceframeworks employed to inform this query, and then

ce to: University of Missouri, Columbia, MO, [email protected]

2013 John Wiley & Sons, Ltd.

move the discussion to exploring how previousresearch informs what we know about cooperativegovernance. We are particularly interested in evaluatingsuppositions regarding agent-driven organizationsmade or implied in the work of Fulton and Larson(2009), Bijman (2005), Craig (1993), and Hind (1999).Fulton and Larson (2009) emphasize agent dominanceoccurs because of CEO hubris and ineffective boardoversight in the face of excessive diversification andcomplexity. Bijman (2005) indicates member heteroge-neity paralyzes the board, shifting control from theboard to management. Craig (1993) groups cooperativestakeholders into conflicting subtaxonomies that destroythe cooperative from within over time. This dynamic ishypothesized to result in agent control as membersubgroups battle internally. Hind (1999), on the basisof limited but in-depth analysis of 10 cooperative cases,concludes that cooperatives progress through a life cycleduring which member heterogeneity increases.Increases in member heterogeneity render member goalcongruence more difficult to achieve, allowing for the

CEO TENURE AND COOPERATIVE GOVERNANCE 219

aspirations of their agents to be more easily met. Thus,organizational age of the cooperative may be an indica-tor of the degree of agent control.

In contrast, corporate governance scholars suggesta shift from principal to agent control may be moreclosely related to the duration of CEO tenure ratherthan organizational age. Dey and Liu (2011) show,empirically, that CEOs are successful in negotiatingless board oversight over the duration of their tenure.Hermalin and Weisbach (1998) propose CEOs maybe successful in negotiating less oversight through adecrease in board independence.

Given evidence generated from corporate gover-nance findings, we seek to examine (i) whether cooper-ative data supports the notion that long-tenured CEOsare able to negotiate less monitoring and (ii) if a shiftin control is observed, what mechanisms are employedto facilitate a shift toward decreased monitoring activity.We introduce results of empirical work derived from asurvey of the largest 1000 agricultural cooperatives inthe USA. To the best of our knowledge, our study isthe first to use survey data to examine whether and bywhat mechanisms the cooperative board may decreasetheir monitoring activities, leaving the organization atrisk for becoming agent driven.

Testing whether board governance reveals a differ-ence in agent control within the cooperative may giveus greater insight regarding the relative power of theCEO position in the cooperative business structure.Institutionally, there does not appear to be a formallysanctioned mechanism by which CEOs are grantedgreater control over the duration of their tenure withthe cooperative. For example, cooperative CEOs arenot typically granted voting rights on the board or nom-inating rights for new director candidates. In addition,cooperative CEOs very rarely occupy the position ofchairperson of the board. Therefore, we are particularlyinterested to determine which, if any, mechanisms mayallude to a method by which long-tenured CEOs wieldgreater control in governance. A lack of significantassociation between CEO tenure and board compositionor procedure would suggest cooperative CEOs mayhave relatively less influence over their boards ofdirectors than their corporate counterparts. This findingmay be consistent with the dynamics of representationaldemocratic boards: Management tends to have minimalopportunity to shape the board or influence its composi-tion. On the other hand, a significant associationbetween CEO tenure and board composition mayindicate the cooperative is at greater risk of becomingagent driven as CEO tenure increases. The same holdsfor procedure.

Copyright © 2013 John Wiley & Sons, Ltd.

THEORY

To investigate the impact of CEO tenure on coopera-tive governance, we develop hypotheses derived fromcorporate governance research. We rely primarily ontwo predominant theories used in corporate gover-nance research: agency theory and stewardship theory(Hung, 1998).

Agency theory suggests owners may possesspreferences distinct from those of their agents (Jensenand Meckling, 1976). From an agency perspective, theboard’s primary role, as fiduciary, is to monitor andcontrol agents. Boards possessing independence fromthe CEO are viewed as optimal. Active, effectivemonitoring ensures agents will act in the best interestsof shareholders and improves company performance.Insufficient monitoring may offer agents more free-dom to pursue selfish objectives.

Agency theorists suggest long-tenured CEOs maybecome entrenched for three reasons (Hill and Phan,1991; Shen, 2003). First, CEO influence may increasewith tenure as a result of a good performance trackrecord. Second, long-tenured CEOs may impact boardcomposition. A long-tenured CEO may influence thenomination of a greater number of incoming boardmembers, resulting in a board that is loyal and sympa-thetic to the CEO. Third, CEOs may increase theirrelative power as they gain control over procedureand internal information systems. Control over infor-mation systems and procedures may afford CEOs theability to withhold relevant information or influencethe board agenda.

Another agency perspective highlights the costsand benefits of decentralization (Aghion and Tirole,1997; Baker et al., 1999). More lenient board controlmay strengthen CEO entrepreneurship and lead tomore innovation and/or growth of the cooperativefirm. These authors argue that formal authority is morelikely to be delegated for decisions that (amongothers) are sufficiently innovative that the principalhas not accumulated substantial prior expertise orcompetencies.

Stewardship theory suggests agents are allegiant tothe organization, genuinely seeking to further organiza-tional goals rather than private motives (Muth andDonaldson, 1998). Therefore, the board’s role is toempower managers. The board empowers managementby lending technical expertise and insider knowledgeregarding operations. Thus, insider boards that arecommitted to the firm are optimal. Insider boards arehypothesized to be associated with greater effectivenessbecause of their superior knowledge of the firm. In

Manage. Decis. Econ. 34: 218–229 (2013)DOI: 10.1002/mde

M. L. COOK AND M. J. BURRESS220

contrast, independent boards may have no incentive toensure organizational success and inadequate knowledgeof the firm. Therefore, independent boards may be morelikely to utilize their position to enhance their owninterests to the detriment of the firm and shareholders.

With regard to CEO tenure, stewardship theoristspropose long-tenured CEOs may have greater commit-ment and firm-specific expertise, leading to enhancedperformance with respect to their duties. In addition,long-tenured CEOs who have dedicated their careerto shaping the firm and its strategy may identify theirpersonal success and satisfaction with the success ofthe firm. Donaldson and Davis (1991) suggest longtenure promotes a ‘merging of individual ego and thecorporation, thus melding individual self-esteem withcorporate prestige’.

In the cooperative setting, researchers claim manage-rial entrenchment is of particular concern (Bijman,2005; Hind, 1999). Cooperatives judged as havingbecome dominated by management are often referred toas agent driven or staff centered (Murray, 1983). How-ever, most research on agent-driven cooperatives is theo-retical or supported by individual case studies (Fultonand Larson, 2009; Hind, 1997). There is little empiricalresearch to determine to what extent cooperatives areagent driven. Furthermore, if long-tenured CEOs areindeed entrenched, it is not clear which mechanismsmanagement has used to effectively negotiate greatercontrol of the cooperative organization. In addition tomaintaining a superior track record, agency theoristsargue CEOs may negotiate greater control by influencingboard composition or procedure (Hill and Phan, 1991).

Board Composition

Do cooperative CEOs have as many opportunities toinfluence board composition as their corporate counter-parts? Although corporate CEOs have historically beeninvolved in nominating new directors (Hill and Phan,1991; Shivdasani and Yermack, 1999), disclosure andindependence requirements as dictated by NASDAQand the Securities and Exchange Commission havecurtailed CEO involvement in director nomination(Chhaochharia and Grinstein, 2007). Spencer Stuart(2010) reports the CEO is the source of 12% of corpo-rate board member nominations.

A cooperative CEO may influence the compositionof the board by suggesting new candidates to membersof the board, encouraging specific members to run forthe board, weighing in on nomination criteria, managingenrichment programs for the development of potentialboard members, or suggesting candidates to serve as


outsider directors or in an advisory capacity. Despitethese informal avenues, there is little evidence tosupport the general proposition that cooperative CEOsinfluence board composition. Cooperative managementis unlikely to play a dominant, formal role in nominatingboard members for election.

Table 1 reviews agency theory expectations regard-ing corporate governance constructs and the associ-ated degree of monitoring. From an agency perspective,we might hypothesize CEOs would strive to negotiateless monitoring during their tenure. Although the abilityto include outside directors on the cooperative boardmay be limited by statutes and bylaws, we would expectfewer financial experts and outside directors serving onthe board of those firms with entrenched, long-tenuredCEOs. This would result in a board with less indepen-dence. Director tenure and age are more difficult tointerpret. On the one hand, we may expect older boardmembers with long tenures to become complacent,acting as a rubber stamp and allowing management togain control of the organization (Cornforth, 2004).However, a long-tenured board member is not likelyto have had their nomination influenced by a CEOwho did not hold office when the director wasnominated. The nomination of younger board memberswith tenures shorter than that of the CEO may be morelikely to have been influenced by the CEO (Dey andLiu, 2011; Hill and Phan, 1991). Finally, agency theorywould posit that larger boards may be less effective intheir monitoring capacity (Jensen, 1993; Yermack,1996). A large board may (i) hinder coordination andcommunication; (ii) lead to ‘diffusion of responsibility’instigating second-order free riding; (iii) allow insuffi-cient time for directors to voice their position reinfor-cing the notion that directors should contribute onlysparingly to board deliberations; and (iv) result in slowdecision making or inaction (Dalton et al., 1999;Eisenberg et al., 1998; Lipton and Lorsch, 1992; Poteeteand Ostrom, 2004; Yamagishi, 1986).

From a stewardship perspective, we might hypoth-esize CEOs would strive to include more financialexperts on the board for the benefit of the company(Table 2.). However, CEOs may not seek to includeoutside directors on the board, as stewardship theoryholds that nominees with greater knowledge specificto cooperative operations would be of greater benefitto the firm. In addition, stewardship theorists argueyounger board members with longer tenures are ofgreatest benefit to the firm. Thus, a CEO acting inthe best interest of the firm would strive to influencecomposition of the board toward younger directorsand to lengthen director tenure. Younger directors


Table 1. Agency Theory Expectations Regarding Governance Constructs and Degree of Monitoring

Construct Variables expected Association Interpretation

Boardcomposition

Financial experts serving asdirectors

+ Increased financial expertise may indicate better-equipped monitors

Director tenure � Complacency may increase with tenure+ Longer-tenured directors are less likely to have had their nomination

influenced by CEODirector age � Complacency may increase with age

+ Older directors are less likely to have had their nominationinfluenced by CEO

Board size � Larger boards may be less effective monitorsOutside directorships + Independent directors may be less likely to be commandeered by

CEOFormalcommittees

Existence of audit committee + Audit committees may indicate greater board oversightExistence of members relationscommittee

+ Member relations committees may facilitate producer control

Existence of executive committee + Executive committees may indicate greater board oversightProcedure Full board training + An increase in board training may indicate better-equipped monitors

Monthly boardmeetings + A board that meets monthly mayhave a greater opportunity toeffectively perform their duties

Board highly involved in strategydevelopment

+ Greater board involvement in strategy development may signalgreater control by principals

Frequent executive sessions + Frequent executive sessions may indicate greater board oversight

Table 2. Agency and Stewardship Theory Expectations Regarding the Relationship between GovernanceConstructs and CEO Tenure

Construct Variable Agency theory Stewardship theory

Board composition Financial experts serving as directors � +Director tenure Indeterminate +Director age Indeterminate �Board size + �Outside directorship � �

Formal committees Existence of audit committee � IndeterminateExistence of members relations committee � IndeterminateExistence of executive committee � Indeterminate

Procedure Full board training � +Monthly board meetings � �

Fewer due to insider boardBoard highly involved in strategy development � +Frequent executive sessions � �


are perceived as having greater expertise because oftheir recent education (Muth and Donaldson, 1998). Inaddition, younger board members may be more willingto take profitable risks. Boards with longer averagetenure may be more cohesive, allowing them to makedecisions in a timely manner. In addition, a longer tenuremay be an indicator of high degrees of commitment andloyalty to the firm. Finally, stewardship theorists wouldagree with agency theorists, hypothesizing smallerboards to be more efficient and responsive (Muth andDonaldson, 1998). The notion that larger boards are lessapt to be commandeered by a powerful agent is primarilyassociated with resource dependence theory (Pfeffer andSalanick, 1978). Resource dependence theorists hold


that the diversity of expertise present within a largerboard may reduce managerial domination. However,the successful, larger boards may achieve monitoringand evaluation through specialized committees. For thisreason, we examine the presence of formal committees(Zahra and Pearce II, 1989).

Formal Board Committees and Procedure

From an agency perspective, principals of organizationsthat have become agent driven would be less indepen-dent and less involved in monitoring. Indicators of adecrease in monitoring may include the lack of audit,executive, and member relations committees. Audit



and executive committees may act as mechanisms tostrengthen board oversight. A member relations commit-tee is a mechanism to increase producer feedback andensure producer voice and control. However, of the com-mittees investigated, the executive committee may be themost collaborative in nature: Executive committeemembers may be actively engaged in decision makingwith the CEO between board meetings. In addition,board size is more likely to be correlated with the exis-tence of an executive committee than other committees.

Executive sessions where the board has an opportu-nity to discuss matters without management presentwould be a mechanism by which the board, asprincipals, can fulfill their monitoring function. Thus,a decrease in the number of executive sessions maybe associated with a decrease in monitoring and amore agent-driven organization.

A board that meets less frequently or is not engagingin training may be an additional indicator of a lack ofmonitoring. Lipton and Lorsch (1992) suggest a boardneeds to meet approximately 100 h annually, roughlyequivalent to once a month, to effectively perform theirduties. An untrained board may be less equipped toperform. Finally, a board highly involved in strategydevelopment is likely to be more assertive, exercisinggreater control and monitoring than an uninvolvedboard. A board highly involved in strategy makingstands in stark contrast to the board Mace (1971)describes as simply going along with management’sproposals. However, of the procedural variablespresented, involvement in strategy making would havethe lowest association with monitoring, possiblyindicating a greater degree of collaboration.

Given that stewardship theory hypothesizes lowerlevels of board independence and an active agent to beassociated with more efficient boards, there is no spe-cific reason to suggest stewardship theory should favora greater number of committees or routine executivesessions. However, stewardship theory may suggest ex-pert boards, well acquainted with the organization, maynot need to meet as frequently. As Vafeas (1999) sug-gests, independent boards may need to meet more oftento educate non-executive directors. Finally, stewardshiptheory would agree with agency theory in suggesting abetter-trained board and a board highly involved instrategy development would be most effective.

METHODS AND PROCEDURES

We utilize data from a mail survey of US cooperativesto investigate whether CEO tenure impacts board


composition, committee structure, or board procedures.The sample frame consists of 2252 US farmer, rancher,and fishery cooperatives listed in the United StatesDepartment of Agriculture Cooperative Statistics data-base. We estimate the top 1000 cooperatives conductat least 90% of US agricultural cooperative businessvolume. Thus, we sample the top 1000 agriculturalcooperatives, maintaining percentages of cooperativesby functional categorization (e.g. service, supply, andmarketing) similar to those in the population. Allcooperatives have a non-member CEO, except for onecooperative. To categorize, we determine whether agreater portion of 2009 cooperative revenue is attributedto marketing, supply, or service receipts. Categorizationby sales receipts characterizes approximately 52.8% asmarketing cooperatives, 41.2% as supply cooperatives,and 5.9% as service cooperatives. We sort eachfunctional category by total sales then total assets todetermine ranking. From this ranking, we survey 529marketing cooperatives, 412 supply cooperatives, and59 service cooperatives. We survey cooperative boardchairs. Chairs are selected by their peers. Their leader-ship position and organizational memory provide awell-informed director perspective. We receive 461survey responses. They are representative of the sample.

We perform categorical data analysis to determinewhether there is a relationship between CEO tenureand board composition, formal committee structure,or process. The sample mean of CEO tenure equals10.38 years. For categorical analysis of CEO tenure,cooperatives were divided into two categories (i)cooperatives with CEO tenure less than the samplemean of 10 years and (ii) cooperatives with CEOtenure greater than or equal to the sample mean of10 years. In corporate governance literature, signifi-cant changes in several aspects of board compositionhave been found to take place once CEO tenureextends beyond 10 years (Dey and Liu, 2011). There-fore, using the 10-year mark also allows us to relateour findings to the corporate governance literature.

Control variables include cooperative size mea-sured by the log of total sales and the log of employees(Carpenter and Westphal, 2001) and industry effects.Industry dummies developed from the United StatesDepartment of Agriculture cooperative statisticsindustry classifications include marketing cooperativeindustries, which include cotton, dairy, fruit and vegeta-ble, grain and oilseed, livestock, nut, poultry, dry beanand peas, rice, sugar, fish, and miscellaneous marketing;supply; and service industries including cotton ginning,artificial insemination, storage, transportation, andgeneral service. Two hundred sixty one marketing



cooperative chairs responded, 184 supply cooperativechairs responded, and 15 service cooperative chairsresponded. The survey was conducted in late 2010.

The Board Chair was chosen to respond as it isassumed he or she represents the interests of thecollective group. The survey did not emphasize prefer-ence indications and consequently did not explorevaguely defined property right issues allowing for theboard chair–collective group assumption. The degreeof vertical integration among US cooperatives is quiteextended and complexity of institutional, industry, andorganizational form extensive.

For each relationship tested, we report chi-squarevalues and odds ratios. The chi-square value comparesthe observed frequency in each category with thefrequencies that would be expected by chance. Thisallows us to determine whether there is a significantassociation between CEO tenure and the categoricalvariable in question. The odds ratio illustrates the sizeof the effect. The ratio is reported as the odds of anevent occurring in one category as compared withanother. Therefore, an odds ratio of 1 indicates the oddsof a particular outcome are equal in both categories(Field, 2005).

RESULTS

We report results regarding the relationship betweengovernance constructs and CEO tenure according tothe following categories: (i) variables showing noassociation; (ii) variables showing a slight association;and (iii) variables showing a significant association.

Variables Showing No Association

We begin by reporting those variables for which cate-gorical analysis indicated no significant associationwith the length of CEO tenure.

Financial experts serving as directors. No signifi-cant association was found between CEO tenure andwhether at least one financial expert serves on thecooperative board. Twenty-one percent of cooperativeCEOs with tenure 10 years or greater have at least onefinancial expert on the board. Twenty-five percent ofcooperative CEOs with tenure of less than 10 yearshave at least one financial expert on the board.

Director tenure. For categorical analysis of direc-tor tenure, cooperatives were divided into those withaverage director tenure less than the sample mean,10 years, and those cooperatives with average directortenure greater than or equal to the mean. No significant


association was found between CEO tenure and direc-tor tenure. The average director tenure is roughlyequivalent to average CEO tenure. Thirty-five percentof cooperative CEOs with tenure 10 years or greaterhave a board with average director tenure of 10 yearsor more. Thirty-one percent of cooperative CEOs withtenure of less than 10 years have a board with averagedirector tenure of 10 years or more.

Director age. For categorical analysis of directorage, cooperatives were divided into those with averagedirector age less than or equal to the mean, 52 years,and those cooperatives with average director agegreater than the mean. No significant association wasfound between CEO tenure and average director age.Forty-seven percent of cooperative CEOs with tenure10 years or greater have a board with an average direc-tor age above 52 years. Forty-three percent of cooper-ative CEOs with tenure of less than 10 years have aboard with an average director age above 52 years.

Board size. For categorical analysis of board size,cooperatives were divided into those with board sizeless than or equal to the mean, nine directors, andthose cooperatives with board size greater than themean. No significant association was found betweenCEO tenure and board size. Twenty-six percent of co-operative CEOs with tenure 10 years or greater have aboard consisting of more than nine directors. Twenty-eight percent of cooperative CEOs with tenure of lessthan 10 years have a board consisting of more thannine directors.

Variables Showing Slight1 Association

Outside directorships. There was a slight associa-tion between CEO tenure and whether outside directorsserve on the board w2(1) = 3.17, p< 0.10. This seems torepresent the fact that, on the basis of the odds ratio,CEOs with shorter tenure were 2.50 times more likelyto have at least one outside director on the cooperativeboard. Thus, as CEO tenure increases, cooperativeboards may become less independent. Three percent ofcooperative CEOs with tenure 10 years or greater haveat least one outside director serving on the board. Sixpercent of cooperative CEOs with tenure of less than10 years have at least one outside director serving onthe board.

Existence of an audit committee. There was aslight association between CEO tenure and whetherthe cooperative’s governance structure included anaudit committee w2(1) = 2.72, p< 0.10. This seems torepresent the fact that, on the basis of the odds ratio,boards of cooperatives with CEO tenure of less than



10 years were 1.38 times more likely to have an auditcommittee as part of their governance structure. Thirty-five percent of cooperative boards whose CEO’s tenureis 10 years or greater include an audit committee as partof their governance structure. Forty-three percent ofcooperative boards whose CEO’s tenure is less than10 years include an audit committee as part of theirgovernance structure.

Full board training. There was a slight associationbetween CEO tenure and whether the cooperativeconducted full board training w2(1) = 3.00, p< 0.10. Thisseems to represent the fact that, on the basis of the oddsratio, boards of cooperatives with CEO tenure of less than10 years were 1.44 times more likely to engage in at least1 h of full board training each year. Sixty-eight percent ofcooperative boards whose CEO’s tenure is 10 years orgreater engage in full board training. Seventy-five percentof cooperative boards whose CEO’s tenure is less than10 years engage in full board training.

Variables Showing a Significant2 Association

Monthly board meetings. There was a significantassociation between CEO tenure and frequency ofboard sessions w2(1) = 12.76, p< 0.001. This seemsto represent the fact that, on the basis of the odds ratio,boards of cooperatives with CEO tenure of less than10 years were 2.29 times more likely to meet at leastonce a month. Seventy-one percent of cooperativeboards whose CEO’s tenure is 10 years or greater meetat least 12 times a year. Eighty-five percent of cooper-ative boards whose CEO’s tenure is less than 10 yearsmeet at least 12 times a year. These results suggestmonitoring of the CEO in terms of the number ofboard sessions held each year tends to decrease withCEO tenure.

Frequency of executive sessions. There was a sig-nificant association between CEO tenure and frequencyof executive sessions w2(1) = 6.51, p< 0.05. This seemsto represent the fact that, on the basis of the odds ratio,boards of cooperatives with CEO tenure of less than10 years were 1.80 times more likely to hold at least fourexecutive sessions a year. Eighteen percent of coopera-tive boards with CEO tenure 10 years or greater hold atleast four executive sessions a year. Twenty-eightpercent of cooperative boards with CEO tenure less than10 years hold at least four executive sessions a year.These results suggest that monitoring of the CEO interms of the number of executive sessions held eachyear tends to decrease with CEO tenure.

Existence of a member relations committee. Therewas a significant association between CEO tenure and


whether the cooperative governance structure includeda member relations committee w2(1) = 7.94, p< 0.05.This seems to represent the fact that, on the basis ofthe odds ratio, cooperatives with CEO tenure of lessthan 10 years were 1.93 times more likely to includea member relations committee as part of their gover-nance structure. Seventeen percent of cooperativeboards with CEO tenure 10 years or greater include amember relations committee as part of their gover-nance structure. Twenty-eight percent of cooperativeboards with CEO tenure less than 10 years include amember relations committee.

Existence of an executive committee. There was asignificant association between CEO tenure andwhether the cooperative’s governance structure in-cluded an executive committee w2(1) = 9.81, p< 0.05.This seems to represent the fact that, on the basis ofthe odds ratio, boards of cooperatives with CEO ten-ure of less than 10 years were 1.91 times more likelyto have an executive committee as part of their gover-nance structure. Sixty-three percent of cooperativeboards with CEO tenure 10 years or greater includean executive committee as part of their governancestructure. Seventy-six percent of cooperative boardswith CEO tenure less than 10 years include an execu-tive committee as part of their governance structure.

Board highly involved in strategy development.There was a significant association between CEO tenureand the board’s level of involvement in strategy devel-opment w2(1) = 4.54, p< 0.05. This seems to representthe fact that, on the basis of the odds ratio, boards ofcooperatives with CEO tenure of less than 10 years were1.52 times more likely be highly engaged in strategydevelopment (ranking ≥5 on a scale of 1–6). Thirty-four percent of cooperative boards with CEO tenure10 years or greater report being highly involved instrategy development. Forty-four percent of cooperativeboards with CEO tenure less than 10 years report beinghighly involved in strategy development.

CONCLUSIONS

In corporate governance literature, significant changesin board composition have been found to take place inorganizations where CEO tenure extends beyond10 years. Changes in corporate board composition areidentified as a mechanism by which corporate CEOsnegotiate less monitoring (Dey and Liu, 2011). We findlong-tenured cooperative CEOs also seem to negotiatelower levels of monitoring. However, the mechanismsby which cooperative CEOs experience a decrease in



monitoring is primarily due to formal committees andprocedural aspects of governance, rather than boardcomposition (Table 3.).

One aspect of board composition showed a slightassociation with CEO tenure: Long-tenured CEOshave fewer outside directors serving on the board.Yet, these compositional results are overwhelmed bythe formal committee and procedural variables associ-ated with CEO tenure. Specifically, we find boards oflong-tenured CEOs tend to meet less often, engagein less board training, and have fewer executivesessions. In addition, boards of long-tenured CEOsare less likely to include formal audit and memberrelations committees as part of their structure. Theseresults suggest long-tenured CEOs experience lessboard monitoring.

We caution against speculation that long-tenuredCEOs and their boards experience lower levels ofmonitoring because they have simply worked togetherlonger and (i) are highly familiar with each other andwith cooperative operations or (ii) board membershave become complacent given their long tenures.Our results do not support either hypothesis. Specifi-cally, we find no association between CEO tenureand director tenure or age.

Among variables that investigate levels of collabo-ration between the board and management, in additionto simple monitoring, we find some evidence tosupport the hypothesis that the relationship between

Table 3. The Relationship between Governance ConAssociation and Odds Ratio

Construct Variable

Board composition Financial experts serving as directorsNoneNoneNoneNoneOutside directorships

Formal committees 2.50 times more likelyExistence of audit committee1.38 times more likelyExistence of executive committee1.91 times more likely existence of mrelations committee1.93 times more likely

Procedure Full board training1.44 times more likelyMonthly board meetings2.29 times more likelyBoard highly involved in strategy de1.52 times more likelyFrequent executive sessions1.80 times more likely


long-tenured CEOs and their board is less collabora-tive. Specifically, we find boards of long-tenuredCEOs are less likely to be highly involved in strategydevelopment and less likely to include an executivecommittee as part of their governance structure.Decreases in levels of monitoring and collaborationseem to support the hypothesis that cooperatives withlong-tenured CEOs may be at risk for becoming agentdriven.

And yet, there may be additional, plausible expla-nations for the incidence of more lenient boardsamong long-tenured CEOs. First, boards of long-tenured CEOs may adopt more of a stewardshipapproach, rather than an agency approach, to fulfillingthe role as board members. In other words, boardswith long-tenured CEOs may perceive that the CEOhas acquired superior firm-specific knowledge overthe length of his or her tenure. Thus, the boardchooses to capitalize on this knowledge for the benefitof the principals. A stewardship approach to gover-nance may explain why boards of long-tenured CEOsare less likely to include outside directors, engage infewer board meetings, and conduct fewer executivesessions.

Second, we cannot rule out the possible influenceof stricter government regulations regarding boardcomposition and procedures. Shorter-tenured CEOsin our sample were appointed subsequent to theSarbanes–Oxley Act of 2002. Even though the large

structs and Cooperative CEO Tenure: Strength of

AssociationBoards with CEO tenure

<10 years

Director TenureDirector AgeBoard Size

Slight, p< 0.10

Slight, p< 0.10

Significant, p< 0.05embers Significant, p< 0.05

Slight, p< 0.10

Significant, p< 0.001

velopment Significant, p< 0.05

Significant, p< 0.05



majority of cooperatives in our sample are not obli-gated to abide by Sarbanes–Oxley or similar stan-dards for publicly traded companies, shorter-tenuredCEOs may be more apt to be influenced by the evolv-ing norms embodied in this legislation. If shorter-tenured CEOs also have some influence over theirboards, this may explain why boards of shorter-tenured CEOs are more likely to include outside direc-tors, employ an audit committee, engage in training,meet at least once a month, and have frequentexecutive sessions. One Sarbanes–Oxley relatedvariable that did not show significance, however,was the number of financial experts serving on theboard.

This study advances the important subject of agentversus principal control in patron owned coopera-tives. As scale economies, scope economies, andorganizational complexities increase, the governanceburden on representational fiduciaries is predicted toshift toward the CEO and senior staff because of accu-mulated data and knowledge. This study does notdirectly address this question but informs relatedconcerns of CEO tenure and the degree of monitoringexercised by the governing body. It also highlights theneed for more testable hypotheses and definitivestudies. The corporate governance literature is anexcellent place to start. But, as the results of this studysuggest, governing cooperatives created by a uniqueset of circumstances, institutional ambiance andguided by a strong set of organizational principlescommands a unique approach to honing mechanismsof governance.

FUTURE RESEARCH

The hypothesis that cooperative CEOs are able tobargain for less monitoring based on changes inboard composition is not supported by the currentdata. However, we do find support for the hypothesisthat CEOs may utilize procedural and formal com-mittee aspects of board governance to negotiatedecreased monitoring. These findings are plausibleconsidering CEOs of cooperatives have limited for-mal avenues to influence board composition. Theseresults indicate further research into this topic maybe warranted. Thus, we propose lines of inquiry forfuture research.

First, if CEOs in fact utilize procedural and formalcommittee mechanisms as identified in this dataset toreduce monitoring, can reversal of these mechanismsreduce the risk of the CEO dominating the board?


For example, if the CEO gains greater latitude throughpolicies that reduce board meeting frequency, canboards regain monitoring capacity through policiesdesigned to increase meeting frequency? These prox-ies may be too simplistic. And yet, awareness of theeffect procedural mechanisms have on monitoringmay induce boards to reconsider policy. Considerthe board payment structure: A per-meeting fee mayact as an incentive for board members to meet morefrequently, whereas an annual retainer may act as anincentive for boards to meet less frequently. Furtherresearch may wish to explore whether those coopera-tives that switch from a per-meeting fee to an annualretainer experience a decline in board meeting fre-quency. This would help determine whether there isa correlation between board payment structure andmeeting frequency. If meeting frequency is subopti-mum, can frequency be improved by adopting theappropriate payment structure? For example, woulda per-meeting fee effectively induce board membersto have more meetings thereby increasing monitor-ing? And, if a reduction in the frequency of boardmeetings is a mechanism by which the CEO gainscontrol, can CEO control be decreased by increasingmeeting frequency?

Second, in our future research, we plan to investi-gate the impact of decreased monitoring on coopera-tive performance. Agency theory hypothesizes thateffective monitoring activity that serves to align theagent’s interest with the principals’ improves firmperformance. And yet, it is not clear from this currentstudy whether CEOs that seem to have negotiated lessmonitoring have had a positive or negative impact oncooperative performance. In a future publication, thetwo sets of performance measures, financial andnon-financial, will be analyzed. CEOs with greaterinfluence may use their influence to refocus theboard’s activities thereby (i) developing an exem-plary, ‘expert’ board or (ii) bargaining for a compli-ant, ‘rubber stamp’ board (Baker et al., 1999). Inaddition, long-tenured CEOs may have developed acredible commitment to pursuing cooperative objec-tives rather than selfish interests. In this case, highlevels of monitoring may have a cumbersome, nega-tive impact on cooperative performance, whereaslow levels of monitoring may allow the CEO greaterflexibility to respond to opportunities in a timelymanner. The relationship between CEO influenceover the board and cooperative performance meritsfurther inquiry.

Throughout this paper, we have assumed the rolesand functions of agents and principals are well



defined. However, defining principal and agent rolesmay be an organizational design issue to challenge.The traditional assumption is that the board of

Figure 1. Percent of cooperatives with at least one financialexpert serving on the board.

Figure 3. Percent of cooperatives with average director ageabove 52 years.

Figure 2. Percent of cooperatives with average directortenure 10 years or greater.


directors is or represents the principal and the CEOis the agent. Consequently, the flow of control isfrom principal to agent. However, Eilers and Hanf(1999) posit that the reverse might be more correct,particularly in marketing cooperatives (they do notexamine supply cooperatives in their study). Whena manager offers a contract to a farmer, he or shemay be acting as a principal. When the farmeraccepts a contract with the cooperative, he or shemay be acting as an agent. Considering such rolereversals might inform cooperative governancedesign and stimulate research on this complex andvaguely defined relationship. Mixed methodapproaches might add a depth of understanding thatwould inform researchers’ ability to more rigorouslyaddress these issues. This initial study appears toonly whet our appetite with regard to perhaps oneof the most understudied yet important area of coop-erative governance (Figures 1–12).

Figure 4. Percent of cooperatives with more than ninedirectors.

Figure 5. Percent of cooperatives with at least one outsidedirector serving on the board.


Figure 7. Percent of cooperatives engaging in full boardtraining.

Figure 8. Percent of cooperatives with boards in session1 day a month or more.

Figure 9. Percent of cooperatives holding at least fouexecutive sessions each year.

Figure 10. Percent of cooperatives with a member relationscommittee.

Figure 11. Percent of cooperatives with an executivecommittee.

Figure 6. Percent of cooperatives with an audit committee.


NOTES

1. By slight association, we refer to the p-value that indi-cates there is less than a 10% chance this result occurredby random coincidence.

Copyright © 2013 John Wiley & Sons, Ltd. Manage. Decis. Econ. 34: 218–229 (2013)DOI: 10.1002/mde

r

2. By significant association, we refer to the p-value. Forthis section, all associations included have a p-value ofless than 0.05, which indicates there is less than a 5%chance this result occurred by random coincidence.

Figure 12. Percent of cooperatives with boards highlyinvolved in strategy development.


REFERENCES

Adams RB, Hermalin BE, Weisbach MS. 2010. The role ofboards of directors in corporate governance: a conceptualframework and survey. Journal of Economic Literature48(1): 58–107.

Aghion P, Tirole J. 1997. Formal and real authority in orga-nizations. Journal of Political Economy 105(1): 1–29.

Baker G, Gibbons R, Murphy KJ. 1999. Informal authorityin organizations. The Journal of Law, Economics, &Organization 15(1): 56–73.

Bebchuk LA, Fried JM. 2003. Executive compensation as anagency problem. Journal of Economic Perspectives 17(3):71–92.

Bijman J. 2005. Network and hierarchy in Dutch co-operatives: a critical analysis. International Journal ofCo-operative Management 2(2): 16–24.

Carpenter MA, Westphal JD. 2001. The strategic context ofexternal network ties: examining the impact of directorappointments on board involvement in strategic deci-sion making. The Academy of Management Journal 44:639–660.

Chhaochharia V, Grinstein Y. 2007. Corporate governanceand firm value: the impact of the 2002 governance rules.Journal of Finance 62(4): 1789–1825.

Cornforth C. 2004. The governance of cooperatives andmutual associations: a paradox perspective. Annals ofPublic & Cooperative Economics 75: 11–32.

Craig JG. 1993. The Nature of Co-operation. Black RoseBooks: New York.

Dalton DR, Daily CM, Johnson JL. 1999. Number of direc-tors and financial performance: a meta-analysis. TheAcademy of Management Journal 42(6): 674–686.

Dey A, Liu X. 2011. CEO tenure, board composition andinnovation. SSRN e-Library.

Donaldson L, Davis J. 1991. Stewardship theory or agencytheory: CEO governance and shareholder returns.Australian Journal of Management 16(1): 49–64.

Eilers C, Hanf CH. 1999. Contracts between farmers andfarmers-processing co-operatives: a principal-agent ap-proach for the potato starch industry. In Vertical Relation-ships and Coordination in the Food System, Galizzi G,Venturini L (eds). Physica: Heidelberg; 267–284.


Eisenberg T, Sundgren S, Wells MT. 1998. Larger boardsize and decreasing firm value in small firms. Journal ofFinancial Economics 48(1): 35–54.

Field AP. 2005. Discovering Statistics Using SPSS. SAGE:Thousand Oaks, CA.

Fulton ME, Larson KA. 2009. The restructuring of theSaskatchewan wheat pool: overconfidence and agency.Journal of Cooperatives 23: 1–19.

Hermalin BE, Weisbach MS. 1998. Endogenously chosenboards of directors and their monitoring of the CEO.American Economic Review 88(1): 96–118.

Hill CWL, Phan P. 1991. CEO tenure as a determinant of CEOpay. The Academy of Management Journal 34(3): 707–717.

Hind AM. 1997. The changing values of the cooperative andits business focus. American Journal of AgriculturalEconomics 79: 1077.

Hind AM. 1999. Co-operative life cycles and goals. Journalof Agricultural Economics 50: 536–548.

Hung H. 1998. A typology of the theories of the roles ofgoverning boards. Corporate Governance: An Interna-tional Review 6(2): 101.

Jensen MC. 1993. The modern industrial revolution, exit,and the failure of internal control systems. Journal ofFinance 48: 831–880.

Jensen MC, Meckling WH. 1976. Theory of the firm:managerial behavior, agency costs and ownership struc-ture. Journal of Financial Economics 3: 305–360.

Lipton M, Lorsch JW. 1992. A modest proposal for improvedcorporate governance. Business Lawyer 48(1): 59–77.

Mace ML. 1971. Directors: Myth and Reality. HarvardBusiness School Press: Boston, MA.

Murray G. 1983. Management strategies for corporatecontrol in British agricultural cooperatives. AgriculturalAdministration 14(1): 51–63.

Muth M, Donaldson L. 1998. Stewardship theory and boardstructure: a contingency approach. Corporate Governance6(1): 5–28.

Pfeffer J, Salanick GR. 1978. The External Control ofOrganizations: A Resource Dependence Perspective.Harper & Row: New York.

Poteete AR, Ostrom E. 2004. Heterogeneity, group size andcollective action: the role of institutions in forestmanagement. Development and Change 35: 435–461.

Shen W. 2003. The dynamics of the CEO-board relation-ship: an evolutionary perspective. The Academy ofManagement Review 28(3): 466–476.

Shivdasani A, Yermack D. 1999. CEO involvement in theselection of new board members: an empirical analysis.Journal of Finance 54(5): 1829–1853.

Spencer Stuart Board Services. 2010. 2010 Spencer StuartBoard Index (25th edn). Spencer Stuart: Chicago.

Vafeas N. 1999. Board meeting frequency and firm perfor-mance. Journal of Financial Economics 53(1): 113–142.

Yamagishi T. 1986. The provision of a sanctioning systemas a public good. Journal of Personality and SocialPsychology 51: 110–116.

Yermack D. 1996. Higher market valuation of companieswith a small board of directors. Journal of FinancialEconomics 40(2): 185–211.

Zahra SA, Pearce JA. 1989. Boards of directors andcorporate financial performance: A review and integrativemodel. Journal of Management 15(2): 291–334.


Documents

The Impact of CEO Tenure on Cooperative Governance