Who Controls the Board in Non-Profit Organizations? The Case of Private Higher Education Institutions in Colombia

How should the governance system in a non-membership non-profit organization be designed? This organizational form has no shareholders; instead, donors provide funds. Thus, at the organizational level the board of directors could have all the power. Under this legal form, who controls the board? If too powerful, boards could misuse resources or distract the organization from its foundational goals. We examine the case of private higher education institutions (HEIs) in Colombia and the balance of power in university governance systems which feature this organizational form. Most HEIs in our sample have a kind of assembly of representatives as the governance body with the highest authority and able to appoint and control the board. We specifically discuss the assemblies’ reason for being, structure and functions in private HEIs in Colombia. We analyze a total of 204 HEI governance structures and find governance arrangements with the characteristics of an assembly of representatives in 154 (75.9%). Our analysis highlights features in some of these governance bodies that could lead to overly powerful assemblies (e.g., founder donors with tenure for life). Clearly, a proper balance of power is required to avoid rent-seeking behaviors or the pursuit of harmful private non-monetary benefits from assembly members as well as boards.

3 powerful boards could potentially misuse organizational resources, adopt rent-seeking 5 behaviors to favor directors over the organization's stability and welfare (e.g., disproportionate compensation, excessive travel allowance or other kinds of perks consumption), pursue private non-monetary benefits harmful to organizational performance, or distract the organization from its mission or foundational goals.
In this paper, we use the case of private HEIs in Colombia, which are non-membership non-profit organizations by law, to discuss the balance of power in the university governance system under this organizational form. While public HEIs in Colombia are established with government funds, funding for private HEIs is provided by private donors (individuals, families, non-profit or for-profit organizations). The local regulatory framework determines the public HEIs' governance system, with a board as the governance body with the highest authority.
However, private HEIs have almost total autonomy to determine their governance system. The majority create a kind of assembly of representatives as the governance body with the highest authority and able to appoint and control board members. We specifically examine these assemblies, their reason for being, structure and functions.
As a governance body to control the board, the assembly of representatives in private HEIs in Colombia is a kind of counterpart to the shareholders' assembly in for-profit firms. It seems to protect the interest of society as a whole, preserving HEIs' foundational goals and monitoring the proper use of HEIs' financial resources. We analyze the governance structure of a total of 204 non-profit private HEIs and find governance arrangements with the characteristics of an assembly of representatives in 154 (75.9%). Our analysis also highlights features of these governance bodies that could lead to overly powerful assemblies. For example, founder donors with tenure for life can interfere in the day-to-day running of the HEIs. Hence, a proper balance of power is required to guarantee that assemblies as well as boards are able to fulfill their functions and to avoid the pursuit of harmful private nonmonetary benefits or rent-seeking behavior by assembly members.
This paper contributes to the corporate governance literature, and specifically to the university governance literature, by highlighting the relevance of the balance of power between different governance bodies in private HEIs. We discuss the emergence of assemblies of representatives as a special arrangement of private HEIs in Colombia and the governance body with the highest authority at the organizational level. This adaptation could represent a response to a weak institutional environment at the country level. Colombia ranks low in terms of law enforcement (La Porta et al., 1997;Djankov et al., 2008) and offers relatively low legal protection (Gonzalez et al., 2014(Gonzalez et al., , 2015. We examine the functions of the assembly of representatives and its structure to facilitate the fulfillment of those functions. Discussion regarding the balance of power between different governance bodies in HEIs has been developed in other countries. Mignot-Gerard (2003) analyzes the increased relevance of executive boards in universities in recent decades and the consequences of strengthening central management for the balance of power between executive boards, faculties, and deliberative bodies. Although her focus is specifically on French 5 We follow the economics and finance literatures in using the term rent seeking. According to Morck and Yeung (2004), rent seeking is a negative sum game that has the highest return from the viewpoint of each individual but that destroys value for society as a whole. It represents a behavior that allows individuals to get pecuniary private benefits against the common interest ( Barclay and Holdemess, 1989), implies exploiting organizational resources to achieve this goal (Anderson and Reeb, 2004), could cause large inefficiencies (Shleifer and Vishny, 1989), and harms economic growth because it affects productivity and innovation (Murphy, Shleifer and Vishny, 1993). universities, she also considers cases in the Netherlands, the United Kingdom, Norway and Sweden, highlighting differences in governance arrangements in HEIs, depending on the context. Hence, the study of this issue in Colombia, an emerging country with specific institutional characteristics, sheds light on the impact of context on specific corporate governance arrangements at the organizational level.
The rest of this paper is organized as follows. In the next section we highlight the connections and complementarities of the theoretical governance frameworks in higher education and corporate contexts. We then describe the data we hand-collected to support our analysis and theoretical discussion. After that, we discuss the purpose of a board or council and an assembly of representatives in a private HEI. We then analyze the main functions and structure of the assembly of representatives. Next we examine the Colombian case, where the majority of private HEIs have established an assembly of representatives. The conclusion follows.

Higher education governance and corporate governance
Studies of higher education governance on one hand and corporate governance on the other lead to the recognition of two theoretical frameworks with connections and complementarities. It is useful to make these links explicit for a broader and clearer understanding of the dynamics of governance in HEIs. Dobbins et al. (2011) offer a comprehensive review of the literature in higher education governance and state that environmental pressures at global and domestic levels are reshaping, transforming, modernizing and "marketizing" higher education governance at the country level. These changes are evident in the emergence of several higher education governance models. Clark (1983) provides a first major classification of these models, stressing three different models whose combination determines the context in which the HEI operates (Reed et al., 2002). The state-control model assumes universities are state-operated institutions, and hence the state coordinates the majority of functions within them; the academic self-rule model suggests weak management, collegial control by professoriate or academic staff and a high degree of self-regulation; the market-oriented model asserts that HEIs function properly when operating as economic enterprises responding to demands from regional or global markets (Dobbins et al., 2011). Maassen (2002) asserts that in the first decades of the nineteenth century the Continental European states made themselves responsible for the regulation and funding of higher education, developing a state-control steering strategy. However, the redistribution of authority and power in society has led to alternative approaches and combinations of Clark's three models, with the state still a prominent actor. Maassen (2002) also stresses that in countries like the United States, Canada, Australia and the United Kingdom the market-oriented model or "academic capitalism" is the dominant model. Magalhães et al. (2013) analyze higher education governance as an aspect of national policy and as a contextual factor. They cite the effect of European Union policies on country-level policies and reforms in England, France, Germany, Italy, the Netherlands, Norway Portugal and Switzerland. These authors assert that European higher education and research agenda contribute to opinion formation and hence determine the legitimacy of country-level reforms and policies.
Some case studies at country or institutional levels highlight the response of HEIs to demands from local or global markets. Nguyen et al. (2016) and Rungfamai (2016) consider Vietnamese and Thai HEIs, respectively, to study the adoption of strategies responding to contextual factors to improve quality, produce knowledge, enhance 5 institutional competitiveness, attract local and international students, and increase income. Nisar (2015) and Yokoyama (2006) stress the changing views about the role of government funding for HEIs and the promulgation of policies to increase external funding. Hence, these authors analyze the U.S. strategy of paying for performance (Nisar, 2015) and entrepreneurialism in Japanese and U.K. universities (Yokoyama, 2006). All of these studies establish a relationship between changes in context and governance practices at the institutional level.
Conceptual models represent a second major classification of governance models in higher education literature. According to Reed et al. (2002) conceptual models include, among others, collegial, bureaucratic, political, organized anarchy, and, more recently, entrepreneurial, service, enterprise and corporate models. Dobbins et al. (2011) stress the shared governance model (participation and integration of all stakeholders relevant to higher education), the corporate governance model (with an emphasis on the planning efforts and entrepreneurial character of HEIs), and the entrepreneurial model (emphasis on links between universities and markets and funding from industry based on applied research). These models seem to offer answers to the question of who should have the ultimate decision-making power in HEIs.
The higher education governance literature has also addressed the design of the internal governance systems and arrangements to guarantee a balance of power among governance bodies. For example, Shattock (2010) asserts that successful universities try to keep the powers of governance in balance and that good governance makes a significant contribution to university success. In addition, balance of power within university governance can be viewed in multiple ways, e.g., between external members and management, between practicing academics and management, and between university councils and academic boards, among others 6 . This issue has been extensively analyzed in the context of for-profit organizations. The main functions, composition and reasons for being for boards of directors and shareholders' assemblies, as well as the proper design of the internal control environment and disclosure, transparency and accountability practices are at the center of the corporate governance literature (Jensen and Meckling, 1976;Fama and Jensen, 1983;Palepu, 1993, 2001;Klein, 2002;Bainbridge, 2003;Apostolides, 2007;and Adams and Ferreira, 2007, among others). This theoretical framework proposes mechanisms to mitigate agency conflicts as its primary goal. Hence, this paper aims to contribute to the university governance literature from the perspective of corporate governance.
We consider it extremely important to clarify that the use of the corporate governance theoretical framework to study HEIs' governance does not imply either support or disagreement with the market-oriented model. It only implies the recognition of the existence of conflicts of interest and agency tensions within HEIs and the need to mitigate these to guarantee the effectiveness of these institutions and the proper use of their resources. Jensen and Meckling highlight the generality of the agency conflict, asserting that: "It exists in all organizations and all cooperative efforts -at every level of management firms, in universities [HEIs], in mutual companies, in cooperatives, in governmental authorities and bureaus, in unions, and in relationships normally classified as agency relationships such as are common in the performing arts and the market for real estate. The development of theories to explain the form which agency costs take in each of these situations (where the contractual relations differ 6 significantly), and how and why they are born will lead to a rich theory of organizations which is now lacking in economics and the social sciences generally" (1976: 309).
Nor does use of the corporate governance theoretical framework imply neglecting social welfare as the primary responsibility of HEIs, that is, the construction of a better society through the provision of a public good (education). From the higher education governance perspective, Leslie (1975) asserts that the tests of higher education governance are effectiveness and legitimacy. Effectiveness is understood as practices and decisions that enhance the long-term welfare of higher education and society, while legitimacy is understood as the rightness or moral goodness of the decision makers and the meeting of stakeholders' needs and interests. From the corporate governance perspective, Tirole (2001) suggests that corporate governance should take into account firms' different stakeholders in particular and society's welfare in general, and that a sole focus on shareholder value is inappropriate. According to Tirole, managerial decisions that affect funds providers also exert externalities on a number of stakeholders who have a natural relationship with the organization: employees, customers, suppliers, communities, and so forth. Hence, Tirole discusses a broad management mission to maximize the sum of the various stakeholders' surpluses.
According to the discussion above, despite the specificities of HEIs, such as their core mission (knowledge production, dissemination and consumption), their constituencies and their historical modes of governance, there are natural connections and complementarities between higher education governance and the theoretical frameworks of corporate governance. The latter brings to the higher education literature the recognition and discussion of conflicts of interest and the need for mechanisms to mitigate them. Thus we are particularly concerned with the balance of power in the corporate governance system at the institutional level. Previous studies in the higher education literature have considered the role of boards or councils in HEI governance, highlighting their relevance for strategy and accountability (e.g., Dixon and Coy, 2007), but we find a gap in studies dealing with the balance of power among constituents within internal HEI governance systems, a gap we address in this paper.

Data and research design
According to the National System of Higher Education Information in Colombia (Sistema Nacional de Información de la Educación Superior, SNIES), there are 303 HEIs in Colombia, 81 public and 222 private (SNIES, 2016). From the total of almost 2 million enrolled students in 2017, 61.1 percent attend private HEIs. In addition, the total number of academic programs offered in the Colombian higher education system in 2017 is 22.856, with 67 percent of these from private institutions. Law 30 of 1992 regulates higher education in Colombia as a public service and there have been no reforms to this legal framework during the last 25 years. We concentrate on private HEIs because they have almost total autonomy in establishing their governance system, while public HEIs have to adopt a governance system pre-established by law. Given the autonomy of private HEIs regarding their organizational structures, including such matters as governing bodies and internal administration, it is relevant to analyze how the governance systems in these institutions are designed. Hence, we searched and hand-collected the statutes (bylaws) of the 222 private HEIs in Colombia and were able to locate the statutes of 204; that is, 91.9 percent of the total. The remaining 18 HEIs do not publish their statutes on their web pages. We analyzed the statutes one by one, with special emphasis on rules about the composition and functions of the governance bodies.
Guzman and Trujillo (2016) highlight the relevance of the statutes since they contain explicit guidelines for any given activity and entail a legal connotation. Organizations establish the ground rules for the relationship with different stakeholders through their bylaws. The connection with the law gives statutes formality and enforcement, much more so than other documents containing organizational rules. Governance arrangement not reflected in the bylaws can lose validity and enforcement with changes in the composition of the governing bodies and senior management. Practices covered in the statutes have greater stability and can be altered only by reforming the bylaws.
Of particular importance, bylaws define the scope and purpose of the governing bodies in terms of management, counseling, and supervision activities.
The majority of private HEIs in Colombia have a governance body, a kind of assembly of representatives, and this body has the highest level of authority in the governance system. This was true in 154 of the HEI statutes, 75.9 percent of the total sample. With an average of sixteen members, this kind of governance arrangement has various names, such as general ward, general ward of members, assembly of members, general assembly of members, general assembly, council of founders, assembly of founders, and constituency, among others. Statutes define their composition and functions as a governance body. We identify six different kinds of members in these assemblies: founders, adherents or benefactors, honoraries, members of number (just members), members elected by some other governing body, and members chosen from the HEI community. However, the power of founder donors is evident. There are references to founders as members in these assemblies in 92.2 percent of the statutes we analyzed. In addition, adherents or benefactors (mentioned in 57.8% of the statutes), honoraries (mentioned in 31.2% of the statutes), members of number (mentioned in 11.7% of the statutes), and members chosen from the HEIs community (mentioned in 22.1% of the statutes) are elected by the founders or by the majority of the assembly of representatives, thereby ensuring founders' control. Only members elected by some other governing body could not be co-opted, elected or appointed by founders, but this is possible in only 4.5 percent of the HEIs in the sample.
Among founders and other members it is usual to find business people, entrepreneurs, academics, professionals, former HEI presidents, and representatives of religious communities, among others.
Despite the legal connotation of statutes, we decided to compare the composition of assemblies of representatives as described in statutes with actual membership as published on institutional websites. However, this was possible for only 14 of the 204 private sample HEIs. We then decided to formally ask 20 HEI presidents about the actual membership of their institution's assemblies; we got answers from 10 of these. For the 24 HEIs (11.8% of our sample) that we were able to compare, we found consistency between the composition described in the statutes and actual assembly membership. This was expected since compliance with statutes is mandatory by law; noncompliance with rules adopted through bylaws could generate legal problems for presidents and members of the council and could imply sanctions imposed by the National Ministry of Education. Thus, we assume that the nondisclosure of detailed information regarding the actual membership of the assembly on institutional websites does not indicate non-compliance with the statutes. We use the data described above to build our arguments and for the empirical analysis that follows.

University governance bodies
The traditional corporate governance system or structure in listed widely held firms (corporations that have many shareholders) comprises the annual general meeting (AGM or shareholders' assembly meetings) and the board of directors. The AGM is the corporate governance body that facilitates mediation of tensions between the board and the firm's shareholders (Apostolides, 2007) and between majority and minority shareholders. It is a source of accountability that allows directors and top management to inform shareholders about accomplishments, and allows majority and minority shareholders to exert their property rights and protect their interests. But boards also have a monitoring and advisory role in their interaction with top management.
According to Adams and Ferreira (2007) the advisory role is a key function not only in the single board system usual in the United States but also in the dual board system common in Europe, where boards are separated into two governance bodies, an advisory board and a supervisory board. Carrying out the monitoring role, boards must approve corporate strategies and implementation plans, and review and approve relevant operating and financial decisions. To accomplish this advisory role, boards take advantage of members' professional experience and ability to counsel management (Adams and Ferreira, 2007).
Schwartz-Ziv and Weisbach (2013) use public data from Israel to analyze the minutes of corporate board meetings and board committee meetings; they find that boards spend their time supervising management for the most part. However, the balance between the monitoring and advisory roles is a key aspect for the functioning of the governance system. According to Faleye et al. (2011), intense board monitoring mitigates earnings management, leads to lower executive compensation and makes CEO turnover in poor performing firms more probable. However, there is also evidence of weaker strategic advising and greater managerial myopia.
The role of the board in HEIs is very similar, if not the same. Middlehurst (2004) analyzes the changes that U.K. universities have been making since the eighties to increase efficiency and improve performance. She highlights adoption of governance structures closer to those in the corporate sector, with stronger boards called councils composed of independent members who supervise management. According to the Association of Governing Boards of Universities and Colleges (AGB, 2007), in the United States the basic responsibilities of a governing board in HEIs include approval of the mission and purpose of the institution; recruitment, appointment, support and evaluation of the president (chief executive officer); approval of the institution's budget; supervision of the institution's resources and productivity; oversight of annual independent audits; participation in strategic planning and its strategic implementation; and an understanding of the institutional issues and challenges confronting higher education. In Latin America, where universities are granted a high level of autonomy, the situation is similar (Restrepo et al., 2012), and even in heavily state-controlled environments, such as in Georgia or Armenia, HEI governance bodies have to deal with these issues (Dobbins and Khachatryan, 2015).
The critical role of boards in university governance generates a natural concentration of power in this body that could be misused if the governance system does not provide a counterbalance on behalf of different HEI stakeholders. This is even more important in organizations like the private HEIs in Colombia, where donors rather than owners have funded the institution in its early stages, presumably with altruistic motivations. In for-profit organizations, ownership rights allow shareholders to control the board and management, the governance mechanism being the AGM. But board control is diffuse in non-profit organizations. In Colombia, private HEIs are under the supervision of the Ministry of Education, which lacks the resources to exert close supervision of each institution.
Indeed, the legal framework grants total autonomy to private HEIs to establish their own governance arrangements.
Furthermore, Colombia has been characterized as an environment of weak law enforcement and low legal protection.
In this context, the assembly of representatives emerges as a counterbalance to board power in Colombian non-profit private HEIs. Its members represent the donors and society in general as the ultimate stakeholders. This arrangement at the organizational level makes the governance system more robust and is the counterpart or equivalent body to the shareholders' assembly in for-profit firms. As mentioned, the difference is that the assembly of representatives does not represent shareholders but rather embodies the ultimate stakeholder of the non-profit HEIs, that is, society as a whole.
The assembly's purpose should be the preservation of the institution's foundational goals and the safeguarding of its assets and any surplus revenues. Implementation of this governance mechanism mitigates the risks associated with overly powerful boards that could misuse resources, adopt rent-seeking behaviors to favor directors over the organization's stability and welfare, distract the organization from its foundational aims or pursue private non-monetary benefits harmful to the HEIs' academic and financial performance (such as recruitment of friends or acquaintances as professors or recommendations for the admission of students lacking the established requirements). To fulfill its reason for being, any assembly of representatives needs specific functions and a special composition that we discuss in the following section.

Functions
As stated, the assembly of representatives should work for the preservation of the foundational goals of the private HEI and to safeguard its financial resources.  asserts that assemblies take decisions about strategic matters, supervise boards and preserve financial stability and resources in listed companies. To perform its duties properly, this governance body needs to accomplish specific tasks. Following the OECD (2015) recommendations on governance, and taking the assembly of representatives in non-profit organizations as the counterpart to shareholder meetings in for-profit companies, assembly members should receive relevant and material information on the HEI in a timely and regular manner, should participate and vote in assembly meetings, elect and remove members of the board, approve amendments to the HEI statutes, and make decisions about extraordinary transactions, including the disposal of the institution's assets. In addition, members should have specific mechanisms to make boards accountable. For example, boards should evaluate their collective and individual performance regularly, and report the results of these exercises to the assembly of representatives. Furthermore, compensation schemes for board and top management team members should be approved by the assembly.
According to , AGMs or the shareholders' assembly in for-profit organizations have three main functions. First, this is the formal venue to inform shareholders about the performance of the organization and about relevant decisions by directors and managers. Second, these meetings permit shareholder approval of decisions outside the discretion of the board of directors. And third, such meetings allow discussions between shareholders, directors and managers about future actions, strategies and policies for the organization. Catasús and Johed (2007) state that the shareholders' assembly allows shareholders to call management to account for their decisions, and hence exercise disciplinary power over managers. The equivalent body in HEIs should play a similar role by supervising organizational performance and endorsing significant decisions and policies for the institution.
The most common agency problem for HEIs is rent-seeking behavior by board and top management team members that divert financial resources for their private benefit. This group has more concentrated power, which facilitates the adoption of misbehaviors to get private pecuniary benefits. For example, de Vise (2011)  The financial economics literature has shown the relevance of disclosure to mitigate information asymmetries and agency problems. Ho and Wong (2001) measure voluntary disclosure and find that effective corporate governance arrangements increase firms' transparency; for example, the existence of an audit committee leads to higher voluntary disclosure. Bhat et al. (2006) highlight the relevance of governance transparency as well as financial transparency, showing that the former is positively associated with analysts' forecast accuracy regarding for-profit companies. Leuz et al. (2009) find that foreigners invest less in more opaque firms located in countries with weak investor protection, suggesting that funding is difficult to obtain with deep information asymmetries and weak legal environments.
An appropriate environment of control benefits organizations. Control is strongly related to the efficiency of organizations in managing their financial resources and surplus. Upadhyay et al. (2014) show that, monitoring committees to oversee audits, executive compensation, nominations, governance, and executive committees make large boards in complex firms more effective. In the same vein, Xie et al. (2003) find empirical evidence for less frequent earnings management in firms when the audit committee meets frequently and has members with specific skills in finance. The relevance of an audit committee lies in its functions. "The audit committee is responsible for appointment of external auditors, to monitor the internal audit function and preserve auditor independence" (Upadhyay et al., 2014(Upadhyay et al., :1487.

Structure
A balance between HEI insiders and outsiders is necessary to mitigate agency tensions that could emerge within the assembly of representatives. This has been extensively examined in the case of boards. According to Fama and Jensen (1983) insiders have access to information relevant for the decision-making process, while outsiders, who should hold the majority of seats, carry out functions that involve agency problems between insiders and other stakeholders; for example, in setting executive compensation. To mention just two examples, Baysinger et al. (1991) and Duchin et al. (2010) highlight the relevance that insider board members have because of the cost of acquiring information about the firm and the impact this has on decisions about research and development investments. In addition, different studies show that a higher proportion of independent directors in board composition has positive effects on firm value and leads to better-governed companies (Shivdasani and Yermack, 1999;Hwang and Kim, 2009;Chhaochharia and Grinstein, 2009;and Black and Kim, 2012, among others).
The assembly of representatives needs HEI insiders for their knowledge regarding the foundational goals, mission and values of the institution. In addition, insiders have access to information that could be extremely difficult for outsiders to get. However, an assembly of representatives totally shaped by insiders could exacerbate agency tensions. Founder donors could take advantage of their position to nominate close friends to the HEI board, adopting rent-seeking behavior, or they could make mistakes in the decision-making process through lack of knowledge or experience that outsiders could bring. Hence, outsiders are needed to mitigate potential agency tensions and guarantee the balanced functioning of the assembly of representatives. Outsiders in HEIs should be people of good reputation and outstanding professional development and experience, some in the business or public sector and others in the academic sector. In this way, outsiders contribute good judgment to the decision-making process of the assembly of representatives.

Governance bodies in private HEIs in Colombia
To test the theoretical discussion presented above, we revisit the main governance bodies of private HEIs in Colombia, analyzing HEIs' statutes with particular focus on the general assembly or assembly of representatives. As mentioned, we were able to locate statutes for 91.9 percent of the total private HEIs in Colombia. Several findings are noteworthy. First, the majority of private HEIs in Colombia (75.9% of the total sample) have an assembly of representatives (with a variety of formal names). With an average of sixteen members, this governance arrangement has the highest level of authority in the governance system, and perhaps mitigates the risk that a powerful board might distract the institution from its foundational goals or misuse its financial resources. In addition, private HEIs could favor the existence of assemblies to give donor founders recognition within the institution and allow them to have influence in the decision-making process. However, note that in 50 of the HEIs in our sample, the board is the ultimate governance body and directors elect their own successors, a situation that could result in overly powerful boards and all the attendant problems discussed above.
We assume that assemblies could be a response to weak institutional environments and could reflect founders' desire to preserve the institution from directors' negligent or opportunistic behavior. To test this assumption we review the statutes of the twenty top private HEIs in Latin America according to the QS World University Rankings® 2015/16, and find that governance arrangements similar to the assembly of representatives in terms of power and functions are widespread (e.g., at the Instituto Tecnológico y de Estudios Superiores de Monterrey, Pontificia Universidad Católica de Chile, Pontifícia Universidade Católica do Rio de Janeiro, Universidad Austral de Chile, and Pontificia Universidad Católica del Perú). All of these private universities are located in countries that share institutional characteristics similar to Colombia's, including relatively weak institutional environments (La Porta et al., 1997). In addition, we looked at the statutes of the twenty top private HEIs in the world (QS World University Rankings® 2015/16). None of these universities (e.g., the Massachusetts Institute of Technology, Harvard University, Stanford University, California Institute of Technology, University of Chicago, Princeton University, Yale University, Johns Hopkins University, Cornell University, and the University of Pennsylvania), all located in strong institutional environments, had a governance body with a higher level of authority than the board. This reinforces our supposition concerning a relationship between university governance arrangements at the organizational level and the institutional environment.
Second, within the group of 154 private Colombian HEIs that have an assembly of representatives, 93.5 percent have the power to designate a proportion of board seats. On average, the assembly of representatives elects four of a total of nine board members, 45 percent of board members on average. In addition, 50 of these 154 assemblies (32.5%) designate the majority of the board seats. Furthermore, the power of the assembly of representatives is not limited just to the election of some directors. According to the statutes that we analyzed, 93.5 percent of the assemblies supervise compliance with the HEI foundational goals as one of their main functions. Also, 127 assemblies (82.5%) are responsible for approving reforms to the statutes; 125 assemblies (81%) designate the external auditor; 86 percent have to approve the annual financial statements and work to preserve the HEI's financial resources; and 113 assemblies (73%) are the competent body to approve the dissolution of the HEI. These functions highlight the relevance of this governance body as a counterbalance to the power granted to the board.
Third, the control environment does not seem strong enough in HEIs with an assembly of representatives.
For example, sixteen HEIs in our sample (10.4%) define an internal control system as one of their main functions.
But in the statutes we analyzed there is no evidence of an audit committee, and only seven HEIs in the sample have a recruitment and remuneration committee. Furthermore, the committees offer support to the assembly for the recruitment of directors in only two cases. Preservation of HEIs' financial resources is more plausible with a strong control environment and with a good selection of directors. Hence, the control environment in Colombian HEIs and the processes to recruit and remunerate directors and top management team members appear to deserve more attention.
Fourth, the excessive privileges granted to insiders in some Colombian assemblies such as tenure for life or the right to appoint the president of the institution could lead to an imbalance of power favoring donor founders and creating opportunities for misbehavior on their behalf. In 115 assemblies of representatives (74.6%) the statutes stipulate at least two different classes of members, and some grant more power and privileges to donor founders, for example, tenure for life. In addition, 61 percent of these assemblies have among their main functions the appointment of the president of the institution. We argue that this function should be granted to the board and not to the assembly of representatives because presidents are accountable to the board, and the power to remove them should rest with the board. Also, according to the statutes, 31 assemblies (20%) have specific functions regarding academic activities, which could imply an intrusion in the day-to-day operations of the HEI. This fact could be reinforced by the number of meetings held each year. In 57.5 percent of these assemblies, representatives meet once a year; 26.1 percent meet twice a year, which is similar to practices of the shareholders' assemblies in the private sector. However, eighteen (13.4%) of the assemblies in our sample meet four or more times a year, some having six to twelve meetings.

13
Control of the board of directors is not the biggest concern regarding university governance for the HEIs in our sample. An equally important question regarding the balance of power is who controls the assembly of representatives. The current role of this governance body reveals weaknesses in the design of the control environment, including high participation of insiders with excessive privileges like long tenures (even for life). These exacerbate agency tensions and conflicts of interests, and deny the relevance of other HEI stakeholders. A proper balance of power between boards and assemblies could lead to good practices at the organizational level in Colombian HEIs, while overly powerful assemblies could lead to the situation described by Kretek et al. (2013: 56), who state that a "highly asymmetric distribution of information among governance actors, no accountability mechanism as well as a wide range and variety of demanding role expectations is assumed to come with a toothless and inactive rubber stamp board with passive and indifferent board members showing a very low level of engagement and involvement."

Conclusion
We aim to contribute to the university governance literature from the perspective of corporate governance, discussing the design of internal governance systems and arrangements to guarantee a balance of power among governance bodies and thus mitigate conflicts of interest that could emerge from different stakeholders and that could threaten the HEIs' perdurability. The corporate governance literature has extensively analyzed this issue in the context of for-profit organizations, and we apply that approach to higher education governance. We clarify that the use of the corporate governance theoretical framework to study HEIs' governance does not imply either support or disagreement with the market-oriented model. It only acknowledges the existence of conflicts of interest and agency tensions within HEIs and the need to mitigate them to guarantee the effectiveness of these institutions and the proper use of their resources. Hence, we recognize natural connections and complementarities between higher education governance and corporate governance theoretical frameworks, which could shed light on structuring governance systems inside HEIs. Governance mechanisms at the organizational level complement the country-level system. Hence, arrangements within organizations to mitigate agency tensions are more relevant in institutional environments with weak law enforcement and a low capacity for governmental oversight of organizations under their purview. A study of private HEIs in Colombia, non-membership and non-profit organizations, highlights the relevance of a balance of power in the governance system. We argue that the natural emergence in these organizations of a counterpart or equivalent body to the shareholders' assembly in for-profit firms is a natural development to counterbalance the power of the board. We refer to this body as the assembly of representatives and view its primary responsibility as preserving the foundational goals of the institution and safeguarding its assets and surplus revenues or 'profits'.
We study the statutes of private HEIs in Colombia and find several similarities in their organizational governance systems. The majority of the HEIs in the sample have an assembly of representatives. Those without this governance body have powerful boards, a situation that can lead to severe agency problems that could threaten the future of the institution and the accomplishment of its foundational goals in a context of low legal protection and law enforcement. In addition, the statutes of those HEIs that have an assembly of representatives allocate different functions to this governance body, including the appointment of some board seats, supervision of efforts to accomplish the foundational goals, approval of statute reforms, appointment of an external auditor, approval of annual financial statements, and approval of the HEI's dissolution. This set of functions highlights the relevance of the assembly of representatives in the governance structure. However, in some instances the assembly functions are too vague to be effective or sometimes too excessive, causing an imbalance of power favoring insiders or donor founders and once again creating a scenario with agency tensions that threaten the organization.
We conclude by suggesting that the Ministry of Education establish a minimum of governance practices and mechanisms in Colombia for private HEIs, to preserve the common interest and mitigate potential agency problems.
It is almost impossible, and too expensive, to impose a closer supervision of the 222 private HEIs in the system.
Hence, minimum requirements regarding governance practices could help the ministry accomplish its goals and could support the HEIs in fulfilling their proper role and reason for being. An example is provided by Maassen (2002) who highlights the establishment of a supervisory board inside Dutch HEIs. A body to represent external interests and stakeholders within university governance processes is positioned between the Ministry of Education and the executive board of the HEIs. Under this arrangement, each university nominates the members of its own supervisory board, but the Ministry appoints the members. This case adapted to the Colombian context could impose guidelines from the Ministry of Education on the composition of the HEI assemblies of representatives, but could elicit opposition from those who support university autonomy regarding governance structure. What is clear is the need to discuss and design policies to mitigate agency tensions and foster HEIs' effectiveness through a proper institutional governance structure.