4 The Critical Ecosystem Partnership Fund and the Promise of Democracy

This chapter and the next evaluate two partnerships in depth, the Critical Ecosystem Partnership Fund (CEPF) and the Global Invasive Species Program (GISP). CEPF was first launched in 2000, when the bank already had significant experience with global partnerships. It is one of the bank’s largest biodiversity partnerships and remains one of the highest-profile initiatives in this issue area. The World Bank also routinely cites it as an example of the kind of partnering that the institution needs to pursue. Partners renewed their commitments to CEPF for a second five-year phase (2007–2012) after the initial phase ended, demonstrating that they were willing to endorse the initiative with their pocketbooks, not just words. However, by 2011, financing for a third phase was not secure, and it seemed to the World Bank that donor commitments would diminish.1

CEPF is also an excellent case for this inquiry because its mandate states: “Enabling a stronger voice, influence and action by civil societies . . . is the hallmark of our approach” (Conservation International 2011a, para. 4). As such, it represents a good test of the ability of bank partnerships to deliver democratic goods, and, insofar as the state is the traditional authority in managing conservation, this is also a partnership that promises innovation.

As noted earlier, bank partnerships have generally been characterized by delegative relationships between the organization’s senior and operational staff at the executive stage, the bank’s operational staff and a lead partner at the management stage, and the lead partner and national organizations at the implementation stage. What remains to be empirically examined is why the bank creates hierarchical instead of horizontal partnerships, to which bodies it delegates in this hierarchy, and to what effect.

To answer these questions, this chapter proceeds in three parts. First, it examines why CEPF is not a horizontal governing arrangement. An analysis of this partnership shows how the bank’s own bureaucratic form of organization instilled a propensity for the bank to replicate similar hierarchical entities that focus on compliance with repeated rules and procedures. This approach promotes hierarchical and bureaucratic partnerships rather than horizontal and democratic initiatives. The second part examines why the bank chose Conservation International (CI) to manage the partnership. It recounts the history of CEPF, and how CI assumed the lead partner position. Here I show that the bank’s risk-averse culture, combined with CI’s ability to efficiently deliver operational goals through knowledge of bank rules and procedures, made it a rational choice as a lead partner. There are important implications behind this, because it suggests that bureaucratic forms of organization end up empowering certain actors but not others in global biodiversity governance. When the bank promotes the influence of a select few organizations, it undermines democracy and curtails the potential for innovation. The third part looks specifically at the conservation approach that the partnership adopted. By analyzing the projects that the partnership finances, this section shows that CEPF promoted traditional command and control type activities, despite claims to innovation because of its small-grant mechanism.

CEPF fell short of achieving the governance values that partnership was supposed to deliver. The theoretical explanation advanced in the previous chapter discussed the tension between value rationality and goal rationality in mature bureaucracies. Bureaucracy privileges instrumentally rational behavior. The conception of rationality for bureaucratic organizations is an economic one: what means need to be mobilized to reach the desired goals? This chapter examines how, in the case of CEPF, the bank focused on mobilizing the required means to reach the goals of timely disbursements, ease of execution, and adequate reporting. This approach ignored democracy and innovation as the ultimate values and ends of partnership governance. In other words, this is a story of how value rationality takes a back seat to goal rationality when large bureaucracies become partners.

Hierarchy, Rules, and Procedures: Bureaucracy Spreads

CEPF is one of the highest-profile partnerships for the World Bank and one of the most often cited examples in bank literature and official speeches of best practices among its environmental partnerships. It is one of the largest bank biodiversity partnership by dollar volume and extent of operational activities on the ground. It was originally launched in 2000 and was renewed in 2007 with additional financing. Its founding partners were CI, the Global Environment Facility (GEF), and the World Bank, with the MacArthur Foundation and the government of Japan joining soon after its inception. In 2007, l’Agence Française de Développement was the last partner to join CEPF. The partner entry fee was a $25 million commitment for five years. Up until 2012, the closing bracket of my study, the partnership managed a total of $250 million from its partners (Critical Ecosystem Partnership Fund 2010a). The six donor members constitute the partnership’s governing body. The partnership provides small grants to civil society groups to protect biodiversity. In some cases, it also cofunds local projects with private industries such as De Beers in South Africa, and Unilever in the Philippines. As such, it promotes bringing an array of state and non-state actors into global biodiversity conservation.

Executive Stage: The Donor Council and the Working Group

Conservation International still hosts the CEPF secretariat, in Arlington, Virginia. The partnership adopted a pure shareholder mode of governance: noncontributing institutions are not part of the decision-making process. The fund is governed by what the partnership calls a “high-level donor council.” During the period of study, which terminated in 2012, the donor council met twice annually and was advised by a working group that also met two times per year, approximately six weeks before the donor council met. The agenda and minutes of the council meetings were regularly posted on the CEPF website. The members of the donor council were all high-level representatives of their institutions. All but four of the five members were CEOs, chairmen, or presidents. In the first phase of the partnership, 2000 to 2007, they included the president of the World Bank, who was chair of the board; the CEO and chair of the GEF; the CEO and chair of CI; the president of the MacArthur Foundation; and a director-level representative from Japan’s Ministry of Finance. In the second phase, which began in 2007 and ended in 2012, James Wolfensohn remained board chair, attesting to his commitment to the partnership and close ties with CI, and the bank was formally represented by the vice president of the bank’s Sustainable Development unit. Decisions formally made by the donor council included setting the strategic vision for the partnership, approval of each of the ecosystem profiles, approval of the annual spending plan, and setting conditions for new donors to enter the fund (World Bank Independent Evaluation Group 2007). Because of the board members’ profiles and because each of these individuals was responsible for leading multimillion-dollar agencies, the group that had actual influence in setting programmatic and geographic priorities, determining the scope of work, and setting partnership grant and governance conditions was the working group. It is at this executive stage that the first delegation relationship unfolded. The bank’s president delegated to operational bank staff in the working group, which included a task team leader, an environmental specialist, and a biodiversity specialist, who liaised with the GEF.

A 2007 bank evaluation noted that the high-level profile of the donor council members was a mixed blessing. On the one hand, it assured visibility and support for the partnership from each partner organization; on the other, “it is unlikely that the members of the Donor Council have sufficient time to ensure that accountability is being adequately defined, accepted, and exercised down the chain of command and control in the program. . . . This task necessarily falls to the members of the Working Group” (World Bank Independent Evaluation Group 2007, xxii). The report also questioned the ease of earnest communication between operational staff and the highest-ranking officers of partner organizations, in light of the hierarchical distance between senior staff and operations staff. Interviews with bank and CEPF staff stress that the working group did most of the substantive negotiation and rolled recommendations up to their superiors, who met in the donor council. Members of the donor council usually adopted the working group’s recommendations. World Bank staff said that, in practice, the bank’s representative contributed needed context to partnership discussions. For example, the vice president had exposure to a wider number of programs and thus could recommend projects with which CEPF could coordinate its work.2

Unlike those for the donor council’s meetings, the agenda and minutes of the working group’s meetings are not publicly available. It was at the level of the working group that institutional tensions tended to play out. For example, in discussing the work program the World Bank wanted to see CEPF expand into new geographic areas, CI argued for consolidating the conservation results achieved to date in a fewer number of ecosystems. Interviews with CEPF staff revealed a concern with selecting fewer areas, ones where the government was stable, capacity was higher, and corruption was controlled, so that gains in conservation could be sustained. CEPF earmarked roughly 20 percent of the budget for “consolidation projects.” In working group meetings, the bank objected to going back to hotspots such as Maputoland (South Africa), while others, such as Madagascar, remained underfunded. From the point of view of CEPF, these differences reflected differences in priority setting between a political multilateral organization that must answer to member states and an NGO that responds to conservation priorities.3

Management Stage: Bank Operations and the CEPF Secretariat

The relationship between bank operations staff and CEPF has fluctuated throughout the years, often depending on the bank task team leader assigned to the partnership.4 There are often visible tensions between the two organizations. For instance, staff at the World Bank have complained that CEPF “acts like a donor when they are service providers.”5 CEPF has been warned not to initiate contact with any member of the World Bank without going through the proper chain of command. The CEPF office in Arlington, Virginia, has on occasion freely invited bank staff from different departments to partnership meetings or events. Bank management has reproached partnership staff for not following proper communication lines with the bank.6 The bank has requested that the partnership coordinate all its activities through the task team leader.

According to the bank, a considerable disadvantage of having CI host the partnership secretariat is the blurring of institutional lines between it and CEPF. From the executive director to the regional grant managers, CEPF staff occupied offices and cubicles side by side with their CI counterparts. CI’s institutional needs and the CEPF’s role became blended as one. In meetings, CEPF staff often spoke as though they were CI staff.7 The bank’s Independent Evaluation Group (IEG) pointed out these and other “costs” that emerge when partnership secretariats are hosted within partner organizations. The bank expects to retain control over the partnership, and expects the managing partner to act as if it were simply a delivery mechanism. The evaluation group noted “(a) the need to transparently identify and manage the conflicts of interest inherent in host arrangements; (b) the ‘two masters’ problem, in which the head of the program management unit reports to both the governing body of the program and the line management in the host organization; and (c) the threat of ‘organizational capture’ by the host organization” (World Bank Independent Evaluation Group 2011, 61). Yet the bank concludes that benefits outweigh costs, because “benefits include the many systems and support services provided by the host organizations, including human resource systems, recruitment, financial management, procurement, communications, legal support, access to information and knowledge databases, and, in the case of international organizations, the privileges and immunities associated with employment” (World Bank Independent Evaluation Group 2011, 61).

Bank staff also see the shortcomings of partnering with CI rather than national organizations, but they describe themselves as being stuck. “Who else can you give the implementation role to? Our transaction costs are too high,” said a bank staff member, commenting on CEPF. When asked why the bank selected CI in particular, the response was, “Because of all the BINGOs, CI was the best administrator; they report to the World Bank once a quarter and the council works,” and then added, “We are a difficult institution for folks who don’t have triple A financial books.”8

During the period of study, CI not only had six hundred field personnel stationed throughout the globe, but also more than twenty field offices seamlessly integrated into the central financial accounting system of CI in Arlington. CI head office’s financial records were audited by PricewaterhouseCoopers, which also conducted sample field office audits. The 2007 operations manual of CEPF illustrates the payoff of partnering with an institution with “triple A financial books”:

CI uses Oracle Financials as its accounting and human resources software. Oracle’s financial management package is an industry leading integrated set of financial management and accounting applications. CI’s budgeting system is also linked to Oracle and runs on an Oracle interface. CI field offices maintain their financial records in Oracle as well, submitting files monthly for review and consolidation, and allowing users with the appropriate authorities to access financial information globally. CI’s Chart of Accounts includes segments for donor, cost center, function, site, and grant number. CI has established a series of donor and grant numbers to track CEPF funds. (Critical Ecosystem Partnership Fund 2007b, 1)

CI remained trusted despite its failure to prevent a conflict of interest as partnership manager. During the first phase from 2000 to 2004, the bulk of CEPF’s financing was awarded to international NGOs and, most prominently, to CI itself, rather than to local civil society groups (World Bank Independent Evaluation Group 2007). The CEPF Annual Report for 2003 shows that more than 40 percent of the year’s grants in the Asia region were awarded to CI. Another 13 percent went to large conservation IOs, such as WWF and Bird Life International. The conflict created by CI’s double role as fund manager and most frequent award recipient was not formally challenged until the first phase of the partnership had ended. Pressure from the United States on the GEF resulted in the GEF’s refusal to formally endorse a second phase until provisions had been made to address this conflict of interest. The message from the US representative to the GEF council was clear about the inadequacy of CI’s response to the problem. A memo dated August 23, 2006, from Helen Walsh of the US Treasury to Steve Gorman of the GEF Secretariat reads:

Thank you for the comprehensive response to the US concerns about the Critical Ecosystems Partnership Fund. This information is helpful in some ways but raises more questions, and we remained concerned about what we view as serious problems with the fund.

The memo goes on to spell out in extensive detail the many issues that informed the US position: administrative expenses, the delivery mechanism, conflicts of interest, complaint mechanisms, independent evaluation, exit strategy, and delegation of authority. For instance, regarding administrative expenses Walsh said:

We do not agree with the CEPF’s apparent definition of administrative expenses. . . . Our very rough estimate of corporate administrative expenses . . . is 38%”. Regarding the delivery mechanism the US memo forcefully questions the sense of having governments provide funding to the GEF who in turn provides money to the Bank, who provides money to a separately governed Fund, with its own administrative structure, and RITs [Regional Implementation Teams] who eventually disburse money to community groups. Each layer adds extra costs and time and is not efficient.

Regarding conflict of interest, she forcefully added:

The problem is a REAL conflict of interest, not a potential or a perceived one, since CI has made decisions on issues in which it has a financial interests. . . . Grantees who lose out to CI, even in regions where CI is not the coordinator would have to ultimately appeal CI if they believe they were unfairly treated. . . .] Also there should be a clearer definition of comparative advantage which does not favour CI, particularly if it cites a compared advantage that relates to previous experience [emphasis in the original]. (Walsh 2006, 2)

Despite the problems the United States pointed out, and the predictable difficulties this program would have to surmount in achieving good governance outcomes, the World Bank supported refinancing the partnership for a second phase beginning in 2007–2008 without funding caps on CI or other international NGOs to act as grantees of a partnership that was intended to empower local civil society. The bank’s rationale for the decision was that capacity was very limited in some countries and, until local NGOs gained institutional expertise to implement CEPF projects, international NGOs could initiate the conservation work on the ground. In the second phase, the World Bank added a formal stipulation to the partnership operations manual. CI could continue to receive money as a grantee; however, when sums of $20,000 or more were involved, the partnership’s working group would have to approve the grant (rather than the CEPF secretariat); for sums lower than that amount, the regional implementation teams could award CI grant funding (Global Environment Facility 2007). CI would also be able to continue to apply to become a regional implementation team, with the condition that the CI representative on the Working Group team recused herself from the selection process.

The extended delegation ladder diagrammed in figure 4.1 has two additional consequences. The first is substantive: as the principal-agent model predicts, there is mandate leakage in each transaction. The second consequence is financial leakage, as noted in the memo by the US Treasury representative Helen Walsh. Every time a new contract is established with an external institutional agent, there is an additional overhead cost of at least 10–15 percent in carrying out the contract to the specific rules of each principal. CEPF, regional implementation teams, and national and local organizations carrying out the biodiversity activities on the ground all have administrative costs associated with management and implementation. Mandate and financial slack become mutually reinforcing. When more money is spent on covering the institutional overhead costs of various agents, less money is available for supervision by the principals, which in turn creates more room for agency slack by the individual agents.

Bank staff said that, given budgetary restrictions, they could carry out supervisory visits to partnership sites once every nine months, and usually by combining monitoring missions they were carrying out for other projects. The global character of the partnership and distant location of the activities relative to capital cities made supervision missions costly. The restrictions on supervision budgets for the bank lowered the overall level of trust that operations staff have in what CEPF is doing in the field. Bank staff argue that the repeated administrative overhead costs also generated donor fatigue and perhaps distrust, and by 2012 they were uncertain that existing donors would be interested in replenishing CEPF in future phases. As of the start of the second phase, both the bank and the GEF declined to pay any administrative costs. The bank operations team argued that it was no longer justifiable to support an expenditure ratio whereby only half the money assigned to an ecosystem was spent on direct conservation activities and the rest was spent on plans, consultancies, or other indirect inputs. The hope was that CEPF would draw up investment plans and other donors would join in and fund the ecosystem profiles. As of 2012, that had not happened. Although a third phase is being discussed, a few donors are making tentative commitments for a third replenishment.

Implementation Stage: CEPF and Local Organizations

It is important to distinguish that a significant level of shared decision-making exists at the pinnacle of the partnership, but not with beneficiaries on the ground. At the community level, CEPF works to “build constituencies of civil society groups in each hotspot that can implement activities in a manner that collectively achieves the program’s conservation goals” (World Bank Independent Evaluation Group 2007, xvii). Building constituencies to implement a predetermined conservation approach does not empower local decision-making in biodiversity governance (Ribot 2004). The discussion in this section demonstrates how a preoccupation with operational efficiencies limits potential achievements in democratic decision-making.

For each ecosystem, the CEPF team developed strategic directions to guide the grant approval process on the ground. These directions were based on what the partnership considered each ecosystem’s most important weaknesses impeding effective conservation. For example, strategic directions might include strengthening the institutional capacity of local groups for conservation, monitoring hotspot areas, legally creating or enforcing protected areas, or enhancing management capacity in these protected areas. Specific project proposals from civil society organizations would be considered for funding if they fit within the parameters of these strategic directions. For example, a project to conduct a survey of vultures in northern Guinea was approved for a grant under the strategic direction of “Monitoring of hotspot areas.” This is the kind of project activity listed under the command and control approach in chapter 3. In policy terms, then, the traditional conservation approaches were still favored over others that might represent innovation.

After CEPF developed the strategic direction for each hotspot, it selected a regional implementation team whose role was to administer grant proposals. In the first phase of CEPF, these activities were centralized. As the partnership evolved, small grants were coordinated by an “anchor project,” a relatively large grant ($1 million or more) whose purpose was to ensure cohesiveness of the smaller grants within the larger ecosystem’s needs. These anchor projects were headed by international NGOs and, in many cases, CI national offices. In the second phase of the partnership, regional implementation teams were instituted to carry out coordination functions.9 These teams were selected through a call for proposals. A CEPF staff member explained that the preference for this mechanism over an invitation to bid was that the CEPF secretariat would retain stronger supervisory oversight over the specific functions and requirements of each regional implementation team.10 This suggests continued hierarchies and limits on actors who might be agents of innovation.

From the onset, it was agreed by the donors that the regional implementation team could not compete for grants, since to both administer and receive grants would amount to another conflict of interest (Critical Ecosystem Partnership Fund 2007a).11 In the 2007 renewal of CEPF, regional implementation teams were given the authority to approve grants below $20,000. This discretionary allowance enhanced the teams’ role from that of simple administrators to that of small-grant reviewers. This was still a limited role in the delegative chain, since regional implementation teams were enabled to evaluate proposals relative to their fit within an ecosystem’s strategic priorities, but donors in the United States could establish those strategic priorities. Furthermore, the teams were most often national offices of international NGOs (many of them CI’s), which circumvented the potential for national or community participation.

Though the partnership’s literature makes frequent reference to the extensive involvement of local communities, there is no evidence that they were called on to define the biodiversity problem but only to respond to the problem as framed by the fund’s partners. A hierarchical programmatic framework has a chilling effect on end-user voices that, in practice, already deploy local approaches to conservation. The case of the Tropical Andes Ecosystem illustrates well the limits to participation at local levels. CEPF financed thirty-seven projects with $8,314,972 in grants between 2000 and 2009 for conservation activities in the Vilcabamba-Amboró conservation corridor (VACC), which runs from Peru to Bolivia. The corridor is a politically volatile area where hydrocarbon concessions, border disputes with Ecuador, the land claims of Amazonian indigenous tribes, and conservation areas collide. The relationship between CI and indigenous groups in the VACC was not a good one. CI had very little engagement at the grassroots level (Earle and Pratt 2009). Indigenous organizations criticized CI’s attempt to portray itself as consulting them, resented the lack of recognition for their ongoing conservation activities, demanded support to fight the damage that oil and gas concessions were causing to their lands, and rejected the label of “partners” of CI (Rainforest Foundation and First Peoples Worldwide 2004).

Of the $8.3 million awarded to the hotspot, 53 percent was granted to national NGOs in Bolivia and Peru, most of which had prior work experience executing projects for the World Bank, CI, or the GEF. The remaining 47 percent was awarded to international NGOs. In three quarters of those cases, CI itself was the final grantee.12 No local community organizations received grants from the CEPF secretariat or the regional implementation team. Instead, one organization, Fundacíon Puma, was selected to re-grant micro financing to local organizations. If one were to draw a map of the institutions that were strengthened to conserve biodiversity as a result of partnership grants in the Vilcabamba-Amboró conservation corridor, two medium-sized circles would surround the capital cities of La Paz and Lima and a large one would surround Washington, DC13 In all but two cases the international grantees were NGOs (the exceptions were two North American universities), and all the national grantees were NGOs, with the single exception of a private tour operator (Conservation International 2012b). Claims that the partnership involved a wide swath of civil society and local communities must be tempered by this financial picture.

All the national grantees had their head offices in La Paz or Lima and were headed by directors whose résumés were dotted with prior experience as consultants for international donors or as project directors of internationally funded development projects. These national NGOs not only are well networked, they know how to develop a project proposal in donor language and format. Fundacíon Puma, which received $500,000 in grants through CEPF to operate a fund that re-grants financing to smaller community organizations, used part of this fund’s resources to run a project preparation school that included training participants to develop projects using the logical framework tool (Fundacíon Puma 2009).

The tool referred to as “logframe” at the bank and in the international development community was developed by the United States Agency for International Development (USAID) in the 1970s. It consists of a four-by-four matrix that lists project goals, purpose, outcomes, and inputs on one axis and objectives, measures, verifiable indicators, and critical assumptions on the other.14 The logframe approach to project development has been the object of extensive criticism (Earle 2003; Gasper 2000; Kilby 2004; MacArthur 1994).15 The logframe tool favors “objectively verifiable indicators,” which are easier to collect from official sources, have a built-in bias toward quantitative over qualitative sources, prioritize formal interactions that can be captured in measurable descriptions, and value precise description that does not accommodate adaptive change through implementation. In sum, the tool is criticized for its rigidity and inflexibility in responding to local contexts, impeding self-directed learning, and breeding a culture of accountability and attribution that creates perverse disincentives for innovation (Ellerman 2001).

The approach also produces risk-averse behavior. In a conference on evaluation and accountability, Lucy Earle presented the paper “Lost in the Matrix: The Logframe and the Local Picture,” noting that “the culture of accountability and attribution breeds fear of retribution. As a result, managers set less ambitious targets and become risk averse, not daring to take the initiative even if they feel instinctively that alternative approaches towards a development problem might be more effective” (Earle 2003, 11). In a BBC lecture series on trust in the public sector, Onora O’Neill (2002) pointed out that the rigidity of such bureaucratic requirements also undermines trust and social capital among supposed “partners” (figure 4.2). The solution is not to eliminate accountability but to analyze biases and to reassess whose purpose it serves. Is accountability meant to feed into the project to enable learning and adapt interventions accordingly? Or are accountability measures upwardly directed toward bureaucratic managers who need to maintain control of the process (Earle 2003; Kramarz and Park 2016)?

Rules and Procedures

CEPF is necessarily a complex organizational arrangement. The partnership awards grants to thousands of civil society organizations across the entire globe, many in areas where fragile ecosystems coexist with poverty and violence, pressure from extractive industries, and the ongoing push of an industrial-scale agricultural frontier, in countries that have weak institutional capacity at the national and local levels and inadequate national budgets for conservation. These factors are the baseline of implementation challenges for a partnership like CEPF. However, the implementation of activities on the ground is made all the more difficult by the rules and procedures that the World Bank imposes on the partnership’s grantors and grantees, none of which actually have much to do with the goals of promoting real democracy and innovation.

It is helpful to start by clarifying how funding flows within the partnership (figure 4.3). CI had a financial agreement with the MacArthur Foundation, the Agence Française de Développement, and the World Bank. The bank had separate agreements with the government of Japan and the GEF, and channeled their funding as fiduciary trustee (Critical Ecosystem Partnership Fund 2007b). CI was the legal administrator of the funds and made these available to the partnership secretariat.

Although there were six donors that constituted this partnership, the language of the CEPF operations manual was explicit about observing the procurement and reporting procedures of the International Bank for Reconstruction and Development (IBRD). For instance: “In no case shall [partnership funds] be used for activities in contravention of the IBRD Safeguard Policies described in section 1(b) of Attachment 2 to this Agreement (‘Standard Provisions’),” and “Grant funds shall not be expended (i) in the territories of any country which is not a member of IBRD or for goods procured in, or services supplied from such territories” (Critical Ecosystem Partnership Fund 2007a, 178). Safeguards are required in all sustainable development investments. However, the point is that the partnership manuals refer only to World Bank safeguards. No provisions appear in the partnership manuals in regard to any of the other donors’ rules, which confirms the overarching influence of World Bank policies and procedures on partnerships. CEFP adopted the bank’s safeguard policies as its own. It incorporated six operational policies (OP) and two bank procedures (BP): Environmental Assessment (OP 4.01), Natural Habitats (OP 4.04), Forests (OP 4.36), Involuntary Resettlement (OP/BP 4.12), and Indigenous Peoples (OP/BP 4.10). The bank defines operational policies as statements that follow from the institution’s Articles of Agreement, while bank procedures are defined as the specific procedures and documentation required to comply with the operational policies (World Bank 2001a). Similarly, the complaint mechanisms available to local communities are those of the World Bank’s own Inspection Panel and the Department of Institutional Integrity. In summary, the bank’s rules rule the partnership’s business.

The procurement processes of CEPF were remarkably “light.” In an interview with World Bank management, this was reported as an achievement of the partnership in its relationship with the IO.16 Bank projects and other partnerships are required to follow onerous procurement processes.17 Furthermore, the organization’s procurement process is set up for buying trucks and shovels, not for enabling processes with local populations. At its core, the bank is an infrastructure institution. In contrast to procurement, the partnership’s operational and financial reporting processes were extensive. At the implementation level, regional implementation teams conducted a programmatic risk evaluation of the individual grantee organizations and their ability to carry out their projects. Each grantee was required to submit financial reports quarterly and progress reports quarterly or semiannually, with audits conducted annually (based on the level of risk assessed). Risk assessments were graded on a scale and included additional risk points if the grantee did not have adequate reporting capabilities to file progress reports or did not satisfactorily address the safeguards policy in the project proposal. Regional implementation teams also completed an additional financial risk worksheet for each grantee that rewarded local organizations that had previously received a CI grant. A number of other rigorous financial and programmatic risk exposure variables were assessed for each grantee. Like the specific types of requirements noted earlier, the risk variables primarily reflected donor needs rather than project integrity, and they were one way in which even at the local level only those organizations that had mastered bureaucratese were allowed to meaningfully participate in conservation activities.

Regional implementation teams and national grantees were also selected based on their ability to disburse funds quickly and reliably, following appropriate rules and procedures on the ground. For example, a CEPF staff said that in Latin America, CI was chosen as the regional implementation team because “we were looking at who could manage quickly and hit the ground running and know the [project selection] methodology.”18 The most tangible benefit of selecting regional implementation teams on the basis of these criteria is operational efficiencies, which allows funding to be disbursed swiftly and according to a program’s timetable. Working with national and community organizations often involves dealing with a lack of institutional capacity to manage a substantial influx of funds, bottlenecks resulting from multiple reporting procedures to donors, high turnover in human resources, and internal politics (Earle and Pratt 2009). This limits the absorptive capacity of national and local organizations. Carrying balances forward from year to year does not create the kind of financial reports that CEPF wants to show donors being asked to renew their commitment to the partnership for the next tranche of financing. The bank’s operational manual mirrors this concern with aid spending. It states that country shares of development lending are determined by five factors, including “the country’s absorptive capacity, and country performance” (World Bank 2004e, OP 8.6, para. 4).19

At the management level, the CEPF secretariat has had to follow reporting procedures that are specific to the World Bank. The only other donor requirement came from the GEF. Before CEPF could begin work in a given country, the GEF operational focal point appointed by the national government had to submit a letter of endorsement.20 Such a requirement can circumscribe the bank’s influence on a partnership. Depending on their influence, national bureaucracies tasked with implementation or oversight of partnership activities can exert their own authority and institute their own processes, thereby diminishing the bank’s goal rationality. In the case of CEPF in China, Chan, Mert, and Pattberg (2017) show how controversial it was to allow GEF and World Bank funding (through the partnership) to flow directly to NGOs, and how CEPF had to redefine its concept of civil society to adapt to the Chinese governance context, agree to fund local governments, and “reinvent itself as a coordination mechanism between government agencies and activities to support government policies” (90). Because of the primacy of sovereignty in international relations and the role of IOs as state clubs rather than supranational organizations, a partnership like CEPF that channels funding through NGOs instead of national agencies can trigger recipient governments to obstruct the bank’s partnerships when their logic of action undermines national priorities. In a critical assessment of the clash between CEPF’s goals and the Chinese government’s goals, Ralph Litzinger (2006) observes:

While the CEPF imagines itself as a universal project with planetary ambitions, which move around the world empire like, in China it has had to funnel its monies through a range of organizations connected to an increasingly anxious state. What results is a transformed global project, where the meanings of environmental activism, civil society, and grassroots resistance are up for grabs. The CEPF thus provides a window in the production of contested sovereignties in the bold new age of imperial sovereignty. (66)

Although other states and bureaucratic organizations did not exert the kind of influence the Chinese government wielded, the experience of CEPF in China suggests an important area for further research. We need to investigate how institutional bureaucracies of varying influence, interacting at global, national, and local levels and at cross purposes, can undermine or benefit partnerships and stakeholders.21

Other than obtaining the endorsement by the GEF national focal point, the remaining management requirements for the CEPF secretariat have focused on bank procedures. This is an illustration of the pervasive influence of bureaucracies. The World Bank is only one of six donors, all contributing the same amount of money to the partnership. The partnership could have been set up with its own rules and procedures, following its own logic. Instead CEPF was developed to adhere to the World Bank’s rules. In the operations manual of the partnership, a section titled “World Bank Reporting Procedures” stipulates that the secretariat must submit quarterly financial reports that include sources and uses of funds, a projected expense forecast, and designated account statements. The secretariat must use the bank’s forms to report this financial information. In addition, it must present an annual external audit, an annual spending plan, and a financial report (Critical Ecosystem Partnership Fund 2007b).

These rules and procedures have a logic within the World Bank but not necessarily within CEPF, since they are onerous and take attention away from substantive governance values within the partnership. For the bank, they are a response to legitimacy demands from civil society. When NGOs have mounted campaigns against the bank’s social or environmental impact, the bank has responded by adding layers of procedures that are meant to protect it from further reputational losses (Weaver 2008). The bank is caught in a dilemma. The bureaucratic character of the organization prompts it to respond to external criticism by adding procedures that are going to make it perhaps more accountable but certainly less inclusive because, with such conditions in place, few organizations can aspire to become bank partners.

Why Conservation International? Risk Aversion and Civil Society Participation

To understand why CI was selected as the managing partner, we need to go back to the original idea for this partnership and trace the influence of authoritative actors in its design, as well as the forces that shaped its launch and subsequent actions. On the CEPF website, CI’s CEO and chair, Peter Seligmann, is identified as the intellectual author of the partnership. In interviews, both CEPF and World Bank staff have also recounted a close friendship between Peter Seligmann and James Wolfensohn as key to the genesis of CEPF. A special publication by CEPF on its tenth anniversary includes a transcript of a public conversation between Seligmann and Wolfensohn, which narrates the origins of the partnership (Critical Ecosystem Partnership Fund 2010b). The two men describe how they met, became friends, and went on fishing trips together in Jackson Hole, Wyoming. During these trips they talked about developing the relationship between CI and the World Bank, where Wolfensohn was about to become president:

Seligmann: You and I met, Jim, at a dinner, and we became friends. And then I think you were offered the position as the president of the World Bank. We went fishing and were talking about ecological hotspots and about your role and what you were going to be able to do.

The two men describe how Wolfensohn wanted to change the face of the World Bank, which was seen as an adversary by NGOs, and Seligmann stood up to criticism for partnering with the bank:

Wolfensohn: This is a tribute to Pete— . . . he was prepared to stand up there and say, “I am prepared to deal with the World Bank.” You remember, Pete, that there were a lot of people who were critical of you because you did it.

Seligmann: I was invited to a whole bunch of meetings by the NGO sector in that period, talking about 50 years [of the World Bank] is enough, and we should come together to have the Bank disband. . . . Of course I disagreed with that. And I was really attracted, Jim, to your personal interest and passion for doing the right thing. . . . I thought, “Here’s someone whom I could not only be a friend with, but someone whom I could really partner with.” I suggested to you that the Bank put $100 million into this. And you came back to me and said, “Let’s create a partnership with a bunch of organizations.” That was when we decided that CI would put in $25 million, and you said the Bank would match it.

Wolfensohn remained the chairman of the CEPF donor board after his presidency, the only member who did not represent an institution and was not a donor of the partnership. The Wolfensohn Family Foundation, a philanthropic organization started in 1995 by James Wolfensohn, has also made private contributions to CI since he left the bank (Conservation International 2013). Adam Wolfensohn, the son of the bank’s president, worked for CI on carbon finance through avoided deforestation from 2002 to 2003 (Bloomberg 2011). Both father and son have been repeat guests at the annual fundraising gala dinner and hosts of after-dinner parties for CI.22

The personal relationship between the bank and CI has not been limited to the highest levels of the organization. Nor are these types of relationships true only of the bank and CI. However, these close relationships are true only of the World Bank and a handful of international conservation NGOs. For instance, the bank’s former chief biodiversity adviser, Thomas Lovejoy, came to the bank after leaving his position of vice president of WWF. Mohammed El-Ashry was senior vice president of World Resources Institute before taking the job of chief environmental adviser to the president and director of the Environment Department at the World Bank and, later, CEO and chair of the bank-administered GEF. Rank-and-file environment staff have similar professional connections to the big international conservation NGOs. For instance, Karen Luz was a senior biodiversity specialist at the bank before she moved on to the position of deputy director at WWF. Claudia Sobrevilla, also a senior biodiversity specialist in the Environment Department, was previously the chief ecologist at The Nature Conservancy. Kathy McKinnon, a long-term bank employee, the lead biodiversity specialist and task team leader for CEPF, came from a previous position at WWF, where her husband, John McKinnon, also worked for many years. Individuals from both institutional sides have published joint academic papers. Hence there are professional, personal, and intellectual ties that bind both groups together.

That there would be close connections between environmental specialists in the World Bank and those working in other conservation organizations should not be surprising. After all, there is a community of expertise that interacts together. The puzzling observation is that this interaction has been overwhelmingly dominated by relationships between the bank and the five largest international conservation NGOs based in the northeast of the United States. Certainly, there is conservation expertise beyond this geographic radius and number of organizations. However, there is an ideational and institutional fit that propels partnerships with the five largest international conservation NGOs and not other civil society organizations. The history of CEPF shows there are very few individuals who would have the sort of access to James Wolfensohn that allowed for the discussion of such a partnership to begin. By Wolfensohn’s own account, he was not thinking about conservation when he entered the bank. He was concerned with the bank’s image among civil society organizations. Wolfensohn said:

I should add that this [biodiversity] was not the center of my attention at that time; the center of my attention was trying to build some sort of relationship at almost any level on almost any subject with civil society, because we were totally distrusted by just about everybody. (Critical Ecosystem Partnership Fund 2010b)

If conservation was not the president’s motivation, why was CI selected for a partnership? Peter Seligmann is a trusted friend, and trusted organizations make a compelling case for partnerships, particularly for the bank’s risk-averse culture. Peter Seligmann was the right man to pitch a partnership to the incoming president. Not many other NGOs could compete with that trust. The picture that emerges is of a bank that is bureaucratic and imposes rules and procedures to which only a few organizations have the institutional capacity to respond, and of one large conservation power broker who could open a door because he was trusted.

CI continued its ascendancy in the bank. In 2010, the World Bank–administered GEF, which provides billions of dollars in financing for environmental projects, issued a call for proposals in which NGOs could become executing agencies and administer funds. CI (along with WWF-US, the IUCN, and the Brazilian Biodiversity Fund, FUNBIO) was one of only four NGOs to have been awarded that status. GEF funds were originally allocated through three international organizations called “implementing agencies:” the World Bank (which is also the GEF trustee), the United Nations Development Programme (UNDP), and the United Nations Environment Programme (UNEP). These agencies were tasked with identifying projects on the ground and helping countries write proposals for submission to the GEF council. Once financing was approved, IOs would be responsible for disbursing funds, supervising project implementation, and evaluating outcomes. Starting in 1999 the GEF expanded the list of executing agencies to include regional development banks and other IOs. When the GEF opened the opportunity to NGOs, CI formally gained access to what was previously reserved for organizations with IO status in global environmental governance. The formal and informal level of influence that this conferred on CI, WWF-US, the IUCN, and FUNBIO (the only NGO from the global south) cannot be overstated. Conservation has become big business for some NGOs.

Failing Innovation: When Agents Limit Conservation Approaches

Bureaucratic hierarchy, rules, and procedures, as well as the World Bank’s risk aversion, have acted as fence posts for the types of partners that are likely to be selected for global conservation partnerships like CEPF. The value of democratizing conservation governance has been eschewed in favor of bureaucratic goal rationality, which values efficiency over effectiveness. Innovation similarly eludes global partnerships, despite claims to the contrary. The bank’s bureaucratic form of organization often requires partners to jump through numerous hoops.23 This reduces the space for creativity and engenders inertia in policy formulation. In business, innovation theory proposes that users (particularly lead users) often form the basis of the most successful commercial product innovations (von Hippel 2005). If form follows function, then the hierarchical structure of CEPF, and its upward focus in reporting requirements, does not enable users’ feedback to reach the bank, CI, or the CEPF secretariat. Rules, procedures, and risk aversion limit the types of organizations from which the bank will hear. Some of the users with the best information to formulate effective conservation interventions on the ground will never enter the conversation because they have not mastered the language. So the next task is to examine when and how actors that do make it through the funnel of World Bank bureaucratic preferences have a feedback effect on bank partnership outcomes. Specifically, they can use their influence to enable experimentation and risk taking, or they can constrain innovation by forcing other project implementation organizations to focus on a limited repertoire of conservation approaches. A discussion of how they use their influence requires widening the analytical lens to include CI’s influence on policy innovation within CEPF.

As mentioned at the start of this chapter, CEPF manages a small-grant program that provides limited financing to NGOs and local communities to work in biodiversity hotspots. The term “hotspot” refers to areas with especially high levels of endemic species that also face severe habitat loss (Conservation International 2011b). It was a categorization created in 1988 by Norman Myers, a British ecologist. Myers’s goal was to establish a priority list of places that were of highest importance to focus conservation resources and efforts on, which he described as getting the “best protection for the least cost” (Myers 1988; Myers et al. 2000). According to this approach of mapping the Earth, there are now thirty-four such biodiversity hotspots (originally ten, but since revised by Myers and Conservation International).24

In 1989, CI adopted the hotspot approach, and it has since been associated as a CI brand. Russell Mittermeier, who revised and co-authored publications on hotspots with Myers, remained president of CI from 1989 until 2014. Some scholars have argued that CI adopted the hotspot concept because it was a young organization—CI was founded in 1987, after it broke away from The Nature Conservancy—that had limited resources and needed to focus on a few areas in the world. It identified the approach as particularly attractive for “selling” CI’s work to large funders like the World Bank, which valued focused interventions (Smith, Verissimo, and MacMillan 2010). Smith and colleagues (2010) have discussed how flagship concepts are beneficial for NGOs because they act as a rallying point; they focus donor attention on a particular species or type of landscape for conservation and raise the organization’s profile in the public eye. Furthermore, since these flagship concepts become identified with the organizations that promote them, the organizations act as gatekeepers of funds. The hotspot flagship concept is seen as a highly successful branding mechanism in raising resources for CI.

The NGO practice of adopting concepts such as hotspots that are identified in high-profile scientific journals and manipulating them for marketing purposes is clearly controversial.25 It gives the impression of sound science and objectivity, which enables an organization to capture funds even when these approaches are widely contested (Chapin 2004; Litzinger 2006; MacDonald 2010; Smith et al. 2009) First, the natural science behind the hotspot approach is debated. Hotspot designation is contingent on two strict criteria: the locales must contain at least 1,500 species of endemic vascular plants and they must exhibit 70 percent habitat loss. Ecologists, many working for other international conservation NGOs, argue against designating areas as hotspots. They dispute the concept because it does not take into account a number of biodiversity measures such as genetic variance, it skews the representation of species richness by focusing on vascular plants and ignoring taxa of other species, and it relies incorrectly on the use of strict thresholds (Kunich 2003; Murdoch et al. 2010). Thresholds may work well as simplifying criteria, but they are, by their own nature, arbitrary lines that fail to create meaningful differentiation of actual threat (Murdoch et al. 2010).

Second, social scientists have been particularly critical of the hotspot approach because it sets aside social considerations that are key drivers of biodiversity loss, including the legal, social, economic, and political context in which these areas exist. The focus on tightly cordoned-off areas designated as hotspots and declared protected does not address root causes of biodiversity loss (Duffy 2010). These areas are home to some of the world’s poorest populations, which are highly dependent for survival on the natural resources that surround them. When CI counted only twenty-five hotspots, these areas already housed more than one billion people, many suffering from severe malnourishment and subsisting on less than $1 per day (Molnar, Scherr, and Khare 2007). Furthermore, 81 percent of major armed conflicts (resulting in one thousand or more deaths) between 1950 and 2000 have taken place in these hotspots (Hanson et al. 2009).

This is a point that is acknowledged by the bank and noted in its own evaluations of CEPF. The bank’s evaluation department criticized the CEPFs decision to focus on environmental protection without sufficient attention to the poverty-conservation nexus. According to the bank’s report,

There is a growing recognition that protected areas alone are not sufficient for biodiversity conservation. Accordingly, the design of CEPF needs to be augmented, particularly from the point of view of the Bank’s mission, by building a livelihoods-based approach into the individual grant schemes to address human threats in protected area buffer zones and the encompassing production landscapes. Adding such a poverty dimension to the program will entail altering the manner in which investments are planned through the ecosystem profiling process. (World Bank Independent Evaluation Group 2007, xix)

Why is the bank not following through on these internal recommendations and ensuring that the partnerships it supports promote more comprehensive conservation approaches than the traditional command and control recipes? This would have entailed crosscutting coordination between different sectors within the bank, for example, between environment and rural development and between research and operations departments. The job specialization feature of bureaucracies does not facilitate this collaboration. The bank assigns a biodiversity specialist to manage its biodiversity partnerships. Biodiversity specialists within the bank recognize the tensions between poverty and conservation but prioritize the latter, and see the tension as an inevitable trade-off for saving biodiversity. For example, a former bank task team leader of CEPF agrees that conservation has a significant impact on local populations but concludes that “unfortunately, there are no silver bullets in conservation, and most successes involve ‘trade-offs’” (MacKinnon 2011, para. 4).26

The assumption that these resource-dependent communities, living in poverty and political violence, can be effectively or ethically removed from the lands that provide for their livelihoods has been both a dominant and a failed approach in conservation. A meta-analysis of community-managed parks shows consistently lower rates of deforestation than in parks without people, thereby increasing both carbon and biodiversity conservation. Conservation outcomes are most significantly influenced by rules made and acknowledged by local user communities (Porter-Bolland et al. 2011). Democratic rule-making is not only a normative priority in global environmental governance, it is also, many argue, a pragmatic principle for policy effectiveness. The encouraging conclusion of such meta-studies is that local needs are not necessarily at odds with the global conservation agenda (Nelson and Chomitz 2011). However, there is a need for new approaches to address the dual goals of biodiversity conservation and enhancement of human welfare.

Yet CEPF relied on traditional command and control conservation recipes. For instance, between 2000 and 2007 a total of 9.4 million hectares of new or expanded protected areas were attributed to CEPF small-grant investments (World Bank Independent Evaluation Group 2007). CEPF did not fund activities that directly required resettlement of populations. However, the definition of displacement is contested: some claim that only outright eviction from lands should be considered, while others contend that economic displacement is a functional equivalent of eviction (Curran et al. 2009). The CEPF’s chief policy of sealing lands for conservation restricted access to resources, which increased populations’ vulnerability to hunger, cold, and disease. In the absence of real alternative sources of livelihoods, these populations would need to “voluntarily” resettle in another area to survive. Innovative alternatives might assume a comprehensive landscape focus that includes environmental, social, economic, and political contexts rather than hotspots as conceptual points of departure. Furthermore, if innovation presumes going beyond what is traditionally done, then governance innovation would involve giving authority for conservation to local governments, community organizations, or local businesses rather than maintaining authority with the state, international NGOs, and multilateral organizations.27 This approach would require the bank to relax the suite of bureaucratic rules and procedures it expects its local partners to know and follow.

Because of the limits of this partnership in promoting democracy and policy innovation, it is not surprising that some of the biggest critiques of CEPF come from community-based conservationists (Litzinger 2006; MacDonald 2010; Smith et al. 2009; World Bank 2007). Ironically, this is the constituency of beneficiaries that CEPF was specifically designed to incorporate in biodiversity governance. If there was one bank partnership that was specifically geared toward participation and democratization of biodiversity governance, this was it. Ribot (2004) argues that internationally led decentralization initiatives in which locals are supposed to lead often result in international institutions making all the decisions and local populations being assigned the implementation work on the ground. Smith and colleagues (2009) add to this critique and argue that priority areas for conservation are often designated by international NGOs and academics without regard for who will implement activities there; consequently, biodiversity effectiveness suffers and democracy in conservation governance is eluded. The CEPF partnership suffered a similar fate. Smith and colleagues (2009) propose that CEPF’s ecosystem planning through CI-designated hotspots has done a lot more for the fundraising success of the NGO than for actual implementation of conservation activities in situ, which they believe is a result of not letting the local residents lead:

When local institutions are weak, this can cause three problems: First, the need to create a sense of urgency among donors leads to short-term funding and ‘quick and dirty’ projects. Second, NGOs tend to advocate their institutional methodology, rather than allowing local agencies to develop approaches that best match their needs. Third, NGO researchers find it easier to produce articles on broad-scale issues for high-impact journals, which help to build scientific support for new campaigns, than to write papers about research on local issues. (280)

Conclusion

CEPF affirms that “enabling a stronger voice, influence and action by civil societies is the hallmark of our approach. Our support equips civil society groups to conserve their environment and influence decisions that affect lives, livelihoods and, ultimately, the global environment for the benefit of all” (Conservation International 2011a, para. 1). According to its own standards of voice, influence, and widespread participation in governance, CI is a model partner for attaining the World Bank’s substantive goals of democratization. Insofar as it places particular emphasis on community governance rather than on state authority to regulate access to and the use of natural resources, this would also seem to be a model of the other substantive goal of policy innovation through partnerships.

However, this case study shows how the highest-profile and most promising bank partnership in biodiversity conservation ultimately delivered limited substantive gains. The narrow goals of promoting institutional needs ultimately trumped the broader goals of democracy and innovation in global governance. The case of CEPF illustrates how a partnership can deliver very well on operational goals. It achieved timely disbursement of more than $120 million through 1,600 small grants to more than six hundred NGOs and private sector organizations throughout thirty-three countries in fifteen regions while maintaining a systematically high level of financial and project reporting (Critical Ecosystem Partnership Fund 2010a; World Bank Independent Evaluation Group 2007). While the bureaucratic requirements of the World Bank strengthened the partnership’s performance in terms of operational outputs, they also limited achievements in democracy and innovation. They limited the participation of actors and inhibited a more comprehensive repertoire of policy scripts for addressing conservation.

During the 2011 Civil Society Forum, a World Bank- and IMF-sponsored gathering that runs in parallel with the two organizations’ annual meetings, civil society had some concrete requests of Robert Zoellick, who would be soon ending his tenure as World Bank president, and Christine Lagarde, who was then just starting her tenure as managing director of the International Monetary Fund. At a large public meeting they asked the World Bank and IMF leaders to deliver on precisely those goals that partnerships had promised. A speech by Laila Iskandar, the civil society representative at this meeting, summed these up. She said:

Can we invite the World Bank and the IMF to open up policy spaces that promote alternative and diverse approaches? We have a wealth of information that differs from the one you have. We want to share it with you. . . . Try us, try working with us. (4)

In smaller gatherings, civil society’s reactions to the CEPF partnership match this sentiment. Local stakeholders say: “You give us your guidelines, then you work with local organizations that are headed by international people,” and ask, “What is the role of civil society in the governance of a partnership? Should they be represented at the table, or sitting at the table?”28 There is a common thread in these reactions to the way the bank and its partnerships do business. This is not coincidental: bureaucracy spreads. Hierarchy, the primacy of rules and procedures, and risk aversion limit the potential of partnerships like CEPF to democratize global environmental governance.