Coordination Conundrum in the United Nations Development System: Solutions from Self-Managed Organizations

Many of the problems that international organizations (IOs) are tasked with solving are interdependent and require concerted efforts. From poverty to education, trade, climate change, and peace-building, contemporary problems cannot be addressed by actors working alone or by siloed thinking. This interdependence is epitomized in the UN’s Sustainable Development Goals (SDGs). But acting in a concerted manner poses organizational obstacles. In this paper, we focus on one of the most prominent of those obstacles, coordination within the UN development system (UNDS)—the collection of UN entities working on development issues. We view international organizations that are part of this system as problem-solving actors that, to achieve their goals, need to find ways to divide their tasks and ensure coordination of those tasks (Mintzberg 1979; Puranam, Alexy, and Reitzig 2014; Haas 1990). Coordination is understood as “the alignment of interdependent organizational activities to accomplish collective organizational tasks” (Jarzabkowski, Lê, and Feldman 2012, 908) and is ensured via a variety of organizational mechanisms. The design and use of such mechanisms have been a key challenge for the UNDS, a complex collection of independent agencies that aim to assist countries in reaching their development goals. Appraisals of the UNDS have long centered on how its fragmented structure inhibits concerted action toward development goals. Consequently, reform efforts have for decades aimed to improve system-wide coherence, parts of which have moved to introduce mechanisms to coordinate work across agencies. Coordination bodies have been instated at the global headquarters level, the regional level, and the country level. At the country level, where agency programs are implemented together with local governments, the organizational fragmentation has been addressed primarily by strengthening the Resident Coordinator and its office, and the UN Country Team it leads—the bodies that bring together the different entities at the country level. The most recent reform, which began in 2018, deepens this approach, which we argue will not bring about much-needed improvements to UNDS coordination. The strengthening of the Resident Coordinator follows a vertical logic of coordination, which is ill-suited for a system made of fragmented and rather autonomous subunits (Mele and Cappellaro 2018, 2) that follow their own vertical work processes (Lindores 2012).

We propose thus to rethink coordination mechanisms and to do this by drawing on insights from organizational theory and management practice on nonhierarchical organizations. In recent years, several large private, public, and not-for-profit organizations have adopted new operating models based on flattened structures, relying on cross-functional teams with specific tasks and empowered with significant authority. The widely shared claim is that such operating models are better suited for a world of uncertainty and volatility that requires a capacity to deal with fast-emerging challenges. Largely without managers to coordinate work, these organizations have developed alternative coordination mechanisms, many of which echo types elaborated in some of the early scholarly literature on the design of organizations (Galbraith 1973; Mintzberg 1979). We argue that some of those mechanisms could be valuable sources of inspiration for addressing coordination challenges within the UNDS, which, given its composite nature and the absence of a central authority of control and command, faces a hybrid situation of within-organization and between-organization coordination issues. We find less value-added in the work that focuses mostly on coordination between organizations, such as the international relations literature or the governance of global value chains. While the logics and incentives of the UN system differ deeply from those of the private sector, coordination challenges are ubiquitous in large organizations, and the mechanisms that are available for addressing them do not fundamentally differ between private and public organizations. Yet we complement the insights from that literature with the practice of coordination in international humanitarian responses as articulated around the notion of humanitarian clusters (HCs).

In this paper, we proceed first with a discussion of the origins and essence of the ongoing reform of the UNDS. We then turn to the theoretical foundations of our approach and their applicability to the case of the UNDS. We enrich the theoretical discussion on coordination solutions with references to recent applications in nonhierarchical organizational settings. We then illustrate the concrete implications of our approach for UNDS reforms with a focus on the Resident Coordinator system, and more broadly to the UNDS management system, illustrated by the case of Pakistan, a country that has a large UNDS presence and that has been open to reforms. We conclude with our main recommendations for a reorientation of the ongoing reform.

The term “UN development system” refers to the complex web of the more than forty-five UN organizations, agencies, funds, and programs that work on some area of sustainable development. The system has its origins in the League of Nations’ approach to economic, social, and financial cooperation and its central idea of separating technical solutions from the politics of member states (Browne 2019, 158). Spread across more than 1,400 field offices in 162 countries, the entities engage in a wide range of programmatic activities aimed at assisting countries in reaching their development goals. The diversity of the entities and their complex relations is indicative of the daunting task of coordinating their work.

Since the founding of the United Nations, the specialized agencies have enjoyed a high degree of autonomy and control over the elaboration and implementation of their programs. Some agencies preexisted the creation of the United Nations. Many were established through thematic conferences held in the aftermath of the Declaration by United Nations in 1942. Entities have also been created to work on shared issues of concern, such as the UN Entity for Gender Equality and the Empowerment of Women – UN Women (2010).¹ However, such joint work has been the exception to the norm where “duplication and competition remain the dominant characteristic” (Browne and Weiss 2021, 12).

The development activities of the UNDS entities take place in field offices at the country level, which report to both the headquarters of their agency and to the UN Country Teams led by Resident Coordinators. Country Teams are made up of the heads of agencies at the country level who have signed the country-specific UN Sustainable Development Cooperation Framework. Each UN Country Team makeup is different, based on the activities UN entities carry out in each country, and based on each country’s development needs. While some UN entities may employ hundreds or in some cases thousands of personnel in multiple country and field offices within a single country, others might be represented by a single project manager or a small team, or the country-level representative might actually sit at the headquarters level of the UN entity they work for.

It is widely recognized that to support countries in achieving their development goals, the UNDS must be better “fit for purpose.” Appraisals of the system have long centered on its fragmented structure, echoing the central findings of the first major report on the UNDS, the capacity study published in 1969 (United Nations 1969). The report called attention to the lack of control, by governments and within the system itself, as well as the need to better ground development activities in national contexts. Fifty years on, the UNDS is still characterized as a rather incohesive collection of independent entities operating in a crowded field of governance arrangements (Bernstein 2017; Fomerand and Dijkzeul 2018). Reform efforts, including part of the ongoing reform, aim to overcome fragmentation and increase national ownership.

The fragmented structure of the UNDS is problematized with respect to how it inhibits effective and efficient pursuit of development goals. Insufficient cooperation and coordination among agencies has long led to the duplication of work within the system, as well as incoherence of policies across agencies (Connolly and Roesch 2020). Even where the mandates of agencies clearly overlap or intersect, “entities tend to operate alone with little synergy and coordination between them” (United Nations 2006, 19). This is problematic for several reasons. First, it increases the costs of administering the system. Second, it leads to inefficiencies stemming from the duplication of administrative and programmatic work in different agencies. Third, it places a higher burden on the developing country governments that must work with multiple entities. Coordination is not an end in itself, and it may not be necessary to improve coherence in all areas (Lindores 2012). Still, there is widespread agreement that many development outcomes can be improved if UN agencies are able to cooperate, coordinate, and exploit synergies. The area of poverty eradication is an example. In his 2018 report to the General Assembly on the implementation of the Third United Nations Decade for the Eradication of Poverty, UN Secretary-General António Guterres emphasized the need to overcome fragmentation and policy incoherence in order to achieve the goals (United Nations 2018a). Most UN agencies have mandates and activities related to this area, but siloed work has prevented coherent and joint action. More generally, the integrated Agenda 2030 highlights that “selective, isolated support measures no longer suffice” (Weinlich et al. 2022, 3).

There are structural and systemic obstacles to a coherent and coordinated UNDS. The accountability lines of the different agencies are divergent, which gives rise to incentives to pursue individual agencies’ goals (Weinlich and Zollinger 2012). Rules and procedures are insufficiently harmonized across entities, such that the transaction cost of coordinated action is high (Fegan-Wyles 2016; Baumann 2018). Over the past years, the portion of funding that is voluntary and earmarked has increased, posing structural challenges to joint-agency efforts as well as to pursuing the nationally most relevant goals.² These challenges are also highlighted in the UN’s own appraisals, such as in the Quadrennial Comprehensive Policy Reviews (e.g., United Nations 2012b, 2016, 2020, and inform the rationale behind the ongoing reform. To address these problems, new entities, rules, and procedures have been introduced to the UNDS.

The UNDS has never had a central authority. At the global level, the system has instead seen the introduction and transformation of different global-level coordination bodies to govern the relatively autonomous entities. From the outset, as outlined in the UN Charter,³ the Economic and Social Council (ECOSOC) was to “co-ordinate the activities of the specialized agencies through consultation with and recommendations to such agencies.” More recently, agencies themselves have been included in bodies set up to coordinate development activities. This is the case with the UN Development Group, set up in 2007 as a part of the reforms implemented under UN Secretary-General Kofi Annan. In 2013 the secretariat for the UN Development Group moved to not only manage and coordinate the inter-agency working groups but also coordinate mechanisms for different agencies to agree on more substantive, programmatic issues, as well as innovation and leadership.⁴ The group was disbanded and re-created as the United Nations Sustainable Development Group as part of the 2018 reforms, to align with the 2030 Agenda on Sustainable Development. It is now the official network of the entities that make up the UNDS. It is made up of the executive heads of each member UN entity (at the Under-Secretary-General level) and is chaired by the Deputy Secretary-General, on behalf of the UN Secretary-General. The UN Sustainable Development Group can now take quicker action on programming initiatives than what was possible prior to its creation when such initiatives required multiple levels of review.

At the country level, the preferred approach has been to increasingly place decision-making at the country offices, while enhancing coherence and coordination among entities by strengthening the capacities of the Resident Coordinator and its office for it to act as a leader of UN development activities on the ground. The Resident Coordinator System comprises the Resident Coordinator and their office, the UN Country Team, coordination units, and common programming and planning tools. ⁵ In 1977 UN member states adopted General Assembly resolution 32/197 that established the Resident Coordinator System and outlined that “[o]n behalf of the United Nations system, over-all responsibility for, and co-ordination of, operational activities for development carried out at the country level should be entrusted to a single official to be designated taking into account the sectors of particular interest to the countries of assignment. . .” (paragraph 34). However, the Resident Coordinator lacks authority and has struggled to lead and coordinate activities in the face of independent entities that have their own internal lines of accountability. The Resident Coordinator’s horizontal oversight often clashes with the vertical work process of individual agencies (Lindores 2012).

An ambitious attempt to achieve a UNDS fit for purpose is the 2007 Delivering as One, a voluntary country-level reform in management and coordination of UN development activities which draws on the UN High-level Panel on UN System-wide Coherence’s report with the same name.⁶ Initially, eight pilot countries signed up to organize the UNDS at country level according to the four building blocks: One Leader, One Budget, One Program, and, where applicable, One Office. UN activities at the country level were to a greater extent placed under the leadership of the Resident Coordinator, who assumed a greater responsibility for coordinating the work among UN entities and with national governments.

However, evidence suggests that the potential envisaged for Delivering as One was not realized (Mele and Cappellaro 2018; Baumann 2018). While applauded for institutional innovation and some enhanced country-level cohesiveness (Fomerand and Dijkzeul 2018), Delivering as One is also considered yet another half-hearted reform attempt (Browne and Weiss 2016). The independent evaluation of Delivering as One (United Nations 2012a) conducted in 2012 found that while national ownership had increased, cumbersome processes continued to hinder coherence as transaction costs for UN entities and their national partners continued to be high. One of its key lessons is that “Delivering as One could be more accurately described as Delivering as if One, given the fact that each UN organization has its own governance structure, mandate and culture” (United Nations 2012a, 83).

The empowerment of the Resident Coordinator under Delivering as One is considered to have been too modest (Weinlich and Zollinger 2012; Baumann 2018), a perspective that underpins the ongoing reform adopted by the UN General Assembly (UNGA) in May 2018 (UNGA A/RES/72/279). The ongoing reform amplifies the trend of moving decision-making from agency headquarters to the country level while empowering the Resident Coordinators. Whereas the reform is encompassing, covering global, regional, and country levels, a key part of it is how the Agenda 2030 and the Sustainable Development Goals give added impetus to achieve concerted action at the country level. The strengthening of the Resident Coordinator and their office is a cornerstone of the reform, whose work can be divided into two main areas: overseeing the elaboration of the Sustainable Development Cooperation Framework—the inter-agency programmatic work plan at the country level—together with representatives of the national government and UN Country Team members, and the coordination of inter-agency operations.

The Sustainable Development Cooperation Framework (previously UN Development Assistance Framework) is now considered “the most important instrument for planning and implementation of the UN development activities at country level in support of the implementation of the 2030 Agenda for Sustainable Development (2030 Agenda)” (A/RES/72/279). It is based on a Common Country Analysis, essentially a desk review of the country situation involving consultation with all UN entities and stakeholders to determine everyone’s priorities, including those of governments and civil society. Following this, a prioritization retreat is held to bring the priorities and inputs of all stakeholders into the Common Country Analysis, and resulting in shared priorities that are aligned with the National Development Priorities and the Sustainable Development Goals, to produce the UN Sustainable Development Cooperation Framework.⁷

The second area is the coordination of the “UN system-wide support to government,” which continues to be a challenge for Resident Coordinators struggling with a lack of resources and authority to achieve effective coherence of action. Previously, inter-agency working groups directed coordination of activities between UN Country Teams, led by a UN entity with significant expertise in the activities being coordinated, but the latest reform aims to standardize essential capacities across Resident Coordinator Offices, particularly in recruiting five core positions covering (a) strategic planning; (b) economics; (c) data management and results reporting; (d) partnership and development finance; and (e) communications and advocacy, as well as additional coordination specialists in multicountry offices. The enhancement of the Resident Coordinator Office’s capacities intends to ensure their credible leadership of the UN Country Teams and engagement with governments. According to the Sustainable Development Group Chair’s report, this has enabled more effective coordination at the country level and has augmented the capacity of Resident Coordinators to contribute to UN Country Team analysis, programming, and partnerships, and to monitor and communicate results (ECOSOC 2021).⁸

While the reform recommendations formulated above apply to the UNDS in all countries, they need to be tailored to the management of each country. We choose here the case of Pakistan as it is one of eight countries that signed up to pilot Delivering as One in 2007 in order to enhance the effectiveness and coherence of development assistance. Since 2009 the UNDS in Pakistan has operated under three versions of the One UN Program (2009–12; 2013–17; 2018–22), the framework that outlines thematic priorities, goals, and management structures (the Sustainable Development Cooperation Framework in Pakistan). Enhancing the coordination between UN agencies’ programmatic activities and between the UN and the Pakistani government has been central to the reforms. To this end, managerial, oversight, and support bodies have been established.

The High-Level Committee was established in 2007 to oversee the reform and comprised UN country-level leadership, the government, and other stakeholders. Below the High-Level Committee, a number of bodies were placed, including the UN Country Team, executive committees, steering committees, subcommittees, and task forces (United Nations Pakistan 2009, 21–23). Under the third and current One UN Program (2018–22) framework, the management structure has been somewhat altered (see figure 1), yet the premise of consolidating coordination responsibility at the organizational apex persists.

In pursuit of both stronger national ownership and interinstitutional cohesion, the strategic direction of the activities under the 2018–22 One UN Program is set by the Joint UN-National Oversight Committee, as well as its provincial counterpart. The Oversight Committee is co-chaired by the Resident Coordinator and a government representative. Internally, the UNDS is led by the Resident Coordinator, which leads the UN Country Team, composed of the heads of the UN agencies and which plays a key inter-agency coordination role. Providing advice upward and tasked with implementing and monitoring activities are the Program Management Team, the Operations Management Team, the UN Communications Group, and ten Outcome Groups, as well as the Provincial Program Teams. In addition to the formalized bodies, need-based task forces and ad hoc coordination bodies complement the management structure (United Nations 2018b, 96).

Table 1 provides a synthesis of the distribution of tasks and composition of the different units of the management system.

To connect back to the broader discussion on the institutional design of UNDS, table 2 reveals that coordination follows a vertical coordination logic. Most, or all, of these bodies established are made up of agency (deputy) heads or specialists that hold significant formal and informal power over programmatic activities and priorities. While the empirical jury is still out on the merits of this new setup, there is good reason to believe that the strengthening of the Resident Coordinator and its office will not markedly improve country-level coordination. The reform increases its capacities and independence, but does not transfer “real authority to RCs [Resident Coordinators], for example by mandating them with the right to make final decisions on UN matters” (Baumann and Weinlich 2018, 3). It therefore follows the same vertical logic of coordination as previous reform efforts while still not solving the existing mismatch between horizontal and vertical oversight. In their evaluation of the reform implementation thus far, Weinlich et al. (2022) find that while there are some improvements in the UNDS’s collective offer at the country level, it remains incomplete. Examples of improvements are not systematic. Looking at the case of coordination in humanitarian emergency responses indicates that such efforts to coordinate via the implementation of a single, overarching strategy are less effective than looser forms of coordination when the different organizational entities have their own mandates and budgetary guidelines (Clarke and Campbell 2018, 666). Moreover, efforts to improve coordination via the reformed Resident Coordinator System are perceived as an additional layer of bureaucracy, potentially duplicating existing coordination mechanisms and increasing meeting constraints on overburdened teams.⁹ From this perspective, the latest reform is a classic case of institutional layering (Mahoney and Thelen 2010). While layering can be a good way of bringing about change in international organizations in the absence of will or capacity for encompassing reforms, it may also reduce efficiency when new rules are followed without removing or reforming existing layers. Our ambition in this paper is to examine potential adjustments to both the existing and new layers that would improve the current setup drawing from insights from work on the design of organizations. With this, we aim to show how incremental changes can improve effectiveness (Singh and Woolcock 2022, this special collection).

In this paper, following Puranam, Alexy, and Reitzig (2014, 163), we view organizations as “(1) a multiagent system with (2) identifiable boundaries and (3) system-level goals (purpose) toward which (4) the constituent agent’s efforts are expected to make a contribution.” This definition is broad enough to be relevant for organizations of different sizes as well as to a system of organizations such as UNDS. Indeed, it is a multiagent system, with broad but identifiable boundaries (see Browne and Weiss 2021, 13), currently oriented toward the achievement of the SDGs with all the agents’ action oriented toward them. Furthermore, efforts to promote the UN Delivering as One indicate the perception by the multiple agents to be working as an integrated organization. It is now expected that everything contained in the country-level programmatic work plan (UN Sustainable Development Cooperation Framework) is also reflected in the country programs documents of UN entities, which must now demonstrate their contribution to the UN’s overall response.¹⁰

To achieve their goals—the provision of public or private goods or services—organizations, according to Puranam, Alexy, and Reitzig (2014), must address two sets of universal problems: division of labor and the integration of effort. Division of labor pertains both to task division and task allocation, and the integration of effort includes the issues of provision of rewards and the provision of information. Martela (2019) further subdivides the provision of rewards into rewarding desired behavior and eliminating free riding and the provision of information into providing direction and ensuring coordination. While any organization has to deal with all those problems, our focus in this piece is on coordination as it has been identified as a recurrent challenge over the series of UNDS reforms.

Coordination is the foundation for the collective performance of activities in an organization. In essence, it is the process of integrating or aligning “interdependent organizational activities to accomplish collective organizational tasks” (Jarzabkowski, Lê, and Feldman 2012, 908). Malone and Crowston (1993, 90) understand coordination as “managing dependencies between activities,” and the plethora of existing definitions all share a core concern with collective and interdependent work toward achieving a goal (Okhuysen and Bechky 2009, 469).

Given its ubiquity in the design and life of organizations, coordination has been the focus of a large array of research across different fields. We do not attempt here to provide a comprehensive literature survey but consider three strands of research that prima facie would appear as most relevant for the case of the UNDS. The first one is the work in the field of international relations that focuses on interactions between organizations in a fragmented or anarchic context. It covers the pioneering work on regime theory that distinguishes coordination from collaboration (Stein 1982; Snidal 1985) and the later work on regime complex with a focus on coordination and differentiation (Zürn and Faude 2013), as well as the work on the agency of international organizations that focuses on their orchestration capacity (Abbott et al. 2015). While some interesting insights can be gained from this literature, such as the notion of vertical versus horizontal coordination (Zürn and Faude 2013, 127–28), it does not capture the system-wide dynamics of the UNDS, which requires a more classical organizational approach, and remains very sketchy on coordination solutions between matching policies or choices.

Work on coordination within global value chains (GVC) is a second interesting source of insights. The degree of explicit coordination between buyers and sellers is a key determinant, along the degree of power asymmetry, of variation in Gereffi, Humphrey, and Sturgeon’s classical five-fold typology of GVC governance structures (Gereffi, Humphrey, and Sturgeon 2005). Among those five types, the UNDS situation resembles relational value chains, characterized by mutual dependence between buyers, host countries (in the case of the UNDS) and sellers, UNDS agencies, and high level of asset specificity (specific country programs in the case of UNDS). In such a situation, there is a need for a “great deal of explicit coordination” that “is achieved through a close dialogue between more or less equal partners”(Gereffi, Humphrey, and Sturgeon 2005, 88). Trust, reputation and spatial proximity facilitate a fruitful coordination. More detailed work on coordination within GVC mostly focuses on the role of lead firms and on issues of global distribution of capabilities across sites, something that is of little relevance for coordination within the UNDS.¹¹

Third, and most promising, research on organizational design in sociology and management has identified a host of coordination mechanisms that help people align their tasks and activities in pursuit of organizational goals. These mechanisms are the “organizational arrangements that allow individuals to realize a collective performance” (Okhuysen and Bechky 2009, 472). Coordination mechanisms differ in the extent to which they are formalized, centralized, personal, and flexible, and they serve different functions and purposes. While traditional manufacturing organizations have been understood to best rely on decomposing tasks and standardizing their execution, organizations operating under uncertainty and with knowledge-intensive processes fare better with flexible and distributed mechanisms that allow for wide information sharing (Faraj and Xiao 2006; Rico et al. 2008). Thus, coordination mechanisms are contextual, an issue that we will deepen in the next subsection.

To guide our discussion below, we draw on the pioneering work of Mintzberg (1979) to identify a few ideal-typical organizational configurations. The structure of an organization is conceived in its simplest form by Mintzberg (1979, 2) “as the sum total of the ways in which it divides its labor into distinct tasks and then achieves coordination among them.” In hierarchical organizations, with direct supervision mechanisms of coordination, managers are a key mechanism through which coordination is achieved, underpinned by the idea that “one brain coordinates several hands” (Mintzberg 1979, 4). Either when managers are removed or when their decision-making power is distributed to teams and individuals in the organization, other mechanisms are required to coordinate work. They include mutual adjustment and different forms of standardization (processes, work outputs, and worker skills). Mintzberg extends this basic insight into a five-fold typology of structures that includes (a) simple structure; (b) machine bureaucracy; (c) professional bureaucracy; (d) divisionalized form; and (e) adhocracy. Most organizations borrow elements from different ideal types, coming closest to one of them. The current wave of flat, or self-managing, organizations (SMOs) has clear analogies with adhocracy, yet with some clear differences (Martela 2019). Those differences do not concern coordination, though, and we therefore ground our analytical discussion on coordination on Mintzberg’s adhocracy as it offers the most thorough discussion of a repertoire of alternative coordination mechanisms applicable to nonhierarchical organizations.

The context in which an organization works is a key element of organizational design. We consider here two contextual features, complexity and dynamism, that apply to the UNDS. We define a complex environment as one in which tasks are highly interdependent and cannot easily be decomposed such that comprehension of the environment cannot be achieved by “one” person (Mintzberg 1979, 276) and a dynamic environment as one in which rapid and unexpected changes make predictions virtually impossible (Mintzberg 1979, 268; Burns and Stalker 1994, 6).

We understand the context in which the UNDS operates as both complex and dynamic. First, its tasks are interdependent, as illustrated by the interdependent Sustainable Development Goals. Work to reduce hunger by enhancing agricultural productivity cannot occur without a view toward access to water, trade, and protection of ecosystems, whether the interdependent tasks be synergistic or conflictual. Approaching complex development issues requires a variety of specialized knowledge, reducing the ease with which a select set of people can comprehend the work of the organization. Second, the UNDS faces unexpected changes in its environments. These changes include shifting national governments, changes in development priorities at the national level, economic shocks, natural disasters, and disease outbreaks and pandemics.

Complex, as opposed to simple, environments are associated with a wider distribution of decision-making, as they resist “one brain coordinat[ing] several hands” (Mintzberg 1979, 4). When an individual lacks the environmental and technical knowledge required to coordinate the tasks, more distributed mechanisms are required (Mintzberg 1979, 270–72). Dynamic environments resist standardization, and the organization requires more flexibility in its coordination mechanism.

Over the last half century, the argument that traditional hierarchical organizations struggle to perform well in dynamic and complex environments has been put forth by organization and management scholars and practitioners. Calls for new organizational models often center on observations that managerial hierarchies hold organizations back in effectively responding to their environments, demotivate employees, and impede innovation. Attention has thereby increasingly turned to organizations that deviate from hierarchical structures, operating with elements of what is often termed self-management. Recent years have witnessed the emergence of the most radical forms of such organizations, what Lee and Edmondson (2017) call self-managing organizations and Laloux (2014) characterizes as Teal Organizations. Within these, the managerial hierarchy gives way to self-organization wherein rules of behavior, distributed decision-making, and organization-wide access to information enable teams and individuals to adjust and coordinate their work according to an organizational purpose. Their organizational structures connect in important ways to earlier efforts to move beyond traditional hierarchies, such as Mintzberg’s (1979) adhocracy, which we just presented, and Burns and Stalker’s (1994) organic organization.

The way these organizations approach design, including coordination mechanisms, merits the attention of those working on and in international organizations, often burdened by cumbersome processes and, in the case of the UNDS, fragmentation that poses challenges to its coherence and ability to deliver collectively. These organizations and their coordination mechanisms come in many different forms. To learn more about this variety, we draw inspiration both from the organizational and management literature as well as from a small subset of five organizations—Buurtzorg, Haier, GE Aviation, W. L. Gore & Associates (Gore), and AES Corporation (AES). Although we do not claim that those five organizations are representative of the whole sample of nonhierarchical organizations, they are interesting sources of inspiration given their size, the multicultural contexts in which they work, and the variation of coordination solutions that they have implemented—especially when read in conjunction with the organization and management literature. In addition to a problem of representativeness, one may argue that their private legal status makes them a poor fit with the public agencies that make up UNDS. Yet coordination is a ubiquitous challenge irrespective of the nature of the goods or services provided by organizations. As an illustration of the relevance of some of the insights from existing work on private entities for international public agencies, we also briefly discuss empirical findings on coordination solutions implemented in international humanitarian responses that involve a large number of actors with separate mandates and strategic objectives.

Before turning to the specific coordination mechanisms implemented by the five nonhierarchical organizations mentioned above, we briefly contextualize that discussion by providing some information on selected organizational characteristics. Buurtzorg is a Dutch home-care organization, organized around parallel and autonomous teams of nurses that have complete decision-making power and no managers. Haier is a Chinese entrepreneurial platform for smart home appliances in which microenterprises can emerge or be attached and collaborate within the platform or outside as they see fit. GE Aviation is a manufacturer and service provider within aviation. Its plants operate with different forms of teaming structures with fluid roles for individuals. Teams serve different functions, and plants and teams are managed partly through committees in which frontline workers participate. Gore is a manufacturer of products, particularly in fabrics, filtration, medical devices, and electronics. It operates with a dynamic project-based team structure, where individuals put forth projects, lobby for funding, and lead teams that are overall governed by collective decision-making. AES¹² is an energy producer and distributer that used to operate with self-managing teams. The teams were responsible for different areas within a plant, and employees would serve on cross-team and cross-plant task forces.

In AES Corporation, employees could, in principle, make any decision as long as they sought advice from those affected by it or with particularly relevant expertise in the domain (Laloux 2014, 99–103). That meant that the forty thousand employees could respond swiftly to the daily demands of energy production and distribution. The one hundred people employed at the US headquarters did not aim to comprehend the full extent of the environments or the full range of tasks performed across the organization. At GE Aviation’s plants, plant workers take on responsibilities for typical management areas such as product quality, production planning, purchasing, and training (Khurana et al. 2012, 5). Those closest to the work—in this case, the operation of manufacturing machines—can call on engineers or the plant manager to support them in making changes to the product design or the working process. When employees hold decision-making power, they do not have to go through traditional layers of approval, which is attractive for organizations operating in complex environments and where outputs are not standardized but have to be adapted to specific (and changing) circumstances (Martela 2019, 16). Not only does this allow for more swift and relevant decision-making; it also reinvigorates employees, giving them a sense of ownership and belonging in the organization.

Organizations operating with organic structures move away from standardizations and toward structures that enable flexible work processes and decision-making. Haier, operating in the fast-paced electronic home goods market, is organized around the interface between autonomous teams (microenterprises) and customers through user community online platforms (Frynas, Mol, and Mellahi 2018). The systematic interactions between autonomous teams and users underpins Haier’s flexibility, relevance, and tailoring of solutions. Buurtzorg’s teams of nurses are highly flexible units whose nurses can respond to their environments. While home-care providers typically standardize their work processes, Buurtzorg nurses are able to adapt their activities to the needs of patients. Sometimes that involves the more straightforward task of ensuring that the patient takes their medication. Other times it means organizing activities that usually fall outside the realm of home-care organizations but that nonetheless play a central role for patient well-being (Laloux 2014, 62–66).

With respect to the human dimension, advocates of self-management emphasize the improved conditions for those working in the organization who, vested with decision-making power, reinvigorate organizations. This echoes the UN Secretary-General’s report on the need to shift the UN management paradigm: “…there is a need to unite staff behind a common purpose and to remember that every job at the United Nations […] is a unique chance for an individual to contribute to the wellness of humankind. The Organization must re-energize the passion to serve the world” (UNGA A/72/492, 9).

Under such new conditions, the mindset and roles of leaders need to evolve significantly. Laloux (2014) emphasizes three key roles for CEOs in his version of self-managed institutions (Teal model): (a) be the ambassador for the organization, represent it to the external world; (b) be a space holder for new work practices; (c) be a role model with a particular responsibility to listen to the purpose of the organization. Performing those roles requires individuals with mindfulness, authenticity, and humility.

When an organization moves away from direct supervision as a coordination mechanism, solutions may turn out to be thorny issues. Whereas a clear organizational purpose makes coordination easier, it is clearly not sufficient as soon as the number of workers extends beyond a limited few. As an illustration, while insisting on the importance of identifying a purpose, the well-known holacracy model (Robertson 2015) identifies coordination roles as central in each of the multiple circles that form the structure of a holacratic organization. As for our case of the UNDS, the breadth of the purpose “contribute to the well-ness of humankind” reinforces the importance of the choice of specific coordination mechanisms.

Another standard solution to coordination, standardization, a key pillar of bureaucracies, has limited bite for organizations working in a complex and dynamic environment. What remains are mechanisms of mutual adjustment. Mintzberg (1979, 162) identifies four of them: (a) liaison positions; (b) task forces and standing committees; (c) integrating managers; (d) matrix structures. Among those, the first two figure prominently in our small set of organizations and beyond.

Liaison positions channel information across departments. A person in this position holds no formal authority and therefore cannot make decisions, and functions more like a formalized communication channel (Mintzberg 1979, 162–63). An example is what Ancona, Backman, and Isaacs (2019, 79) call the “enabling leaders” at Gore. At Gore, organization-wide coordination is achieved inter alia via liaison positions. The enabling leaders engage in organization-wide communication to make sure that each part of the organization is aware of what the others are doing. Employees (or associates, as they are called at Gore) in these positions are commonly people with more experience in the company, but they are more like coaches than traditional managers. But while they do not manage, they tend to have a broader view of what is going on in and outside the organization, thereby providing and channeling information and connecting relevant people—“they can see opportunities to create value and can spot ‘structural holes’ that need to be filled. In some cases, they connect entrepreneurs to end users; in others they provide connections to similar or complementary projects within the firm” (Ancona, Backman, and Isaacs 2019, 79).

Task forces and standing committees are “institutionalized meetings.” That is, they represent a formal part of the structure of the organization when a select set of people meet regularly to discuss issues of shared concern. While a task force is a temporary and task-specific formation, standing committees are more permanent bodies (Mintzberg 1979, 163–64). At AES Corporation, work was coordinated with the use of both temporary task forces and more permanent committees spanning multiple teams or the entire organization. These were made up of all types of employees. All employees were expected to contribute 20 percent of their work hours to task forces and committee work, which coordinated work such as maintenance, safety, human resources, or investment budgeting (Laloux 2014, 89–90). Not only does this increase the diversity of employees’ tasks; it also provides them with a broader view of the organization and its interconnected parts and activities.

Organizations often blend different mechanisms of coordination. As an illustration, at GE Aviation, coordination follows a combination of different solutions. First, team members in a plant take on different roles, such as responsibility for product quality, production planning, purchasing, or training. Second, teams elect a member for a leadership role, which they hold for two-year periods. These elected leaders “take charge of Line of Balance . . . which meant that he or she was responsible for managing the flow of material through production, including assigning members of the teams their tasks each day” (Khurana et al. 2012, 5). This is a coordination role, but it does not fit either the liaison position (because they have formal authority) or the integrating manager position (because they deal with one team/department). Third, workers participate in management by sitting on committees. These representatives are also elected by teams. The committees consist of elected team members, plant leadership, and human resources representatives. There are also specific committees for hiring, wage grades, promotions, et cetera. The committees are important ways of channeling information across the organization “since representatives from each production team could communicate the actions of the committees back to their teams. And their representatives could voice concerns or opinions emerging from the team in a plant-wide context” (Khurana et al. 2012, 5). These look more like Mintzberg’s standing committees.

Although it could be instructive to delve into the details of solutions adopted by the five organizations in our sample, this would go beyond the scope of this paper. Instead, we emphasize a few key aspects from the discussion above so as to assess the potential relevance (or irrelevance) for the situation of UNDS.

1. None of the solutions mimic a top-down style of ensuring intra-organizational coordination. Even when coordination is in the hands of team leaders, those leaders tend to act as facilitators rather than integrators or implementors of a strict strategy. Their role may serve the development of a shared culture.

2. Not only people that are formally or hierarchically higher up in the organization serve as leaders or facilitators. A much broader set of employees have roles and responsibilities relating to a broader organizational oversight, which facilitates coordination.

3. All the solutions tend to rely on fluidity of team composition or of distribution of roles. Job rotation helps develop a better understanding of the tasks and roles throughout the organization and hence ultimately makes coordination more natural.

4. Interpersonal relationships are key to smooth information transmission and coordination of activities, with experienced people (independent of job titles) fostering connections and communication.

5. There is some correlation between the extent of interdependence of tasks and the formality of coordination mechanisms, yet extensive interpersonal relations can be an alternative to formal setups.

Recent work on the practice of coordination among international humanitarian organizations reveals the effectiveness of those solutions in contrast to “coordination by command” attempts (Stephenson 2005, 338). Coordination has been a key challenge in large international responses to disasters with hundreds of organizations striving to provide quick relief. Since 2005, following the Humanitarian Response Review, coordination relies on so-called humanitarian clusters (HCs) that are tasked with sectoral coordination in the main domains of water, health, and logistics. Coordination between clusters relies on intercluster coordination mechanisms under the overall guidance of the Humanitarian Country Team (HCT), composed of the heads of operational agencies and responsible for a unitary approach to coordination (Clarke and Campbell 2018). Whereas leaders of humanitarian clusters have been tasked with multiple functions, ranging from mere facilitators to implementors and integrators, empirical work highlights that the most effective leaders have acted as facilitators (Ruesch et al. 2022) and trust builders (Stephenson 2005). Interpersonal relations are key to effective information sharing leading to voluntary alignment across organizations (Stephenson 2005; Clarke and Campbell 2018).

Coordination in the UNDS relies on a large set of mechanisms, both at the overarching level and at the specific output level. We provide a brief synthesis in the appendix and focus here on the key pillar of the Resident Coordinator and their office.

As highlighted in section 2 above, the Resident Coordinator system is the pillar of coordination in the UNDS. Member states established it in 1977 through the adoption of General Assembly resolution 32/197, which stated:

However, the independence endowed upon UN specialized agencies at the founding of the United Nations meant that Resident Coordinators lacked any authority over most UNDS entities and their staff. This lack of tangible authority over the work of UN entities and their staff continues today, and coordination is therefore reliant on relationship building between the Resident Coordinators and UN heads of agencies in the UN Country Teams. Recent reforms have affected this relationship in two major ways.

First, the 2018 reforms introduced the requirement to make the Resident Coordinator independent of any specific UN entity—while they had previously been by default the Resident Representative of UN Development Programs in a country, skewing the interests of the Resident Coordinator when they met with local government actors to seek funding and/or support for projects. This includes a new appraisal procedure directly under the authority of the Secretary-General,¹⁴ which has established for the first time a certain level of tangible authority that Resident Coordinators have over UN Country Teams. Second, the reforms have increased the capacities of Resident Coordinators with the standardization of resources across Resident Coordinator Offices. The enhancement of the Resident Coordinator Office’s capacities intends to ensure their credible leadership of the UN Country Teams and engagement with governments.

In sum, it is clear that the most recent reforms tend to focus on improving the ability of Resident Coordinators—single individuals, as stated in GA resolution 32/197—to coordinate UNDS action at the country level. While they have contributed to some positive change, those reforms fall short of addressing the key difficulty facing the Resident Coordinators: the lack of formal authority over members of the UN country teams, an obstacle that is unlikely to change in the near future.

Thus, drawing from the experiences of the private organizations and the Humanitarian Country Teams described in the previous section, we argue that reforms should be reoriented away from this “vertical” mindset. Acknowledging that the incentives faced in the private sector differ from those in multilateral organizations, and that development work occurs in less volatile environments than humanitarian emergencies, we specifically highlight how lessons from both can serve as inspiration for a reorientation of UNDS coordination solutions.

First, the UNDS should move its focus on coordination from being solely on the “top” (e.g., the Resident Coordinator and their office) to the coordination functions that team members can perform either on task forces or in liaison positions or in other roles. Resident Coordinators can then serve in a role closer to that outlined by Laloux (2014), as discussed above, and prioritize the fostering of a shared organizational culture. While they can oversee the development of the strategic direction (i.e., the elaboration of the Sustainable Development Cooperation Framework together with the national/regional government), their primary role is to ensure that organizational principles are upheld. In contrast to the current system, there would be many fewer “line managers,” with leadership exercised by senior staff along with some other roles. The Resident Coordinator system would mainly serve as a strategic convener with governments and individual agencies, rather than as an introspective coordinator.

Second, team members performing coordination functions can take several forms. W. L. Gore is interesting with respect to having some more experienced employees taking a “broader view” and thereby serving in liaison positions. AES Corporation’s model is also interesting with its expectation that employees spend 80 percent of their time in their main role and 20 percent contributing to task forces or committees they themselves find interesting. These could either be thematically based committees, or all those different “expert” functions now served by the RC office, UNCT, and the coordination bodies. Such participation aids coordination by connecting people with multiple parts of the organization and providing them with a better overview of the interconnections of parts and activities. It may also help prevent siloed thinking by increasing the disciplinary diversity of these bodies (see Weaver et al. 2022, this special collection).

Third, job rotation should feature as a principle of team composition, as it helps develop a better understanding of tasks and roles throughout the organization and hence ultimately makes coordination more natural.

Fourth, systems should be in place that ensure the smooth and broad provision of information, both via formal communication platforms and channels and the fostering of interpersonal relationships.

As the world is facing the daunting challenges of sustainability and better life for a large part of the human population, making the UN development system fit for purpose has been a key priority. Accordingly, there has been recent effort to significantly reform the current structure weakened by fragmentation, overlap, and insufficient local grounding of activities in national contexts. In this paper, our focus has been on changes that aim at improving coordination with a focus on the Resident Coordinator and their office, and the UN Country Team they lead—the bodies that bring together the different entities at the country level. We argue that the strengthening of the Resident Coordinator both in terms of independence from the UNDS agencies and in terms of capacities is unlikely to bring much improvement as it follows a “vertical logic” of coordination that is ill-suited to a system made up of fragmented and rather autonomous subunits. We thus propose to revisit the changes brought by the latest reform, drawing upon the experience of flattened organizational structures relying on nonhierarchical coordination solutions. Our recommendation follows a “horizontal logic” of coordination with a redistribution of coordination resources within the system and a Resident Coordinator acting mostly as a strategic convener, thus minimizing the negative impact of adding a new layer of bureaucracy on top of the preexisting system.

Cédric Dupont is professor of international relations/political science and co-chair of the TASC (Thinking Ahead on Societal Change) platform at the Graduate Institute of International and Development Studies in Geneva. His previous research has focused on international governance and international regulation, the politics of international economic relations, and international negotiation processes more broadly. Ongoing work includes three major projects. The first one, near completion, develops a systems perspective on international investment arbitration (forthcoming book with Oxford University Press). It highlights the sources of tensions stemming from the difficulty of reconciling the objective to protect foreign investors, on the one hand, and the desire to adopt policies that serve society as a whole. The second explores when and how governments link trade agreements to a series of flanking domestic policies intended to address economic disruptions and spillover effects on the environment and society. The third, more tightly connected to this article, explores the ability of international organizations to address rapidly changing environments. It examines to what extent and how international organizations sense (scan, analyze, interpret) their environments and subsequently reconfigure their means of action to address opportunities and challenges identified at the level of the bureaucracy’s staff. Astrid Skjold is a PhD student in the International Relations and Political Science Department at the Graduate Institute of International and Development Studies in Geneva, and a research associate at the Global Governance Centre. Her PhD project explores the dynamics and consequences of emergency claims in climate politics and analyzes the conditions under which emergency politics take different shapes. She has contributed to the research project “Self-management solutions for the 21st century United Nations,” led by Professor Cédric Dupont, and is currently working on the project “De-blackboxing the Production of Expert Knowledge in Global Governance,” led by Professor Annabelle Littoz-Monnet, both hosted at the Global Governance Centre. She holds an MA in international relations and political science from the Graduate Institute and a BA in international relations and economics from the University of Birmingham.

The authors have no competing interests to declare.

Special thanks to Cecilia Cannon for her assistance on coordination mechanisms within the UNDS and Velibor Jacovleski for his continuous feedback on the broader research project on “Self-management solutions to the 21 st UN” with funding from the Swiss National Foundation (grant CRSK-1_195365). We also thank the editors of this special issue and Silke Weinlich for their detailed comments on a previous version of this paper as well as participants to the workshop on the Future of Multilateralism and Global Development.

UN Country Teams (UNCT)

• UN in-country inter-agency mechanism for coordination and decision-making among UNDS entities in a country, for an integrated response to SDG-related priorities in a country

• Led by Resident Coordinator, made up of country-level heads of agencies

• Co-responsible for the Common Country Analysis (CCA) and the Sustainable Development Cooperation Framework (Cooperation Framework)

• Heads of agencies are responsible for devising/implementing their entity’s part of the Cooperation Framework

• Dual accountability: UN country heads of agencies are accountable to own agency’s regional director and the Resident Coordinator

Resident Coordinator

• Independent of any specific UN entity since 2018 (previously UNDP resident representative)

• Coordinates the UNCT, co-responsible with UNCT for the CCA

• Co-responsible with UNCT and national government for the Cooperation Framework

• Responsible for the One Programmes (inter-agency coordination of operations)

• Since 2020, Resident Coordinators are accountable to the UN Secretary-General; appraisal process includes feedback from the UN Sustainable Development Group Regional Directors, with input from heads of agencies in a country

Country Analytical Support Team (CAST) (formerly Regional Peer Support Group)

• Supports the UNCT and Resident Coordinator in conducting the CCA

• Convened by the Regional Development Cooperation Office from interested UNDS entities (Regional Economic Commissions, regional and headquarters entities on development, humanitarian, human rights, political, peace and security)

• Helps identify/encourage UN agencies to contribute to CCA

• Resident Coordinator submits the CCA to the CAST for quality assurance, prior to sharing with national stakeholders for strategic prioritization

• Disperses after the CCA; may be called on to support future analyses

Data and Analysis Repository

• The CCA process produces/updates the Data and Analysis Repository

• Ongoing country updates to serve the UNDS and to shorten the time required to conduct Common Country Analyses (ongoing up-to-date information and data)

• UNCT meets at least annually to review the Repository and to identify new data needs, analyses, or other outputs

• Includes links to other data platforms and sources

National/UN joint steering committees

• Co-chaired by Resident Coordinator and senior counterpart government representative

• Composed of key partners in the Cooperation Framework; meets at least annually

• Ensures strategic direction and oversight of Cooperation Framework; alignment with national, regional, and international development plans; links with other processes; monitors progress; steers implementation of Cooperation Framework; reviews One UN Country Results Report; supports resourcing for Cooperation Framework/development

• Resident Coordinator presents the One UN Country Results Report, evaluation reports, and evaluation management response and action plan

Cooperation Framework results groups and thematic groups

• Created for each Cooperation Framework strategic priority to improve internal coordination and ensure a coherent UN system approach to each strategic priority

• Composed of UNCT entities, chaired or co-chaired by UN heads of agencies and relevant government counterparts; meet every two months

• Aligned with existing government-led working structures to encourage UN coherence, incorporating relevant national and international partners

• Results groups, led by the Resident Coordinator, develop UN joint work plans for the Cooperation Framework, identify opportunities for closer inter-agency collaboration (e.g., through joint programmes), monitor and report on joint outputs, and input to CCA

• UNCTs are encouraged to establish working mechanisms (thematic groups, advisory capacity)

Joint programmes and joint work plans

• Cooperation Framework results groups may identify need for increased joint delivery through joint programmes—sets of activities contained in joint work plans (plus related Funding Framework), involving two or more UN entities contributing to the same Cooperation Framework outputs

• Reflect Cooperation Framework outputs, all related key UN development contributions delivered jointly or by individual entities, to maximize synergies and avoid duplication

• Can be annual or multiyear, detailing needed/available resources and funding gaps (pooled or other funding mechanisms can fund joint programmes)

• Other stakeholders, UN missions, and humanitarian actors can be implementing partners, for greater UN coherence and to reduce transaction costs

UN Development Programme Representative

• Supports countries in their efforts to realize the 2030 Agenda, assisting the Resident Coordinator and UNCT to deliver integrated and multidimensional approaches to the SDGs through country platforms and other tools

• Helps identify relevant UN entities based on needs for achieving the SDGs

Monitoring and Evaluation Groups

• Coordinates monitoring and evaluation of Cooperation Framework among UNCT entities

• Composed of M&E representatives from each UNCT entity

Resident Coordinator’s Office

• Resident Coordinator’s office now resourced with a team of four to five support staff to facilitate coordination among UNCT entities for core operations (finance, strategic planning, communications, partnerships, evaluation and monitoring)

Committees to coordinate operations among UNCT entities

• Committees to coordinate operations among UNCT entities can be chaired by a UN entity with relevant expertise