In Proper Organization We Trust: On Extrapolation From Proper Organization Proxies

Who Trusts the Global Union Movement These Days?

On its webpage, the Swedish Union to Union federation proudly presents and defends its democratic coordinating of the global union movement:

The work is carried out in existing independent trade union organizations. It is precisely the large international network of free, democratic trade unions and their global federations that makes trade development cooperation possible. ⁵

Union to Union representatives tell us how they have, over time, experienced a decrease in Sida's readiness to trust the specific domain of the union movement with its traditional democratic coordination procedures. And how, in its place, Sida representatives have come to express a propensity to place their trust in the ideal of an autonomous “proper organization” with a clear mandate and responsibility. This shift in the trust pattern seems to have turned the once unanimously acclaimed democratic structure of global unions into an obstacle to decision-making on the part of the donor, and increasingly also a threat to the legitimacy of aid projects carried out in this domain.

Sida's archive on Union to Union shows a lengthy discussion between the parties centering on the donor's difficulty of assessing risks and results, stemming from what are described as “deviating features” of Union to Union (deviating in the sense of departing from the rational, principal–agent standard form), and how this debate has spurred a large number of controls, such as spot-check reports, organizational assessments, audits and evaluations, over the years. Union to Union is an example of an inclusive federative organization that has long struggled to defend its ways of organizing. One of its representatives stated, for example, that:

Transnational union organizing carries costs. It cannot be avoided… But they [the donors] don’t realize … what is completely lost [in their view and assessment], are the [values of the] coordination aspect.

Here, our interviewee argues that while there is a general understanding in the union domain that it is at coordinated global and regional union meetings that decisions about project aims must be made, and that this is key to good results, it is more seldom that donors understand the value of this type of coordination these days. Instead, they question the ability of the federative organization to govern actions in the direction of results. As stated by Union to Union's former secretary-general (about to leave her post in protest, when interviewed):

They [Sida staff] have come to the conclusion that we don’t fit in as a frame organization… our rules and our movement are not … well … we’re an odd bird.

Sida's appraisal report on Union to Union (Sida, 2018a) requests, for example, that the federation “decrease the number of links in the contract chain to ensure that most of the funds get as far as possible, and to reduce transaction costs.” As mentioned in Chapter 4, Sida also asks for a “clear structure” of contracting parties that accentuates the roles of a clear principal and agent, as more organizations and levels involved throughout the report are discussed as potential drivers of “unnecessary transaction costs” (p. 17). For Union to Union representatives, this feedback from Sida has been difficult to grasp and accept since they see their particular way of organizing as their success factor. Union to Union's (now former) secretary-general stated again:

I mean, one would think that a union… representative democracy must be the most important aspect. Making sure that is in place. But that’s not what is most important, it’s the administrative processes and routines. It’s the ticking of the boxes… The perfect systems. […] And for me, in the midst of this, it’s incredibly painful.

Sida, in turn, relies on and refers to previous consultant reports that call for clarity and simplification of the organization. In the Sida appraisal report on Union to Union (Sida, 2018a), this consultancy advice is repeated in numerous places (such as the below), leaving the impression of a deteriorating trustor–trustee relation:

Union to Union has long had difficulties in providing a clear picture of the structure, governance and control of the operations. In addition, roles and responsibilities have been unclear. (p. 9)

Finally, the management of the own contribution has differentiated from other frameworks' handling and caused the lack of clarity in accounting and follow-up of the operations. (p. 14)

Isomorphic Pressure to Conform to an Organizational Ideal?

Although not yet commonly applied in development aid studies (Moe Fejerskov, 2016), neo-institutional theory offers an explanation to why, despite the good arguments found in contingency theory for defending differences in organizational forms and processes, there is indeed a gold standard in development aid project management these days (Alexius & Vähämäki, 2020). Neo-institutional theory is a rich source of knowledge on how strong ideas are produced, travel, and become institutionalized, that is, the ideas are being taken for granted and adopted as generally good and correct (Czarniawska & Sévon, 1996; Furusten, 2023). A key concept in this theoretical tradition is legitimacy and a core theme is that organizations adapt and respond to the expectations of their institutional environment. ⁶ They do so in order to raise their chances of efficient operations and long-term survival, outcomes of external legitimacy gained from adjusting to expectations in their institutional environments.

And while some of the additional controls requested of Union to Union by Sida may be explained by actual mishaps on Union to Union's part, we suggest that, for the most part, the added control may be due to so-called isomorphic pressure on decision-makers in the donor role (Meyer & Rowan, 1977). The concept of isomorphism, which is key to neo-institutional theory, was first introduced by Meyer and Rowan back in 1977 (Meyer & Rowan, 1977) in their seminal article on how formal organizational structures are often set up as shared “myths and ceremonies” in search of external legitimacy – rather than set up for functional reasons such as to increase internal efficiency. Meyer and Rowan argued that, in the longer run, these processes of adjustment lead organizations of different types to conform (with or without changing their legal form) to similar organizational structures, processes, and technologies in their search for legitimacy (Brunsson, 1994; Furusten, 2023; Meyer & Rowan, 1977). The concept of company-ization, for instance, denotes a process whereby public and civil society organizations take on ideal-typical forms and traits (structures, processes, and ideologies) first associated with the business enterprise (Brunsson, 1994) as these traits have spread to other institutional environments as widely accepted norms. A key insight from neo-institutional theory is hence that aid organizations may become increasingly similar, at least on the surface, as they conform to “rationalized myths” in society about what constitutes a legitimate organization in the aid field.

In line with this insight on the power of shared norms, and as discussed in previous chapters in this volume, the spread of market fundamentalism in recent decades has moved the ideals and practices of performance management to the top of the public agenda. The introduction of the new public management (NPM) doctrine (Hood, 1991; Pollitt, 2007) introduced auditing practices, performance standards, and an increased focus on efficiency and “value for money” through a range of seemingly apolitical tools, which are in many cases “more statements of political faith than empirically demonstrated findings” (Parker & Gould, 1999, p. 114). This development has prompted states to emphasize financial motives and targets in their governance (Alexius & Cisneros Örnberg, 2015; Tarschys, 2006), and the fact that aid organizations are increasingly established as auditable objects (Bromley & Powell, 2012; Power, 1997) is an illustrative example, where acts of auditing that follows certain general standards shape both perceptions of the organizations' performance, as well as their internal processes (Ek Österberg & de Fine Licht, 2021). However, following Bromley and Meyer (2015, introduction), we conclude that the irony is that, in order to realize the ideal of autonomous actors, organizations are highly dependent on their institutional environments:

… organizations are constructed to be proper social actors as much as functionally effective entities. They are painted as autonomous and integrated but depend heavily on external definitions to sustain this depiction.

Thus, as an empirical illustration of isomorphic pressure from the institutional environment, while Union to Union tries to defend a single global union movement aid system with its particular features and modes of coordination, this view is not shared by the Sida representatives. Rather, the Sida representatives experience an external pressure in the form of assessments and audits that emphasize the potential risks embedded in a multitude of contracting parties with “unclear legal status.” Hence, when looking at the situation from the donor's standpoint, we see external norms pressing the aid bureaucrats to shape and support “proper organizations,” at the same time as doing so often conflicts with the official policy aims of variation and diversity in aid implementation.

The “Proper Organization” Ideal

The uncertainties that arise from the many interacting organizations, fluid boundaries, and unpredictable dynamics in development aid operations (Rutter et al., 2020, discussed in Chapter 2) must be responded to, and considering the fluidity and dynamics at stake, many donors long for structure and stability in these relations. Boundaries must be fixed, messiness tidied up. Looking at the theory on relational uncertainty, it becomes reasonable to want a given format, an organizational standard that can be used in attempts to reduce the four sources of relational uncertainty as identified by Knobloch and Solomon (1999) and described in more detail earlier in this volume (see Chapter 2): clarifying the definition of the relationship as well as its goals, norms, and how it is evaluated. As organization scholars, we acknowledge that these key aspects of relational uncertainty are closely related to key elements of the formal organization (Brunsson & Sahlin-Andersson, 2000): membership, decision-making authority, rules, monitoring, and sanctions. The organizational elements of membership and decision-making authority relate to the attempts to reduce the relational uncertainty as concerns the definition of the relationship (e.g., Who are the parties in this relationship? and Who has decision-making authority, and how is responsibility allocated among these parties?). The rules element responds to the need to reduce the uncertainty concerning the relationship's goals and norms. And lastly, the elements of monitoring and sanctions respond to the relational uncertainty concerning how the relationship is evaluated.

By “proper organization,” we refer here to the empirical observation that aid project results today seem to be increasingly described as dependent on organizational elements and capabilities rather than on project features, contextual factors, or key individuals, for example. Although there are more organizations operating in the aid field, and the complexity and variety of actors involved has increased dramatically (as described in Chapter 2), we argue that there is an observable general institutionalized ideal that can be seen among these organizations, a shared norm of what counts, socially, as a trustworthy, legitimate partner in development aid. Core to this ideal is the idea that good results stem from sticking to a standard format for modern actors – actors that are purposeful, autonomous, and rational (Brunsson & Sahlin Andersson, 2000; Fredriksson, 2023). As discussed in Chapter 4, in development aid, this standard format means, for example, that organizations should have a proper results-based management (RBM) system and use certain management technologies.

When donors hesitate to stand up and defend the particularities of a federative democratic structure or collegiality, for example, the respective unions and universities have a harder time being trusted as unions and universities per se. Their domain-specific organizational features and modes of coordination that were once their “unique selling point” (and, according to the official political ideal of diversity in implementation, still are) have in practice often become a “sore spot” of confusion and potential mistrust: “If only they would step in line and become proper organizations!” Consequently, rather than presenting themselves as universities, car producers, unions, sex educators, or public agencies, we find that potential recipients feel pressure to present themselves first and foremost as autonomous actors with clear boundaries and a clearly defined purpose and objectives, as well as a hierarchy with rational decision-making power and generally approved structures, processes, technologies, and methodologies that should, ideally, be validated or certified by external parties such as consultants, auditors, or other governance experts (more on that below) (Bromley & Meyer, 2015; Brunsson & Sahlin-Andersson, 2000).

We believe that one reason why we see an increasing use of proper organization proxies (POPs) today is that we live in an organization society where, as discussed in Chapter 3, organizations are expected to play the lead roles in many complex social settings (Bromley & Meyer, 2015). By looking at it in this light, we are able to make sense of why so many aid bureaucrats are so busy dressing their legal persons in the ideal structures and processes for sustained legitimacy and survival in the complex and uncertain system of development aid.

Many donors are handling more money today but with fewer administrative resources, and this has likely also contributed to the “how” (doing things right) having taken over from the “what” (doing the right things, i.e., mastering both domain-specific and thematic knowledge). Our tentative findings also suggest that there has been a shift over time, where general expertise on how to manage aid projects has become increasingly emphasized and valued by organizations in the donor role. This view is expressed by a senior union representative, with many decades of experience:

What I’ve discovered is that there is absolutely no… or, very little, knowledge within Sida about the unions today. Frankly, I’m chocked!

The former secretary-general of Union to Union similarly stated in an interview that, despite results evaluations having been favorable with respect to the organization's cause and outcomes, she felt that the donor bureaucrat responsible was more interested in the cost-effectiveness of organizing operations in a certain standardized way. The secretary general also explained that she, personally, felt a need to resign from her post because she felt that too much time was being spent on streamlining organizational structures and processes and pinpointing potential administrative risks, when neither weak ownership and coordination nor that money could disappear were the really alarming problems. Rather, in her view, what was really at stake was the lack of civil and human rights.

Another illustration of this mismatch between what organizations in the recipient role believe should be trusted about their operations and modes of organization, and the actual trust pattern of donor representatives, is provided by a program officer from one of the Swedish agencies who stated that:

We are in fact public agencies; we have no personal interest in this. We don’t make money on it […] Sometimes we feel that perhaps Sida should trust us more than they do.

As several agency representatives we interviewed experienced, being a public agency, legally on par with Sida and tasked with carrying out government decisions, is not enough to be trusted to carry out aid operations. And, here again, we find that the donor representative responsible becomes frustrated because, in order to make proper decisions on financing these days, additional regulation and control requirements are required. The head of unit responsible for the Swedish agencies at Sida explains:

It is my experience that other Swedish agencies don’t understand Sida’s core competence. They think that they know this stuff, know how to work in developing countries, more or less like we do. They don’t understand that Sida contributes anything special. We are just a hurdle they have to clear to get the money. They would prefer to get the aid money directly, and the fact that Sida holds the purse strings frustrates them.

This quote suggests a power play and offers the insight that, in practice, both the aid donor and the aid recipient can experience a lack of trust and understanding from their counterpart, concerning context-specific forms, contributions, and expertise. A consequence of the increasing demand for generally acknowledged organizational standards is that domain-specific or thematic expertise becomes less sought after, less valued, less used, and less trusted (Alexius & Vähämäki, 2020).

Unsurprisingly from a neo-institutional perspective, our findings indicate that most organizations that have recently acquired financing from Sida have learned to not deviate from the norm by persisting with accounts of complex, domain-specific information (since, this can increase the donor's uncertainty levels). Rather, most organizations attempt to conform to the institutionalized norms by producing and passing on simplified and standardized information about their organizing that can serve as a source of trust and a proxy for results (Hoey, 2015; Porter, 1996). The recipient's ability to contribute to positive results is thus to a large extent based on general organizational features and management skills, knowledge of certain project modalities, the use of RBM, etc. An almost taken-for-granted formula then centers on rational decision-making and the use of “proper” management techniques.

When looking more generally at cases of low-conflict relations between aid organizations, we find organizations in the recipient role that go to great lengths to conform to the “proper organization” ideal. It is, for example, apparent that in the field of development aid, so-called “strategic partner organizations” (SPOs) occupy a highly regarded position of status among civil society organizations. One of the case organizations we studied, Riksförbundet för sexuell upplysning (RFSU) (the Swedish Society for Sexual Education) reinvested a fair amount of the profits gained from its fully owned business (mainly selling condoms and other contraceptives) in the organization's ability to conform to the many detailed requirements to become an SPO (Alexius & Segnestam Larsson, 2019). RFSU has gone to great lengths and taken on much additional administration starting in 2007 when it embarked on the application process with the ultimate aim of attaining the attractive status of a Sida SPO (at that time called a “frame organization”), which it did attain in 2009, with its first funded projects as an SPO starting in 2010. For this to happen, two new controllers had to be hired at RFSU, consultants and auditors had to be consulted, and countless hours of administrative work was put into the application process. As RFSU's secretary-general commented:

We scurried around like scalded rats the first year. It felt like there were audits upon audits, so many new systems and processes … a big leap for us indeed.

No less than 29 appendices and over 60 application documents later, as stipulated by the application guide, both the secretary-general and head controller at RFSU concluded that, had it not had access to its own company revenue, RFSU would never have been able to follow through and complete the process. The concept of isomorphism helps us to make sense of why RFSU expended all this time and money in an effort to conform to all of these externally defined requirements.

Extrapolating Results From Proper Organization Proxies

In Chapter 4, we presented an account of how the management dreams of simplification and future control have lingered in the field of development aid for over six decades now. And how, in pursuit of these dreams, responsible aid bureaucrats remain highly preoccupied with different approximation practices. As also explained in Chapter 4, the task of approximating results in a project's activity and output phases is difficult and often conflict-ridden. Even when information is available, it is not always beneficial for the legitimacy of the project since approximations may be perceived as insufficient or confusing, which can in turn increase the perceived uncertainty at hand. Over all, we believe that this helps to explain why the kinds of input proxies described below – the POPs – have become so attractive.

In general, results should be secured as soon as possible, ideally even before a project has begun. At this early, preproject stage, organizational structures and processes come in handy, as do separate third-party standards and assessments and management technologies and methods that can all be in place ex ante, before the project decision is taken, and from which future project work and results are then extrapolated. This means that a positive trend is projected from the POPs to the future results. The mathematical concept of extrapolation thus draws our attention to how aid bureaucrats can derive a sense of certainty from tangible organizational input proxies in the preproject state. And while projects are assessed continuously throughout the project period, we find the preproject phase to be particularly emphasized by organizations in the donor role. In fact, in several cases, we found a comparative lack of interest from the donor (Sida) when it came to learning and following up how the different management technologies are used by the recipient and whether and how they affected the results achieved. As a head of unit at Sida pragmatically concluded:

It would have been much more difficult if we’d focused more on the assessments of what they [the recipients] actually do.

To sum up, it seems that referring to POPs gives the bureaucrat responsible a greater sense of certainty, despite the actual uncertainty of the future results.

In Proper Organization We Trust

In recent years, there have been official calls for less NPM-like management and more so-called “trust-based” management in the Swedish public sector. Nevertheless, the trust patterns identified among the aid bureaucrats we interviewed indicate that many of the same old NPM tools and governance processes remain important as trust-enhancing objects and rituals in interorganizational aid relations. The identified trust patterns may therefore help to explain why NPM modes of management and governance linger on in development aid. Formal governance, measurements, and control do not always hamper trust. Depending on the situation, it may even be the other way around: formal governance may actually enhance trust.

Mollering (2006) argues that although there are many similarities to interpersonal trust processes, there tend to be additional factors assessed in interorganizational trust processes. These factors emphasize the impersonal rather than the personal and often focus on bureaucratic procedures and routines. In line with Mollering's conclusions, we found that donors prefer to place their trust in ideal-typical traits and features of what we call the “proper organization” (Alexius & Vähämäki, 2020, 2021). In fact, the actual sources of trust used by donors all seem to point in a similar direction and contribute to an organizational standard for trustworthy aid organizations (Brunsson & Jacobsson, 2000; Segnestam Larsson, 2011).

We have discussed approximation practices in aid organizations, practices which in previous literature have been claimed to contribute to OMD (Natsios, 2010). In many of the examples above, aid bureaucrats have expressed frustration over the fact that their organizations have become more preoccupied with approximation practices than with the content or understanding of their projects. According to Natsios (2010), this type of “mission drift” is a clear risk factor for OMD. For some (perhaps newcomers, outsiders, or individuals who seriously believe that the policy dreams are indeed attainable), these practices might be perceived, at least, as an obsessive seeking of measurements to substantiate facts and replicate the messy, dynamic reality. Most of the senior aid bureaucrats we interviewed say, however, that they understand the need for these approximation practices. And, as one of our informants told us, approximations may even have “saved” the organization in the short run. To put it crudely, the organizations fear that they would no longer be in operation if they had not adapted to certain institutionalized approximation practices.

But what happens in the longer run, when, in order to be an eligible aid recipient, the organization is faced with many POPs? A representative from Swedish mission council states in a report by Wohlgemuth and Ewald (2020) that recipients perceive that “narrative reporting requirements from Sida have eased,” but in their place, “financial reporting requirements have increased” and remain a challenge (p. 23). For organizations in the recipient role, demonstrating both benevolence (a willingness to adhere to a certain management standard to make life easier for the donor) and ability in terms of general management knowledge and skills can represent a shortcut to a higher trustworthiness assessment (Meyer et al., 1995). But a heavy emphasis on POPs may also hinder organizations from being innovative, and some recipients feel that the risk-taking is left up to them and not the donor (Wohlgemuth & Ewald, 2020).

As seen in this chapter (and as will be further discussed in Chapter 7), bureaucrats acting on behalf of organizations in the recipient role may assume that references to their organization's particular institutional context, domain-specific knowledge, and actual complex results will enhance their legitimacy and trustworthiness (Alexius & Vähämäki, 2020). In practice, however, despite official praise, it turns out that donors are not as keen on trusting these sources as recipients think. We found that bureaucrats in the donor role generally opt for general management expertise (such as organizational and RBM expertise), partly because they find it easier to comprehend at a distance (see also Chapter 7). The use of general standards also enables comparisons that donors hope will help them to improve efficiency through monitoring.

To sum up, actual complex information about results on the ground tends to confuse rather than to qualify and hence generally does not lead to a higher trust assessment from a donor. We also conclude, in line with previous studies (e.g., Alexius & Vähämäki, 2020; Vähämäki, 2017), that simply having a certain results technology or organizational structure in place is already interpreted as a result in itself. As one of the senior management consultants from the consultancy company Niras put it:

If it’s not documented, it doesn’t exist. If the results aren’t in a report, there are no results.

Since performance measurements and control requirements form the heart of the trusted ideal, this presents us with a possible paradox. Despite calls in recent years for less of NPM-like rational management methods and more trust-based management, the trust pattern we have identified indicate that what most aid officials actually lean on and trust are the same old NPM and governance tools. To cope with the uncertainties at hand, donors tend to place their trust in impersonal sources, such as general control systems, management technologies, and specific formal structures and processes. It is also likely that they actively hide or downplay the importance of personal relations due to a fear of legitimacy loss (see Chapter 8 and Alexius & Vähämäki, 2020). Paradoxically then, the identified pattern indicate that, despite the official critique, formal control technologies are highly valued as prominent sources of legitimacy and trust that can reduce the perceived uncertainty.

Due to the social messiness and dynamics present, it is seldom the case that decision-makers in the field of development aid are able to predict project outcomes. Still, they are expected to attempt to do so. Senior aid bureaucrats may be well aware that management tools such as RBM and the theory of change will – alas – not serve to fully reduce the uncertainty at hand, but most stand firm in their belief that these techniques should nevertheless be used, for legitimacy reasons, to support decision-making and the general public in assessing whether aid funds deliver results and whether taxpayers are thereby helping to tackle the grand challenges of the world. There are typically no quick fixes for development problems. But when faced with substantial uncertainty, donors can at least make sure that money flows to projects in organizations that are considered proper with reference to acknowledged standards. In the face of uncertainty, donors are able to justify their decision-making, by referring to POPs.