Competitive/comparative governance mechanisms beyond marketization: A refined concept of competition in education governance research

Field of contestants

The first dimension of the entanglement of competition is the field of contestants and the competitive space of relations between these contestants. As demonstrated in the section on the concept of competition in educational governance research, competition-based governance takes place among states, institutions, and individuals (and at levels in between), and these contestants conjointly make up the field in which competition takes place. An important characteristic of the field of contestants is its temporal and spatial features, including how the contestants have come to be exactly that, whether through formal or non-formal, recent or historical processes. At state level, most states not only compete on the global capitalist free market, but are also enmeshed in various transnational or federal organizations, associations, and agreements that may involve particular interests in establishing a competitive field among the participating states. For example, the OECD has installed a space of competition via PISA (Grek, 2009), through which some nation states are articulated as “comparable” and thereby in direct competition with each other. Similarly, intra-national regions or municipalities are sometimes enrolled in mutual competition by the state (Krejsler, 2018), from which new hierarchies and other relations among them emerge.

Institutions may also be enmeshed in various competitive spaces and thereby become contestants. These competitive spaces may be global and include all institutions of a particular kind, or they may be stratified and only include the top league or bottom of the field. Different competitive spaces may also intersect. For example, universities appear as simultaneously embedded in both the global competition installed by rankings and in national competition situations related to government funding (Naidoo, 2018; Robertson, 2017b). In addition, institutions may also cascade the competitive situation downward to other institutional levels, for example to faculties, departments, or degree programs (Madsen, 2019b). Thereby, the contestants may become nested in several layers of competition. These cascading and nesting processes may have contributed to the separation of faculties, departments, or degree programs into individual administrative entities in the first place.

At the individual level, pupils and students are embedded in various competitive communities, including the classroom (Knudsen and Christensen, 2018), special elite programs (Keddie, 2016), and the national space for allocating US students to colleges through the admission test called SAT ¹ (Espeland and Sauder, 2016). However, especially at the individual level, there may be differences in the student population that affect the situations in which contestants compete (Dixon-Román, 2017), for example in relation to gender, race, ethnicity, migration history, socio-economic background, educational level of parents, and so forth. Similarly, the composition of students in schools may affect the attractiveness of these schools for parents, and thereby affect their possibilities to compete in a (more or less controlled) free market of school choice (Bunar, 2010; Gorur, 2013).

All these particularities of the competing entities make up the particular entanglement wherein competition takes place and from where competition obtains its ontology in a specific case. The formal and informal, more or less settled spaces of competition and the inherent differences that make competition an impossible game for some to win, as well as the relations of different competitive situations nested within each other, make a difference to how competition appears in educational governance. The layout of the specific competitive spaces and their effects on the emergence of separate measurable entities are important to include in an analysis of competition.

The rules of the game

As indicated above, the awareness of the field of contestants is already widespread across educational governance literature that discusses competition. A less emphasized dimension in this literature concerns the rules of the game of the particular competition in question. The rules of the game are determined by the particular competition design and determine the affective economies that this design allows for. Here, my use of the word “design” does not imply (nor reject) that one or more actors have intentionally laid out a great master plan for a competition arrangement. The word “design” merely refers to the materialized competition arrangement as it has evolved through specific processes. Different designs, whether these are a product of public policy instruments (Lascoumes and Le Gales, 2007), or of the capitalist model of free markets, affect the ontology of competition in a particular context. Both the competition criteria (Fourcade and Healy, 2017), the hierarchical organization of contestants (Madsen, 2019a), and the fabrication of particular temporalities (Decuypere and Vanden Broeck, 2020) play an important role here.

Let me illustrate this with an example. Wendy Espeland and Michael Sauder (2007, 2016) have studied how US Law School rankings affect the schools. ² In US Law School, the competition design involves an annual ranking, which hierarchically orders entities from 1 to n, regardless of the distances between the scores of the entities. The difference between position 43 and 44 may represent a huge difference in scores, or merely an insignificant difference. The criteria of the ranking are based on a composite measurement, where the reputation of the school is given a weighting of 40%, the selectivity of the students 25%, the placement success of the graduates 20%, and the faculty resources 15%. Espeland and Sauder’s studies show that the exact ranking position, even though this may reflect insignificant measurement differences, directly affects the choice of schools for potential students. As the number of well-qualified students selecting the school is a part of the ranking criteria, the specific position in one ranking affects consecutive rankings. Similarly, another part of the ranking criteria is calculated on the basis of a reputation survey that is highly affected by a school’s previous rank. Espeland and Sauder (2007) highlight this aspect of the measurement as a matter of how rankings can become “self-fulfilling prophecies.” The rankings of US Law Schools is thus an example of competition design that involves upward or downward multiplier effects. As Espeland and Sauder describe, Law Schools positioned in the top tier have a higher chance of scoring well in the next ranking and may be able to attract higher amounts of funding, which improves their chance of performing well on a range of other parameters in the ranking. These schools ride on a wave of prosperity. On the other hand, Law Schools at the bottom of the ranking list might find it difficult to attract well-performing students, high status academic staff, and funding. The competition design has built-in mechanisms that keep the hierarchy among Law Schools intact. These are the rules of this particular game.

Thus, the rules of the game can include dynamics that make some contestants more or less likely to win or lose. They can even include designs where there is no winner. While competition in a leisure context is usually focused upon the victorious winner and the prize, and in a market context is focused upon the profit of the competitive firms compared to the not-so-competitive firms, competition in an education governance context is in many cases primarily concerned with the losers (although there might also be cases where the competition is a “race to the top”; Robertson and Komljenovic, 2016: 598). In public education, competition often involves the identification of weak entities that “need” reform, as also pointed out in relation to PISA (Grek, 2009; Lingard et al., 2016). This is where we can begin to discuss the boundaries between competition and other phenomena. The attitudes of the contestants are affected differently in situations where the hierarchical organization does not involve sorting the contestants on a numerical continuum of performance but merely dichotomizes good and bad performance (Madsen, 2019a).

The rules of the game of competition enable affective economies to circulate through the field of competition. The affective here refers to pre-individual or collective affective states of being that circulate through individual bodies and subjectivities, and from where individual emotions are subtracted (Clough, 2007: 3; Whatmore, 2006: 604). When competition enables affective economies, this means that affectivity is exchanged and capitalized (Brøgger, 2016: 84) in order to motivate contestants to greater efforts. In a traditional capitalist market design, where businesses are competing for profit and only the sky is the limit, affective economies may mobilize actors to invest their time and resources in an entrepreneurial effort in order to achieve a prosperous future. A comparison of comparable but different schools may also invite lower-performing schools to self-improve by creating temporal desires for potential futures visualized in the image of the higher-performing schools (Lewis, 2018). But in competition designs where there is no real opportunity to stand out positively, only a risk of being identified as a problem case, affective economies seem to involve the distribution of guilt and a sense of crisis that can catalyze changes and open a window for reform (Lingard et al., 2016). Competition design or rules of the game and their entanglements with particular affective economies affect the behavior of the contestants and thereby the ontology of competition in a particular context.

The competition objective

The third and final dimension of competition that I propose for my analytical framework is the competition objective. The competition objective depends on the field of contestants and on the rules of the game, including the hierarchical organization of the contestants. It also depends on the stakes in the competitive situation and whether these are direct effects internal to the competitive space or indirect effects imposed by actors external to the competitive space. These mechanisms affect the directional orientation of the contestants.

We can distinguish between horizontal, vertical, and temporal directionalities. Horizontal orientation is typical for highly marketized modes of competition, where institutions such as universities compete against each other for a larger share of the (eventually) limited amount of international (and in some cases also national) fee-paying students, and where success or failure in the market has immediate effects on the university as a business (Blackmore, 2009: 857; Robertson, 2017b: 303). Significantly, horizontal orientation is also typical for other forms of competition deeply entrenched in traditions of education, including academic recognition (Naidoo, 2018). Vertical and temporal orientations are typical for competition situations installed by a government, or by the senior management of an education institution at lower hierarchical levels, or by a teacher in the classroom. The effects of success or failure at a lower hierarchical level are in these cases not immediate, but enforced by actors at the higher hierarchical level. An example of a competition situation entangled with vertical relations and orientation is the allocation of public funding based on performance measures (Naidoo, 2018: 607–608). An example where temporal orientation is strong is the performance measurement embedded in quality systems, where institutional units need to show progress compared to their own previous performance (see the example in the following section). These competitive orientations, whether horizontal, vertical, or temporal, or a complex entanglement of different orientations, critically affect what is at stake for the competing entities, and thereby how the affective economies of competition appear in a given context. For the competing entities, a vertically oriented type of competition is not so much about besting one’s peers, but rather about avoiding an externally imposed reform from above, such as a closure or take-over of a school (Krejsler, 2018: 400), or an externally imposed regulation (Madsen, 2019a). Thereby, vertically and temporally oriented competition will often result in a struggle to become “good enough” to ward off the ever-looming threat of reform, rather than a competition to achieve the highest possible result.

In educational governance literature, the different stakes and orientations, and thereby objectives, materialize as different vocabularies associated with the concept of competition. In competition with immediate internal effects and horizontal orientation, words like marketization, commodification, privatization, students as consumers, profit, investment, positional advantage, and elites are often used (Gorur, 2013; Keddie, 2016; Naidoo, 2018; Robertson, 2017b; Robertson and Dale, 2013; Shore and Wright, 2015a; Tomlinson, 2010, 2012). These concepts have obtained quite a settled role in characterizations of contemporary modes of educational governance, and indicate that the normal conception of a competitive situation is a zero-sum game where the objective is to be more successful than one’s competitors. However, in relation to competition with indirect effects imposed externally, a very different set of concepts is deployed. These include audit cultures with a purpose of management and control (Shore and Wright, 2015b: 24), performance standards and benchmarking imposed to identify weaknesses or problems (Henry et al., 2001: 92; Lawn, 2011: 264–265), and an orientation toward efficiency and risk management (Dale, 2017; Naidoo, 2018: 606). These concepts imply a vertically and temporally oriented competition context where the objective is to improve one’s own performance and satisfy the higher administrative levels. Here, the ontology of competition is altered from a situation of rivalry between actors of a particular kind (nation states, institutions, individuals) in the pursuit of profit, status, and prosperity, to a governance instrument of identifying entities in “need” of reform.

Competition as a governance instrument is used to improve efficiency and austerity in the use of public money (Bear and Mathur, 2015; Henry et al., 2001: 84). It sometimes involves the creation of new markets (Robertson, 2017b), and in these cases the vocabulary of marketization is relevant. But sometimes, it does not quite involve real marketization, but rather managerial competition. New Public Management scholars distinguish between marketization instruments in which sectors are marketized and the public sector is rolled back, and managerial instruments that can involve market-based performance measurement but do not reduce the role of bureaucracy (Andersen et al., 2017: 42; Pedersen and Löfgren, 2012: 437; Pollitt and Bouckaert, 2011: 10). Comparable measurements are one type of managerial instrument that involve competition. They are deployed to individualize the responsibility for a lack of efficiency and austerity to institutions, institutional units, or individuals downward in the (formal or informal) governance hierarchy (Miller, 2001: 381; Miller and Power, 2013: 583–584) via the use of performance contracts and service and quality management (Andersen et al., 2017: 42; Pedersen and Löfgren, 2012: 442–443). Their competition objectives are not immediately meaningful, nor high-stakes, but become so by virtue of the overall reform purpose of efficiency and austerity, and by virtue of the sanctions that may accompany a bad performance.

The deployment of different instruments varies across countries. According to Danish public administration scholars John Storm Pedersen and Karl Löfgren, managerial instruments of governance make up the governance instruments deployed in the Danish education sector in recent decades. This trend is typical for universalistic welfare models, where “the chief mechanism for allocation of resources is not the market” because the consumption of public goods and the payment for it are decoupled, and where the achievement of efficiency and quality is prioritized over the use of particular marketized modes of governance (Pedersen and Löfgren, 2012: 446). Thus, the ontology of competition emerges partly out of its entanglement with particular governance traditions and state forms. As these vary across Europe and the world (Pollitt and Bouckaert, 2011), the specificities of the objective of competition (which is sometimes defined by particular governance relations) are crucial for an analysis that draws on a concept of competition in education governance. A competitive situation may be composed of complex combinations of both marketization and managerial comparison relations, which make the objectives even more important to disentangle.

The graduate unemployment rate in national policy

The graduate unemployment rate was developed as part of a national policy from 2014 aimed at a reduction in graduate unemployment. The policy targeted degree programs from selected areas of study with a relatively high level of graduate unemployment and introduced a cap on the number of graduates that these programs were allowed to produce. Thereby, the policy interfered with the longstanding Danish tradition of delegating management decisions such as enrollment numbers to the universities and other higher education institutions themselves. The tradition of delegating decisions to higher education institutions was embodied in the 2003 University Law, where the institutions became organized as self-owning juridical and economic entities separate at “arm’s-length” from the ministerial organization and governed through contractual relations and financial incentives rather than direct orders (Wright and Ørberg, 2017). The policy on graduate unemployment relieved the institutional management bodies of their autonomy in this area and reassigned decisions regarding enrollment numbers to an algorithmic governance procedure, even though in practice this operation was supported by manual calculations and estimations. The graduate unemployment rate instantly became an important measure in the higher education sector with tangible consequences for the relative sizes of the student cohorts especially in humanities departments (as the humanities were particularly affected by the policy) and thereby also with consequences for national funding allocation.

More specifically, the unemployment rate was used to identify higher education areas of study (as defined in a list of groups of degree programs) with “systematic and striking excess unemployment” (Uddannelses- og Forskningsministeriet, 2014) and assign them a cap size of 90%, 80%, or 70% of previous enrollments. The unemployment rate works as a benchmark, where areas of study with more than two percentage points above the average unemployment rate across all Danish higher education institutions are assigned enrollment caps. Thereby, the policy instrument established a competitive field across the entire higher education sector in Denmark. Measurement against the national average across all higher education programs in the country enforces a strong field of competition across a very diverse group of programs that were not previously considered comparable. However, the market-based measure of the unemployment rate, which is interpreted politically as an indicator of the labor market supply-and-demand equilibrium of various areas of study, imposes commensurability across diverse programs.

The relative measurement implies particular dynamics. On the one hand, the dynamics of competition work like a quagmire, where lower unemployment rates from degree programs with high numbers will cause the average (and thereby the benchmark) to move downward, making it difficult to ever leave the morass of relatively high numbers. The trajectory of the competition design makes no promise of a fairer future in which the lower-performing programs can catch up. On the other hand, there is no real winner situation in this competition design, as the degree programs scoring better than the average do not obtain any benefits. The maximum score of 0% unemployment, which is implied in the construction of the unemployment rate, prevents degree programs from performing exceptionally well. As a result, the unemployment rate benchmark installs a competitive situation where only the losers stand out. Their chances of getting out of the quagmire are low, no matter how hard they try. The design thereby allows for an affective economy in which a negative affectivity is directed toward these areas of study, particularly in the humanities.

The graduate unemployment rate in institutional quality assurance

Besides being the object of a national policy that caps the enrollment numbers for degree programs with a “systematic and striking excess unemployment,” the unemployment rate has also been adopted as a key figure in the quality assurance systems of most Danish universities. The quality assurance systems of Danish universities are subject to accreditation by The Danish Accreditation Institution (2018). The Danish Accreditation Institution requires universities and other higher education institutions to monitor their degree programs via various key figures or performance indicators of which the unemployment rate has become one, in order to identify problems and take action to solve them (The Danish Accreditation Institution, 2013). This practice of accreditation is an important context of the competitive situation, as it makes certain program leaders keen to show that they are willing to solve problems, whereby the key figure raises the stakes for programs.

Quality assurance is a kind of audit process aimed at improving the quality of a public service (in this case education) in order to improve the efficiency of public investment in education, in other words to improve educational outcomes. Interestingly, in the context of quality assurance, degree programs (rather than the broader areas of study) are assigned the role of the contestants. The programs’ status as contestants implies that in this context, it is not the labor market demand for graduates that is being measured, but rather the performance of a particular educational design and content. The unemployment rate of a program is configured as a function of its design and content, and the unemployment rate as a performance measure. In this configuration of the unemployment rate, the horizontal orientation toward other degree programs becomes stronger, though not so much in the form of rivalry, but more in the form of an interest in best practice examples within and across Danish universities.

The design of the quality assurance systems and the use of key figures involve a threshold which marks the dichotomous difference between succeeding and failing. In the case of the graduate unemployment rate, the threshold at most universities is set at two percentage points above the national average, which is the same threshold as the one used to cap enrollment numbers in the national policy. This means that degree programs with a higher unemployment rate than the threshold are in fact competing against this average rather than against the other programs. As a result, this benchmarking mode of competition installs what I would call a comparative attitude oriented toward best practices, rather than a competitive attitude among the programs.

The threshold is used to divide key figures into green and red (and sometimes yellow), where green numbers represent normal performance and red numbers problematic performance. In the annual or bi-annual reports on the key figures, which are routine procedures in the quality assurance systems, red numbers require special attention, reporting, and action planning, while green numbers do not receive a lot of attention. Thereby, the dynamics of the quality assurance system involve a distinction between normal green key figures and extraordinary red key figures, and produce an affective economy of the distribution of urgency and risk among those programs that appear with red key figures. The green programs can be complacent and continue as normal, but do not receive any special reward for their performance. Thus, the competition design is merely about avoiding a particular situation—not about achieving something extraordinary.

Furthermore, besides the national average, the competition is oriented toward senior management and The Danish Accreditation Institution, as well as toward previous performances. During my fieldwork, I experienced the following situation where the vertical relation within the university and its intersections with temporal orientations played out:

Before a meeting at one of the universities I followed in my fieldwork, I met up with a Head of degree program. On our way to the meeting, he talked in an agitated manner about the disturbing events of the preceding week. What had happened was that the University Rector had been very close to closing down his and seven other programs due to their continuous red key figures. Only through an intense presentation of the latest statistics, which pointed toward an improvement in the key figures, were the head of program and his superior able to convince the Rector to maintain the program. The Head of program speculated that the Rector was forced to initiate such drastic moves because he needed to prove his decisiveness and capability to take action to The Accreditation Institution. He was quite relieved that the closure of the program had been warded off. (Observation notes from informal talk with Head of program, May 2017)

In this informal conversation, the entanglement of the quality assurance system with The Danish Accreditation Institution appears to matter in the competition in key figures, in that it installs a vertical relation of a need for consequences if a program does not meet the key figure threshold. The competitiveness of the program is not oriented horizontally toward other programs, not even primarily toward the national average, but temporally inward, toward an improvement of the program compared with previous performances. This vertical-temporal orientation implies that the competition is about satisfying senior management rather than beating the other programs. Obviously, the incidence provoked a temporary disturbance, but the incidence can also be seen as the condensed materialization of a general affective economy of urgency and risk that was imposed on degree programs as a result of red key figures. Even though the closure had been warded off for now, I had the impression that a general sense of urgency had been enforced by the situation. At other universities, the sense of urgency did not appear as strong, as the heads of program at these universities did not see an immediate threat of closure in their departments. Here, the affective state of resignation was more prevalent, due to the feeling that competition factors were beyond their control.

The graduate unemployment rate and students

The affective economy of the unemployment rate also affected the humanities students whom I encountered during my fieldwork. For these students, it seemed as if the high unemployment rates of their programs would affect their positional advantage in the labor market negatively. The negative impact on their positional advantage did not so much appear as an effect of the content and design of the programs, and thereby the skills and knowledge of the students. The negative impact rather appeared to be an effect of the graduate unemployment rate itself. In the media, which mediates the distribution of the graduate unemployment rate to the students, the fine calculative distinctions between programs are generally lost. Thus, the graduate unemployment rate here becomes a binary marker used to single out programs with a high unemployment rate. To the students, the unemployment rate had thus become a separator of programs into high and low utility programs, and thereby of students into graduates more or less worthy of employment. The high unemployment rate has become a negative label for most humanities programs and thereby for the students on them.

This binary separation of programs and students has imposed a competitive space on the students that does not induce humanities students to compete against their peers (neither within nor outside the humanities). Instead, it has installed a collective categorization of these students as at risk of unemployment. The use of the unemployment rate as a comparative measurement instrument that can identify weak entities and thereby allow for effective risk management is thus extended from program level to the disaggregated level of individual students. The categorization of some students as at risk of unemployment has distributed anxiety and pessimism among the students, and these affectivities seem to occupy a great deal of students’ time and attention—attention that could otherwise have been directed at their studies. Furthermore, it was quite common among these students, including several of the students that I interviewed, that they felt the need to counter the positional disadvantage associated with the humanities with a stressful amount of extra activities, such as job experience, internships and so forth. These activities have become a mode of self-reform to the students, and a way of getting out of the risk category.

Summary of the case study

In summary, the competitive struggle for better employment rates in this particular case acquires its ontology through the entanglements of accreditation, organizational hierarchies, key figures produced by particular calculative practices, thresholds and color coding, national averages, comparative attitudes, attention toward the problematic, best practices, a need to prove oneself decisive and prepared to take action, temporalities of improvement or decline, the loss of detail mediated by the media, and affectivities of risk and urgency, resignation, or anxiety and pessimism. The entanglements turn the ontology of competition into a governance instrument and the act of competing into a fight to avoid being reformed, which in the case described took the drastic form of threatened program closure. The affective economies of this configuration of competition involve a utilization of the sense of crisis at the bottom of the competitive field as it paradoxically mobilizes self-reform both at university, program, and student level.

Consequently, the comparable unemployment rate does not create a free market competition. The modes of governance embedded in both the national policy and the quality assurance systems imply an enforced bureaucratization of regulatory decisions and restrictions, rather than an unleashing of the programs onto a free market where they can freely attract students by flashing their low unemployment rates and other quality indicators. Hence, the government and the university management turns to the well-known command-and-control instruments (Pedersen and Löfgren, 2012: 445) that allow them to make political and managerial decisions. Therefore, even though comparable measures are deployed as part of the state and university governance of Danish degree programs, we can only roughly speak of a competitive field among them, and if so to refer to a competition to avoid the bottom of the field and the reform consequences that follow from ending up there.