Industrial action involving teachers’ unions in Norway and Denmark: comparing bargaining trends and outcomes

Industrial action involving teachers’ unions in Norway and Denmark: comparing bargaining trends and outcomes

Eyvind Elstad*

Department of Teacher Education and School Research, University of Oslo, Oslo, Norway


The purpose of this article is to contribute to an improved understanding of (1) changes in bargaining models between employer and employees, and (2) differences in these changes in Norway and Denmark. The aim is to explain divergent outcomes of collective bargaining regarding teachers’ working hours in Norway and Denmark. Social media is assumed to reinforce grass-roots rebellion level as well as the internal decision-making structure within the unions. The theoretical framework draws on situational analysis in order to understand how exercising of power at different levels influences the negotiations between employer and employee as well as the outcome. Drawing on situational analysis, the research method has a comparative approach, comparing two case studies, one from the Norwegian context and one from the Danish. The Danish employers’ organisation succeeded to perform major changes after a lock out, whereas the same ambition failed in Norway because of a teacher strike supported by social media. By using a comparative method, based on analyses of the two cases, the author aims to explain possible mechanisms of negotiation. The findings, presented in the discussion, show that new circumstances concerning the work of the teacher unions are disrupting the prevailing industrial negotiation model. The force of social media and newspapers in evoking public sympathy for teacher unions in Norway is an important explanatory factor regarding outcome differences in Norway and Denmark.

Keywords: bargaining; teacher union; Norway; Denmark; social media.

Citation: NordSTEP 2016, 2: 31165 -

Copyright: © 2016 Eyvind Elstad. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License, allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.

Published: 28 June 2016

*Correspondence to: Eyvind Elstad, ILS, University of Oslo, P.O. Box 1099 Blindern, NO-0317 Oslo, Norway, Email:


The Scandinavian countries have long had a tradition that employees have significant influence in centralised bargaining regarding collective agreements on salary and working hours (Byrkjeflot, 2001). The term Nordic or Scandinavian bargaining (or negotiating) model (which is shielded as a brand by The Court of Patent Appeals) refers to inter-party collaboration between employer and employee organisations (and, in some instances, the state) which is based on influence, joint decision-making and mutual relational respect (Brandal, Bratberg, & Thorsen, 2013). The idea of the inter-party collaboration is to reduce the negative effects of inherent conflicts of interest which in some instances can disrupt processes (Moene & Wallerstein, 2002). Negotiations take often place in connection with collective and centralised tariff agreements (Ingle, Willis, & Fritz, 2014) and through agreements on working conditions (e.g. the overall framework agreement which establishes fixed periods of truce during the lifetime of agreements between employer and employee organisations). In addition to a centralised agreement on collective salary and working conditions, local negotiated settlements provide salary supplements where necessary to satisfy the needs of individual companies in relation to recruitment (Barth, Moene, & Willumsen, 2014; Holden, 1988). This basic model for finding solutions to conflicts has been regarded with both astonishment and interest, given that the influence of Scandinavian trade unions has contributed to a relatively low level of workplace conflict and a relatively high standard of living among industrial workers (Lindbeck, 1997).

The bargaining model from the industrial sector has been of significance for the issue that will be studied in this article: collective negotiations regarding teachers’ working hours. However, is the archetypal bargaining model based on joint decision-making, influence and mutual respect a good description of the nature of bargaining processes in which Danish and Norwegian teacher unions and employers are involved? In recent years, the countries which participate in the European Union (EU) have witnessed an accelerated process of integration, which has affected the process of collective wage bargaining on wages (Dumont, Rayp, & Willemé, 2006). A shift in bargaining power from labour to employers is especially present in some EU countries (Abraham, Konings, & Vanormelingen, 2009; McCann, 2010).

Another trend is the advent of social media which may reinforce mass mobilisation and grass-roots rebellion by communication (Tufekci & Wilson, 2012). Grass-roots mobilisation is a consequence of bargaining. These trends are partly converging and diverging. It is an open question whether the potential use of social media heralds a new age for teacher unions’ execution of power or not. The purpose of this article is to contribute to an improved understanding for (1) changes in the bargaining models between employer and employees, and (2) differences in these changes in Norway and Denmark. Both countries share a Scandinavian bargaining model, but Denmark is a member of EU and Norway is not. The differences between the Danish and Norwegian situations emerge also via the structure of interaction and influence of social media mobilisation in negotiations regarding working hours can be of significance for the distribution of power between the teacher unions and their employer counterparts.

The contexts

The conditions that allowed the role of the unions to become increasingly powerful must be seen in connection with long periods in which social democratic parties have been in power (Moene & Wallerstein, 1995). Social democratic governments have historically been significant for the development of the Scandinavian bargaining model and the welfare state (Falch & Rattsø, 1997) in that the state functions as a third party in the negotiations (Iversen & Stephens, 2008). Generous welfare systems established through a comprehensive public sector have had significance for the establishment of the inter-party collaboration that is a feature of the Scandinavian bargaining model (Barth et al., 2014; Moene & Wallerstein, 2002). The centralised bargaining model remains a powerful force, but in connection with negotiations around working conditions for teachers the inter-party collaboration has to some extent been placed under pressure (Danmarks Lærerforening, 2013). Whether these changes herald a new age for bargaining in a Scandinavian context is an open question. If so, this tendency induces new challenges for teacher unions. In this part, I compare Norwegian and Danish contexts; however, the same tension induced by changes in the bargaining model between employer and employees is present in a Swedish context (Piazza, 2005; Wiborg, 2013).

The working hours and salaries of teachers in Norway were formerly determined through negotiations with the state, but in which the local authorities and county councils were the actual employers. The Conservative Bondevik II government (2001–2005) emphasised decentralisation of decision-making and for this reason in 2004 transferred the responsibility of bargaining salaries and working hours regulations to a local-authority’s corporate body: Kommunenes Sentralforbund (KS). In Denmark, an equivalent organisation has been given responsibility for bargaining with primary and lower secondary teachers, while the Danish state conducts negotiations with the upper secondary teachers (Kommunernes Landsforening, abbreviated KL). What is interesting is that the results have differed so greatly between Norway and Denmark.

The demands for change are part of an international trend which emphasises the decentralisation of decision-taking (i.e. head teachers have management rights), tighter control over teachers’ work by means of test-based accountability systems and marketisation principles for schools (Ingersoll, 2006). In Norway and Denmark, these reform ideas have met opposition from the teaching unions. In Denmark, the Danish state held negotiations in 2012–2013 with Gymnasieskolernes Lærerforening (GL), in which GL’s leader surrendered to employer’s demands for significant reforms to working hours. The organisation of Danish local authorities entitled KL encountered greater resistance in their negotiations with the union of primary and lower secondary school teachers. In this situation, KL chose to impose a lockout of all primary and lower secondary school teachers. The lockout lasted for a month in 2013. The Danish government stepped in with an imposed public mediation which gave victory to the employers. ‘We have lost the battle. KL has got things exactly as they want them. And it is grotesque’, stated the leader of the teaching unions (Pedersen, 2013). The changes in Denmark inspired the leader for the negotiations on behalf of the Norwegian employer organisation KS to state that: ‘The results attained by KL are something we would like to see in Norway as well’ (Almendingen, 2013).

Norwegian agreements regarding working hours apply for 3 years at a time. At each negotiation, the employer body KS has wished to change the system from centralised regulation of teachers’ working hours to local agreements. KS wishes to reinforce the management powers of head teachers at the cost of a centralised negotiated agreement which regulates teacher’s working hours; their desire is that the number of actual teaching hours should be set by the individual school. Centralised regulation of working hours as opposed to local self-regulation by individual schools has been a contested question in negotiations regarding working hours after the millennium, but the teacher unions have so far succeeded in maintaining the status quo. The period 2005–2013 saw a red–green coalition government in Norway and KS (which was led by an individual who had an association with the Labour Party) avoided a head-on confrontation in the question of changing teacher’s working hours regulations.

In the autumn of 2013, Norway elected a new government led by the Conservatives and the far-right Progress Party. In addition, KS had a new leader from 2012 (who belonged to the Conservative party). In the negotiations regarding working hours regulations that began in the autumn of 2013, this conflict took on a new edge. KS built its strategy on demands to change the current working hours regulations: (1) the teachers’ working year was to be extended ‘to 45 weeks’ (in other words, an increase of up to 6 weeks presence during the working year), (2) the number of teaching hours was to be set by the individual school (local solutions). ‘Greater flexibility’ was a core goal, at the cost of previous hard-won national principles for working hours which KS regarded as rigid.

The pressure for change reinforced by international trends and the strong backing of KS’ political leadership fired up the expectations of the employers’ body that a breakthrough would occur in the negotiations with the teaching unions. The Norwegian teaching unions were unwilling to give in to the demands of KS and the negotiations broke down several times. The Norwegian negotiation system is such that an independent body – the National Mediator – is tasked with mediation in conflicts of interest between parties in the workplace when negotiations between them break down. The National Mediator put forward a proposed compromise solution involving 7.5 hours presence in the school every working day. The largest (166,000 members) professional organisation Utdanningsforbundet (UF) recommended acceptance of this suggestion, while three small teacher unions1 rejected the National Mediator’s suggestion. When 73% of the members in the largest union UF rejected the mediator’s suggestion, the teachers went out on a strike which lasted 60 days (UF, 2015). The members’ rejection of the National Mediator’s suggestion that had been recommended by the leadership can be regarded as a grass-roots rebellion. The teachers’ strike resulted in thousands of pupils being unable to go back to school after the summer holidays in August 2014. Strikes are often damaging to both parties in a workplace conflict. Each of the parties in this particular conflict attributed responsibility to the other.

The consequence of the grass-roots rebellion was a complete victory for the teachers’ interests in the dispute regarding working hours regulations. The employers’ negotiation leader had to accept that ‘we would have preferred a different kind of result’ (Skrede, Olsen, & Auestad, 2014). Why did this tug of war between the Norwegian teaching unions and the employers end with a complete victory for the teachers in the question of a more fixed working day and greater management rights for head teachers? The result was by no means a foregone conclusion. The logic of negotiation is that conflicts of interest discover a compromise solution in order to avoid strife (Dixit & Nalebuff, 1991). Why did the employer side gain no kind of success in relation to their strong desire to loosen up a ‘rigid and centrally managed practice in the Norwegian school system’ at a time when a liberal–conservative government was in power in Norway? The result of the Norwegian tug of war between the employer and employee camps was the exact opposite of the corresponding struggle between the Danish teachers and their employer counterparts. How did this contrast arise? Explanations of these phenomena need to take account of the different effects of power depending on how the decision-making processes occur within the representative organisation and within the organisation’s external relationships (how the public perception of a workplace conflict can affect the way in which political parties and the government of a country will relate to a conflict). I will therefore attempt to understand potential forms of power progression in the light of theories of the use of power in the interaction between unions and their employer counterparts. Firstly, I will present theoretical bargaining models before discussing the teaching unions’ battle to establish centralised regulations for working hours in Norway and Denmark. Secondly, I will discuss these two cases with the lenses of the theoretical framework.

Theoretical framework

Game theory as theoretical lens

When workplace interest groups stand in opposition to each other we speak of a conflict of interests: an increased attainment of one party’s goals will be achieved at the cost of the other party’s goals, except where a win–win situation is attainable. We typically find such a conflict of interests in connection with working hours agreements and tariff agreements. The core of the conflict of interests will resemble the Battle of the Sexes game (Fudenberg & Tirole, 1991), in which no clear single solution will stand out.2 However, this game is too parsimonious. Going from this simple model towards more content in the model’s explanatory adequacy can make the researcher better equipped to analyse complexity. The model must contain the essential in an explanatory mechanism, and ‘if the fundamentals have been captured, the work has started and can go on. Variables can be added, relations modified, and results interpreted in other contexts’ (Rapoport, 1959, p. 371).

One addition is linked to time dimension. The parties to the conflict may find it helpful to enter into time-limited agreements. The conflict of interests is normally regulated by legislation and social norms which define the framework in which the conflict can be handled within a set time frame. This can occur through entering into binding but time-limited agreements (for instance that the agreement is to regulate aspects of the conflict and that re-negotiation will take place within established forums, Elstad, 2002).

Assertiveness and power

Another addition is assertiveness and its potential countervailing effect via grass-roots rebellion. Assertiveness is a party’s tendency to actively defend and pursue its own interests during bargaining. Ames and Flynn (2007) assume a curvilinear relation between assertiveness and effectiveness: above some level of assertiveness, the costs loom too large for the assertive party involved in bargaining. Furthermore, at chronically low levels of assertiveness leaders at party’s bargaining can undermine the party’s goal achievement. Main effects for the overall perceptions of leadership: assertiveness is positively linked to instrumental outcomes to a certain point, but perceivers weigh costs heavily after the certain point. Assertiveness may influence the perception of relational trust among the parties. When a party perceives the costs too heavily, a counter-hegemonic movement might fuel grass-roots rebellion. Social media appear able to reinforce the effect of grass-roots mobilisation in conflicts, and grass-roots support for a proposed position will have the effect of strengthening a party’s leverage in negotiations.

For the parties in an employment situation, a period of truce for the time governed by the agreement will generally be a social norm. During such periods of truce the parties will often display an apparent consensus. Models based on assumptions of rational choice can be specified in a negotiation theory based on situational analysis (Popper, 1976). Such an analysis is based on a model of the problem situation as well as the assumption that the participants behave more or less rationally. A move closer to the employers’ preferred outcome entails an equivalent distancing from the employees’ preferred outcome and the opposite. Both parties can benefit from consensus as long as both of the parties gain something from the agreement (Elstad, 2002). If there are no such gains, the agreement will fail, assuming that the decisions of the parties are taken on rational grounds. The parties can however exercise restraint and uphold the agreement even if one party could benefit from withdrawing from the agreement. How the parties will rank the various outcomes of the negotiations depends on the parties’ perceptions of reality, their values and on their respective institutional frameworks and their inter-party interaction in relation to situations of conflict. If both parties choose a tough strategy the result can be a strike and/or lockout (Rabinovitch & Swary, 1976). How society chooses to resolve a workplace conflict (such as by an imposed mediation which in Norway often favours the status quo), will affect the direction of the parties and their choice of future strategy.

Situational analysis

The assumption of rationality on which situational analysis is based is controversial in its absolute form. For instance, one party can benefit from the other party believing that it is governed by inflexible feelings or principles (Dixit & Nalebuff, 1991; Frank, 1988).3 This can affect the motivation of the opposing party in negotiations or it can have the result of mobilising the motivation of the group’s own members. The institutional framework can be of significance for the degree of success of the parties in exercising power. A ballot is an example of something that can affect the other party’s perception of the resolve behind threats of the consequences and duration of conflict. Bargaining theory has a problem of indeterminacy in terms of predicting consequences (Elster, 2007), but our expectation is that grass-roots support for a proposed position will have the effect of strengthening a party’s leverage in negotiations (Strunk & Grissom, 2010). For this reason, the obvious relative strengths of the resources between employer and employee will not necessarily predetermine the dynamics of negotiation in a workplace conflict (Schelling, 1980). In such a case, the reactions of a third party that are affected by the dispute can affect the dynamics of a workplace conflict.

Involving a third party

A workplace conflict involving teachers will typically affect a third party – pupils and their parents. The third party can also include politicians and the public at large (in that politicians can perceive the benefit of media positioning when a workplace conflict is in the media spotlight). The parties will thus find it sensible to attempt to influence the third party’s general perception of a work-related conflict in order to avoid the blame of public opinion (Hood, 2011). This involves one party reinforcing their public communication in order to convince people that the other party deserved the blame. Arguments put forward in press interviews, posts on social media and stands distributing brochures at (striking) schools and in public places are all typical public-relations activities for a union. However, the power of one party’s committed communication about why the conflict arose and about the justification of its position will typically have a curvilinear progression: if the other party and the public perceive the argument as unreasonable the party’s position in the public assessment of sympathy or antipathy will be weakened (Ames & Flynn, 2007). As such, one party will gain an advantage through having a suitable strong assertiveness in its public communication. Ames and Flynn found that assertiveness have a curvilinear relation in multiple samples and with multiple measures (2007). My theoretical expectation is thus as follows: a party that is either chronically low or chronically high in its assertiveness may be regarded in a more negative light than a party that is moderately assertive (Fig. 1). It follows from this that if one party provokes the opposing party’s members in such a manner as to evoke harmful feelings, this can heighten the other party’s determination in the conflict.

Fig 1

Fig. 1.  Assertiveness is a party’s tendency to actively defend and pursue its own interests during bargaining. (Figure drafted after Ames & Flynn, 2007, p. 309).

My expectation is also that if one party manages to convince people that the other party carries the blame for the inconvenience caused to the third party, the pressure on the other party will be increased. A strike or lockout will thus erode public goodwill in proportion to the perceived distribution of blame (Hood, 2011). Conflicts that have reached a stalemate can therefore be resolved because one of the parties moderates its position on account of the level of pressure from a third party. In other words, the public perception of a workplace conflict will influence the willingness of a party to maintain an unyielding position. An extension of this argument is that the result of imposed mediation can be affected by the mediator’s perception of the workplace conflict. In the latter instance, relevant legal practice will also play a role.

Figure 2 depicts symbolically the bargaining problem. The feasible set of the bargaining is the convex envelope in Fig. 2. Any solution in the bargaining set is a possible candidate for agreement. The vertical axis is defined in terms of a satisfaction index for the teachers’ union (where the working hour issue is an important entity), and the horizontal axis is defined in terms of a satisfaction index for the employers (where the realisation of more flexibility is an important entity). The line T’T’ depicts the teachers’ reservation value for consent. The line A’A’ is the employers’ reservation value backed by means of power. Point Y is the initial outcome.

Fig 2

Fig. 2.  Feasibility and distributional properties of power: a bargaining problem. In looking for agreement, the parties might adjust on the status quo point Y.

The postmodern metaphor of the divorce between employer and employee

The theoretical framework outlined above must be said to be an example of Occam’s razor. On the basis of a simple model, an attempt is made to explain as much as possible with as little as possible. However, Zygmunt Bauman’s concept of liquid modernity (Bauman, 2000) is interesting for understanding how a rapidly changing order might undermine all notions of durability in this kind of modelling reality. Liquid modernity is, therefore, Bauman’s conception of how the world today denies solidity that it once struggled assiduously to create and maintain. This assumption challenges the meaning of modernisation as long lasting structures, and by applying modernisation to development, it is possible to address the postmodern metaphor of the divorce between employer and employee transformed into casual relationships (Bauman, 2000, 2013).

Empirical methodology: two case studies

The research method has a comparative approach, comparing two case studies. The construction of a theoretical model on the basis of the case material is a rational reconstruction of how institutional features are embodied in the basic logic of systems and can influence the actions of the players. The dynamics are thus explained by the structures inherent in the institutional framework. An important step in the analysis consists of constructing models that capture core elements in the interactions between teacher union and employer organisation. The content of the models is believed to consist of reasonable assumptions about employee employer interaction.

Firstly, I examine the negotiations regarding regulations on Norwegian teacher’s working hours 2013–2014, and secondly, I investigate negotiations regarding regulations on Danish teacher’s working hours 2013. These cases are compared for understanding the possible mechanisms of negotiations.

Case 1: negotiations regarding regulations on Norwegian teacher’s working hours 2013–2014

Over the years, KS and the teaching unions have reached various kinds of collaborative agreements. One example is a document which ‘specifies common ambitions and values on which the inter-party collaboration between union officials, managers and head teachers must be based’ (UF & KS, 2011, p. 3). This can be regarded as an explicit expression of an implicit contract based on a mutual strategy of consensus. Such implicit contracts are maintained until one party finds that it can make significant gains by breaking the contract (Elstad, 2002). It is very probable that such a breach will damage loyalty and thus pollute the climate of collaboration between the parties for a while. From a political perspective, it is also undesirable to have cold fronts with strongly divergent opinions regarding the exercise of the teaching profession, in that issues of quality assurance in the Norwegian school system and the key role of teachers for educational improvement are matters of importance for the various parties. There are thus rational justifications for these parties to display an apparent consensus during the bargaining process. It is also a social norm to abstain from making negative statements about the other party while negotiations are in progress (The National Mediator, 2012).

Negotiations regarding teacher’s working hours were restarted in November 2013 but broken off in February 2014. The teaching unions regarded their counterpart, KS, as too inflexible in a number of key questions, such as the demand for increasing the teachers’ working year from 39 weeks a year to 45 weeks (point c in Fig. 1). The teachers’ professional identity is an important component of teacher unions’ collective bindings (Søreide, 2008) and political activism (Hartney & Flavin, 2011). This demand provoked grass-roots members and contributed to awakening a latent dissatisfaction and resistance (UF, 2015). Likewise, the parties were widely separated on the question of centralised working hours regulations as opposed to local agreements. The parties agreed to reopen negotiations in February 2014 in connection with the national tariff agreements. Once again, the negotiations broke down, but after the National Mediator entered the scene, the largest union, UF, went along with accepting 7.5 hours presence at school each working day. There are reasons to believe that fears of a Danish-type situation with no working hours agreement (and perhaps a lockout) may have been a contributory factor. The small unions, however, refused to accept the solution proposed by the National Mediator. Analytically I interpret this proposal as perceived by these unions as socially insufferable (for instance, point c in Fig. 1) and interpreted as under the line T’T’ in Fig. 2 (for instance, equilibrium 2). KS believed that if UF accepted the negotiated result the small unions would follow suit. UF’s decision was to put the negotiated mediation proposal to a ballot, at which the central committee in the leadership of UF and UF recommended that members should vote to accept the mediation proposal: ‘All in all we believe that we have achieved everything that is possible’ (UF, 2014, p. 6). However, a grass-roots rebellion spread through the social media prior to the ballot and reinforced a feeling of dissatisfaction (UF, 2015). By an overwhelming majority, the members rejected the proposal that had been negotiated by their own leaders. UF’s leadership was greatly weakened, both internally and externally. A strike became reality.

The membership of UF mobilised a very large volume of reader comments in newspapers and arguments in social media (UF, 2015). Attempts to strengthen UF’s bargaining position (Kiander, 1991) in the battle of interests were clearly moved from closed bargaining rooms to public media. In Fig. 3, this phenomenon is entitled the snowballing effect. While UF chose an offensive strategy in media-based communication, KS deployed a strategy which ‘to far too great an extent [surrendered] the media depiction of reality to our opponents’ (KS, 2014). KS lost the battle for public sympathy (Ipsos MMI, 2014). Even though it was the teachers who through their strike prevented Norwegian schools from starting up in August/September, the blame was largely directed towards KS in media coverage of the teacher strike. Very few politicians (even from the government parties) expressed public support for KS. While the strike was going on, new negotiations were taking place behind the scenes. KS made several compromise proposals which were rejected. ‘We (the KS negotiators) eventually realised that it was impossible for UF to send anything out to its members (by way of a ballot) without an assurance that longer working hours in the school were not going to be imposed’ (Grande & Haug, 2014, p. 12). The UF leadership, which had been voted down by its own members, had nothing to offer on the question of more compulsory presence in the school. The UF bargainers’ lack of freedom had a strategic value and turned the game to an advantage. During the strike, UF’s intransigence denied KS an opportunity to come back with an additional compromise offer. UF’s intransigence became a strategic move: it altered during week 36 the beliefs and actions of KS in a direction favourable to UF. I find the explanation in ‘the principle that, in bargaining, weakness may be a strength’ (Schelling, 1980, p. 52).

Fig 3

Fig. 3.  Snowballing effect (from m to n). Starting a grass-roots movement may incur costs on the first protesters to themselves (from 0 to m). However, after the point m the presence of positive externalities may increase the average benefit of the grass-roots movement. This is entitled the snowballing effect (Figure drafted after Elster, 2007, p. 395).

A further explanation is the curvilinear effect of assertiveness (Fig. 1). The members of UF were furious about the earlier KS’ demands for a 45-week working year and the negotiated compromise of 7.5 hours of specified working time. These proposals can be regarded as socially insufferable for the teachers (point c in Fig. 1). After weeks of informal contact and further compromise proposals, KS was forced to acknowledge defeat. The teachers received statements of support in newspapers and social media despite the inconvenience inflicted by the strike on an innocent third party – the pupils. Two months after the teachers began the strike, KS chose to give in: ‘It is the pupils that provide the explanation’ (DN, 2014, p. 2). Pressure from politicians and public opinion was too great. Equilibrium 1 remained.

Case 2: negotiations regarding regulations on Danish teacher’s working hours 2013 Negotiations between Danish upper-secondary-school teachers and the Danish state

Firstly, I examine the negotiations between Danish upper-secondary-school teachers and the Danish state. Secondly, I investigate the negotiations between Danish ‘Folkeskole’ teachers and KL.

GL was involved in negotiations with the Danish state regarding teaching duties and working hours regulations. The Danish Finance Minister was personally involved in these negotiations and placed pressure on GL. In contrast with UF, which allowed its members to vote on the negotiated proposal, GL’s members (about 14,000 members) express their interest by a Committee of Representatives which votes on the members’ behalf. ‘There was no – and I mean quite literally no – room for negotiation. This was raw power. … That night (8 February 2013) the choice was between a lockout – presumably without a penny for the young teachers – or a financial settlement’ (Leschly, 2013). In this situation, the GL leader chose to accept the latter solution, equilibrium 2 in Fig. 2). The negotiated agreement was not put to a ballot. As such, the subsequent grass-roots movement was not in a position to influence the agreement that was made. The result was that the leader of GL encountered a lack of confidence and did not stand for re-election. These events can be regarded as an example of a lack of an effective grass-roots influence. The feature I will draw attention to here is the different structure of the proceedings: GL lacks the two-stage process that is a feature of the decision-making at UF.

Before March 31 2013 Danish teachers at the ‘Folkeskole’ had been scheduled on the basis of a collective working hours agreement, which entails the number of teaching lessons. This agreement (which was in force for a period of 2 or 3 years) was a result of bargaining between DLF – Danmarks Lærerforening – and KL. DLF represents teachers at public primary and lower secondary schools (about 67,000 members), while KL employs the ‘Folkeskole’ teachers. December 2012 DLF and KL presented their requirements for a collective bargaining rounds. KL wanted to remove all working hour rules while DLF wanted to preserve the principles of earlier agreement. On 27 February 2013, KL declared that the collective bargaining had broken down and notified a lockout of 1 month before entering into force. On 1 April 2013, members of the DLF were locked out for about a month. A school lockout means that Danish ‘Folkeskole’ teachers were not allowed to show up at the school and did not receive any salaries in order to put pressure on DLF. The Danish government decided to intervene in the conflict on 25 April 2013, and the act was adopted by the Danish Parliament. This ended the conflict. The act gave unanimous support for KL.

The teacher stands alone in terms of the tasks involved in teaching pupils in Danish schools, but there is a politically declared goal that teacher collaboration should form part of the solutions to the challenges in Danish schools (Egelund, 2009). Large-scale international surveys such as PISA show that Danish 15-year-olds perform relatively stably over a period (roughly in line with the international average, OECD, 2013a) at the same time as Denmark is amongst those countries in the world that spend the most on schooling (OECD, 2013b; Shrewbridge et al., 2011). The politicians would like to have greater returns for the money that is allocated to running the schools (Aftale, 2013). Part of the strategic initiative is to change the conditions relating to teachers’ working hours with the intention that teachers should collaborate more. It is expected that school teachers in primary and secondary schools should be more observant in terms of following up the progress of pupils and should collaborate more across disciplines.

The finance ministry has engaged McKinsey to identify measurement parameters in order to find out whether the new large-scale school reform is working as intended (Vangkilde, 2010). The Danish authorities also expect school principals to emerge as clearly marked work leaders (Ministry of Education, 2014). The rules are to be simplified and management is to work towards clear goals. Performance-based salary contracts are to be introduced for school leaders. There is to be transparency about school results. A national body of learning consultants is to go through the schools’ results development and offer advice to those schools that do not attain the expected results. A more efficiently run school system should free up resources for future welfare reforms (KL, 2013). Costs relating to special education and to text books are to be cut. Teachers are to do more teaching, prepare lesson plans, mark the pupils’ work at school and above all collaborate more across disciplines, amongst other things to encourage inter-disciplinary study. The concept of teacher collaboration is being presented as a particularly significant mainstay for the Danish school reform (The Ministry of Education, 2014). Some of the flexibility that is associated with the teaching profession will be reduced in that teachers will be committed to a normal working day from eight to four. Teachers’ work-life is to be normalised. Before this reform Danish teachers spend less time teaching than in other countries (OECD, 2013). And Denmark is the OECD country investing the most of its wealth in education. In short, the reform bears similarities to corresponding school reforms in a number of countries in which target management with results control and decentralisation of decision-making are features.

These changes come in the wake of a month-long and deeply seated employment conflict in 2013 between the primary and secondary-school teachers’ unions and KL. The leader of KL says that forcing the issue ‘was a price one has been willing to pay’ (Ekeberg, 2014, p. 18). Opinion polls taken 1 year after the conflict show that a majority of teachers said in the spring of 2014 that they lacked the motivation to carry out the new school reform and that many were considering leaving the profession (Faurfelt & Reymann, 2014). Teachers now have to teach every week about 2 hours longer than before (equilibrium 2 in Fig. 2). In the short-term, this leads to a reduction in the required number of teachers, which has in turn contributed to increased unemployment amongst teachers. In January 2016, 73% of the municipalities experienced lack of teachers (Danmarks Lærerforening, 2016).

The changes in employer policy in regard to Danish teachers have been far more comprehensive than has been the case in Norway. Denmark can thus be regarded as an extreme case for the policies that appear to be desired by employers in the other countries, clearly in Norway. The complexity of how teachers are motivated by environmental factors should be strongly stressed. We definitely need more research in order to be able to say something about how teachers perceive the new employer policies. The director of the employers has stated that ‘we believe that this is something that we actually will get past, once people see that it is actually possible to create a better school within this new framework’ (Ekeberg, 2014, p. 18). It is indeed possible that time heals all wounds. It is possible that teachers’ calling and commitment are so strongly implanted into the professional code that memories of the bitterness of the work-related conflict will have fewer consequences for how teachers work individually and together with other teachers. It is however also possible that playing the role of victor is a risky strategy.


Education is not so much being debated and contested, as a limited range of workplace factors such as the length of the working day. For this reason, successful industrial action may lead to employees feeling empowered, but when it comes to the overall direction and makeup of the educational system that feeling may be somewhat chimerical. The unions in Norway and Denmark have traditionally occupied a stronger and more influential position than is the case in many other countries. This means that unions are often involved in a cross-party collaboration together with the employers. A cross-party collaboration entails mutual accommodation based on mutual respect, with a potential benefit for both parties in the shape of periods of low levels of conflict and participation in bottom-up processes which lead to employees feeling empowered. Even though there is a past history of cooperation between the parties, there is a tendency towards new circumstances surrounding the work of the teaching unions that disrupt the ideal of the Scandinavian model. The Danish examples illustrate that teachers’ hard-won interests have been put under pressure by an offensive employer. The Danish KL felt so confident that they performed a lock out. Are differences regarding membership in the EU not mentioned as an explanatory factor? Or was a backstage agreement with the government negotiated? More research is needed. It is too soon to assess the long-term effects of this trend. Teacher strikes in neighbouring countries (Iceland and Estonia) also point towards the same tendency that unions are being obliged to use their most powerful weapon if they are to have any hope of influencing the situation. From this point of view, the events in Norway appear astonishing and almost paradoxical: union grass roots win in all respects in a heavy-weight and demanding tug of war with the employers. Any explanation will require assessments of causality and comparisons of cases in itself provides a weak basis of evidence. The discussion below will therefore take the form of assumed explanatory mechanisms4, for which it will be impossible to test the relative strength of one explanation vis-á-vis the other plausible ones.

As with all similar studies, this study has certain limitations from a methodological, as well as a conceptual, perspective. I acknowledge these limitations and argue that they contribute to a foundation for future studies. Firstly, case studies represent only examples and do not allow us to justify our claims. More social research is needed to address the complexity of interaction dynamics between employees and employers and their links to institutional arrangements they belong. Secondly, another shortcoming is the limited use of cases examined.

I suspect that the structure of the unions’ decision-making processes can be of significance for the potential power of the unions. Union internal decision-making processes can comprise one or several phases. In 2012, GL carried out negotiations through leadership decisions. UF carried out ballots. In the case of UF, the leadership was of the opinion that ‘we have accomplished all that is possible’ through negotiation. In addition, UF’s standpoint was that ‘your vote is not an expression of whether you regard this as a good or bad agreement. Your vote is about whether we should accept the agreement’ (UF, 2014). The senior leadership regarded the rewards as greater than the disadvantages. The grass roots in UF had a different perception. My interpretation is that two-stage decisions may be of significance for power potential in negotiations; and that an indication of this is that KS’ negotiation leader recognised that a compromise solution involving greater physical presence at the school was not something for which it was possible to obtain the members’ acceptance. This reinforced the transactional position of UF. The reactions of UF members in the social media, newspapers and in protest messages demonstrated an emotional strength in the way in which the membership reacted to the agreement and to KS as an opponent. From KS’ side, demands for an extension of the working year and an increase in attendance requirements can appear bold with the idea that strong demands can lead to a compromise solution somewhere in the middle. A possible interpretation is that these very bold demands may have led to an emotional mobilisation on the part of the teachers (Fig. 3). In other words, I understand this phenomenon to be an indication of the curvilinear effect of assertiveness: high levels of assertiveness can have a disproportionate impact on employee perception (the b–c segment of Fig. 1).

In addition, employer’s demands that are perceived as socially insufferable (the b–c segment of Fig. 1) increase the resolution of the resistance. In the Norwegian instance, the teachers’ battle in the social media, newspapers and through protest messages helped awaken sympathy in public opinion (the increasing part of Fig. 3): several opinion polls have shown that the teaching profession has gained a better reputation during the strike (Respons Analyse, 2014), while the reputation of KS has strongly suffered (Ipsos MMI, 2014). Although KS became economically better off during the strike, the strike did hurt KS’s reputation more than it did with UF’s reputation. These mechanisms in turn contributed to putting KS under strong pressure to give in to the teachers in the tug of war. All in all, these mechanisms help explain the paradoxical phenomenon that weakness can be a strength in negotiations. However, more research is needed for better understanding.

As mentioned earlier, the unions’ institutional framework can have been significant in terms of how these presumed mechanisms work. GL (like some other unions) did not employ ballots, while UF did so. The latter two-stage process, involving first negotiations with union leaders and then a ballot of members, can be advantageous to unions in certain confrontational situations. The former leader of GL found the heavy pressure of the lockout threat to be a decisive factor in accepting the employer’s demands. This threat turned out to be reality for the Danish lower secondary teachers. The moment the GL leader gave in to the pressure, there was no return. An equivalent potential lockout threat may have been in the minds of the UF leadership when they accepted the mediator’s suggestion which subsequently was rejected by the members. The corrective to the leaders’ acceptance came by the way of the grass-roots rebellion in the cases of both UF and GL, but it was only in the UF case that the grass-roots rebellion could exercise any influence over the negotiation process. In the case of the Danish lower secondary teachers, the union leaders and the members stood side by side.

The threats of strike or lockout would have been able to influence how the parties bargaining about working conditions perceived the institutional framework. A lockout has so far never been used in salary or working hours disputes relating to teachers in Norway and it is difficult to say how real the possibility of a lockout was in the spring of 2014. Strikes and lockouts are the most extreme tools available to the parties. The possibility of an imposed mediation may affect the way the parties evaluate this question. Norway and Denmark have rather different customs in the question of what constitutes the basis of an imposed mediation. Norwegian dispute settlements have traditionally been associated with status quo solutions (Østbakken & Schøne, 2007). In connection with dispute settlements in Denmark, the mediating authority seems to be more independent of custom (Danmarks Lærerforening, 2013). The suggestion was made following the mediation regarding the working conditions of the lower secondary teachers that KL and a majority of the Danish political parties had made a tacit pact to find the resources to create a basis for a whole-day school system by forcing Danish teachers to work more. Such a suggestion has not been documented, but differences in the traditions relating to dispute settlements in Norway and Denmark may influence how the parties perceive the consequences of steering a locked workplace conflict towards an imposed mediation settlement.

The Norwegian teacher strike could not have been stopped by a half-hearted compromise solution that would once more have been voted down by the teachers, resulting in a further weakening of KS’ bargaining position. In the end, the KS’s bargaining leaders were forced to acknowledge that the teachers as the opposing party were not able to meet with them in any kind of compromise. This mechanism can be seen as an illustration of the potential power of pre-commitment in negotiations (Dixit & Nalebuff, 1991). On the other hand, it also shows that negotiations can lead to situations that are extremely difficult to resolve (Schelling, 1980). All in all, these tendencies, seen in negotiations over teachers’ working conditions in Norway and Denmark, illustrate that the archetypal bargaining model based on joint decision-making, influence and mutual respect is no longer a good description of the nature of bargaining processes in which teaching unions and employers are involved. Whether or not this heralds a new age for Scandinavian unions remains to be seen. It is my view that the grass-roots rebellion forms a significant explanatory factor for the developments in the Norwegian negotiations around working hours regulations. This grass-roots rebellion was reinforced by communication through social media.

This raises new questions about the significance of social media for mobilising the membership (Enjolras, Steen-Johnsen, & Wollebæk, 2013) and suggests that there is a substantial need for more research about the significance of social media for empowerment. It is widely acknowledged that mass mobilisation as an entity can create a strong force which overcomes the disadvantages of individuals in confrontation with a powerful opponent (of which the mobilisation in the former East Germany in November 1989 could be taken as an example, Elster, 2007). There are formidable differences between the social systems in the former East Germany and in Norway, but it may be that a similarity is the power of collective mobilisation to trump short-term immediate disadvantages.


I would like to thank two reviewers and Sigrid Jørgensen for their valuable comments.


Abraham, F., Konings, J., & Vanormelingen, S. (2009). The effect of globalization on union bargaining and price–cost margins of firms. Review of World Economics, 145(1), 13–36. Publisher Full Text

Aftale. (2013). Aftale mellem regeringen (Socialdemokraterne, Radikale Venstre og Socialistisk Folkeparti)., Venstre og Dansk Folkeparti om et faglig løft af folkeskolen. Retrieved from

Almendingen, B. (2013). Har lært av danskenes lockout. Retrieved from

Ames, D.R., & Flynn, F.J. (2007). What breaks a leader: The curvilinear relation between assertiveness and leadership. Journal of Personality and Social Psychology, 92(2), 307–324. PubMed Abstract | Publisher Full Text

Barth, E., Moene, K.O., & Willumsen, F. (2014). The Scandinavian model – An interpretation. Journal of Public Economics, 117, 60–72. Publisher Full Text

Bauman, Z. (2000). Liquid modernity. Cambridge: Polity.

Bauman, Z. (2013). The individualized society. New York: Wiley.

Brandal, N., Bratberg, Ø., & Thorsen, D.E. (2013). The Nordic model of social democracy. London: Palgrave Macmillan.

Byrkjeflot, H. (2001). The Nordic model of democracy and management. Bergen: Fagbokforlaget.

Danmarks, Lærerforening. (2013). Lockout of teachers in Denmark. Copenhagen: DLF.

Danmarks Lærerforening. (2016). Lærermangel: Skoler har problemer med at finde lærere. Retrieved from

Dixit, A., & Nalebuff, B.J. (1991). Thinking strategically. The competitive edge in business, politics, and everyday life. New York: W.W. Norton.

DN (2014, September 2). Derfor vant lærerne fram [This is why the teachers won]. Dagens Næringsliv, pp. 12–13.

Dumont, M., Rayp, G., & Willemé, P. (2006). Does internationalization affect union bargaining power? An empirical study for five EU countries. Oxford Economic Papers, 58(1), 77–102. Publisher Full Text

Egelund, N. (2009). TALIS: Lærere og skoleledere om undervisning, kompetenceudvikling og evaluering – I et international perspektiv [TALIS: Teachers and head teachers on competence development and evaluation in an international perspective]. Copenhagen: Styrelsen for Evaluering og Kvalitetsudvikling af Folkeskolen.

Ekeberg, E. (2014). Skoleeksempelet. Klassekampen, pp. 14–18.

Elstad, E. (2002). Towards a model of strategic actions in the classroom: Games theory as research heuristic. Scandinavian Journal of Educational Research, 46(1), 65–82. Publisher Full Text

Elster, J. (2007). Explaining social behavior. More nuts and bolts for the social sciences. Cambridge: Cambridge University Press.

Enjolras, B., Steen-Johnsen, K., & Wollebæk, D. (2013). Social media and mobilization to offline demonstrations: Transcending participatory divides? New Media & Society, 15(6), 890–908.

Falch, T., & Rattsø, J. (1997). Political economic determinants of school spending in federal states: Theory and time-series evidence. European Journal of Political Economy, 13(2), 299–314. Publisher Full Text

Faurfelt, R., & Reymann, C. (2014). Skolereform dræber lærernes motivation. Retrieved from

Frank, R. H. (1988). Passions within reason: The strategic role of the emotions. New York: WW Norton.

Fudenberg, D., & &Tirole, J. (1991). Game theory. New York: MIT Press.

Grande, A., & Haug, A.K. (2014). Derfor vant lærerne fram. Dagens Næringsliv, pp. 12–13.

Hartney, M., & Flavin, P. (2011). From the schoolhouse to the statehouse: Teacher union political activism and US state education reform policy. State Politics & Policy Quarterly, 11(3), 251–268.

Holden, S. (1988). Local and central wage bargaining. Scandinavian Journal of Economics, 1(1988), 93–99. Publisher Full Text

Hood, C. (2011). The blame game: Spin, bureaucracy and self-preservation in government. New York: Princeton University Press.

Ingersoll, R.M. (2006). Who controls teachers’ work? Power and accountability in America’s schools. Boston, MA: Harvard University Press.

Ingle, W.K., Willis, P.C., & Fritz, J. (2014). Collective bargaining agreement provisions in the wake of Ohio teacher evaluation system legislation. Retrieved from Provisions manuscript_AEFP.pdf

Ipsos, MMI. (2014). Profilundersøkelsen. Norske etater og organisasjoner 2014. Oslo: Ipsos MMI.

Iversen, T., & Stephens, J.D. (2008). Partisan politics, the welfare state, and three worlds of human capital formation. Comparative political studies, 41(4–5), 600–637. Publisher Full Text

Kiander, J. (1991). Strike threats and the bargaining power of insiders. The Scandinavian Journal of Economics, 93(3), 349–362. Publisher Full Text

KL. (2013). Fakta. Folkeskolereformen. Copenhagen: KL.

KS. (2014). Streik og ansvar [Strike and responsibility]. Retrieved from

Leschly, G. (2013). Formandens beretning 2013. Retrieved from

Lindbeck, A. (1997). The Swedish experiment. Journal of Economic Literature, 3, 1273–1319.

McCann, D. (2010). The political economy of the European Union. Cambridge: Polity.

Ministry of Education. (2014). Den daglige skoleledelse. Retrieved from

Moene, K.O., & Wallerstein, M. (1995). How social democracy worked: Labor-market institutions. Politics & Society, 23(2), 185–211.

Moene, K.O., & Wallerstein, M. (2002). Social democracy as a development strategy. Princeton, NJ: Princeton University Press.

OECD. (2013a). PISA 2012 Results in Focus. What 15-year-olds know and what they can do with what they know. Paris: OECD.

OECD. (2013b). Education at a glance 2013. Denmark. Paris: OECD.

Osborne, M.J., & Rubinstein, A. (1994). A course in game theory. New York: The MIT Press.

Østbakken, K.M., & Schøne, P. (2007). Den norske forhandlingsmodellen i et likelønnsperspektiv. Report no. 85. OSLO: Institutt for samfunnsforskning.

Pedersen, O.A. (2013). Lockouten er over i Danmark [The lock-out is over in Denmark]. Retrieved from

Piazza, J.A. (2005). Globalizing quiescence: Globalization, union density and strikes in 15 industrialized countries. Economic and Industrial Democracy, 26(2), 289–314. Publisher Full Text

Popper, K.R. (1976). The Logic of the Social Sciences. In T. W. Adorno (Ed.), The Positivist Dispute in German Sociology (pp. 87–104). New York: Harper and Row.

Rabinovitch, R., & Swary, I. (1976). On the theory of bargaining, strikes, and wage determination under uncertainty. Canadian Journal of Economics, 9(4), 668–684. Publisher Full Text

Rapoport, A. (1959). Uses and limitations of mathematical models in social sciences. In A. Diekman & P. Mitter (Eds.), Stochastic modeling of social processes (pp. 7–38). Orlando, FL: Academic Press.

Respons, Analyse. (2014). Results: Project 20143801. Oslo: Respons Analyse.

Schelling, T.C. (1980). The strategy of conflict. Boston, MA: Harvard University Press.

Schoenbrun, D. (1966). The Three Lives of Charles de Gaulle. New York: Athenaeum.

Shewbridge, C., Jang, E., Matthews, P., & Paulo Santiago, P. (2011). OECD Reviews of Evaluation and Assessment in Education: Denmark 2011. Paris: OECD.

Skrede, K.M., Olsen, S.V., & Auestad, G.E. (2014). Ikke ett sekund for tidlig. Retrieved from

Strunk, K.O., & Grissom, J.A. (2010). Do strong unions shape district policies? Collective bargaining, teacher contract restrictiveness, and the political power of teachers’ unions. Educational Evaluation and Policy Analysis, 32(3), 389–406. Publisher Full Text

Søreide, G.E. (2008). Teacher union and teacher identity. Nordic Studies in Education, 3, 193–202.

The National Mediator. (2012). Information about Mediation at the National Mediator’s Office. Oslo: The National Mediator.

Tufekci, Z., & Wilson, C. (2012). Social media and the decision to participate in political protest: Observations from Tahrir Square. Journal of Communication, 62(2), 363–379. Publisher Full Text

UF. (2014). Tariffrevisjonen 2014: Uravstemningsdokument – KS-området. Oslo: UF.

UF. (2015). Lytt til medlemmene! Rapport fra særskilt evaluering av hovedtariffoppgjøret på KS-området. Rapport 1/2015. Oslo: Utdanningsforbundet.

UF & KS. (2011). Ledelse og tillitsvalgte sammen om den gode utdanningsledelse. Oslo: UF.

Vangkilde, J. (2014, February 24). McKinsey får nøglerolle i skolereformer [McKinsey gets a key role in school reforms], Politiken. Retrieved from

Wiborg, S. (2013). Neo-liberalism and universal state education: The cases of Denmark, Norway and Sweden 1980–2011. Comparative Education, 49(4), 407–423. Publisher Full Text


Dr. Eyvind Elstad is a professor of education at the University of Oslo. He is involved in several research projects on teacher education, newly qualified teachers and educational technology.


1Norsk Lektorlag (4,800 members), Skolenes Landsforbund (6,500 members) and Musikernes Fellesorganisasjon (8,000 members).

2This game has two pure strategy Nash equilibria and one mixed strategy Nash equilibrium. But these equilibria are deficient. Some sophisticated solutions of this game are discussed in Osborne and Rubinstein (1994).

3‘Charles de Gaulle used the power of intransigence to become a powerful player in the arena of international relations’ (Dixit & Nalebuff, 1991, p. 14–15). Schoenbrun (1966) assumes that du Gaulle’s intransigence did hurt the interests of France over time.

4Mechanisms are ‘frequently occurring and easily recognizable causal patterns that are triggered under generally unknown conditions or with indeterminate consequences’ (Elster, 2007, p. 36).

About The Author

Eyvind Elstad
Department of Teacher Education, University of Oslo

Department of Teacher Education, pprofessor

Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Related Content