Abstract
Public sector industrial relations in Denmark are normally perceived as relatively consensual, and as a ‘model employer’ country with a strong collective bargaining tradition it is one of the countries where unilateral regulation could be least expected. However, in 2013, a lockout without any prior strike or strike-warning in the bargaining area for primary and lower secondary education only, came to an end through legislative intervention. The article includes three main arguments. First, the government and the public employers took these drastic steps because various factors created a rare ‘window of opportunity’ for them. Second, the reason a Norwegian industrial conflict in 2014 with a very similar point of departure ended very differently was first and foremost that the Norwegian process was not embedded in politics and policy reform to the same extent as the Danish process. Third, the Danish case shows that Denmark might not have escaped the trend towards unilateralism seen across Europe.
Introduction
The recent economic crisis has in general caused changes in public sector industrial relations across Europe. These changes include, according to Bach and Bordogna (2013), a challenge to the traditional perception of public sector industrial relations (IR) being sheltered from international market pressures; a revival of unilateralism; a recentralization of wage determination; a weaker role for public sector trade unions; and a weakening of the distinctiveness of public sector IR compared with the private sector. However, public sector IR remains nevertheless distinctive compared with private sector IR and variations in the crisis impact across countries exist.
Denmark is one of the countries where the crisis impact has been relatively weak and has not fundamentally changed core public sector IR institutions (Hansen and Mailand, 2013; Mailand, 2014). These IR institutions place the Danish IR model closer to the ‘model employer’ than the continental ‘sovereign employer’ (Bach and Bordogna, 2011). In the latter, employment relations are unilaterally determined by the government, and collective bargaining is absent or severely restricted. The former is based less on isolating the public sector from conventional processes of employment relations, such as collective bargaining, but more on promoting the public sector as an example to other employers.
The important role of collective bargaining in the Danish public sector IR model is supported by a relatively equal balance of power between employers and trade unions compared to most other European countries. Moreover, relations between the parties are relatively consensual and major industrial conflicts are rare (Dølvik, 2007; Due and Madsen, 2009). The importance of collective bargaining and the power position of the trade unions exemplifies the Danish public sector IR model – together with the other Nordic public sector IR models – as critical cases (Flyvbjerg, 1991), in the sense that unilateral state actions with regard to core IR issues such as wages and working time are least likely in these IR models. If they do take place here, they could be expected potentially to take place everywhere.
The 2013 bargaining round was one of the rare occasions in which a major conflict took place. The conflict was found within a major public sector bargaining area (primary and lower secondary education) and was ended only by governmental intervention. What was outstanding in relation to this event was, in a Danish context, not the intervention in itself. Government interventions (with the necessary support from a parliamentary majority) in collective bargaining rounds, after both negotiation and arbitration have failed, are legally possible in Denmark and have taken place before. What was very remarkable in this case was the process leading up to the intervention, not least because the conflict took the form of a lockout without any prior strike or strike-warning. Not only in Denmark, but also everywhere else where collective bargaining takes place in the public sector, it is unheard of that public employers decide to lock out employees without a prior call for a strike – and in several countries it is not even legally possible for the public sector employers/governments to lock out employees at all.
The conflict was connected to remarkable demands from employers to reduce the sector-level collective agreement on working time down to a few lines, and a winding-up of all existing local agreements on working time for teachers in the Folkeskole 1 (municipal employers’ demands) and in most post-15 education institutions (state employers’ demand). The aim, it was argued, was to strengthen the management prerogative and facilitate the implementation of a large-scale reform of the Folkeskole. Only after failed arbitration, three and half weeks of lockout, and finally parliamentarian intervention, were the employers’ demands met. 2
So how is it that public employers, which on several occasions have concluded path-breaking deals with trade unions, and have never been seen as especially tough, suddenly used the hardest tool in the bargaining toolbox? Have similar processes taken place in other Scandinavian countries and can these cases improve our knowledge about public sector IR more generally? The research questions will address these issues. The research questions are: (1) What factors explain the public employers’ actions? (2) Why did an industrial conflict in Norway on the same issue for the same group in the following year end up with the employers giving in to employee demands? (3) What can we learn from these conflicts with regard to the autonomy of collective bargaining in contemporary public sector labour market regulation?
The methods used are, first, 36 semi-structured and (mostly) face-to-face interviews with Danish public sector key bargaining persons representing the social partner organizations. Of these 36 interviews, 18 focused especially on bargaining in the education areas (see Appendix); second, analyses were conducted of documents such as the collective agreements, policy papers, debate papers, internal documents and email correspondence of the included Danish social partner organizations; third, two semi-structured interviews were carried out with Norwegian chief negotiators in the teaching area, as well as analyses of the social partners’ newsletters and newspaper articles from the Norwegian press. Where nothing else is stated, the source of the analyses is the interviews.
The section after this introduction describes the basics of the Danish industrial relations model (the IR model) and especially the public sector part of it. The third section includes a process-description and analysis of the very conflictual bargaining processes for the gymnasiums and for the Folkeskole. The fourth section analyses the later arbitration phase, the lockout and the legislative intervention. The fifth section compares the Danish situation with the teachers’ industrial conflict in Norway in 2014. The sixth section includes a discussion of the results and answers to the research questions.
The Danish IR model in the public sector
About 32% of the Danish workforce is employed in the public sector. This comparatively high share has remained relatively stable over the past 20 years despite increased outsourcing (Ibsen et al., 2011). Research of the pre-crisis public sector has found New Public Management (NPM) reforms to have been introduced in Denmark, but these have been described as ‘moderate’ and as not having bypassed or sidelined trade unions (e.g. Ejersbo and Greve, 2005; Ibsen et al., 2011).
The IR model
The Danish public sector IR model as described can be seen as a Nordic version of the model employer IR model. More specifically, the Danish IR model is characterized by relatively limited legislation, bipartite collective agreements at all levels with high coverage rates, (ad hoc) tripartite social dialogue, an extensive system for employee involvement and relatively strong trade unions.
The collective agreements in the public sector cover wages and all issues of working and employment conditions, and a developed co-determination system, with co-determination committees on occupational as well as local level, has been established. Social partners establish general wage scales and terms and conditions at the overall level (state, region or municipalities), which are then integrated into individual agreements for different occupations. However, wage reforms since the 1990s have introduced local-level wage bargaining. This allows for individual or group supplements at the administrative unit/workplace. Yet another important feature of the Danish public sector IR model is a declining and limited number of civil servants with special statutory employment protection. Civil servants are today, in general, covered by collective bargaining, and not by unilateral regulation (Due and Madsen, 2009).
Collective bargaining covers no fewer than 98% of the employees in the state sector (Due and Madsen, 2009: 360). No statistics exist for the regional and municipal sector, but the collective bargaining coverage is estimated to be at least as high as in the state sector.
However, these high percentages do not imply that collective bargain is the sole important type of regulation of pay and conditions. Legislation plays a role, most importantly when it comes to employment conditions (terms of notice, etc.), holiday regulation, leave of absence due to childbirth and working environment issues. Moreover, in the higher parts of the job hierarchy, individual agreements often supplement collective agreements.
All three main bargaining areas – state, regions (health) and municipalities – have a three-tier structure, where the first two (highest) tiers are closely related (see Table 1) (Hansen and Mailand, 2013). The first tier is ‘cartel bargaining’, which normally takes place every second or third year. During these bargaining rounds, the state, the regional and the municipal employers respectively bargain with cartels (coalitions) made up of representatives of trade unions. The second tier is organizational bargaining (individual unions), which takes place more or less simultaneously with the sector-level bargaining. Here the individual trade unions themselves bargain on all occupation-specific parts of wages, pensions and working conditions within a decided economic framework. The local level is the third bargaining level. This has gained in importance due to the partial decentralization mentioned above (Hansen, 2012). As a general rule, it is the trade union related shop stewards who bargain. Bargaining issues include wages, working time, training and policies for senior employees. This tier will not be analysed in the present article.
Levels, bargaining tables and actors in the Danish public sector IR model.
The bi/triennial collective bargaining rounds
Since the focus of the present article is on the sector-level bargaining round, it is worthwhile to describe the framework around it. Here only three of the most basic features will be described.
First, there exists a hierarchy between the three main bargaining areas mentioned (state, regions and municipalities). Although they are formally independent of each other, de facto, the state area is the lead bargaining area for the other two. One of the most important effects of this hierarchy is that it is difficult for the social partners in the municipal and the regional sector to forge more costly agreements or which in content deviate from the agreements in the state sector, unless the issue is something specific to the municipal or the regional sector.
Second, one of the bargaining partners, the government (more precisely the Minister of Finance) has a double role as both negotiator and legislator. This has several consequences. One of them is that if the government fails to achieve a bargaining demand during a collective bargaining round, the double role provides it with the opportunity to attempt to push it through the political arena (unless the issue is dealt with exclusively in the collective bargaining arena, such as pay). But more relevant for the case in focus here, if the social partners fail to come to an agreement during the bargaining process, the National Arbitrator cannot facilitate agreement and an industrial conflict has not made one of the social partners give up, it is the government who – if they have support from a majority in parliament – draw up the legislative intervention. If the industrial conflict has been in the municipal or regional sector, this is less controversial. However, if the state sector has been involved, the government role as both bargainer and legislator could be seen as problematic. Public employers thus hold a strong power position and can make the balance of power somewhat lopsided vis-a-vis the trade unions.
Third, if the social partners fail to strike an agreement, it is legal for both trade unions and the public employers to initiate an industrial conflict – a strike or a lockout. This is contrary to the situation in several other European countries, where it is either illegal for both social partners or illegal for only employers to do so. The latest industrial conflicts prior to 2013 in the public sector in connection with the collective bargaining rounds took place in 1985, 1995, 1999 and 2008. Most of these have been related to specific occupational groups.
Fourth, although not necessarily fruitless, since they can be politically effective, strikes are in general less efficient in the public than in the private sector. It is impossible through strike action to render a public institution bankrupt, at whatever level – it just saves money during the strike. Together with the double role of the Ministry of Finance, this implies that the balance of power between employers and employees in the public sector is more unequal than in the private sector.
2013 bargaining round: Different teachers, different bargaining processes
The bargaining processes in the education areas (second tier) are linked to the general agreements agreed during cartel bargaining (first tier), as described above.
The 2013 bargaining round at the cartel level was – like in 2011 – influenced by the crisis, but not very dramatically. In both the state and the municipal area, the employers had remarkable demands, but gave up on most of them, probably to isolate the trade unions in the education area, where the most important demand was that concerning working time.
Hence, the result of both the state and the municipal sectors’ bargaining were agreements with few changes and little drama.
The employers’ demands in the education area
Contrary to the cartel bargaining, the organizational bargaining in the education area turned out to be very dramatic. As mentioned, the public employers’ aim was a winding-up of all existing local agreements on working time for teachers in the Folkeskole (municipal employers’ demands) and in most post-15 education institutions (state employers’ demand) in order to strengthen management prerogative, and in the case of the Folkeskole, also to facilitate and finance the implementation of a large-scale reform of the Folkeskole. 3
According to the employers, the aim was not to make the teachers work longer, but for them to spend more time in the classroom with the pupils. This was a long-standing desire on the part of the employers, stemming, among other things, from (1) the PISA studies, which showed mediocre performance of Danish students despite relatively high funding; (2) studies showing that Danish teachers were spending relatively few hours in the classroom compared to teachers in other OECD countries; and (3) the belief in a positive correlation between hours in the classroom and the quality of the education. The Danish Union of Teachers (DLF) contested the employers’ claims, arguing contrarily that reduced hours in preparation would reduce the quality of education, and that the national advocacy organization of the Danish municipalities, Local Government Denmark (KL), and the government were applying an outdated teaching concept when they concluded that Danish teachers in the municipal sector were only teaching 16 hours per week on average. The real figure was, according to DLF, 25 hours per week.
In the Folkeskole area, steps towards a more flexible and decentralized and less bureaucratic working time regulation had been agreed upon during the 2008 collective bargaining round. KL recognized this as a step in the right direction, but found it insufficient. In the case of the gymnasiums, an agreement had almost been reached with the Danish National Union of Upper Secondary School Teachers (GL) during the 2011 bargain round, but this failed at the last minute, causing immense frustration in the Ministry of Finance.
To reach their goals, the government and KL wanted full management prerogative on teaching. The new working time regulation regime should then feed into a larger scale school reform with more flexible and class-based teaching and a longer school week.
The public employers were well prepared. Already in late 2011 they had established a joint working group to prepare the negotiations. One of the controversial issues during the bargaining round was the allegation that the working group had decided not to compromise, because the bargaining process could be concluded with legislative intervention to secure the employers’ main demands. This has been denied by both the government and KL, who nevertheless refused public access to the documents of the working group.
The gymnasiums – the state bargaining area
The bargaining process in the gymnasium area 4 was planned to end in early February – and so it did. After a lengthy standstill in the negotiations, GL agreed to waive their claim for the right to bargain on working time, and for the phasing out of the special senior conditions, which was also one of the employers’ demands. In return, they received a substantial wage increase and a (limited) fixed framework (‘fense’) to secure planning and avoid an excess teaching workload. In justifying the decision to strike an agreement, GL’s general secretary explained that the union would have lost their bargaining right in any case, because the Ministry of Finance would have been willing to initiate an industrial conflict on the issue, which GL could not have won. By accepting ‘the unacceptable’ during the bargaining phase, GL obtained a substantial economic compensation.
The explanation sounded reasonable, but there were a couple of problems. The bargaining committee of GL had not approved the decision of the chairman (who was also the chief negotiator) to make the deal with the employers, despite it being the chairman’s contention that he had informed them what he intended to do prior to the final negotiations and they had not stopped him (Mailand, 2013b). Because GL, in the final phase, was part of the Danish Confederation of Professional Associations’ (Akademikerne) bargaining process, only a very small GL delegation was physically present at the last bargaining meeting. Furthermore, telephone communication failed – intended or unintended – during the last meeting, which contributed to a situation in which the bargaining committee was informed only through the mass media, the morning after the agreement had been reached during the night.
A small majority of GL’s board voted for an acceptance of the agreement the following day. However, having signed a bargaining agreement with Akademikerne, GL could only escape the agreement if a qualified majority of the members in all organizations on whose behalf Akademikerne had bargained, voted ‘no’ in the following membership referendum. In late March it was clear that although 85% of GL’s members had voted ‘no’, the overall result was a solid ‘yes’ among the member organizations of Akademikerne. Hence, GL’s agreement could not be annulled.
The Folkeskole – the municipal bargaining area
Until the agreement between the Ministry of Finance and Akademikerne/GL was signed, not much had been happening at the bargaining table in the parallel negotiations between LC 5 and KL. These negotiations had to be concluded before 1 March 2013, if an arbitration process was to be avoided. Prior to this date, KL was reluctant to present any written proposals about how they imagined working time regulation was to take place in practice if full management prerogative was to be applied. Then, shortly after the agreement was signed on 9 February, DLF was offered a similar deal. However, LC made it clear that they needed a compromise, and not only compensation as the offer to GL included. However, a very different/better deal for LC than the one GL had agreed to was not a very realistic scenario, first, because the hierarchy in the bargaining model as described includes that the state sector sets the trend and only small variations from this are allowed, and second, because the number of teachers in the Folkeskole is so much higher than in the gymnasiums. A better deal for the former would therefore end up being very expensive for KL.
During the latter half of February, a few bargaining meetings between DLF and KL were held. During these meetings LC proposed a number of models which – to some extent – met with KL’s wish for enhanced management room for manoeuvre. However, they still included two features which were unacceptable to KL. First, and most importantly, they all included some form of teaching maximum or teaching preparation factor. Second, they all included the special conditions for senior employees.
With the bargaining partners’ positions still far apart, and with the perception by KL that no movement had taken place on the part of LC, and their dissatisfaction with the latest reactions from LC to KL’s proposals, KL decided unilaterally to declare a breakdown in the bargaining process on 27 February. LC wanted to continue the bargaining process to the last minute, saying that the effort and number of bargaining meetings had been very limited. Still, KL refused to make another attempt. Accordingly, the attempt to strike an agreement would then have to continue under the leadership of the National Arbitrator.
Summing up, in the negotiations with LC, KL acted much like the Ministry of Finance did in relation to GL: they established very narrow parameters for agreement, which had to include full management prerogative and a winding-up of the special senior conditions. Within these parameters, there were hardly any opportunities for compromise on the content, only opportunities for compensation.
Lockout and government intervention
The rules of the National Arbitrator prescribe that they have one month to find a solution which the negotiators can accept. If they do, the proposal will afterwards be sent to be accepted among the social partner organizations involved. If the National Arbitrator fails to convince the social partners within the deadline, they can still postpone industrial action twice for 14 days.
Since LC had also failed to reach an agreement for a number of smaller post-15 educational institutions within the state area on the same issue, two separate but similar arbitration processes were taking place: one with KL and one with the Ministry of Finance. In neither of these processes did the arbitrator succeed in getting the parties close to an agreement.
The public employers had asked for a ‘normalization’ of the teachers’ working time, in order to illustrate and facilitate their management prerogative aim. At the end of the bargaining process, after having failed to convince the employers to accept working time regulation from a number of other collective agreements (+ maximum hours for teaching/preparation factor and special senior rights), LC suggested using the legal framework for the civil servants, which was both an occupational group more fitting to the teachers’ situation than the previous suggestions, and having an agreement with a more limited regulation framework – as the employers wanted. However, again, the teachers’ suggestion included a preparation factor and the special senior rights, and was therefore unacceptable to the employers.
After an odd and fruitless final act with verbal fighting in front of the television cameras and a new model being suggested after the arbitrator had given up, the lockout was unavoidable. The arbitrator had not used their right to postpone the conflict, since they found the parties to be too far apart. No compromise was within sight.
Hence, the lockout was put in force from the 2 April: 56,000 teachers in the Folkeskole and 17,000 teachers from the vocational schools in the state area were locked out.
During an industrial conflict – at least in Denmark – it is the employee side that has to bear the direct economic burden. This is also the case during a lockout. DLF’s strike fund would have lasted for approximately 10 weeks, but by initiating a loan system with a right to tax reduction, the trade union was able to extend the conflict for much longer.
Nevertheless, already during the first week of the lockout, the chairman of DLF called on the government to end the conflict with legislative intervention. For a union leader, this is an unusual step. It was taken because he expected the lockout to be ended through government intervention in any case – and one which would lean towards the employers’ demands. With a fast intervention, DLF could save some of their strike fund. As the days went by, DLF’s demand for the government to intervene got support from more and more unions.
However, the other unions’ support for DLF had been limited during the whole process. The bargaining cartel in the municipal sector had postponed the signing of the general agreement in the municipal areas and there had been trade union protest about ‘lack of respect for the collective bargaining model’, and a number of demonstrations were held – the last one included 40,000 dissatisfied trade unionists, parents and other citizens. But one of the two trade unions who could have called a ‘sympathy strike’ – the Trade Union of Pedagogical Staff (BUPL) – rejected such an initiative from the other relevant trade union, FOA. And this was the only attempt taken to initiate a strike (Mailand, 2013a).
This lack of support from other trade unions was partly due to a wish to conclude their own agreements ‘without’ trouble. However, it also reflects a division within the trade union movement on the issue. Some – especially private sector unions – found that the very special working time regulation in the Folkeskoles and the special senior conditions were not a case worth fighting for. Moreover – spontaneously, or at the suggestion of their political contacts in the Social Democratic Party – a number of trade union leaders announced late in the process that there had been no violation of the Danish model of collective bargaining.
Neither LC, nor KL and the Ministry of Finance, changed their positions during the lockout. After three and half weeks, two of the three parties in the government found that it was time to intervene, if the lockout were not to have too great an effect on the final examinations of both the Folkeskole and the vocational education sector. The government had, well in advance, secured its backing from the opposition. Hence, after a speedy two-day process in parliament, the legislative intervention came into force on 25 May, and the pupils and students could return to school.
The main features of the intervention were:
Full management prerogative on working time regulation (but still a 37-hour working week).
Working time ‘fense’: working time normally to be scheduled during normal working hours on weekdays. Overtime pay will be paid in connection to some activities placed outside normal working hours.
Annual norm: the total working time of teachers is still calculated annually, and not monthly, as DLF wanted.
The special senior conditions are to be phased out gradually.
Wage compensation: the teachers will be compensated with nearly 300 million Danish kroner (€40 million) in total. The compensation was calculated as the value of the special senior conditions.
Projects on cooperation, trust and better working environment worth 20 million Danish kroner (€2.7 million).
Further education: 1 billion Danish kroner (€130 million) for the further education of teachers. However, these were already included in the government proposal for primary school reform. Hence, they cannot be included as part of the compensation to the teachers.
In sum, the intervention met the employers’ main demands, and the compensation was limited and mainly related to wages. Calculated per teacher it was substantially lower than what the gymnasium teachers got. DLF complained about the calculation of the compensation, which they found too low. Moreover, they found the working time ‘fense’ inadequate. KL was in general satisfied with the intervention, but would have liked an even more limited working time ‘fense’.
The 2014 Norwegian teachers’ strike
Norway has a large public sector of about the same size as Denmark’s: the public sector accounts for 33% of total employment. Moreover, the Norwegian public sector IR model is similar to the Danish one in many regards, and the two models show more similarities with each other than they do with Sweden and Finland. Among the similarities is the presence of a National Arbitrator and occasional government interventions when the negotiating parties cannot strike an agreement and mediation has taken place without success (Stokke and Seip, 2008), and the possibility for the public employers to use lockout as a conflict weapon.
The Norwegian bargaining process
In the school area, as in Denmark until 2014, working time was divided into three categories, namely teaching hours, common activities at the school and free hours (including preparation). As in the Danish case, the employers’ wish to change the working time regulation in the school area had been on the agenda for some years. In 2006, two years after the employers took over responsibility for the primary and lower secondary school area, Kommunesektorens Organisasjon (KS) tried to strengthen the management prerogative, but they only managed to convince the trade unions on minor changes. However, during its preparation for the 2014 bargaining round, KS agreed internally to push harder for changes than in 2006.
KS’s counterpart on the trade union side was Utdanningsforbundet (Union of Education Norway, UF), which is the largest of three trade unions in the education area. KS and UF exchanged demands in November 2013 and started the bargaining process in the school area immediately after. The employers’ demands in the Norwegian case were in many regards similar to the Danish ones and it is very likely that the Norwegian public employers to some extent had been inspired by – inter alia – the demands of their Danish colleagues in KL. The employers’ aims were to increase the management prerogative (to increase the quality of management), increase cooperation between the teachers, and develop the teachers’ competences through increased participation in further training. To obtain these goals, the most important of the employers’ demands were: (1) a winding-up of the division in working time between the three categories, which, according to KS, included a de facto power of veto for teachers who did not want any change; (2) the opportunity for the school management to demand teachers to be present at the school 7.5 hours per day; and (3) the extension of the maximum number of working weeks from 38 to 45 weeks per year.
UF’s demands pointed in a somewhat different direction. They asked for (1) a reduction in teaching hours; (2) time to be allocated for supervision and support for junior teachers; (3) initiatives to ensure that the teachers can focus on core tasks; (4) better opportunities for the teachers themselves to decide how they can best support the pupils’ vocational and social competences; and (5) a specification of the municipalities’ responsibility to provide sufficient resources and room for manoeuvre in pedagogical management in the schools (Utdanningsforbundet, 2013).
The parties planned to complete an agreement before Christmas, which would have been before the general wage negotiations in the public sector started. However, with demands far apart, major changes being asked for by the employers, and little movement on either side of the bargaining table, the parties could not reach an agreement. Consequently, UF decided to leave the negotiation table in late January. All of the three above-mentioned employer demands were unacceptable to them.
Following the breakdown, the bargaining process was moved to a higher bargaining level. It became part of the general wage negotiations in the public sector, which took off in early April. To be part of the general negotiations complicated the process further, but also put the parties in the school area under pressure to reach an agreement. However, it was nevertheless still de facto UF and KS who were the main bargaining partners. Again, troublesome negotiations took place and at the end of April – when the ‘old’ agreement terminated – a breakdown in the negotiations was announced. During this phase of the negotiations KS had renounced their desire to liquidate the central working time agreement for teachers totally, but UF was still dissatisfied with the demand for more flexibility and management prerogative regarding the use of the working time (Sæther, 2014).
The Norwegian arbitration process and the strike
After this second breakdown in the bargaining process, the attempt to strike an agreement was passed on to the National Arbitrator, who had a deadline in late May to make the parties agree and avoid industrial conflict. The negotiating parties did reach an agreement in late May after intense meetings in the final phase lasting several days. Evaluating the process, both KS and UF found their counterpart much more open to concessions and constructive dialogue in the arbitration phase than in the previous phases. The agreement included a continuation of different categories for working – including preparation hours and teaching hours on the same level as previously. Nevertheless, the agreement opened the way for more management influence over the teachers’ working time. The agreement followed KS’s demands primarily in other ways too. Most importantly, it made it compulsory for the teachers to spend 7.5 hours in the school daily, but only if the physical working conditions allowed for it. Additionally, the teachers could be asked to work 10 days longer than the ‘teaching year’ – previously, the maximum was four days (Seip, 2014; Utdanningsforbundet, 2014a).
The agreement represented clearly only minor changes in the agreement it was replacing when compared to the result of the similar Danish process, and the chief executive of the Danish Teachers Union commended the agreement (Utdanningsforbundet, 2014b). UF’s reason for recommending the agreement to their members was that the teachers were not going to teach more than previously, that the agreement included better opportunities for preparation and follow-up tasks and that a local arbitration structure was a step forward. However, the members of UF were of a different opinion and 73% voted against the agreement − 67% of UF members participating in the vote. The interpretation in the trade union afterwards was that it was the proposed compulsory 7.5 hours per day presence at the school which was the most important reason for the ‘no’ vote.
As a consequence of the outcome of the election, a strike came into force on 1 July. The strike was gradually extended. In the first phase, it included fewer than 1000 teachers. Arbitration was attempted in July, but without any success. In the second phase from 1 August, 1500 teachers had come out on strike. In the last phase, which started on 21 August, shortly after the beginning of the teaching period, 8000 teachers were on strike. The strike meant cancellation of teaching hours on a large scale, and increased media coverage.
The negotiating parties were clearly under pressure (Petersen, 2014; Seip, 2014) and contact between the parties was quite frequent during the strike. An implication of the clear ‘no’ vote was that a substantially different proposed agreement had to presented to the union members unless a humiliating and fruitless second ‘no’ should be the outcome of the election and the route to legislative intervention initiated. It had also to address the dissatisfaction with teachers’ compulsory presence at the school. Hence, it might have been KS who were under the most pressure in this phase. Importantly, KS excluded lockout as a possibility, even though it was legally possible (Ritzau, 2014), the reason being, according to the interviewees, that KS found the political price – in terms of the reasons from and their relations with the trade unions – too high. Observing and learning from the Danish experience played a role in this judgement. By excluding lockout as a possibility, KS could either reach an agreement with the trade unions or hope for ‘employer-friendly’ government intervention as had taken place in Denmark in April 2013.
The role of the government in the process is not completely clear. It might be that the Norwegian right-wing government – who seemingly had kept they hands off the process until the strike broke out – informally played a role in pushing for further relaxations of the employers’ demands after the strike broke out. In that case, not having the employers ‘on their side’, as in Denmark, might have contributed to KS’s decision to exclude the use of a lockout.
The outcome in Norway
After 11 weeks, the strike concluded on 1 September. According to the new agreement, teachers cannot be asked to be present at the school if local agreement on this cannot be reached. If local-level consensus cannot be reached, the old centrally negotiated agreement will still be valid. Moreover, local agreements can only be negotiated in relation to three working time related issues: the number of working days, compulsory working time and the annual number of working hours. Finally, KS gave up their demand for teachers to be compulsorily present 7.5 hours daily at the school, so that it did not become part of the final agreement. What KS did get from the agreement were small steps in the direction of more local-based and flexible working time regulation as they wanted.
What is striking about the Norwegian process when compared to the Danish equivalent is first and foremost that although KS’s demands from the outset were much less far-reaching than the Danish employer demands, most of the Norwegian employer demands had to be abandoned before an agreement could be signed. The following section will address this and try to explain why the outcome of the two bargaining processes were so different from each other.
Comparing the two teachers’ conflicts
The Danish and the Norwegian teachers’ conflicts focused on a similar issue – working time – but when the processes are compared, the differences clearly outnumber the similarities.
Apart from the obvious difference between a strike and a lockout, at least three major differences between the processes in the two countries are important to note.
First – and most importantly – the Norwegian bargaining process was not embedded in politics and policy reform to nearly the same extent as the Danish one. Therefore, government pressure on the negotiating parties was much weaker in Norway than in Denmark. Moreover, and also related to the degree of politicization, the Norwegian government did not – at least not openly – support the municipal employers as the Danish government did. Finally, the Norwegian government was not directly a negotiating party as was the Danish government (in the gymnasiums and vocational schools). It seems that the Norwegian government only played a direct role in the last phase of the process, if at all. As a consequence of all this, the balance of power was less lopsided in the Norwegian than in the Danish case.
Second, although inspiration may have come from Denmark, this did not include the lockout, because the Norwegian employers from the outset excluded lockout as a possible step. This decision facilitated a negotiated agreement and contributed also to a less unequal balance of power than in Denmark.
Third, the Norwegian employers were open to compromise in relation to the management prerogative, not only compensation. This is unlike the Danish employers, who were open to economic compensation, but were unwilling to strike a compromise on the management prerogative. Again, this facilitated a negotiated agreement in the Norwegian case.
Among the similarities, three stand out as especially important. First, in both countries, the process was initiated by employers’ long-standing desire to strengthen the management prerogative on working time, and in both countries the recent events were not the first time that employers had attempted to reach this goal. Second, the media played an important role in both cases and the negotiating parties tried to use the media strategically. Third, the National Arbitrator had a formal role in both countries, but the Arbitrator’s attempts were not crucial in the processes. They could not be blamed for the failure to reach an agreement in the Danish case, nor commended for the successful agreement in the Norwegian case.
Conclusions and discussions
In the introduction three research questions were raised. The first research question asked what were the explanations for the Danish public employers’ actions. The unusual actions cannot be explained by a single causal factor, as already indicated by the question. It was a combination of various factors that together created a ‘window of opportunity’ (Kingdon, 1984) which the public employers would not let pass. This was so, even though the public employers knew they had to pay a price for it (teachers not supporting the implementation of school reform, and the risk of losing voters were part of this price). Some of these explanatory factors are interconnected.
Historical factors: First and foremost, as mentioned above, the wish to get rid of the Folkeskole teachers’ specific bargaining rights on the use of working time was a long-standing wish on the part of KL, disregarding the fact that the 2008 agreement had been praised by KL when signed, and positively evaluated in 2011. For the Ministry of Finance, the humiliating failure to change the gymnasium teachers’ working time regulation in 2011 also gave them a reason to prepare well and harden up for the next bargaining round in 2013.
Economic factors: The economic crisis had put the public budget under pressure, and the centre-left government was following an austerity path – although a mild one. This influenced their own and KL’s approach to the school reform and collective bargaining in the education areas. Moreover, the crisis generally weakened the trade unions, including their willingness for strike action and other protests, and therefore facilitated tougher employer approaches during collective bargaining rounds.
Organizational factors: An analysis of several types of stakeholders carried out by KL in 2011 showed a widespread wish for KL to act tougher in employment issues (Mailand, 2012a). Following the analysis, but not only connected to this, a number of key persons who had previously been involved in collective bargaining were replaced. Simultaneously, the agency in the Ministry of Finance responsible for bargaining was merged with another agency. Also here, a number of key persons responsible for bargaining were replaced during 2012 (Mailand, 2012b). This facilitated the process for the employers.
Political factors: Contrary to what most would expect, recent Danish history shows that it takes a social-democratic government to introduce substantial welfare and labour market reforms against the will of the trade union movement. A right or centre-right government would most likely have to face much stronger opposition from the trade unions then the social-democratic government faced. Moreover, the government and nearly the whole opposition agreed on the aim of the proposed schools reform and the public employers’ working time demands, which naturally facilitated the process for the public employers. Furthermore, opinion polls before and after the bargaining round showed that the Danish population – which from early on had been ‘worked on’ by both the public employers and DLF through the media – was of a similar opinion (although they found the public employers not sufficiently willing to make compromises).
The coming together of all of these historical, economic, organizational and political factors contributed to creating the ‘window of opportunity’ that the employers would not let pass. Therefore, the explanation for the public employers’ rare behaviour is multi-causal. However, without yet another political factor – the government’s decision to embed the attempt to wind up existing working time regulation in the larger high-profile Folkeskole reform – it is unlikely that the public employers would have been willing to pay the political price for the battle. Hence, this decision can be seen as the most important factor and the trigger of the events. Furthermore, as described above, the lack of a link to a similar reform in Norway was the most important difference between the industrial conflicts in the two countries.
The second research question was comparative and asked why an industrial conflict in Norway on the same issue for the same group the following year, ended with the employers relenting on their demands. Here it is possible to point to three interrelated factors. The Norwegian bargaining process was not embedded in politics and policy reform to nearly the same extent as the Danish one, which, inter alia, implied that the Norwegian government did not support the municipal employers as the Danish government had done, resulting in a less lopsided balance of power in the Norwegian case than in the Danish case. Moreover, the Norwegian employers excluded lockout as an option, because the predicted political price was too high. Therefore, the Norwegian employers were open to compromise in relation to the management prerogative, not only compensation. This was unlike the Danish employers, who were open to economic compensation, but were unwilling to strike a compromise on the management prerogative. Nevertheless, it is unlikely that it was a coincidence that the Norwegian public employers raised similar demands to their Danish colleagues only a year later. Although the Norwegian employers rejected lockout as an option, the Danish employers’ actions were still in part a source of inspiration and the Danish process could to some extent be seen as a trigger for the Norwegian process.
The third research question asked what could be learned from the two Nordic cases regarding the autonomy of collective bargaining in contemporary public sector labour market regulation. On a general level, the cases illustrate a continuation of a long-standing tendency of the employers to take a proactive role in initiating changes in IR (Bach and Della Rocca, 2001). Moreover, although the Danish public sector IR model has not been changed fundamentally either by the economic crisis or the 2013 conflict, the latter illustrates that even in a ‘model employer’ country with a very strong public sector collective bargaining tradition, the autonomy of the collective bargaining is relative. Collective bargaining in the public sector might not anymore be allowed to stand in the way of larger political reforms. Denmark might therefore not have escaped the aforementioned ‘revival of unilateralism’ (Bach and Bordogna, 2013) seen across Europe in public sector IR. However, the Danish 2013 conflict also indicates that unilateral action is contingent – in this case dependent on a window of opportunity created by several factors. Moreover, in Norway a lockout was never a serious option, since the employers found the political price too high. Hence, it seems that collective bargaining is still a cornerstone in Nordic public sector regulation of wages and working conditions and can be expected to remain so for the foreseeable future.
Footnotes
Appendix: List of Danish and Norwegian interviews focused especially on the teaching area
Acknowledgements
Thanks to Nana Wesley Hansen and Jonas Felbo-Kolding for useful comments on an earlier draft of this article.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.