A service of the

Download article as PDF

This article is part of What We Can Learn from the Nordic Model

For over a century, automotive manufacturing has been a major source of stable, well-paying jobs for workers of all skill levels. It is economically, socially and politically important. It also exemplifies the challenges of structural change, like an ageing workforce and increasing automation. In 2007-08, these long-running changes collided with the global financial crisis. Especially in countries with relatively high labour costs, like the Nordic countries, the crisis was a “stress test” of states’ systems for responding to structural change.

The most extreme case for testing those limits may be that of Saab Automobile AB. The Swedish car manufacturer employed over 3 000 people in the 55 000-inhabitant town of Trollhättan, Sweden, and indirectly created thousands of additional jobs in the region. In December 2011, after a two-and-a-half year fight to save it, the company and subsidiaries closed for good. Thousands were out of work overnight in a town and region that had already been hard hit by layoffs in the preceding decade. However, it seems that the town has avoided the major economic, political and social disruptions that many predicted. Much of the credit for the ensuing stability is due to aspects of the Nordic model.1

The state did not heavily subsidise the industry, as in the United States or Germany. It continued to support unemployment insurance and finance infrastructural investment. Most of the response was coordinated through transition companies (Omställningsorganisationer): permanent, private-sector outplacement service companies created, funded and overseen by the social partners. The Public Employment Service (Arbetsförmedlingen, or PES) backstopped these efforts by helping those who needed more help, but most did not. It was a highly coordinated, multi-faceted response to a crisis that allowed likely necessary changes to occur and enhanced long-term competitiveness. Barth, Moene and Willumsen write that active labour market programmes and moderated employment protection are important parts of the Nordic model.2 The case of Saab and Trollhättan shows one way that societies can harness the productive side of creative destruction while mitigating the harm it causes.

Evidence for this assessment comes from surveys of political attitudes, voting behaviour and economic performance data. It also includes primary evidence from interviews with officials representing the company, labour unions, firms tasked with helping the unemployed find new jobs, and government officials, from the local level on up to the European Union.3

The crisis in context

The automobile industry at the start of the financial crisis was a major part of Sweden’s economy. With Saab Automotive and Volvo Cars as anchors, the country produced 361 237 vehicles in 2007, a 0.5% share of the 72.1 million vehicles manufactured worldwide.4 Of those, 87% were non-commercial vehicles. When the financial crisis struck in 2007, Sweden’s automotive industry employed 1.7% of the country’s workers and generated 1.9% of its GDP.5 The European Automobile Manufacturers’ Association estimated that 140 000 people worked in the Swedish automotive industry, including upstream and downstream producers.6 The percentage of the Swedish labour force directly employed by the industry was the third-highest among OECD countries. The industry comprised 10.9% of manufacturing jobs, which is the second-highest among OECD countries. It was the country’s largest export industry, responsible for 9.6% of Sweden’s total exports in 2007 and 15% of its exported manufactured goods.7

The impact of the economic crisis went well beyond Saab. Volvo Cars and its subcontractors laid off over 3 000 workers in 2009, before parent company Ford sold it to Geely Holding, a China-based investment firm. Executives at Volvo Group, a commercial vehicle producer, feared having to lay off up to 8 000 people in Sweden and 25 000 people worldwide. Even in crisis planning, they had not imagined a drop so severe.8

The car industry’s role in Swedish national identity added to the pressure. In an interview, Leif Johannsson, former CEO of Volvo Group, described the strong and ongoing “emotional” response to his decision in 1999 to sell (Swedish) Volvo Cars to (American) Ford Motor Company.9 Popular Swedish author Fredrick Backman marks characters’ personality types and builds core tensions in his books around brand preferences. Survey data reflects this culture. Averaging roughly ten years of data leading up to the crisis, more Swedes trusted Saab (51.2%) and Volvo (61.5%) than “large business” and parliament (both below 40%).10

Trollhättan was a microcosm of this car culture. Saab defined the town’s identity. People talk about “Saabanda”, the Saab spirit that rendered workers tightly bound to the Saab family. Then-owner General Motors had announced plans to close the company in 2009, but a new owner, Viktor Muller, brought renewed hope. Despite the well-known and long-standing financial troubles at the company, it was crushing for workers, for Trollhättan and for many Swedes when, on 19 December 2011, Muller announced that he had filed for bankruptcy.

The closure meant that 3 239 workers, including subcontractors, were out of work. An additional 509 workers had been laid off in the restructuring before the bankruptcy. Roughly 2 000 of the workers lived in Trollhättan, comprising 8.4% of the municipality’s employed inhabitants. The month after Saab closed, unemployment in Trollhättan hit 21%, the highest of all of Sweden’s 290 municipalities. There was every reason to believe that Saab’s closing would be devastating for Trollhättan.

Coordinating a response to crisis: drastic times, normal measures

In the face of the crisis, governments elsewhere intervened strongly in their domestic automotive industries. The German government provided major incentives for new car purchases, provided bridge loans and guarantees, and funded the shifting of millions of workers to part-time work. In the US, the government copied Germany’s incentives programme, provided bridge loans and ultimately nationalised General Motors in order to save it. The crisis easily could have led to similarly drastic measures in Trollhättan, too. In the end, this did not happen. The simplest summary of Sweden’s collective response to the crisis at Saab is that it approached the extenuating circumstances with a standard, if intensified, set of tools that are key to the Swedish iteration of the Nordic model.

The role of the state

The state’s role in the Swedish economy is minimal, and the 2006 election of a centre-right coalition, the Alliance for Sweden, made that role still smaller. Prime Minister Fredrick Reinfeldt trimmed entitlement spending and cut unemployment benefits. He also openly criticised his European peers for failing to address the overproduction of automobiles in Europe in a more competitive way. Sweden would not buy car companies, his government pointedly emphasised.11

Nevertheless, in December 2008 the government offered 28 billion Swedish kronor (€2.8 billion) worth of aid to the Swedish automotive industry.12 This included 20 billion kronor (€1.99 billion) in credit guarantees to back companies’ loans from the European Investment Bank for conversion to green technology; 3 billion kronor (€299 million) for investment in a state-owned research and development company; and 5 billion kronor (€496 million) for automotive companies facing collapse. Government officials said the goal was to help companies survive the crisis and re-tool to be more competitive going forward.13 It is unclear how much of the money companies took. Some in the industry have suggested the funds were mostly a public relations effort, as the demands for returns on investment were unrealistically high.14 The funds for failing companies seemed targeted at Saab, but the government found General Motors’ business plan untenable and decided against the loan. The government instituted a tax premium of 10 000 kronor (€975) for private purchases of new “eco” cars, but it ended other tax exemptions for the same cars.15 By comparison, the United States gave rebates of $3 500 or $4 500 per car.16 The German government provided a €2 500 subsidy per car.17

There were also important indirect means of support. The state continued pre-existing plans to expand rail service and the highway between Trollhättan and Gothenburg. While initially the goal of those investments was to improve infrastructure for Saab, in the end, the investments would mean that workers could commute more easily from Trollhättan to the much larger labour market of Gothenburg, which housed the revived Volvo Cars and Volvo Group headquarters and production facilities.

The state’s most important economic role in Sweden is that of co-funder of unemployment insurance (UI). As a Ghent system, the state partially supports UI funds (A- kassor) that the social partners administer. Reforms by the Reinfeldt government in 2006-07 shifted the funding burden toward members, reducing publicly financed support from around 90% to roughly 50%. They decreased the benefits, tied fees to the overall unemployment rate of the fund and ended the tax deduction for union and membership fees. The changes not only decreased benefits but also increased the cost sixfold for many members.18 As a result, membership in the funds dropped ten per cent in 2007.19 Those who remained members could receive up to 60% of their former salary (or 67% if responsible for a child) for the first 200 days, and 50% for the next 100 days.20 A daily ceiling on benefits, however, meant that well-paid workers would earn considerably less than 60%. To offset that, some unions offer supplementary insurance. Regardless of the insurance, however, regulations capped the total benefit at 80% of the previous salary.

For those unable to find employment after a period of 300 working days, the Arbetsförmedlingen, or Swedish Public Employment Service (PES), was the next option. The PES was once the monopoly provider of job lists and placement services in Sweden. In recent decades, its task has evolved to provide job coaching and vocational training to those “farthest from the job market”. For those cases, the PES offers a “Job and Development Guarantee Programme.” It is comprised largely of job-coaching, placement services and advice on starting a new business to job seekers. The programme can continue for 90 weeks. After this period, the PES can decide to provide a more intensive re-training, as it deems appropriate.21

In short, the state’s role was largely indirect and long-term. This stands in stark contrast to moves made by other governments facing similar crises. The government provided little extraordinary support in comparison. The Reinfeldt reforms also meant that even standard means of support, like the subsidisation of UI, was less than it had been in decades. That said, unemployment benefits lasted a considerable length of time.

The social partners

Ultimately, the social partners regulate redundancies in Sweden. Companies must give notice of coming redundancies if it appears that they will lay off more than five people. That generally prompts a negotiation between labour and management to settle on final numbers. While “last in, first out” remains the law in redundancies, the law allows adjustments via collective agreements. Local union leaders are then responsible for making decisions about exceptions to those rules and for negotiating additional severance benefits for those laid off out of turn. As part of the collective agreement, the unions negotiate the length of time that workers receive their salary after having been laid off, which varies roughly in accordance with the amount of time served at the company. The social partners also administer the unemployment insurance system, as discussed above. In the case of Saab, IF Metall, the union representing the blue-collar workers, even had an agreement with two banks that provided workers with continued access to credit at discounted interest rates.

What makes Sweden’s approach to redundancies unique is the role of the omställningsorganisationer, or transition company.22 In the 1970s, white-collar workers in Sweden began to feel the effects of structural change for the first time. The PES had a monopoly at the time on job announcements and searches as part of Sweden’s effort to maintain effective active labour market policies. While the unions and PES worked closely together – union representatives were on the board of the PES – white-collar unions felt like the PES was ill-equipped to help their workers find new employment.

Unions began signing “transition agreements” that established specific companies to help members find work in the case of redundancies. The first, Trygghetsrådet TRS, was established by white-collar unions in 1972 to help workers in the cultural and “ideas” sector. Federal public sector workers created Trygghetsstiftelsen in 1990. Most importantly, blue-collar workers signed on to Trygghetsfonden TSL in 2004, which in effect completed the transition away from the PES as the primary party for job searches and toward a system organised through collective agreements and the transition companies established therein.23 Today, roughly 90% of workers in Sweden are covered by a transition agreement; the two largest companies – TSL and TRR – cover nearly two million workers. One important difference: while others provide the services directly, TSL contracts the services out. The union involved selects the company that will provide the services.

The agreements provide adequate and consistent funding, so the companies are permanent and help any worker laid off due to a “lack of work” (arbetsbrist), rather than just those let go as part of a mass layoff. Each year, TRR assists roughly 15 000 of its 850 000 white-collar members, while TSL helps roughly 21 000 of its 900 000 members. The most common agreement requires that the relevant transition company receive the equivalent of 0.3% of the monthly salary paid by an employer. Employers transfer these monies regardless of whether the transition companies are currently helping members. Part of this money is used for the “top-ups” that the transition companies pay to workers while they are actively working with a counselor. For some, like TSL, this is a lump sum of money that workers get, while for many, including TRR, it is a monthly stipend.

Saving workers: Advantages of the Swedish system in practice

The Swedish system creates a formal system of professional service providers. They have adequate funding to provide the services workers need; none of our interviewees said they needed additional money. They have networks within the sectors they work in, which helps them to better know what the market needs in terms of skills. It also means companies looking for help know who they can call to find it. Counselors have a multifaceted job that includes everything from emotional to technical and financial advice, so the experience they gain is useful.

The Saab case displays these strengths in different ways. TRR and TSL were responsible for the layoffs from Saab’s bankruptcy. The IF Metall ombudsman chose Startkraft as the service provider for TSL.24 The permanence of the service providers allowed the actors to begin cooperating before Saab even closed. In 2009, just after GM announced its plans to close the company, the PES established an informal working group that called itself Omställning för Framtiden, or Transition for the Future. It included TRR, Startkraft, local and regional educational institutions, and, most remarkably, Saab representatives. Jonas Björkman, a manager in charge of TRR’s efforts with white-collar workers, says the group helped the parties involved learn to trust each other. It also allowed them to act more quickly once the bankruptcy was official. That group would eventually coordinate the application for EU funds to support their work. In cases where such programmes are not permanent, this kind of pre-coordination is not possible.

Managers at TRR and Startkraft both emphasised the need to provide constant information to workers. Muller filed for bankruptcy less than one week before Christmas, which made it more difficult. Pre-planning, however, meant that the PES was able to call all dismissed workers to a meeting just after Christmas. Registering with the PES was the first step toward receiving unemployment benefits. In normal situations, that process would be coordinated through the workplace. Bankruptcy made that impossible. The PES brought the necessary paperwork to the meeting and helped the workers fill out the papers on-site. TRR and Startkraft managers also explained to workers how their respective systems would work. The permanent nature of the system meant that managers could transfer professionals from their offices around Sweden to deal with the extra caseload at TRR and Startkraft. The reliable funding also meant that TRR was able to quickly establish a new, bigger office in the middle of Trollhättan.

One vital task was recording the experiences and qualifications of each former employee. Normally that information would be provided and certified by the employer. In this case, all the records were subject to legal protection because of the ongoing legal proceedings. Transition company officials, in interviews, emphasised that their prior knowledge helped in this re-construction process. For example, Björkman at TRR hired a former Saab human resources director with whom he had worked in previous redundancies at Saab to help with that process.

Once the immediate rush was over, the different actors moved forward in their respective roles. In interviews, counselors talked consistently about the need to help workers deal with the emotions of having lost their jobs. Some workers needed help learning to use a computer to find job openings. Others needed technical training or certification of skills they had learned on the job. While the project was larger than previous ones, the system worked relatively efficiently to process an extraordinary situation.


The results from the interventions are overwhelmingly positive, especially given the challenges of the case. TRR reports that 1 448 white-collar workers received some kind of support from TRR. Subtracting those who began parental leave, retired, left the job market for some other reason, were disabled or died, 1 294 were actively seeking work. Of those, 91% found a new job, four per cent founded their own companies, four per cent studied for longer than six months and one per cent chose to leave TRR’s advising process. Of the 91%, 84% found permanent employment. An impressive 78% found positions with incomes equal to, or higher than, their previous employment. The automotive industry, including technical consultants and suppliers, remained the most common employer.

It is not surprising that the results for blue-collar workers are not as positive. Notably, the average blue-collar worker at Saab was 48 years old and had been working for over 25 years. Thirty-seven per cent had left school after ninth grade. Keeping these statistics in mind, the numbers for blue-collar workers also seem strong. According to Startkraft, by mid-June 2013, 81.5% had received a new job, were getting an education or had started their own business. Of the total laid off, 20% were receiving additional training or education, 20% had permanent jobs, 15% had fixed-term contracts longer than six months and 24% had fixed-term contracts for less than six months. Ten per cent had left the job market and eight per cent had transferred to the PES for further help. The most common sectors of employment were automotive manufacturing, engineering and, interestingly, health and medical care. Trollhättan is home to the largest hospital in the region, so healthcare is a particularly strong sector.

Looking more broadly, unemployment in Trollhättan in January 2012, one month after Saab’s bankruptcy, was 21%, the highest of all municipalities in Sweden. By the end of 2014, the end of the transition process, those numbers were back down to 14%, their pre-crisis levels. While this was higher than the Swedish average, this had always been the case. In 2017, unemployment in Trollhättan hit 10.6% and had experienced the sharpest reduction of any Swedish municipality.

Politically, we see a similar trend. Despite being a solidly social democratic town, it had always favoured the conservative side of social democracy. Like the rest of Sweden, the parliamentary constituency around Trollhättan has seen an increase in votes for the far-right Sweden Democrats: an increase from six to nine per cent in 2010 and to 15.1% in 2014. That is definitely a significant increase. But it should be noted that the Swedish Democrats saw an average 7.2 percentage point increase in votes nationwide, from 5.7% to 12.9%.

In other words, Saab’s closure did not devastate Trollhättan. The town still faces serious problems, but Trollhättan today is more or less like Trollhättan in 2007. Maintaining the status quo in the face of such a crisis is arguably a success. One key difference is that the economy of the region is no longer as dependent on Saab. Its economy is more diversified, which should make it more resilient if, and when, the next crisis comes.

Lessons for the future

The challenges of an ageing workforce and increasing automation are a concern throughout the West. The workforce at Saab was relatively old and active in an industry that is undergoing massive automation. The bankruptcy of Saab Automobile had the potential to devastate Trollhättan as well as the entire region. People in the area at the time reported fearing a broader social collapse.

This case illustrates how societies’ responses to structural changes can strongly shape the lasting impact of those changes. In writing about the creative destruction that drives capitalist systems forward, Barth et al. argue that

(t)his political-economic equilibrium path [that the Nordic model represents] has benefitted from consistent policies that have supplemented market forces. Active labour market programs and moderations on employment protection have been important policies to facilitate structural change and reallocation of labour without excessive wage differentials.25

The Swedish model is arguably one of the strongest examples of those policies, highlighting what truly active labour market policies look like. In the case of Trollhättan and the collapse of Saab, it not only prevented potential devastation, it appears to have put the town on strong footing moving forward in an economy that remains heavily export-dependent and must therefore remain highly competitive.

  • 1 E. Barth, K.O. Moene, F. Willumsen: The Scandinavian model – An interpretation, in: Journal of Public Economics, Vol. 117, 2014, pp. 60-72.
  • 2 Ibid.
  • 3 All interviews conducted by M.T. Nance in Gothenburg, Trollhättan, and Stockholm, Sweden; and in Brussels, Belgium, September 2017-July 2018. The interviews lasted one to two hours, were recorded and are on file with the authors.
  • 4 A. Binder: Ward’s Automotive Yearbook, 2008, Ward’s Communications.
  • 5 European Automobile Manufacturers Association, 2008; OECD Economic Surveys: Sweden 2009; OECD Economic Surveys: Sweden 2011.
  • 6 European Automobile Manufacturers Association, op. cit.
  • 7 OECD Economic Surveys: Sweden 2009.
  • 8 Interview with the author. Gothenburg, Sweden.
  • 9 Interview with the author.
  • 10 Poll results available at http://medieakademien.se/. The original question wording is: “Hur stort förtroende har Du för det sätt på vilket följande institutioner och verksamheter sköter sitt arbete?”
  • 11 GM ‘hasn’t taken responsibility for Saab’, The Local Sweden, 23 February 2009, available at https://www.thelocal.se/20090223/17776.
  • 12 All exchange rates are in 2018 euros.
  • 13 Sweden to launch car industry aid package, The Local Sweden, 11 December 2008, available at https://www.thelocal.se/20081211/16278.
  • 14 Interviews with the author.
  • 15 D. Haugh, A. Mourougane, O. Chatal: The Automobile Industry in and Beyond the Crisis, OECD Economics Department Working Paper No. 745, 2010, p. 14. LO, the largest blue-collar union umbrella organisation, called for a premium of SEK 25 000; see LO: En alternative krispolitik mot arbetslösheten, 2009.
  • 16 M. Hoekstra, S.L. Puller, J. West: Cash for Corollas: When Stimulus Reduces Spending, in: American Economic Journal: Applied Economics, Vol. 9, No. 3, pp. 1-35.
  • 17 A. Kaul, G. Pfeifer, S. Witte: The Incidence of Cash for Clunkers: Analysis of the 2009 Car Scrappage Scheme in Germany, University of Zurich Working Paper No. 68, 2012.
  • 18 A. Kjellberg: The Swedish Ghent system and trade unions under pressure, in: Transfer – European Review of Labour and Research, Vol. 15, No. 3-4, 2009, pp. 481-504.
  • 19 For more information, see in particular A. Kjellberg, C.L. Ibsen: Attacks on union organizing: Reversible and irreversible changes to the Ghent-systems in Sweden and Denmark, in: T.P. Larsen, A. Ilsøe (eds.): Den Danks Model set udefra – comparative perspective pa danks arbejdsmarkedsregulering, Copenhagen 2016, Jurist- og Økonomforbundets Forlag, pp. 279-302. See also IAF: Effekterna av den fördöjda finansieringsavgivten, Dnr 2007/514, 28 June 2007.
  • 20 There were work and length-of-membership requirements, as well.
  • 21 Arbetsförmedlingen: The Job and Development Guarantee Programme, 2017.
  • 22 The OECD refers to them as “Job Security Councils”. Lars Walter, who has written the most on them, refers to them in Swedish as omställningsavtal, or transition agreements, which reflects their basis in the collective agreements. See e.g. L. Walter: Mellan job. Omställningsavtal och stöd till uppsagda i Sverige, 2015, SNS Förlag.
  • 23 Ibid.
  • 24 Startkraft is actually a subsidiary of TRR, the white-collar transition company.
  • 25 E. Barth et al., op. cit., p. 71.

© The Author(s) 2018. Open Access: This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

DOI: 10.1007/s10272-018-0748-1