Flexicurity – the Danish Labour-Market Model

Erik Haller Pedersen and Johanne Dinesen Riishøj, Economics

INTRODUCTION AND CONCLUSION

Over the last 15 years, unemployment in Denmark has gradually fallen to one of the lowest levels in the OECD concurrently with an increase in employment. So far, this has not given rise to any large imbalances in the economy – an achievement that has attracted both national and international attention.

The positive development should be viewed against the background of e.g. the comprehensive reforms of the labour market that were launched in the mid-1990s. Fine tuning of labour-market policies has continued up to the present time. The result is the special Danish labour-market model referred to as flexicurity^[1]. The model combines a flexible labour market with social security and active labour-market policies. This article describes and evaluates the Danish labour-market model.

Flexicurity has already been the subject of intensive research, and it is not possible to discuss all contributions in this article. The general consensus seems to be that the flexicurity model has contributed to the decline in unemployment during the period, but also that it is expensive for the public purse. In particular, a stronger focus on labour-market policy incentives, such as increased use of proven incentive schemes, seems to have had an effect. However, the decline in unemployment should be considered a precondition for the political approval of the many labour-market reforms.

Although the term flexicurity had not been invented at the time, all the elements of the model were present for a period up to the early 1970s. Only after the rise in unemployment following the two oil price shocks in 1973 and 1979, did Denmark begin to take a more socio-political approach to labour-market policy. This process was reversed with the labour-market reforms of the 1990s, but it is an open question whether the flexicurity model in its present form will prove to be robust if the country reverts to a situation where unemployment rises significantly. The historical development illustrates that flexicurity is not a panacea for labour-market policy.

One of the advantages of the Danish labour-market model is that it has probably contributed to a more positive general view of globalisation in Denmark than is the case in many other countries.

The Danish labour-market model has been the subject of considerable international interest. However, the model is an integral part of the Danish welfare model, and as such it is difficult to implement in other countries with different labour-market traditions and a different institutional framework.

A DYNAMIC LABOUR MARKET

The Danish labour market has seen extraordinary development over the last 15 years. Since the kick-start of the Danish economy in 1993, unemployment has dropped from more than 9 per cent to approximately 3 per cent in 2007, cf. Chart 1. It is the lowest unemployment figure in Denmark for more than 30 years and the lowest in the EU. Especially the rate of long-term unemployment has gone down. Youth unemployment has declined in step with general unemployment, but remains relatively high in other countries, cf. Chart 2. During the period, overall employment increased to the highest level ever.

Remarkably, the substantial drop in unemployment has not so far resulted in a significantly higher wage-increase rate. This reflects the decrease in structural unemployment that is now presumably among the lowest in the OECD. This is the result of active labour-market policies as well as other factors. An increased understanding of the conditions of the fixed-exchange-rate policies has anchored inflation expectations at a level close to 2 per cent p.a. Together with more decentralised wage formation and increased opportunities to relocate production, this has to a greater extent brought Danish wage increases on a par with those of other countries. Decreases in unemployment typically have a delayed impact on wages, and recent developments with incipient wage acceleration show that all good things come to an end.

The general public's perception of unemployment has changed in line with the decline in unemployment. Previously, when unemployment was high, it was a widely held view that a high rate of unemployment was an inherent defect in our economic system that we had to live with while seeking to remedy its worst negative social consequences. Today, the opinion tends to be that the right financial conditions and an appropriate economic policy provide for a more permanent reduction of structural unemployment. Obviously, this is no guarantee against unemployment temporarily going up some time in the future, e.g. as a result of unsustainably high wage increases and the ensuing loss of competitiveness.

The Danish labour market is characterised by a relatively low average number of working hours: just under 1,600 hours per year. On the other hand, the participation rate is high by international standards, especially for women. It is a typical pattern across countries that a low number of individual working hours implies a high general participation rate. In addition to the number of working hours and participation rates, the total input of work in the economy is determined by the population aged 20-64 years. The labour force is mainly recruited from that age group. The combination of the three factors can be measured by the total number of working hours per capita per year. In such a comparison, Denmark is above the EU15 average, but below the total OECD average. Overall, the total number of Danish working hours in relation to the size of the economy, is thus not particularly high by international standards.

Naturally, besides the total number of hours worked, actual performance during those hours, is crucial. The allocation of labour is important to ensure that the human capital available is utilised in the best possible way. In a changing world this calls for regular adjustments, and in this area the Danish labour market stands out.

250,000 jobs disappear every year in Denmark, but even more jobs have been created in recent years, cf. the Confederation of Danish Industries (2006). In addition, the rate of job turnover is high. More than 600,000 people change jobs every year. The average seniority on the Danish labour market is only 4.8 years, the lowest in the OECD, cf. Chart 3. Part of the explanation is the extensive lay-off opportunities, e.g. in connection with seasonal production fluctuations. In this situation employees receive unemployment benefits for a short period and then return to their original jobs. This, however, does not significantly alter the overall picture.

One drawback of high job turnover may be a reduction of the enterprises' incentive to invest in employee education. In Denmark, this is solved by increased government involvement in the co-financing of supplementary training, etc. Denmark has one of the highest rates of supplementary training, so in practice the potential problem has turned out not to be significant.

FLEXICURITY

The upswing in the Danish economy since 1993 has been accompanied by a number of reforms of the Danish labour market. Combined with continued adjustments up to the present time, this has resulted in the special Danish labour-market model referred to as flexicurity. The model combines a high rate of labour-market flexibility with a social safety net and active labour-market policies and has attracted international attention in recent years, cf. e.g. OECD (2004) and (2005) and the European Commission (2007).

The flexicurity model comprises three main components: liberal rules for the hiring and dismissal of employees (the "flexi"' part), a relatively fine-meshed economic safety net in case of unemployment (the security part) and active labour-market policies, cf. Chart 4. The latter area in particular has seen significant changes in the last 15 years. The most important measures since 1993 are listed in the Appendix^[2].

A characteristic of the Danish labour market is a considerable degree of regulation arrived at through agreements between the social partners, rather than legislation. Presumably, such a system assumes a fairly high rate of unionisation. Danish trade-union membership comprises about 75 per cent of the labour force. This level has been more or less constant for several decades, but with significant shifts between different unions due to the change in education patterns.

Flexibility
The basic system of agreements between employers and employees dates all the way back to the September Agreement of 1899. This agreement adopted the principle of the employer's right to manage and distribute work, while employees obtained the right to be members of a trade union. Ever since that time, the employer's right to hire and fire employees has been relatively liberal compared to most other OECD countries, cf. Chart 5. Employers also have a wide scope to redeploy employees within individual enterprises, which facilitates adjusting the number of employees to match demand. Recent legislation, e.g. EU rules concerning collective dismissals, is typically incorporated into current collective agreements, but EU regulation of the labour market does generally represent a challenge to the Danish collective agreement system.

A high level of job security makes it more costly for an enterprise to reduce its labour force. This is an immediate advantage to those who already have a job. On the other hand, the inability of enterprises to quickly reduce the labour force makes them more reluctant to hire new staff. This hits young job seekers particularly hard. Thus, a distinct insider/ outsider conflict is created in countries with a high level of job security.

It is unclear whether a high level of job security in itself influences unemployment, but it unequivocally reduces the labour-market dynamics. This entails a risk of actually protecting jobs in sectors in decline while impeding socially desirable redeployment of the labour force to sectors with higher added value. At the same time it becomes more difficult for individual enterprises to withstand fluctuations in supply and demand. Box 1 seeks to quantify the flexibility of the Danish labour market.

MEASURING LABOUR-MARKET FLEXIBILITY	Box 1
High labour-market flexibility is evident e.g. in quick adaptation of employment to changes in the production of an enterprise. The quicker the adaptation, the smaller the cyclical fluctuations in simple productivity measured as GDP per employee. The employees of an enterprise normally represent knowledge and education that are neither immediately replaceable nor replaceable free of charge. The enterprise, therefore, typically does not adapt the number of employees to brief production fluctuations. Consequently, it is a normal cyclical phenomenon that productivity fluctuates with the economic cycle, cf. Chart 6.
ANNUAL GDP GROWTH AND GDP PER EMPLOYEE	Chart 6
Source: Statistics Denmark and own calculations.
Average volatility in annual GDP growth over the last 35 years is estimated at 1.9 per cent, while volatility in GDP growth per employee is 1.1 per cent. The 0.8 per cent difference reflects the flexibility with which employment adapts to production fluctuations. An international comparison shows that the adaptation estimated using the above method takes place more quickly in Denmark than in most other countries, cf. Chart 7. This is a specific indication of the relatively high level of flexibility in the Danish labour market. The fast employment response to production changes is not the only criterion indicating whether the labour-market policies are working well. In a flexible labour market, employment should adapt to the labour force, resulting in a low unemployment rate.
EMPLOYMENT ADAPTATION TO PRODUCTION	Chart 7
Note: The bars measure the dispersion of annual GDP growth less the dispersion of productivity growth in the period 1970-2006. For Germany the period is 1992-2006. Source: Own calculations.
1  See Kiander and Virén (1998) for further discussion.

Flexibility is not just a question of low job security. During the 1990s, more flexible rules on the organisation of working hours throughout the year were incorporated in the collective agreements, and wage formation became more decentralised. As a result, wages are to a higher extent fixed on the basis of the circumstances of individual enterprises and the performance of individual employees. This principle is applied to a lesser degree in the public sector.

Questionnaire surveys show that the Danish labour market with higher job turnover implies less fear of unemployment than in countries with high job security, cf. the European Commission (2006). Under these circumstances, part of the explanation is that Denmark has been ahead in the economic cycle, and a strong economic position reduces the fear of unemployment.

Security
Increased flexibility with respect to the organisation of working hours is a typical demand on the part of employers. On the other hand, employees are ensured relatively favourable compensation in the event of unemployment. The unemployment benefit and employment service system dates back to the beginning of the 20th century, but found its present form in the late 1960s. The union insurance system is contribution-financed, but with substantial government co-financing. The government finances the costs of marginal unemployment. Unlike the situation in a number of other countries, membership of an unemployment fund is voluntary in Denmark, and the admission requirements are relatively lenient.

The degree of compensation for persons entitled to unemployment benefits is 90 per cent of the previous income for the lowest income brackets, but with a fixed cap. The average degree of compensation is just over 50 per cent and has been declining over time. In the early 1990s, the average degree of compensation was almost 60 per cent. Thus, for high-income groups the unemployment benefit system is not particularly attractive, but it may be supplemented with private supplementary insurance. Overall, the finely-meshed safety net that is often attributed to the Danish unemployment benefit system is only evident for low-income groups, while high-income groups typically suffer substantial income losses in the event of unemployment. This illustrates that the Danish unemployment benefit system includes a significant market-determined element. Proposals have been made to increase the degree of compensation in general, but also to reduce the unemployment benefit entitlement period of four years, which is relatively long in Denmark, and, as is the case in a number of other countries, to scale down unemployment benefits over the period of unemployment.

The reforms of the 1990s were aimed at the unemployment benefit system, among other things, based on the principle of retaining the unemployment benefit level^[3] while increasing the incentives for the unemployed to seek and take jobs. Among the most significant changes, the effective period of unemployment benefit entitlement was reduced to four years. Previously, unemployed persons were able to regain the right to unemployment benefits by participating in activation projects, and therefore the period of entitlement was actually unlimited.

The active period used to be preceded by a passive period during which the unemployed were under no obligation to participate in supplementary training, upgrading of qualifications, etc. The passive period was gradually reduced and was discontinued completely as of 1 July 2003. Now the unemployed are met with requirements from the first day of unemployment. If they do not succeed in getting a regular job during the four-year unemployment benefit entitlement period, they are transferred to cash benefits at a lower benefit level that is dependent on the family's financial circumstances. Cash-benefit recipients, except the most vulnerable, also have a right and obligation to activation.

The youth unemployment programme from 1996 deviated from the principle of not touching the benefit level. The reform was aimed at young people under the age of 25 without a qualifying education, who were entitled to unemployment benefits and who had been unemployed for more than six months. These young people used to be entitled to either unemployment benefits or cash benefits at levels significantly above student grants, but this changed with the new scheme aimed at young people that required them to get an education at 50 per cent of the unemployment benefit level, find a job or get reduced cash benefits. From 1999, these requirements applied to all young people under the age of 25. These reforms may be the most obvious success of the flexicurity model and contributed strongly to the fact that the youth unemployment rate in Denmark is among the lowest in the world. It also illustrates the importance of creating the right incentive mechanisms.

Active labour-market policies
The entitlement to compensation in the event of unemployment is countered by the obligation to actively seek a job and to participate in job-related activities. The reforms have resulted in activation of unemployed persons earlier in the period of unemployment, more frequent assessments of their availability and sanctions in case of non-compliance with the rules.

The aim of the active labour-market policies is to ensure that unemployed persons are actually available for work and to promote upgrading of qualifications through job training and education. Their availability is checked through guidance, job plans, meetings with caseworkers and other administrative procedures as well as sanctions in case of non-compliance, e.g. if the unemployed person fails to appear for compulsory meetings. For unemployment benefit recipients, sanctions against non-compliance with the availability rules are usually imposed by the unemployment fund concerned. The level of sanctions against unemployed persons varies from one unemployment fund to the next. The unemployment funds with the highest rates of unemployment are typically the most reluctant to impose sanctions, cf. the Danish Directorate of Labour (2006).

If the unemployed person does not succeed in getting a job, an activation process is initiated. This includes private or public job training with wage subsidies, educational activation or other activation. The majority of the persons in activation receive educational activation, but their share has gone down in recent years. Only approximately 10 per cent are in private job training. Recipients of cash benefits are activated more often than recipients of unemployment benefits in whose case administrative procedures are more commonly used.

The Danish Economic Council (2007) has presented an empirical analysis of the labour-market policies, including the activation part. The conclusions of this, and a number of other analyses, are fairly discouraging. It is typically found that only private job training has any significantly positive effect on employment. It is a statutory requirement that employment of persons in activation should result in a net increase of the number of employees to ensure that ordinary jobs are not just replaced. A survey by the Danish Labour Market Authority (2005) shows that in practice the replacement effect is fairly limited.

Analyses of educational activation, which is a priority area, indicate that it is both expensive and inefficient in terms of improving the job opportunities of unemployed persons. This is the conclusion of the Danish Economic Council (2007) and the Ministry of Labour (1999), among others. To the extent that courses and other forms of supplementary education result in the unemployed subsequently seeking a narrower range of jobs, educational activation may actually be harmful to their chances of finding a job.

Ideally, activation is about upgrading qualifications, but it has some side effects. The activated person's job search will typically decline during the activation period, and as a result he or she may remain unemployed longer than necessary. Obviously, this sustaining effect is an unintentional effect of activation.

The prospect of activation affects the behaviour of all unemployed persons, not just those in activation. Activation is time-consuming, and payment in the form of cash or unemployment benefits is lower than wages earned in a job, so the unequivocal incentive is to get a job at a regular wage if at all possible. The literature refers to this as a motivation or threat effect. For theoretical modelling of these effects, see Andersen and Svarer (2007), and Zhou (2007).

It seems well-documented that active labour-market policies contain an element of threat vis-à-vis the unemployed, and that in practice this effect is a significant channel through which the policies work. For an overview of studies, reference is made to Bjørn et al. (2004). According to an estimate by the Danish Economic Council (2007) the threat effect reduces the unemployment period by just under a week. This translates into a social gain of kr. 2 billion per year. Others have found similar effects, cf. Rosholm and Svarer (2004), and Svarer (2007).

Costs of labour-market policies
Compared to other countries, Denmark spends a fairly large amount of resources activating the unemployed, and the number of persons in activation has almost doubled since 1994 despite the fall in unemployment. A requirement to participate in activation after a certain period of unemployment exists only in few countries.

Even at the current low rate of unemployment, in 2006 the resources spent amounted to 4.26 per cent of GDP, including expenses related to both active measures and passive benefits, i.e. unemployment benefits and early-retirement benefits, cf. Chart 8. This is the highest level in the OECD. Expenditure on passive benefits has declined concurrently with the fall in unemployment, while expenses related to active measures have remained more or less unchanged in relation to GDP since the mid-1990s.

Comparisons across countries are affected by the fact that while in Denmark expenses related to early-retirement benefits are included in the costs of labour-market policies, expenses related to pensions are generally excluded. In some countries, the retirement age is considerably lower than in Denmark, so that in certain cases the retirement age in those countries is lower than the Danish early-retirement age. In other countries, the system is very much based on private schemes that are not included either. Adjustment for these factors is not possible.

THE DANISH LABOUR-MARKET MODEL – AN EVALUATION

Up to the mid-1970s, unemployment was as low as it is today, but it increased steeply in the wake of the oil price hikes in 1973 and 1979. The term flexicurity was not invented in the 1970s, but all the elements of the model were present when growth in unemployment began to accelerate. The rules of employment and dismissal were as liberal as they are today, the unemployment benefit coverage level was approximately 75 per cent, while the unemployment benefit entitlement period of 2½ years was shorter than today. Finally, active labour-market policies had been introduced in the 1960s in order to increase geographical and professional labour mobility. Then as now the key issue was to find the right approach to a tight labour market. Even so, the downturn of the 1970s hit the Danish labour market hard, and the active labour-market policies were watered down in favour of the aim of simply compensating the unemployed.

The labour-market policies pursued prior to the launch of the reforms should be seen in the light of the fact that unemployment was considered to be a system error, and therefore activation was very much regarded as a social measure. The introduction of early-retirement benefits in 1979 and the extension of the transitional allowance in 1994 were based on the same line of thought. Both schemes contributed to draining the labour market of labour in a deliberate attempt to reduce registered unemployment. The rise in the unemployment rate had to be stopped. The leave schemes (child-minding leave, educational leave and sabbatical leave) tended to have the same effect. They were first extended in the mid-1990s, only to be phased out over the following years. On the other hand, the maternity leave was extended. This illustrates that the reform process was not fully planned in advance. Instead it was to some extent a matter of trial and error in step with the decrease in unemployment.

The labour-market reforms shifted the focus of the unemployment system towards a labour-market policy aim. As described above, the effective unemployment benefit entitlement period was reduced, the opportunities to regain the right to benefits were discontinued, the availability and mobility rules were tightened, the sanctions were made more stringent and the activation obligation was tightened. So overall, it is very much a case of increased use of the 'stick' rather than the 'carrot' approach. However, the use of schemes with a mixed labour-market and social policy aim has also increased. By way of example, the number of persons in flexijobs and light jobs has grown substantially. Here, the aim is to get persons who are not completely fit for work into jobs.

Empirical surveys of the Danish labour-market policies clearly indicate that while results have been positive, the economic costs have been high, and it is not evident that the most cost-intensive policies have the greatest effect.

Increased use of labour-market policy sanctions interacts with general social policies. The stricter the labour-market policy methods used, the more attractive social benefits without an obligation to be available for work become, even at a slightly lower benefit level. Thus, flexicurity has a direct impact on the number of persons with a loose or no attachment to the labour market. As a result of the development since 1993, marginalised persons are increasingly receiving other forms of public benefits than unemployment benefits. Thus, the decline in registered unemployment compared to harmonised unemployment, cf. Chart 9, not only reflects the rise in employment. The number of persons of working age receiving public benefits remains high and is at the same level as in the early 1990s.

At relatively favourable benefit levels for low-income brackets, the flexicurity model together with general social policies contributes to a high minimum wage level in Denmark. This may also have to do with the fact that in Denmark the minimum wage is not fixed by law the way it is in many other countries, but is agreed directly between the social partners. The high minimum wage level acts as a barrier to the labour market for people with poor qualifications. In fact, in an international perspective the entire Danish wage structure is compressed both before and, even more so, after tax. This is in line with the popular Danish attitude in favour of income distribution to avoid excessive inequalities of income in society. Denmark has the lowest inequality of disposable incomes in the OECD.

The labour-market reforms have contributed to a reduction of structural unemployment. However, actual unemployment has fallen even more and must be considered to be below structural unemployment at the present time. Looking ahead, the pressure on the labour market will be aggravated by a demographically-determined decline in the labour force of 80,000 up to the year 2020. This means that the labour market is under pressure from both the demand and the supply side.

It is uncertain whether the flexicurity system will be able to survive a renewed general rise in unemployment. Experience from the 1970s and 1980s shows that rising unemployment typically leads to political pressure to ease requirements in relation to the unemployed. At the same time there is a risk that expenses for the cost-intensive activation policies will increase further.

THE DANISH WELFARE MODEL AND FLEXICURITY

The Danish welfare model is based on a market economy where the central government defines the general framework but does not manage private production in detail. Enterprises that are unable to match the competition are allowed to go out of business, and direct government subsidies to enterprises are limited. At the same time there is an extensive social safety net with massive redistribution of income between different social groups and for individuals over their lifetimes. The welfare model may be viewed as a number of implicit insurance contracts between different social groups which protect people, not jobs. There are tax-financed social safety nets not only covering unemployment, but also insuring against sickness, disability, loss of economic capacity, old age, etc.

The welfare model assumes a high willingness to pay and thus popular acceptance of a high tax burden. To achieve such acceptance it is probably essential that the distribution of government expenditure on production of public services and the redistribution of income are considered to be fair. This assumes that everyone who is able to do so contributes to the national household, and that no benefits for major groups are seen as improper.

The demographic changes whereby the group of elderly people is growing and the age groups from which the labour force is predominantly recruited are becoming smaller, pose a major challenge to the Danish welfare model. Labour-market policies play a key role in this context. If we are to preserve and preferably increase our prosperity, we still need a high labour supply in the years to come. The flexicurity model helps to ensure this, but it is hardly enough in itself. The tightening of the early-retirement scheme, including the gradual raising of the retirement age by two years from 2019 and the subsequent indexation of the retirement age to life expectancy will boost the labour supply, but only in the long term.

In addition, importing foreign labour may also play a part in securing sufficient labour supply in the short term, but it is no miracle cure. Importing labour leads not only to an increase in output, but also in the number of people on whom it is to be distributed. At the average participation rate of the Danish population, real income per capita will only rise if the imported labour is productive, i.e. contributing above-average income.

In terms of the effect on government finances, importing foreign labour typically contributes only half as much as the effect of getting an unemployed person a job. This is because unemployment benefits are saved only in the latter case, and because a separation allowance is granted for temporary jobs. This reduces tax payments for foreign labour.

Furthermore, it is essential whether importing foreign labour also involves the relocation of the family of the persons concerned. Calculations based on the DREAM model show that fiscal-policy sustainability deteriorates through increased immigration from less developed countries and is slightly improved as a result of immigration from more developed countries, including Eastern Europe. The fundamental assumption of the calculation is that the immigrants are similar to those already in Denmark in terms of participation rate and drain on public services. Generally, there is a significant correlation between the immigrants' home country and the extent to which they get jobs rather than transfer payments, cf. Table 1. To some degree, this pattern also applies to descendents. Thus, if immigration is to support the labour market and the Danish welfare model in general, it should be targeted at groups with a high participation rate.

The evaluation of the flexicurity model is far from clear-cut. For an example of a predominantly positive evaluation, reference is made to Jørgensen and Madsen (2007), where the introduction states that "Flexicurity as a political strategy promises to make an end to the old conflict between efficiency and equity". Another example at the other end of the spectrum is CESifo (2007) in which practically all positive flexicurity model contributions are rejected as being "largely a myth".

In the EU context, flexicurity has been made an element of the Lisbon strategy, the aim of which is to develop the EU into one of the most dynamic regions in the world, cf. the Ecofin recommendations (2007). The associations of European employer and employee organisations, respectively, also recommended the model recently. However, flexicurity is an integral part of the Danish welfare model, and as such it cannot be implemented unchanged in other countries with different traditions and a different institutional framework, which is also emphasised by Ecofin.

Basically, the development of the flexicurity model may hinge on the character of the Danes, cf. Alger and Cahuc (2006): "This paper argues that the efficiency of the Danish Flexicurity Model relies on strong public-spiritedness which is absent in many other countries …. this may hinder the implementation of the Danish recipe. More generally, this analysis suggests that public-spiritedness is a key ingredient in the possibility for a society to implement efficient public unemployment insurance. To that regard, a country may be unlikely to succeed in its labour market reforms without a comprehensive policy affecting civic behaviour of its citizens."

Alger, Yann and Pierre Cahuc (2006), "Civil Attitudes and the Design of Labor Market Institutions: Which Countries can Implement the Danish Flexicurity Model", Université Marne la Vallée, CEPREMAP, OEP, IZA Bonn and CREST-INSEE, CEPR Discussion Paper no. 1928.

Andersen, Torben M. and Michael Svarer (2005), Flexicurity, Department of Economics, University of Aarhus, CEPR, CESifo, IZA and CAM (Unpublished paper).

Andersen, Torben M. and Michael Svarer (2007), The role of workfare in striking a balance between incentives and insurance in the labour market, Working Paper no. 9, University of Aarhus.

Danish Directorate of Labour (2006), Report on benchmarking of unemployment funds (in Danish only).

Danish Labour Market Authority (2005), Survey of potential replacement of ordinary jobs when using activation with wage subsidies. Report, the Labour Market Committee

Danish Ministry of Labour (1999), Status of the labour-market reforms (in Danish only).

Danish Ministry of Labour (2000), Effects of the activation measures (in Danish only).

Danish Ministry of Employment (2005), Flexicurity – Challenges to the Danish model (in Danish only).

Bjørn, N. H., L. Geerdsen and P. Jensen (2004), The Threat of Compulsory Participation in Active Labour Market Programmes for Unemployed, Preliminary version of research overview, www.sfi.dk.

Blanchard, Oliver (2007), A Review of Richard Layard, Stephen Nickell, and Richard Jackmans' Unemployment: Macroeconomic Performance and the Labour Market, Journal of Economic Literature, Vol. XLV, June.

CESifo (2007), The EEAG Report on the European Economy. By L. Calmfors, G. Corsetti, M. P. Devereux, S. Honkapohja, G. Saint-Paul, H-W. Sinn, J-E. Sturm and X. Vives.

Confederation of Danish Industries (2006), In Search of Best Nordic Practice, A Case Study on How to Adjust to Globalisation.

Economic Council (2007), The Danish Economy, Spring.

Ecofin (2007), ECFIN/EPC(2007)REP/54182, September.

European Commission (2006), European Employment and Social Policy, Special Eurobarometer, October.

European Commission (2007), Commission Communication on Flexicurity, MEMO/07/256.

Gaard, S. and Mads Kieler (2005), Two decades of structural reform in Denmark: A review, Danish Ministry of Finance, Working Paper no. 16.

Jørgensen, Henning and Per Kongshøj Madsen (2007), Flexicurity and Beyond, Finding a new agenda for the European Social Model, DJØF Publishing, Copenhagen.

Kiander, Jaakko and Matti Virén (1998), Employment Growth and Labour Market Flexibility in OECD Countries, Government Institute for Economic Research, Helsinki.

OECD (2004), Employment Outlook

OECD (2005), Employment Outlook

OECD (2007), Employment Outlook

Rosholm, M. and M. Svarer (2004), Estimating the Threat Effect of Active Labour Market Programmes, IZA DP 1300.

Svarer, M. (2007), The Effect of Sanctions on the Job Finding Rate: Evidence from Denmark, University of Aarhus, Manuscript, April.

Zhou, Jianping (2007), Danish for All? Balancing Flexibility with Security: The Flexibility Model, IMF Working Paper no. 36.

APPENDIX: IMPORTANT LABOUR-POLICY MEASURES SINCE 1993

1993
Cash-benefit recipients obtain the same right and obligation to activation that unemployment benefit recipients have had since 1978.

1994 Labour-market reform I
Discontinuation of the right to regain the right to unemployment benefits by activation

Period of entitlement to unemployment benefits fixed at 7 years

Possibility of transitional allowance for the 50-59 age group

Improved access to take leave

1995
Tightening of the availability rules and possible sanctions

Right and obligation to activation after 4 years of unemployment

Admission to transitional allowance scheme stopped

1996 Labour-market reform II
Reduction of the unemployment benefit entitlement period to 5 years with full effect from 1998

Earlier offer of activation and tightening of sanctions

Programme aimed specifically at combating youth unemployment: After six months of unemployment, young people under the age of 25 without a qualifying education, who were entitled to unemployment benefits, obtained a right and obligation to activation at an allowance corresponding to 50 per cent of the unemployment benefits. If a young person rejected the activation offer, he or she would lose the entitlement to unemployment benefits and be transferred to reduced cash benefits.

Unemployment benefit recipients aged 25 or more obtained a right and obligation to activation after 2 years of unemployment

1998
Obligation for unemployed persons to take jobs outside their own field after 6 months of unemployment

More in-depth availability assessment of cash-benefit recipients

Cash-benefit recipients registered by the employment service

1999 Labour-market reform III
Reduction of the unemployment benefit entitlement period to 4 years with full effect from 2001

Earlier offer of activation

Limitation of access to take leave

All young people under the age of 25 obtained the right and obligation to activation after 6 months of unemployment

Early-retirement reform with a view to reducing admission to the scheme

2000
Educational leave discontinued

2001
Tightening of availability rules

2002
Introduction of start help benefits and reduction of introductory benefits. Granted to unemployed persons who have not been in Denmark for 7 out of 8 years

Phasing-out of child-minding leave

2003 Labour-market reform "More people in jobs"
Unemployed persons must be available for work from the first day of unemployment

Earlier offer of activation

Standardisation and tightening of the availability rules for recipients of cash and unemployment benefits

Tightening of sanctions against unemployed persons who are not actively seeking jobs or who fail to appear for interviews, etc.

Reduction of the cash benefit level for certain groups

2006
Requirement for 300 hours of work during a 2-year period for spouses who both receive cash benefits

2007 Welfare agreement
Increased activation and tightened availability assessment

Raising the early-retirement age by two years from 2019 to 2022, raising the retirement age by two years from 2024 to 2027 and indexation of age limits after 2025 based on the remaining life expectancy of 60-year-olds

Discontinuation of the extended entitlement to unemployment benefits for persons aged 55-59 and granting of a right and obligation to activation for persons aged 58-59 in line with other unemployed persons.