Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00758.x Vol. 45, No. 2, April 2011, pp. 111–113
Editorial Introduction: The Nordic Welfare States – Revisited spol_758
111..113
Bent Greve The Nordic welfare state has now been changed in such a way that it raises the question whether a distinct Nordic model is still in play. Perhaps the only reason a Nordic model is still in play is that recalibration and restructuring have taken place to a large degree in most other countries around the world in the wake of the financial crisis. The clear focus on generosity, gender equality, full-employment and high level of public sector spending is now less clear than it used to be just five to ten years ago in the Nordic countries where it seemed that the Nordic model had prevailed; however, services are still a central element in the Nordic model. For an overview of the situation in the region, see the regional issue from 2004, where it was concluded that Nordic countries seemed, despite internationalizations, able to have a distinct welfare model (Greve 2004). Changes and types of changes are the focus of several of the articles in this regional issue, which also tries to embrace different service sectors (labour market, childcare and family policy), types of income compensation (unemployment benefits and pension) and the different Nordic countries (e.g. Iceland, Denmark, Sweden, Norway and Finland). The articles deal with comparative analysis and case studies of individual countries. Despite the fact that the articles, due to the space constraint of the regional issue, do not cover all the details, the story collectively in this analysis is clearly an indication of how the Nordic model has been changed dramatically over the last five to ten years, albeit differently in different fields where the distinctness, for example, in family policy, still prevails. The analysis is also an indication that some changes develop incrementally; in fact, analysis over a short time-span is not able to show clearly what is happening. This is a reminder that methodological analysis needs to be clearly aware of the time-span of the analysis, as just a few years might be insufficient to ensure a fully informed analysis about changes. This is also the risk of single case studies; however, when collected together with other studies, as in this regional issue, they can still present information on the ongoing transformation of the Nordic model. Address for correspondence: Bent Greve, Department of Society and Globalisation, Roskilde University, 4000 Roskilde, Denmark. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
An example of an area where Nordic countries are, seemingly, still distinct is the pension system, which Philip Haynes shows in his cluster analysis of income for older people in the OECD countries. The clear distinction is especially prevalent in the area of income protection for women, whereas the same does not seem to be the case in relation to men. This is presumably due to the more distinct pattern of female labour market participation in Nordic countries. The article also shows that Nordic countries are not always the same, and Sweden and Denmark seem to be the closest pairing. Employment policies were central to Nordic countries long before it became fashionable to talk about active labour market policy. In his article, Sven Jochem discusses Nordic countries’ labour market policy and concludes that distinctness can still be found in certain areas, whereas in others a convergence has taken place. Active labour market policy has thus become central for many EU countries, and Nordic countries have by now a more decentral wage bargaining system. The unemployment level is no longer lower for all Nordic countries than in the rest of Europe, while at the same time Nordic labour markets are still very flexible. In the article by Kvist and Greve, it is argued that, despite universality still being in place, a multi-tiered welfare state including the use of occupational and fiscal welfare has gradually transformed the Nordic model. This is the conclusion of a single case study using core areas including policies for the unemployed, the ill and the old, and family policy in Denmark as the analytical background for the conclusion on the development. The universality it still at play in core areas of the welfare state, however, in relation to both old age and income replacement for the sick, private collective and private individual initiatives have become still more central in relation to the welfare state’s development. Furthermore, there have also been direct benefit cuts, although more limited, but this combined with the gradually and seemingly ever-increasing role of the labour market and individual coverage imply that the distinctness of the model has been severely eroded over the last ten to 15 years. An example of an area with some distinct elements is Nordic childcare policies, as shown by Eydal and Rostgaard. However, the introduction of cash-for-care schemes could risk moving the focus on a gender equal approach away from its original pathway and at the same time installing a new division between more classical distinctions in welfare state analysis; for example, the high and low income groups, where the low income group, after the introduction of cash-for-care schemes, take care of their children at home to a larger degree than other income groups. So despite the fact that the Nordic model promotes a dual earner/dual carer model and more equality in care for children, new dividing lines can be witnessed. This is, for example, the case in relation to the length of leave, where fathers in low income families take shorter time spells than other fathers. Despite the commonalities among the countries there are also diversities as, for example, mothers in Norway and Finland have a longer period away from the labour market than in the other countries. A country analysis focusing on family policy is by Kimberly Earles, in which she argues that, despite change in the Swedish family policy, it has 112
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
remained socially democratic at its core. This is despite the introduction of private for profit childcare providers and home care allowances. The argument being that, despite some changes, there has been the introduction of a gender equality bonus in the parental leave system, and also two ‘father’ months to help in ensuring gender equality. The childcare system has, further, over the last 20 years been expanding and has reached a level where it is able to include all children in need of care, and this still at an affordable price. A country analysis showing a profound but gradual change is the one presented by Merja Jutila on the change in public responsibility in Finland over the last 20 years. Benefits have not been raised in line with real income change, public sector spending has been reduced in some areas but not all, and tax-cuts have diminished the ability to finance the welfare state. Welfare state retrenchment has thus taken place, however, mainly due to the gradual and slow change in the Finnish welfare state. A possible reason for using 1990 as a critical juncture was the strong impact on Finnish society of the change in Eastern Europe. Perhaps this also helps to explain why there have been cutbacks first, followed by a more gradual and subtle change in the model. Despite the fact that Nordic labour markets still function and show a high degree of flexibility and, further, that active labour market policies are central, there has, as shown by a case study by Jørgensen and Schulze, been an erosion of the partners’ central role both in the preparation and implementation of labour market policies. This loss of power, especially for the trade unions, also changes the classical way the labour market partners can be active in the system, pointing towards a system more in line with a lobbyist tradition than a corporatist approach. In conclusion, profound changes have taken place and the Nordic model is no longer so distinct and special as it has been in a historical perspective.
Reference Greve, B. (2004), Editorial Introduction: Are the Nordic Welfare States Still Distinct? Social Policy & Administration, 38, 2: 115–18.
© 2011 Blackwell Publishing Ltd.
113
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00759.x Vol. 45, No. 2, April 2011, pp. 114–130
Are Scandinavian Countries Different? A Comparison of Relative Incomes for Older People in OECD Nations spol_759
114..130
Philip Haynes Abstract Recent research has cast some doubt on the consistency of models that isolate Scandinavian countries as different to other nations in terms of egalitarian welfare states. For example, it has been argued that if occupational welfare polices have their contribution to national welfare better accounted for in comparative models, or if there is a focus on specific aspects of welfare coverage, then Scandinavian welfare states look less different to other European countries. Using case-based methods that try to balance the demand for generalizations and the qualitative integrity of the case, this research examines the specific case of income protection for the older population post-retirement, to see if Scandinavian countries emerge as consistently different when compared with other OECD nations. Scandinavia can be clearly identified in a separate cluster, but it is not alone and other countries also share its similarities. In particular, Scandinavia emerges in a cluster that provides more income replacement and protection for women. In part, this is reinforced by the high rate of labour participation by women in Scandinavian countries. However, there is little evidence of Scandinavian countries retaining noticeable differences and divergence in income protection policies for men.
Keywords Income inequality; Older people; Cluster analysis Introduction Greve’s detailed definition and deconstruction of occupational welfare (2007) leads to a complex analysis of the possible social and individual benefits of occupational welfare and their combined effect on the behaviour of the state and public sector. Traditionally, occupational welfare has been thought of as welfare benefits obtained through employment for entitlements such as child care, pension provision, health services and sick pay. However, Greve shows the interaction and entwined nature of state and occupational welfare, to the Address for correspondence: Philip Haynes, School of Applied Social Science, University of Brighton, Falmer, Brighton BN1 9PH. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
extent that the separation of such forms of delivery for the purposes of social policy and economic analysis is not straightforward and perhaps empirically impossible. Public employment is one key example of the complex entanglement of occupational welfare and state welfare. This is employment that is fully funded or part-funded by taxation, examples being civil servants, teachers, military personnel, etc. It has recently been argued that, in some countries, public employment gives more generous occupational benefits in terms of pensions, sickness and leave entitlements (OECD 2007: 4) This is, in part, to offset higher wages paid for similar jobs in the private sector (especially for highly qualified, professional workers, like economists, lawyers and scientists). It has been argued that occupational welfare increases individual motivation to work and accompanies a flexible and adaptable labour market. It may also be linked to increases in the proportion of women working, as women feel that they have to work in order to become full citizens and to benefit from all entitlements and welfare opportunities, and also if they find that essential support such as maternity leave and child care are provided as benefits by employers. The most obvious link between occupational welfare and old age is the income transfer over the life course in the form of a pension that has been acquired as protected and dedicated savings through the adult life, with locked in contributions from employer and employee. These benefits can only be taken as a lump sum and/or a regular income upon retirement and when leaving the workforce towards the end of the life course. Occupational pension provision is different to state income protection entitlement (or state pension) paid in old age and not necessarily linked directly to payments during one’s working life. This form of income protection in older age protects those without occupational pensions. In developed welfare states, for the bottom quartile of the income distribution, income level is rarely predicted by a simple association with employment record (OECD 2009). This is because those with intermittent work records or work records in low paid and casual jobs without adequate occupational pension schemes would be pauperized in old age if they had to live off the little, if any, occupational pension entitlement that they might have built up. For this group, state-funded income benefits in old age are an important policy strategy for preventing poverty. State income protection for those with inadequate or no occupational pension is necessary if their absolute and relative needs are to be met. Women are particularly at risk of being dependent on this type of state income protection, especially in countries where the percentage of women of working age in employment is relatively low. So a key concern is that a move to occupational welfare may disadvantage women with a minimal employment record and a history of low pay and low occupational pension contributions. In contrast, for those with stable and secure work histories such as male skilled workers, occupational entitlement delivers a more direct relationship between one’s working life and the level of income available in old age. An additional complexity occurs in countries that have a state-funded scheme, in addition to occupational pension schemes, which can appear to be defined by lifetime contributions in taxation or similar. Often, state protection © 2011 Blackwell Publishing Ltd.
115
Social Policy & Administration, Vol. 45, No. 2, April 2011
to individuals in old age is not paid from separate contribution accounts. It has been argued that these ‘pay as you go’ state-funded pension schemes (with benefits not paid from a separate saving and investment account, but paid from current government expenditure) have broken the direct link between work and income in old age. This is because such state schemes increasingly have a tendency to pay a standard amount beyond the normal retirement age (Hill 2007), regardless of the individual’s historical taxation and payment record. An example would be a person who has made tax-related payments (such as UK national insurance) during his or her working life without a separate employer-based occupational pension, who then finds his or her income benefit payments in older age are identical to the income support benefits paid to those who have not been employed during their adult working life. So in the UK there is a concern that there is no discernable benefit of paying national insurance and it has, in effect, become a form of general taxation rather than an individual entitlement linked to the level of later individual benefit payments. The relationship between occupational welfare and state-public welfare cannot only be understood solely with reference to pensions and income support benefits. In some countries, occupational services in old age are also significant in addition to income payments made in occupational pensions. Examples would be occupational housing and health entitlements retained after retiring from the occupation. Again, these occupational advantages may be more likely in public employment than private employment (for example, through public service such as the military, etc). A notable comparative difference between nation states is the variation in how much pension provision is either organized primarily by the state or by direct occupational payments (or some mixture of both methods) (OECD 2009; Greve 2006: ch 2). Analysts note that countries where income protection for older people and those in retirement has a relatively higher proportion underwritten by the state-public sector also have a tendency to have more ‘pay as you go’ schemes, where the state essentially underwrites current payments to the elderly through current taxation. In comparison, in a country with relatively less total dependence on state-public provision, the proportion of older income derived from occupational schemes with separated market investments reduces the immediate link with taxation and fiscal payments. A concern about the growing proportion of the older population in OECD countries has fuelled a political move to reduce pay as you go schemes and move to occupational schemes with more limited and predictable consequences (i.e. defined contribution rather than ‘defined benefit’). Cesaratto (2006) has argued that this policy assumption may erroneously ignore the external benefits of well-funded and relatively generous state-organized schemes. This includes stimulating consumer-based economic activity and reducing the risk-adverse propensity to save amongst older people. These external benefits are important in the current economic downturn. Historically, it has been argued that the Scandinavian welfare states of northern European countries have a more comprehensive coverage of risk through universal services and a greater ability to deal with income inequality 116
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
through the life course (Esping-Andersen 1990; 1999). Greve (2007) has questioned the extent of this difference, arguing that if occupational benefits and income and benefits net of taxation are included in more complex analysis, then these countries begin to converge more in terms of overall similarities with other European countries. Soede et al. (2004) and Soede and Vrooman (2008) have explored the proposition that Esping-Andersen’s welfare state regimes and the universal and comprehensive nature of Scandinavian provision can be confirmed for the specific policy area of the income protection of older people. Soede et al. (2004) concluded that Sweden, Denmark and Finland formed a key ‘social democratic’ group, with Norway slightly outside this group. However, the 2008 study concluded that the Scandinavian countries formed different regime groups: Finland was in a corporatist group with a number of other central and southern European countries; Norway was in a ‘modest’ group; and Sweden and Denmark were in a ‘mandatory private group’. While Soede’s recent work is comprehensive, reducing many variables to some innovative ideal type clusters, its method puts emphasis on the influence of categorical variables that define the structure and policy priority of a country’s policy regime. The purpose of this article is to use a different method to Soede to further examine comparative average incomes, pensions and benefit policy structures for older people in OECD countries during the previous decade, to conclude whether Scandinavian countries are qualitatively different to other major economies.
Methods The research method applied in this article uses principals from complex systems theory (Byrne 1998; Haynes 2008). This is an epistemology that recognizes a high level of interaction between different variables which influence welfare provision. This results in the impossibility of predicting with certainty the independent variables that cause one national state to have relatively different welfare outcomes to others. As a result, comparative studies using national quantitative data need to be considered as the qualitative comparison of unique cases where no two countries will ever be identical, but the key becomes searching for patterns of similarity and difference. Quantitative data therefore provides evidence of how countries cluster and set in relation to each other, rather than modelling exact and precise combinations of variables that predict future social and economic outcomes for countries. Likewise, complex systems approaches emphasize the interaction of social and economic variables with policy activity and that deterministic, causal relationships are at best only weakly specified and subject to redefinition over time. Recent methodological work in this respect has placed an emphasis on case-based methods when considering national comparisons and using quantitative case-based methods like cluster analysis, correspondence analysis, Qualitative Comparative Analysis (QCA) and fuzzy set analysis (Byrne and Ragin 2009; Hudson and Kuhner 2010). The aim is to find quantitative rationale for placing countries into similar sets or clusters where membership © 2011 Blackwell Publishing Ltd.
117
Social Policy & Administration, Vol. 45, No. 2, April 2011
is subject to similarities rather than precise allocation. Therefore, members of any one set or cluster can still be examined as individual cases for consideration of their unique and individual characteristics in addition to their links with the other members of the set. Hudson and Kuhner (2010) have examined the different comparative methods for examining welfare state typologies and argued for the need to avoid algorithms that average scores in a single model and, therefore, lose the ability to see key differences on some variables when groups of countries are placed together for a generalized multiple component score. They argue for the importance of being able to identify welfare states as weak in some areas while strong in others, and that this is more useful than knowing states are above average performance across all areas of welfare. They give examples from comparisons that focus on ‘protective’ and ‘productive’ welfare. The method in this article is to explore the complex interconnection of key variables while preserving the integrity of cases. This is achieved by using cluster analysis and then to clarify the differences between cases (countries) with QCA. Uprichard (2009) and Byrne (2002) have argued for the appropriateness of cluster analysis to explore the structure in multi-variate analysis while preserving the integrity of real cases. Three cluster models are developed and then explored further with QCA: one general model is used to examine the overall policy impact for older people; a second to examine impact on women; and the third to examine the policy impact on men. Hudson and Kuhner (2010) have argued that comparative models do look different when aspects of social policy are entered and or removed. In the same way, this article argues that gender effects on comparative models are under-researched and might result in different patterns of similarity and difference between countries. This article focuses on one specific aspect of the welfare state – the provision of pensions and income support to older people. The variables used in the models are identified in Table 1. In the QCA the variable thresholds are set at the median point. The hypotheses to be analysed are thus: • Are Scandinavian countries different to other OECD countries in their provision of income to their older population and the level of income inequality? • Do Scandinavian countries form a separate cluster set that defines their coverage for women as compared with other OECD countries?
Results Table 2 shows that Scandinavian countries all score below the mean OECD average (0.29) for the Gini Coefficient of income inequality in older age after government taxes and transfers. Nevertheless, there are several other European countries from western and eastern Europe that also achieve this equality. In all of these countries there is no change in the Gini Coefficient of income inequality in older age, between 2000–2005, apart from in Norway, 118
© 2011 Blackwell Publishing Ltd.
© 2011 Blackwell Publishing Ltd.
Average age of labour force exit, women, 2002–2007 Net present value of pension wealth women, 2005, as a percentage of gross earnings of an average production worker Gross present value of pension wealth women, 2005, as a percentage of gross earnings of an average production worker Difference gross/net women Weighted average pension wealth, women, 2009 Percentage of women in employment, 2003, as a percentage of working age population
Average age of labour force exit, men, 2002–2007 Net present value of pension wealth men, 2005, as a percentage of gross earnings of an average production worker Gross present value of pension wealth men, 2005, as a percentage of gross earnings of an average production worker Difference gross/net men Weighted average pension wealth, men, 2009
Model 3: impact for men LFExitMen NPVmenPen GPVmenPen DifferenceGPV-NPVmen WAvPenmen
Average age of labour force exit, 2002–2007 Difference in gross present value and net present value of pension wealth, where expressed as a percentage of gross earnings of an average production worker, 2005 Gini Coefficient of income inequality for those aged 65 and over, 2005 Percentage of GDP spent on income benefits for older people, 2005 Private pension fund assets as a percentage of GDP, 2005 Public expenditure on pensions as a percentage of GDP, 2005
Extended name
Model 2: impact for women LFExitWomen NPVwomenPen GPVwomenPen DifferenceGPV-NPVwomen WAvPenWomen PerWomenEmployed
GiniCoef PercentG PrivPenF PubExPen
General model 1 Average L DiffGPVN
Abbreviated name
Variables used in the analysis
Table 1
Social Policy & Administration, Vol. 45, No. 2, April 2011
119
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 2 Gini Coefficients of income inequality, after taxes and transfers; sorted by most unequal income distribution for older population, 2000–2005 Working age pop, mid-2000 Mexico Korea USA Portugal Turkey Japan Greece France Italy Spain Australia Austria Ireland Poland Switzerland Canada Germany UK Norway Belgium Finland Iceland Netherlands Luxembourg New Zealand Hungary Sweden Czech Republic Denmark Slovak Republic OECD average OECD st dev
0.47 0.30 0.37 0.38 0.42 0.31 0.31 0.28 0.35 0.31 0.31 0.27 0.32 0.38 0.27 0.32 0.30 0.34 0.28 0.27 0.27 0.29 0.27 0.26 0.33 0.30 0.24 0.27 0.23 0.27 0.31 0.05
Older population 2000
mid-2000s
0.56
0.56 0.40 0.40 0.38 0.37 0.34 0.33 0.31 0.31 0.31 0.30 0.28 0.28 0.28 0.28 0.27 0.27 0.27 0.25 0.24 0.24 0.24 0.24 0.23 0.23 0.22 0.22 0.20 0.20 0.20 0.29 0.08
0.37 0.38 0.36 0.37 0.31 0.33 0.26 0.31 0.24 0.33 0.25 0.29 0.26 0.26 0.29 0.22 0.29 0.24 0.25 0.25 0.28 0.23 0.22 0.18 0.2
Difference with working age pop, mid-2000 -0.09 0.10 -0.03 0 0.05 -0.03 -0.02 -0.03 0.04 0 0.01 -0.01 0.04 0.1 -0.01 0.05 0.03 0.07 0.03 0.03 0.03 0.05 0.03 0.1 0.08 0.02 0.07 0.03 0.07 0.02
Change older pop 2000– mid-2000 0 -0.03 0 0.02 0.04 0 0.02 -0.05 0.01 -0.04 0.05 -0.03 0.01 -0.01 -0.01 0.02 -0.03 0.05 0 0.01 0.02 0.05 0.01 0 -0.02 0
Source: OECD online database, www.oecd.org.
where there has been some marginal increase in inequality. All the Scandinavian countries have less income inequality in old age than in their working age populations, but again several other countries share this characteristic. Similarly, table 3 shows that all the Scandinavian countries are below the OECD 2000 mean average relative poverty rate for those aged 65 and over, although Norway is close to the mean with a score of 12.4 per cent (OECD 25; average 13.3). It is important to note that the median income will vary much 120
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 3 Relative poverty rates, aged 65 and over, 2000 Ireland Portugal Mexico USA Greece Australia Japan Turkey Italy UK Norway Switzerland France Finland Austria Germany Sweden Denmark Luxembourg Hungary Canada Poland Czech Republic Netherlands New Zealand OECD 25
35.5 29.2 28.4 24.6 24.3 23.6 21.1 16.4 15.3 14.4 12.4 11.2 10.5 10.4 9.2 8.5 7.8 6.1 6.1 5.2 4.3 4.3 2.1 1.6 0.4 13.3
Source: OECD 2009, www.oecd.org. Note: Poverty rates are measured as the proportion of individuals with equivalised disposable income less than 50 per cent of the median income of the entire population.
between countries, so for example while relative poverty is low in Hungary, the average income will be considerably less than northern European countries, and Norway has a high average income level compared to other countries because of its oil wealth. Figure 1 shows the results of the first cluster model. The first dendrogram illustrates the sequence of construction of overlapping cluster groups. The second dendrogram shows how the variables contribute in the clusters to the separation of the cases. In the first dendrogram there are three main clusters and four country outliers not in clusters. The first cluster at the top of the dendrogram is southern and central Europe (Austria, France, Greece, Italy, Poland, Belguim and Germany). The second cluster – in the middle – is international and diverse (Australia, Canada, the Netherlands, Iceland, Ireland, New Zealand and the USA). The third cluster is predominantly © 2011 Blackwell Publishing Ltd.
121
Social Policy & Administration, Vol. 45, No. 2, April 2011 Figure 1 Hierarchical cluster analysis – Model one
European (Portugal, the Czech Republic, Hungary, Luxembourg, the Slovak Republic, Finland, the UK, Denmark, Norway, Spain and Sweden) with the addition of Japan. This third cluster contains all the Scandinavian countries. Within the structure of this cluster, Denmark and Norway are in close proximity as a subset and then they join the UK, Finland, Spain and Sweden. At 122
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 4 QCA truth table for Scandinavian group in model 1, using median thresholds
Spain Sweden Finland UK Norway Denmark
v1
v2
v3
v4
v5
v6
v7
1 1 1 1 1 1
1 1 1 1 0 0
0 0 0 0 0 0
1 1 0 1 1 1
0 0 1 1 0 1
1 1 1 0 0 0
0 1 1 1 1 1
Notes: v1 = DepRatio. v2 = PubExPensionsGDP. v3 = DiffGPVNPV. v4 = AverageLFE. v5 = PrivPenFundAssetGDP. v6 = PercentGDPOldBen. v7 = Gini Coefficient.
the bottom of the first dendrogram in figure 1 are the four ‘outlier’ countries that do not converge into the three major cluster groupings: Switzerland, Turkey, Korea and Mexico. The second dendrogram in figure 1 shows the effects of the variables on the model. Public expenditure on pensions and the percentage of GDP spent on income benefits are closely related and these, in turn, are associated with the ratio of older people to the working age population. The difference in gross pension as compared to net pension has a more distant relationship with these variables but is nevertheless partly linked to them. The other group of variables are linked separately in their influence on constructing the case base model. The average age of exit from the labour market is closely related to the Gini Coefficient of income inequality in older age, while private pension funds as a percentage of GDP only have a very limited attachment to the other variables. The QCA ‘truth table’ in table 4 shows the strong similarities but also the case differences in the cluster group that includes the Scandinavian countries. A truth table assigns variable information about each case into two binary categories: above threshold (1) and below threshold (0); here the threshold is the median score. Cases can then be compared by their pattern of 1 or 0 scores across the table. Finland is different to the other Scandinavian countries because of its large private pension fund assets that give it a similarity point with the UK (over 70 per cent of GDP). Denmark also has over the threshold score on this variable, given its private pension fund assets are 32 per cent of GDP. Finland has a lower than average age of labour force exit when compared to the others. Denmark and Norway have lower proportions of GDP spent on public pensions than the other countries. © 2011 Blackwell Publishing Ltd.
123
Social Policy & Administration, Vol. 45, No. 2, April 2011 Figure 2 Hierarchical cluster analysis – Model two: women
We can conclude that in the general model the Scandinavian countries are identifiable together but while they have key similarities, they also have some specific and important differences. In the second model (figure 2), the cluster analysis is designed to see if Scandinavian countries can be identified as having different policy outputs for women when compared to other countries. The dendrogram shows four core cluster groups. The first group is international, largely non-European, but including the UK and Ireland (Australia, Canada, the UK, the USA, Ireland, New Zealand, Japan and South Korea). 124
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 5 QCA truth table for Scandinavian group in model 2, using median thresholds
Switzerland Portugal Sweden Iceland, Norway Netherlands Finland Denmark
v1
v2
v3
v4
v5
v6
1 1 1 1 0 0 0
1 1 0 0 1 1 0
1 1 1 0 1 1 0
1 1 0 1 1 0 1
1 0 1 1 1 0 1
1 1 1 1 1 1 1
Notes: v1 = LFExit Women. v2 = NPVwomenPen. v3 = GPVwomenPen. v4 = DifferenceGPV-NPVwomen. v5 = WAvPenWomen. v6 = PerWomenEmployed.
The second group is central and eastern European (France, Germany, Belgium, the Czech Republic, Poland and the Slovak Republic). The third group includes the Scandinavian counties with several other European countries (Finland, Switzerland, Portugal, Norway, Sweden and Denmark). A fourth group includes southern European and eastern European countries (Austria, Hungary, Greece, Spain, Italy and Turkey). Mexico and Luxembourg are outliers. The QCA truth table in table 5 gives some insights into the key differences between the third cluster group of Scandinavian and European countries with regards to income support policies for women. All the countries in the cluster group have above average participation rates for women in employment. All of the Scandinavian counties apart from Finland have above average weighted pension wealth for women. Sweden and Finland have below average net minus gross present value of pension wealth for women when compared to Norway and Denmark, and the reverse relationship is true for below average gross present value of pension wealth. This indicates that tax transfers are more important towards achieving income equality objectives for older people in Norway and Denmark. Similar to the taxation of employmentbased income, income tax deducted from the income of older people’s pensions varies considerably from one country to another and is dependent on national and local income tax policies and practices. Sweden and Norway have higher average ages of labour market exit for women when compared with Finland and Denmark. When looking at the actual indicators for each case for the four Scandinavian countries, Finland’s marginally different performance is notable (table 6). It has the lowest proportion of women working, © 2011 Blackwell Publishing Ltd.
125
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 6 Performance of income support policy outputs for women, Scandinavian countries
Percentage of women in employment, 2003, as a percentage of working age population Labour force exit female, 2002–2007 Net present value of pension wealth women, 2005, as a percentage of gross earnings of an average production worker Gross present value of pension wealth women, 2005, as a percentage of gross earnings of an average production worker Difference gross/net women Weighted average pension wealth, women, 2009
Norway
Sweden
Denmark
Finland
72.87
72.83
70.47
65.70
63.23 11.70
62.91 12.10
61.31 10.50
61.03 14.70
9.50
11.40
8.40
13.30
2.20 12.00
0.70 12.50
2.10 15.60
1.40 10.80
the lowest age of labour market exit, the lowest weighted average pension wealth, but the highest net present value and gross present value of pension wealth for women. In the third model (figure 3), the cluster analysis is again designed to see if Scandinavian countries can be identified as different. In this model, the aim is to explore policy output differences for men when compared to other countries. The dendrogram shows three core groupings and a pairing of Denmark and Iceland. In addition, Luxembourg is an outlier. The Scandinavian countries cannot be distinguished as a separate group. The first group is a large and diverse mixture of European countries joined by Australia and Canada. Sweden and Norway are in this cluster and have no proximity as a subset. The second group is European, including Finland. The third group is international, including Ireland, with the UK and the USA in a subset pair. The variable clustering indicates that net present value of pension wealth for men, 2005, and gross present value of pension wealth for men, 2005, work together in forming the cluster definitions. The average age of labour force exit for men between 2002–2007 contributes little to the definition of clusters, and this is confirmed when the model is re-run without this extra variable and there is very little change in the dendrogram output. Given the convergence of Scandinavian countries into other European groupings, the analysis is not pursued with a truth table.
Discussion This article produces different cluster sets than those of Soede et al. (2004) and Soede and Vrooman (2008). This is unsurprising, given the different method and variable input and the total number of countries examined. 126
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Figure 3 Hierarchical cluster analysis – Model three: men
Soede’s recent work used the method of Categorical Principal Components Analysis (CatPCA) with some rotation, with an emphasis on whether a country had a particular policy characteristic or not, for example, whether first tier pensions are means tested. CatPCA is a statistical method that creates an argument that some cases share the same overall variable characteristics based on the frequency of shared categories. In conventional Principal Components Analysis (PCA), group membership is defined by loadings (correlations) with the components and the variables used to make these decisions are ordinal or interval numerical scales, but with CatPCA the logic for group membership © 2011 Blackwell Publishing Ltd.
127
Social Policy & Administration, Vol. 45, No. 2, April 2011
comes from the combined classification of category of membership for each variable. For example, a category might be nominal, either a country has a regulation requiring all employees to have occupational pensions, or it does not. In Soede’s work, some quantitative variables were used, but these were recoded into categories based on threshold points (similar to the method used in QCA). This has the overall effect of making the model more influenced by the designation of policy characteristics rather than quantitative comparisons of relative income levels. Soede and Vrooman’s 2008 model is therefore impressive for its sophistication with dealing with complex differences between policy and delivery structures (34 variables were used), and the simpler model used in this article does not attempt such a comprehensive coverage of policy attributes. But the use of categorical variables can make the clusters more arbitrary with less ability to observe the ‘fuzzy’ nature of cluster group definition, as can be observed with hierarchical principals of cluster analysis. Hierarchical cluster analysis is ‘fuzzy’ in the sense that the analysis starts with isolating a large number of pairs or small clusters and then reduces the differences between these groupings or pairs, so as to propose further combinations. Soede and Vrooman’s 2008 model is more sophisticated in its detailed policy observations from European countries because it focuses on fewer counties – 23 rather than the OECD 30 used in this article. This maximized the potential for the model to discriminate within European countries and raised the likelihood of getting a good fit with ideal type regime classifications. The model in this article uses a more exploratory inductive approach to key quantitative data and places less emphasis on the theoretical definition of the cluster sets identified. Hierarchical cluster modelling is used that demonstrates the evolving and ‘fuzzy’ definition of cluster sets, rather than using a hypothesis generation of trying to place countries into regime-based ideal type theoretical groups. The method is consistent with Uprichard (2009: 135): ‘it is more accurate to think of these as fuzzy sets which may overlap with each other, to a lesser or greater extent’. The key methodological point here is that complexity theory demonstrates that it is impossible to ever have a complete picture of how nation states compare but only a partial understanding of how they might be similar and different, and that these aspects are dynamic and constantly changing over time and never static. Despite these important methodological differences in the approach of this exploratory article when compared to Soede’s comprehensive model, there are some key points of analysis that are supported by both approaches. These points relate to the differences of the Scandinavian countries in how they compare as cases within that group; but on the issue of where the Scandinavian countries are placed in relation to other OECD countries, the differences in this article when compared to Soede and Vrooman’s are in part because of their methodological starting points, as discussed above. In terms of how the Scandinavian countries compare to each other, the general findings of this article are similar to those of Soede and Vrooman (2008). Denmark and Sweden are identified as the closest pairing. Soede and Vrooman’s work on policy structures and attributes analyses this as being primarily due to the presence of private schemes that employees are obliged 128
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
by government to contribute to. In Finland and Norway the role of state- and public-funded schemes is still more central to the overall provision. Finally, this article makes a contribution to an aspect of comparative analysis that is under-researched and not well-developed in previous comparative research on regime theory, i.e. the gender aspect of income support in older age. The data analysis in the second model confirms that the Scandinavian countries have some key similarities. Women have a high participation in the labour market and a tendency to leave the labour market later when compared to other OECD countries, and an above threshold weighted average pension wealth (with the exception of Finland on the last two points). Norway, Sweden and Denmark are most closely related as confirmed by their subset cluster in the first tier of the hierarchical analysis in model two, figure 2.
Conclusion The analysis in this article suggests that Scandinavian countries can continue to be viewed as social democratic, comprehensive and egalitarian (EspingAndersen 1999: 78) in terms of their welfare provision when compared to other OECD countries. These countries perform above the OECD average in measures of inequality of income such as the Gini Coefficent of older age net income, and relative poverty in older age when compared to their internal national income distribution. In terms of the most generalized reasons for these positive outcomes, the link with occupational welfare seems to be the key, given the relative late exit from the workforce in those countries, the high proportion of women of working age in employment and the move towards occupational pension schemes in Sweden and Denmark (as identified by Soede and Vrooman 2008). Analysis of gross and net present value of pension wealth as a percentage of gross earnings of an average production worker adds little understanding to the comparative analysis undertaken in this article. These variables analyse internal compositional effects within national states, and do not in themselves offer much insight into cross-country comparison. This is particularly noted in the attempt to create a model for a comparison of men’s income in old age where these variables do not add any insight into a synthesis of how countries compare with each other. This confirms the formidable challenge of trying to isolate occupational and fiscal effects on welfare outcome as argued by Greve, and others, and the exploratory models in this article are not able to provide much new insight into these highly technical and conceptually complex measurement issues. Further work is needed to see how variables that measure internal compositional aspects of welfare delivery relate to categories of policy type and policy attributes. In this respect, it is important to see this article as a macro exploratory analysis and broad confirmation that Scandinavian countries can claim to be different, and it is not a comprehensive attempt to confirm a hypothesis of previous regime types and/or how various complex and hybrid policy models perform. Nevertheless, the overall conclusion from this study is that Scandinavian counties are different in their provision of income for older people and in particular for older women, and this is important when seen in light of other © 2011 Blackwell Publishing Ltd.
129
Social Policy & Administration, Vol. 45, No. 2, April 2011
research findings on income provision for older people and comparative policy such as Soede and Vrooman’s seminal study. In Scandinavia, the move towards occupational welfare and pensions seems less likely to disadvantage women because of their high participation in the labour market. In countries where women are encouraged and incentivized to take traditional caring roles over and above employment and where participation in the labour market is low, occupational welfare systems may pauperize women (Haynes et al. 2010). Other countries would do well to watch with interest how the Scandinavian countries are evolving and restructuring their policy models and their implementation as the proportion of the older population increases in all OECD countries.
References Byrne, D. (1998), Complexity Theory and the Social Sciences, London: Macmillan. Byrne, D. (2002), Interpreting Quantitative Data, London: Sage. Byrne, D. and Ragin, C. (eds) (2009), Case-Based Methods, London: Sage. Cesaratto, S. (2006), A Tale of Two Pension Reforms: A Sraffian View. In B. Greve (ed.), The Future of the Welfare State: European and Global Perspectives, Aldershot: Ashgate. Esping-Andersen, G. (1990), The Three Worlds of Welfare Capitalism, Cambridge: Polity Press and Princeton, NJ: Princeton University Press. Esping-Andersen, G. (1999), Social Foundations of Post Industrial Economies, Oxford: Oxford University Press. Greve, B. (2006), Is there a Demographic Time-bomb? In B. Greve (ed.), The Future of the Welfare State: European and Global Perspectives, Aldershot: Ashgate. Greve, B. (2007), Occupational Welfare: Winners and Losers, Cheltenham: Edward Elgar. Haynes, P. (2008), Complexity Theory and Evaluation in Public Management. A Qualitative Systems Approach, Public Mangement Review, 10, 3: 401–19. Haynes, P., Hill, M. and Banks, L. (2010), Older People’s Family Contacts and Long Term Care Expenditure in OECD Countries a Comparative Approach using QCA, Social Policy & Administration, 44, 1: 67–84. Hill, M. (2007), Pensions, Bristol: Policy Press Hudson, J. and Kuhner, S. (2010), Beyond the Dependent Variable Problem: The Methodological Challenges of Capturing Productive and Protective Dimensions of Social Policy, Social Policy and Society, 9, 2: 167–80. OECD (2007), Public Sector Pensions and the Challenge of an Ageing Public Service, OECD Working Papers on Public Governance, 2007/2, Paris: OECD, http://www.oecd.org/ (accessed 1 October 2010). OECD (2009), Pensions at a Glance 2009: Retirement-Income Systems in OECD Countries, Paris: OECD, http://www.oecd.org/els/social/pensions/PAG (accessed 1 October 2010). Soede, A. and Vrooman, C. (2008), A comparative typology of pension regimes, ENEPRI, http://www.enepri.org (accessed 1 May 2010). Soede, A. J., Vrooman, P. M. and Segre, G. (2004), Unequal Welfare States, The Hague: SCP. Uprichard, E. (2009), Introducing Cluster Analysis: What can it teach us about the case? In D. Byrne and C. Ragin (eds), The Sage Handbook of Case-Based Methods, London: Sage.
130
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00760.x Vol. 45, No. 2, April 2011, pp. 131–145
Nordic Employment Policies – Change and Continuity Before and During the Financial Crisis spol_760
131..145
Sven Jochem Abstract The Nordic countries serve as models for successful employment and labour market policies. In this article, Nordic employment and labour market policies are analyzed from a comparative point of view. It is argued that Nordic employment and labour market policies have lost some of their distinctive features. Active labour market policies, for example, are now at the centre of policy priorities in many countries of the EU. And in some other respects, the Nordic countries have converged towards political patterns characteristic for states in central Europe, for example, de-centralized patterns of wage bargaining and the partial lack of corporatist concertation. During the current financial crisis, specific patterns of crisis management can be observed in Scandinavia that make these countries distinct from many other countries in Europe.
Keywords Employment policies; Nordic model; Financial crisis; Comparative analysis Introduction In many respects, the Nordic countries serve as models. Beyond the broad field of social security, the Nordic model of employment is the one mostly referred to in the literature. In Scandinavia, full employment, i.e. low open unemployment and high labour force participation or employment rates, is a central political goal – politically independent of incumbent parties. At the beginning of the 1990s, the Nordic countries suffered from a banking crisis and a steep decrease in employment (or increase in open unemployment) (Jochem 2000). However, in the aftermath of this Nordic downturn, governments in Scandinavia gradually restored full-employment – albeit with differing success rates. Until the outbreak of the current global financial crisis, Scandinavian countries were able to successfully recalibrate their welfare states and consolidate public budgets. Address for correspondence: Sven Jochem, University of Bamberg, Feldkirchenstraße 21, D-96045 Bamberg, Germany. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
It is argued in this article that, on the one hand, the Nordic model has lost some of its traditional peculiarities in labour market issues, i.e. centralized wage bargaining has gradually vanished and active labour market policies, once dominant in the Nordic countries, have been gradually extended to other OECD countries. On the other hand, however, there is still a distinctive Nordic pattern in employment policies, such as high rates of public sector employment or lifelong learning policies intended to enhance labour force skills and economic growth. This article is divided into six sections. After a short introduction, an overview of labour market profiles and employment developments in Scandinavia compared to other OECD countries is provided. In the following two sections, Nordic employment and labour market policies are analyzed respectively. In the fifth section, Nordic crisis policies targeted to combat the current global financial crisis are presented and comparatively discussed. The final section concludes.
Nordic Employment and Labour Market Profiles In Scandinavia, the high integration of the population into the labour market is exceptional. As can be seen from the figures in table 1, the Nordic countries rank high in regard to employment–population ratios. Iceland, with 84.2 per cent employed compared to the population (aged 15–64), is at the top of the range of the classical OECD countries. Only in the UK or Switzerland do employment ratios reach Nordic levels. Finland is the country of the Nordic ‘family of nations’ (Castles 1993) with the lowest ratio, being rather similar to countries in the mid-range, such as Germany or Japan. At the bottom of the range, Italy, Belgium and France represent examples of the ‘conservative welfare state’ (Esping-Andersen 1999), where ‘welfare without work’ still seems to be a correct description for comparatively low employment rates (Eichhorst and Hemerijck 2008). Additionally, we can observe how impressive the increase in Nordic employment was after the economic crisis in the early 1990s. All five countries were able – from an already high level – to further increase employment. With regard to the spread of part-time employment, it becomes obvious that the ‘Dutch Miracle’ (Visser and Hemerijck 1997) is built on the pillar of part-time employment expansion. In the Nordic countries, part-time employment is lower, as in the Netherlands. In the Netherlands – and in most other OECD countries – part-time employment is concentrated mainly on the female labour force. In this respect, Finland and Iceland are prominent examples of overall low part-time employment and a (relatively) low share of female part-time employment. Therefore, the low employment ratio for Finland should be interpreted from this perspective. In this country, full-term employment is dominant, even for women. Full employment in the Nordic countries is more or less the attempt to integrate most people into the labour market. As the figures for the standardized unemployment rates show, low unemployment is a feature of the Danish, Norwegian and Icelandic path (until the outbreak of the global financial crisis). In Finland and Sweden, however, open unemployment is in 132
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 1 The Nordic employment profile Employment/ population ratio
Australia Austria Belgium Canada Denmark Finland France Germany Greece Iceland Ireland Italy Japan Luxembourg Netherlands New Zealand Norway Portugal Spain Sweden Switzerland UK USA Mean
Standardised unemployment rate
Long-term unemployment
Part-time employment (percentage share)
1994
2008
2008
1994
2008
Total 2008
Female 2008
66.0 68.4 55.7 67.0 72.4 60.7 58.4 64.5 54.1 78.5 51.9 51.5 69.3 60.2 63.9 68.0 72.2 64.0 47.4 71.5 75.6 68.7 72.0 64.4
73.2 72.1 62.0 73.7 78.4 71.9 64.6 70.2 62.2 84.2 68.1 58.7 70.7 64.4 76.1 74.9 78.1 68.2 65.3 75.7 79.5 72.7 70.9 71.1
4.2 3.8 7.0 6.1 3.4 6.4 7.8 7.3 7.7 3.0 6.3 6.8 4.0 4.9 2.8 4.2 2.5 7.7 11.4 6.2 3.5 5.6 5.8 5.6
36.1 18.4 58.3 17.9 32.1 – 38.5 44.3 50.5 15.1 64.3 61.5 17.5 29.6 49.4 32.7 28.8 43.4 56.2 25.7 29.0 45.4 12.2 36.7
14.9 24.2 52.6 7.1 16.1 18.2 37.9 53.4 49.6 4.1 29.4 47.5 33.3 38.6 36.3 4.4 6.0 48.3 23.8 12.4 34.3 25.5 10.6 27.3
23.8 17.6 18.7 18.4 18.0 11.5 13.4 22.1 7.8 15.1 21.0 16.3 19.6 12.7 36.1 22.4 20.3 9.7 11.1 14.4 25.9 22.9 12.2 17.9
71.7 80.4 82.9 67.8 61.7 63.0 79.5 79.9 67.4 71.2 79.0 75.8 70.4 91.9 75.5 72.3 71.7 68.5 80.6 64.6 81.2 76.1 68.1 74.0
Source: OECD 2009, Statistical Annex. Notes: Employment/population ratio = percentage of employed persons compared to population aged 15–64. Long-term unemployment = percentage of individuals unemployed for 12 months or longer compared to all unemployed persons. Part-time employment (percentage share total) = percentage share of part-time employment compared to total employment. Part-time employment (percentage share female) = percentage share of female part-time employment compared to total part-time employment. – = value not available.
© 2011 Blackwell Publishing Ltd.
133
Social Policy & Administration, Vol. 45, No. 2, April 2011
the mid-range for all OECD countries. Of special interest for the Nordic employment model is the comparatively low rate of long-term unemployment. Even from a longitudinal perspective, the decline of longterm unemployment is impressive for the Nordic countries. While in most countries of the OECD long-term unemployment was already high in the early 1990s, in some countries this share has increased even more since then – Germany being a concise example of countries with severe labour market problems. In contrast, long-term unemployment decreased significantly in Denmark, Iceland, Sweden and, most impressively, Norway. Hence, we can conclude that the Nordic welfare states not only provide encompassing social security, they also enable – or at least do not inhibit – dynamic labour markets. Welfare state institutions shape the structure and the dynamics of labour markets. In the Nordic countries, welfare policies frame employment patterns in a specific way. One example is the integration of employees between 55 and 64 years into the labour market (cf. figures in table 2). The highest share is reported for Iceland, and the other Nordic countries are at the top of the range (with Finland having the lowest share of the Nordic countries). Hence, early-exit routes from the labour market to the social security system exist even in the Nordic countries, but they are not used to the extent as in, for example, Belgium, Italy or France. A ‘social investment strategy’ implies the presence of activating labour market measures and encompassing education schemes (Morel et al. 2009). In this respect, the Nordic countries also rank at the top of the OECD countries. One reason for this is the spending on active labour market policies. Denmark and Sweden spend more than any other OECD country on measures that are designed to re-integrate unemployed into the labour market or to upgrade their employment-specific skills, and Norway and Finland are in the mid-range. This ‘social investment strategy’ is complemented by intensive education measures. Not only do Nordic countries spend more money than other OECD countries on education, if we measure the participation of citizens in lifelong learning schemes, the Nordic countries are at the top of the sample, too (Eichhorst and Hemerijck 2008: 16). The final aspect of labour market profiles is wage formation. The policies are analyzed in detail below, however, the figures in table 2 show that wage increases in the Nordic countries were generally, since the early 1990s, in line with those in the OECD countries. Recently, wage increases in Sweden and especially in Norway exceeded the OECD average significantly. In Norway, this development may be explained by wage-driving oil industries in the economy and leapfrogging strategies of social partners in the industrial sectors. Given the data at hand, the Nordic employment patterns are to a certain extent different from most of the other OECD countries. An encompassing integration of the citizens into the labour market, low rates of long-term unemployment, activating labour market policies and intensive investments into lifelong learning schemes are the cornerstones of the Nordic model. 134
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 2 The Nordic labour market profile
Australia Austria Belgium Canada Denmark Finland France Germany Greece Iceland Ireland Italy Japan Luxembourg Netherlands New Zealand Norway Portugal Spain Sweden Switzerland UK USA Mean
Employment ratio 55–64
EALMP
Average annual growth rates of real average wages (percentage)
2008
2007
1990–1995
1995–2000
2000–2005
2006–2007
57.4 41.0 32.8 57.5 57.7 56.4 38.2 53.8 42.9 83.3 53.9 34.4 66.3 38.3 50.7 71.9 69.3 50.8 45.6 70.3 68.4 58.2 62.1 54.8
0.32a) 0.68 1.30 0.29a) 1.31 0.86 0.92 0.77 – – 0.62 0.46 0.16a) 0.48 1.09 0.35 0.56 0.53 0.80 1.12 0.60 0.32a) 0.13a) 0.7
1.0 1.2 2.1 -0.2 0.8 -0.1 1.1 2.1 3.2 – 2.5 -0.7 1.3 1.9 0.3 – 1.2 1.1 1.9 -0.3 1.0 1.0 1.0 1.1
2.1 0.6 1.3 2.0 1.6 1.4 1.3 0.8 2.4 – 2.2 0.8 0.5 1.2 0.0 – 2.2 2.6 -0.5 3.3 0.9 2.6 2.9 1.5
1.2 0.9 0.3 1.1 2.0 2.4 1.4 0.3 2.8 – 2.7 0.3 0.3 1.1 0.4 – 3.2 0.3 -0.1 1.4 1.1 1.6 0.4 1.2
1.1 -0.1 -0.1 2.6 0.3 1.3 1.0 0.0 2.2 – 3.2 0.1 -0.4 1.9 3.3 – 6.0 1.0 0.2 3.7 0.3 1.4 1.0 1.4
Source: OECD 2009, Statistical Annex. Notes: Employment ratio 55–64 = employment ratio of employees aged between 55 and 64 compared to respective population. EALMP = expenditure for active labour market policy as a percentage of GDP. a) = fiscal year 2007–08. – = value not available.
Employment Policies – Openness, European Integration and Institutional Inertia Most of the Nordic economies are highly competitive, as comparative investigations report (World Economic Forum 2010). Out of 139 countries analyzed in the study, Sweden, Finland and Denmark rank 2nd, 7th and 9th, respectively, whereas Norway ranks 14th and Iceland 31st. Hence, the Nordic economies are robust and even in respect of growth performance, the Nordic countries are at the top of the OECD countries. © 2011 Blackwell Publishing Ltd.
135
Social Policy & Administration, Vol. 45, No. 2, April 2011
In contrast to the ‘golden age’ perception of the Nordic model, today all Nordic economies are highly integrated into the political frame shaped by the European market project – with the partial exception of Iceland (until the outbreak of the global financial crisis). Hence, the monetary policy-frame embedded into the EU becomes imperative even for the Nordic countries. Quite in contrast to many countries in central, southern or eastern Europe, the Nordic countries were successful in combining economic openness with financial consolidation. After the economic crisis in the early 1990s, governments in the Nordic countries implemented strict fiscal policy rules. These countries entered the crisis with sound public finances – in contrast to most of the other European countries (cf. European Commission 2009b). Therefore, Nordic employment growth during the past decades has rested on the foundation of sound public finances, and comparative data show that this ‘employment puzzle’ was not enabled by increasing income inequality. Since the mid-1990s, inequality in the Nordic countries has slightly increased (OECD 2010: 49). However, compared to other OECD countries, equality of income is still a distinctive feature of the Nordic model. These observations question the well-known ‘trilemma argument’ put forward by Iversen and Wren (1998). According to this argument, Nordic employment policies would enable full employment and equality but would violate against the goal of sound public finances. At least since the mid-1990s, this argument cannot explain Nordic employment patterns as all Nordic governments switched to a framework of sound public finances without sacrificing high employment patterns. Nordic employment patterns rest on a mixture of different economic policies and historic path dependencies (Becker and Schwartz 2005; Becker 2008). Beyond the high share of public employment (cf. below), private sector employment is highly dynamic and has benefited from the strategies of Nordic firms in exploiting strategic niches in the international economy. While Finland and Sweden are contemporary IT champions in the Western world, Danish firms were able to successfully exploit niches in manufacturing, handcrafted products and food production (Becker 2008). Beyond this success story lies the fact that Nordic economies tend to facilitate structural change in a special way – by the highly skilled labour force, dynamic and activating labour market policies, and various structural policies implemented by the governments to enable research and development as well as innovations. The Nordic economies are small and open towards world markets (and global crises). Therefore, the employment success is dependent on the ability of wage bargaining partners to ensure wage policies that do not harm national export industries. During the ‘golden time’, it was possible for Nordic social partners to follow strictly domestic wage bargaining goals, given the (implicit) leeway for monetary policies. Frequent strategic devaluations of Nordic currencies to defend the competitiveness of Nordic products on world markets is an example of this monetary freedom. However, the monetary framework of European integration does effectively impede such strategies. Nordic wage bargaining is judged to be highly corporatist and centralist (Kenworthy 2000; Siaroff 1999). However, since the 1980s, wage bargaining institutions and patterns have gradually changed in Scandinavia (Elvander 2002). Danish social partners reacted to pressures from the bourgeois govern136
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
ment in the mid-1980s by anchoring explicitly national wage developments to those in the main trading countries. The new social pacts ensured wage moderation and enabled labour market reforms. Additionally, the social partners implemented a gradual decentralization of wage bargaining. Today, centralized wage bargaining in Denmark covers basic pay schemes, but main wage negotiations take place at the local level. To ensure moderate wage growth, coordination is intensive in the camps of employers and trade unions, respectively, which circumvents strategies of leapfrogging. As a consequence, wage bargaining in Denmark is highly de-centralized but nevertheless coordinated. The last tripartite social pact from 2007 emphasized this wage bargaining pattern and strengthened some policies of paternity leave and on the job training (EIRO 2007). During the global financial crisis, tripartism was challenged – work-sharing schemes were widely used, but the government was not willing to extend these schemes or to combine them with further educational measures (EIRO 2009a). In Finland, centralized wage bargaining patterns have dominated since the 1970s. Even during the economic crisis of the 1990s, social partners and (oversized) governments could successfully strengthen tripartism, thereby enabling moderate wage growth and encompassing welfare state reforms. Even if the Finish wage bargaining system may be judged as rigid, some flexible wage negotiations at the local level have taken place. Nevertheless, centralized (and long-time) agreements have prevailed until today. In the last bargaining round, concluded in April 2008, most effective wage negotiations took place at the industrial sectors level, wage increases were comparatively high and the collective agreements last until 2010/2011. This departure from the centralized pattern was accompanied by working time changes and the introduction of reforms concerning sabbatical leave schemes. Currently, the financial crisis and the crisis policies implemented by the government are highly disputed between the trade unions and the employers’ federation. The ambition of the government to raise the retirement age is, especially, highly contested (EIRO 2009b). The Norwegian development is dominated by central agreements which fix minimum wage increases and additional wage negotiations at the industrial sectors or local level. During the past wage round, the highest wage increases in recent history were negotiated and the banking sector could negotiate the highest wage increases compared to other industrial sectors. Norwegian manufacturing is faced with declining competitiveness because of higher national wage growth than in the main trading countries and a strong national currency. Nevertheless, even during the global financial crisis, unemployment increased only slightly as the Norwegian economy could rely on the wealth of Norwegian oil industries and an encompassing public sector. Added to which, the government strengthened the rules for temporary layoffs in 2009, thereby reducing lay-offs during the crisis – a scheme that is quite similar to the German system of ‘Kurzarbeit’ (EIRO 2009c). Finally, the Swedish development resembles the Danish one to a certain extent, but in contrast to Denmark, tripartite negotiations have generally failed during the recent past. Bi-partite negotiations could restore the wage leading function of wage bargaining partners in the industry (Industriavtalet © 2011 Blackwell Publishing Ltd.
137
Social Policy & Administration, Vol. 45, No. 2, April 2011
from 1997 to date), but wage competition between different industrial sectors remains intense. It is telling that, despite several attempts, the Swedish social partners could not agree to reformulate the historic compromise from Saltsjöbaden (1938) and neither to agree on tripartite social pacts, such as those in Denmark or Finland. During recent years, wages have increased steadily and undermined the competitiveness of Swedish export industries. The 2010 wage round is still not finished. Many open conflicts between trade unions and employers’ federations occurred, and most importantly, the Association of Swedish Engineering Industries, the largest employer organisation in Swedish industry, decided to leave the agreement (EIRO 2010). The fate of the Industriavtalet is undecided. Given the high uncertainty because of the deep recession and the disputes over labour market policies – and especially the government reforms of unemployment insurance (see below) – the climate between the social partners (and the bourgeois government) is frosty. In sum, wage bargaining institutions of the ‘golden age’ model changed significantly. Only in Finland is the centralized tripartite pattern still observable, albeit under pressure. In the other Nordic countries, decentralization has taken place, and while social partnership is alive in Denmark and in Norway, Swedish social partners are involved in intensive disputes over wage formation institutions, labour market policy and crisis policies. Therefore, Scandinavia is no longer the region of tripartite social pacts. While these policies work elsewhere (Hassel 2009; Natali and Pochet 2009; Siegel 2005), Nordic corporatism has lost some of its archetypical charisma. The employment success of the Nordic countries depends further on the specific tax regimes. High tax burdens result in the prize of encompassing welfare states in Scandinavia (Ganghof 2006; 2007). During the past, rightwing populist parties have challenged the Nordic tax systems in Denmark and Norway. In Sweden, the incumbent bourgeois coalition avoided a strict policy of overall tax reductions but introduced an earned income tax credit scheme in 2007, which reduces the tax burden for low income groups. Given the necessity to finance the public sector (and the crisis packages, cf. below), it is evident that the Achilles’ heel of the Nordic model is still the capacity to tax the population more than in other European countries. During the electoral campaign in Sweden, the Social Democrats claimed increasing taxes were necessary to manage the challenges ahead (and restore the quality of the public sector), whereas the bourgeois parties defended their policies of moderate tax reductions. Another cornerstone of the Nordic employment model, and highly dependent on the tax regime, is the broad share of public employment. Starting in the 1960s, public employment increased in all five Scandinavian countries. This trend was induced by the expansion of public welfare services, a prominent feature of the Nordic model (Sipilä 1997). Due to the complexity of different forms of public sector employment and the difficulty in drawing the borderline between public and private employment spheres accurately, there are only few comparative data available (OECD 2008; Derlien and Peters 2008). Nevertheless, the available figures show that the Nordic countries have still one of the highest public employment shares of all OECD countries – for Norway and Sweden, approximately 30 per cent of total employment is in the 138
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
public sector (2005), while in Germany or Austria it is approximately 10 per cent (OECD 2008). However, due to political decisions, mainly of bourgeois parties in government – and partly due to new dynamics and challenges of intensified European integration (Höpner and Schäfer 2010) – the public sector was deregulated and partially privatized. It is noteworthy that the deregulation and the partial decline of public employment – especially in social services – were also introduced by social democratic parties in Denmark and Sweden during the 1980s and 1990s (Klitgaard 2007). Hence, the public employment share of total employment still seems to be decisive for the Nordic employment model. However, this employment reserve is challenged by European integration, the erosion of tax compliance and the ambitions of bourgeois and even social democratic parties to deregulate the public sector.
Labour Market Policies and Danish ‘Flexicurity’ Active labour market policies are a cornerstone of the Nordic model. Historically, Swedish active labour market policies were an important part of the Rehn-Meidner model of economic policies (Milner and Wadensjö 2001). Especially during the crisis in the 1970s, Swedish governments invested huge amounts of money to activate the labour force. Today, the Danish model of ‘flexicurity’ is the main archetype of active labour market policies in the OECD countries (Dingeldey 2007). However, in all Nordic countries – with the partial exception of Norway – spending on active labour market policies is still above the OECD average (cf. table 2, above). Active labour market policies are nothing new to the Nordic model. However, in recent times the activation approach has changed somehow. While in the ‘golden age’ Nordic model, active labour market policies were framed to enhance regional flexibility of the unemployed, to offer direct wage subsidies or job offers in the public sector, nowadays they are mainly targeted on further training and education (Bonoli 2010). It is argued that these policies improve the matching of demand and supply of labour on the labour market. Critics, however, argue that intensified spending on active labour market policies intervenes into the wage bargaining process and strengthens revolving door effects as participants of active labour market schemes re-enter unemployment schemes after further training. Hence, not only in Scandinavia but in all OECD countries there is a widespread debate about the efficiency of active labour market policies. Furthermore, it is difficult to assess the degree of activation of different policies normally subsumed under the heading of active labour market policies. Bonoli (2010) shows how different policy instruments have different impacts on the activation of unemployed job-seekers. He argues convincingly that the capacity of active labour market policies depends on the interaction of activation measures and passive labour market policies, i.e. how unemployment benefits are designed in each country. This interaction effect mirrors the degree of ‘conditionality’ of labour market policies, as Clasen and Clegg (2007) have termed it. This is the nucleus of the Danish model of ‘flexicurity’. During the 1990s, several reforms increased the activation potential of Danish labour market policies (Schwartz 2001; Madsen 2006). While activation measures were © 2011 Blackwell Publishing Ltd.
139
Social Policy & Administration, Vol. 45, No. 2, April 2011
implemented, they were tightly linked with unemployment benefits. As unemployment benefit is rather generous in Denmark, the unemployment compensation reduces after certain periods of time or when the unemployed person is not willing to co-operate with local labour market authorities (now mostly privatized). In effect, the Danish system includes security as well as conditionality, as the activation effort is directly linked with unemployment benefits. This flexicurity is integrated into a system of low employment protection, which is a historical feature of the Danish labour market, agreed upon already in the historic compromise between labour and capital in 1899. Other Nordic countries, such as Sweden and Finland, oriented policy-making more and more in line with the Danish model, and in Sweden, especially, labour protection has been reduced during the recent decade. These reforms should accomplish the ‘flexicurity’ approach and enhance the flexibility of domestic labour markets. It should be emphasized that the capacity of Nordic active labour market policies to be implemented counter-cyclically depends on the financing of these schemes through taxes. This ensures the expansion of these measures when the economy enters a downturn, layoffs increase and social insurance contributions decrease. This might be one institutional reason that many countries – for example, Germany – could not implement an effective flexicurity package, as active labour market policies are financed out of social insurance contributions and therefore the institutional design aggravates counter-cyclical policy initiatives. The new design of activation measures in the Nordic countries has not altered the traditional organization of national unemployment schemes. In Denmark, Finland, Iceland and Sweden, unemployment insurance is voluntary, while it is obligatory in Norway. In the first group of countries the unemployment schemes are administered by the trade unions (the so called ‘Ghent-System’). This kind of organization is currently highly disputed in Sweden. While the bourgeois government could abolish the Ghent System after several legislative attempts in 1994, the social democratic government after 1994 re-established the traditional scheme, explicitly to protect power resources of trade unions. The current bourgeois government cut public subsidies to the voluntary unemployment schemes, and in consequence many Swedes cancelled their insurance – and their trade union membership. During the financial crisis, many Swedes are therefore without unemployment insurance; and Swedish trade unions have had to witness extreme membership losses during the past five years (Kjellberg 2009).
The Nordic Model and the Financial Crisis The Nordic ‘family of nations’ (Castles 1993) has been hit harder by the global financial crisis than many of the other OECD countries – with the exception of Norway. This is not surprising, as economic policies in these countries are targeted towards openness and the exploitation of globalization. The bust in GDP for 2009 was: for all 27 EU countries on average -4.2 per cent, Finland -7.8 per cent, Iceland -6.5 per cent, Sweden -5.2 per cent, Denmark -4.9 per cent and Norway -1.5 per cent (Eurostat 2010). The high level of openness of Nordic economies enforced this rapid downturn. To date, Nordic banks – 140
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
with the well-known exception of Iceland – are in a sound position (Gylfason et al. 2010). It is remarkable, especially for Swedish banks, that the financial turmoil in the USA had no immediate impact on the balance of Nordic financial systems (except Iceland). Intensive linkages of Swedish and other Nordic banks to eastern European banks and the insecure position of the latter, however, do still impose financial risks (Jochem 2010). The story of the Icelandic bust is a story of its own (Schwartz 2010; Gylfason et al. 2010: 137–66). Political actors in Iceland changed very abruptly their policy stance on the issue of European integration. Iceland is currently preparing for EU membership, despite some problems originating from the financial crisis and popular reservations on membership – a development that was unbelievable only a few years ago. The comparative research on the politics of crisis management is still in its infancy. However, the first comparative accounts are available (OECD 2009; European Commission 2009a; ILO 2009; Bertelsmann Stiftung 2010; Gylfason et al. 2010). Although it is too early to state the consequences of the financial crisis on national economies in the medium and long term, immediate measures following the financial crisis to stabilize financial systems were similar in most countries; however, measures to stabilize domestic demand and to foster national export industries differed. First, most central banks co-ordinated their measures to slash interest rates rapidly. As short-term rates approached zero, not only the Federal Reserve but many central banks in the OECD expanded their balance sheets by ‘quantitative easing’, which included hitherto unconventional purchases of securities. In this respect, even Nordic central banks not integrated into the EMU framework followed the steps taken by the European Central Bank. If necessary, central banks, in close co-ordination with national and European authorities, safeguarded individual domestic financial institutions. Second, monetary policies differed. Member countries of the Eurozone could not use monetary policy tools to stabilize exports, which is the situation for Finland. In Sweden (and of course Iceland), the national currency was devaluated after the financial crisis started. However, whether this will, in fact, stabilize or even foster exports, remains to be seen (Gylfason et al. 2010: 167–96). Third, the global bust in trade immediately influenced domestic employment and consumption. In this respect, Nordic welfare states have the largest automatic stabilizers in the OECD. This means that budget elasticity and fiscal sustainability are the highest in Denmark and Sweden; Finland, Norway and Iceland range in the upper mid-field. Therefore, automatic stabilizers of economic swings are working and are supplemented by discretionary policy packages (FinanspolitiskaRådet 2009; OECD 2009). Fourth, the Nordic countries implemented various stimulus policies. In the EU, most countries implemented policy packages with fiscal stimuli. Preliminary accounts show that the fiscal leeway for such measures is high in all Nordic countries; but only in Sweden and Finland did stimuli packages reach approximately 1.5 or 1 per cent of GDP for 2009, while the figures for Denmark and Norway were lower. The highest stimuli were implemented in Spain, Austria, the UK and Germany (European Commission 2009a: 66–9). © 2011 Blackwell Publishing Ltd.
141
Social Policy & Administration, Vol. 45, No. 2, April 2011
Fifth, most countries in the EU or the OECD invested in the public infrastructure or implemented tax cuts to stabilize household income and consumption. It is noteworthy that some countries even tried to subsidize important export industries – such as the automotive industry in Germany. This policy stance is not observable in the Nordic region. In contrast, investments into active labour market policies or education dominate in the Nordic countries. As a consequence, open unemployment increased sharply in Iceland, Denmark, Sweden and Finland. A working-time policy like the German ‘Kurzarbeit’ is not observable in Scandinavia (with the above mentioned exception of Norway). Swedish social partners, however, agreed on voluntary working time reductions negotiated at company level without wage compensation (EIRO 2009d). Sixth, financing the welfare state in times of economic crisis management becomes a salient issue. The Nordic welfare states are still oriented towards services, and these services are administered by local or regional authorities. It is noteworthy that the Swedish government transferred huge amounts of money to these authorities to circumvent layoffs – a policy that has not been implemented in Denmark until now, despite broad claims from several trade unions. Seventh, it is too early to evaluate contemporary crisis policies. The midterm projections depend on the strategies of central banks and continuing insecurities on global financial markets. However, it is already clear that the Nordic countries – except Norway – have to tackle high unemployment in the near future. The low rates of long-term unemployment, however, seem to be a good indicator in assuming that unemployment may be reduced when the global economy picks up again (and the growth forecasts for the Nordic countries are rather prosperous). The good financial position of the Nordic countries further indicates a great potential to further implement efficient employment policies. The Nordic countries were hit hard by the global recession. Immediate crisis reactions followed the international mainstream and were targeted at stabilizing the national financial systems. Moreover, monetary policy is no longer available to counteract economic downturns, and if so, as in the case of Iceland or Sweden, the medium-term effects are highly contested in the literature. But in one respect, Nordic countries differ still from the rest of the OECD – the encompassing welfare states and invasive tax systems provide huge automatic stabilizers. In the short run, they undermine fiscal stability in economic hard times and imply high taxation pressures for the work force. But at the same token, they stabilize domestic demand via social security transfers. And in the medium term, when the upswing arrives, the automatic stabilizers and especially invasive tax policies provide automatically to consolidate public budgets.
Conclusions The Nordic employment model is changing. Beyond the impact of the financial crisis, certain features of the ‘golden age’ model have changed significantly since the 1990s. While the high share of public employment is in most Nordic countries stable (but under pressure), centralized corporatism is vanishing 142
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
gradually. Recently, these changing wage bargaining patterns allowed for wage growth in some Nordic countries clearly above the OECD average. Active labour market policies, once the centre-piece of the Nordic model, are ‘best practice’ for the EU and, hence, some European governments intend to expand these policies. Nevertheless, most Nordic countries still spend far more money on these measures than most other OECD countries. What is still a distinctive Nordic pattern of employment policies is the high investment in education policies and the attempts to continuously upgrade the skills of the labour force. In the aftermath of the global financial crisis, governments in Scandinavia esteemed the encompassing welfare states with huge inbuilt automatic stabilizers. And in contrast to Central Europe, labour market policies focusing on activation were further expanded. In the Nordic countries, we can observe only a few initiatives to subsidize firms and industries directly, which is different to crisis policies in Germany, for example. Despite increasing rates of open unemployment in Scandinavia, the historical policy pattern remained stable not to subsidize particular firms but to help the unemployed to re-enter labour markets after structural changes of the economy. Given the fiscal solidity and sustainability in the Nordic countries, the leeway to counteract the current crisis is greater than in most other OECD countries. But how the different governments will use this leeway and which policy priorities they will prefer, depends largely on how long the current crisis will impact the small and open economies in Scandinavia.
References Becker, U. (2008), Was ist dran am skandinavischen Modell? Eine vergleichende Betrachtung, Leviathan, 36: 229–48. Becker, U. and Schwartz, H. (eds) (2005), Employment Miracles. A Critical Comparison of the Dutch, Scandinavian, Swiss, Australian and Irish Case versus Germany and the US, Amsterdam: Amsterdam University Press. Bertelsmann Stiftung (ed.) (2010), Managing the Crisis. A Comparative Analysis of Economic Governance in 14 Countries, Gütersloh: Bertelsmann Stiftung. Bonoli, G. (2010), Varieties in Social Investment in Labor Market Policy. In N. Morel, B. Palier and J. Palme (eds), What Future for Social Investment?, Stockholm: Institute for Future Studies, pp. 55–66. Castles, F. G. (1993), Families of Nations: Patterns of Public Policy in Western Democracies, Aldershot: Darmouth. Clasen, J. and Clegg, D. (2007), Levels and Levers of Conditionality: Measuring Change Within Welfare States. In J. Clasen and N. A. Siegel (eds), Investigating Welfare State Change. The ‘Dependent Variable Problem’ in Comparative Analysis, Cheltenham: Edward Elgar, pp. 166–97. Derlien, H. -J. and Peters, B. G. (eds) (2008), The State at Work, Volume 2. Comparative Public Service Systems, Cheltenham: Edward Elgar. Dingeldey, I. (2007), Between Workfare and Enablement – The Different Paths to Transformation of the Welfare State: A Comparative Analysis of Activation Labour Market Policies, European Journal of Political Research, 46: 823–51. Eichhorst, W. and Hemerijck, A. (2008), Welfare and Employment: A European Dilemma?, IZA Discussion Paper No. 3870, Bonn: IZA. © 2011 Blackwell Publishing Ltd.
143
Social Policy & Administration, Vol. 45, No. 2, April 2011
EIRO (2007), New Industry Agreement Marked by Innovatory Elements, http:// www.eurofound.europa.eu/eiro/2007/03/articles/dk0703029i.htm (accessed 25 April 2010). EIRO (2009a), Economic Crisis leads to Extensive Use of Work-Sharing, http://www. eurofound.europa.eu/eiro/2009/03/articles/dk0903021i.htm (accessed 25 April 2010). EIRO (2009b), Widespread Opposition to Government Plan to Rise Retirement Age, http:// www.eurofound.europa.eu/eiro/2009/03/articles/fi0903019i.htm (accessed 25 April 2010). EIRO (2009c), Government launches Measures to tackle Economic Crisis, http:// www.eurofound.europa.eu/eiro/2009/02/articles/no0902049i.htm (accessed 25 April 2010). EIRO (2009d), Agreement on Temporary Layoffs Reached in Manufacturing, http:// www.eurofound.europa.eu/eiro/2009/03/articles/se0903019i.htm (accessed 25 April 2010). EIRO (2010), Largest Employer Organisation in Industry Leaves Industrial Agreement, http:// www.eurofound.europa.eu/eiro/2010/05/articles/se1005019i.htm (accessed 10 November 2010). Elvander, N. (2002), The Labour Market Regimes in the Nordic Countries: A Comparative Analysis, Scandinavian Political Studies, 25: 117–37. Esping-Andersen, G. (1999), Social Foundations of Postindustrial Economies, Oxford: Oxford University Press. European Commission (2009a), Economic Crisis in Europe: Causes, Consequences and Responses, European Economy 7/2009, Luxembourg: European Commission. European Commission (2009b), Sustainability Report 2009, European Economy 9/2009, Luxembourg: European Commission. Eurostat (2010), Wachstumrate des realen BIP, http://epp.eurostat.ec.europa.eu/tgm/ table.do?tab=table&init=1&plugin=1&language=de&pcode=tsieb020 (accessed 20 April 2010). FinanspolitiskaRådet (2009), Swedish Fiscal Policy, Report of the Fiscal Policy Council 2009, Stockholm: FinanspolitiskaRådet. Ganghof, S. (2006), The Politics of Income Taxation, Colchester: ECPR. Ganghof, S. (2007), The Political Economy of High Income Taxation. Capital Taxation, Path Dependence, and Political Institutions in Denmark, Comparative Political Studies, 40: 1059–84. Gylfason, T., Holmström, B., Korkman, S., Söderström, H. and Vihriälä, V. (2010), Nordics in Global Crisis. Vulnerability and Resilience, Helsinki: ETLA. Hassel, A. (2009) Policies and Politics in Social Pacts in Europe, European Journal of Industrial Relations, 15: 7–26. Höpner, M. and Schäfer, A. (2010), A New Phase of European Integration: Organised Capitalisms in Post-Ricardian Europe, West European Politics, 33: 344– 68. ILO (2009), World of Work Report 2009, Geneva: ILO. Iversen, T. and Wren, A. (1998), Equality, Employment, and Budgetary Restraint: The Trilemma of the Service Economy, World Politics, 50: 507–46. Jochem, S. (2000), Nordic Labour Market Policies in Transition, West European Politics, 23: 115–38. Jochem, S. (2010), Sweden Country Report. In Bertelsmann Stiftung (ed.), Managing the Crisis. A Comparative Assessment of Economic Governance in 14 Economies, Gütersloh: Bertelsmann Stiftung. Kenworthy, L. (2000), Quantitative Indicators of Corporatism: A Survey and Assesment, MPIfG Discussion Paper 00/4, Cologne: MPIfG.
144
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Kjellberg, A. (2009), The Swedish Ghent System and Trade Unions under Pressure, Transfer, 3, 4: 481–504. Klitgaard, M. B. (2007),Why are they doing it? Social Democracy and MarketOriented Welfare State Reforms, West European Politics, 30: 172–94. Madsen, P. K. (2006), How Can It Possibly Fly? The Paradox of a Dynamic Labour Market in a Scandinavian Welfare State. In J. L. Campbell, J. A. Hall and O. K. Pedersen (eds), National Identity and the Varieties of Capitalism. The Danish Experience, Montreal: McGill-Queens University Press, pp. 321–55. Milner, H. and Wadensjö, E. (eds) (2001), GöstaRehn and the Swedish Model at Home and Abroad, Aldershot: Ashgate. Morel, N., Palier, B. and Palme, J. (eds) (2009), What Future for Social Investment?, Stockholm: Institute for Future Studies. Natali, D. and Pochet, P. (2009), The Evolution of Social Pacts in the EMU Era: What Type of Institutionalization?, European Journal of Industrial Relations, 15: 147–66. OECD (2008), Employment in Government in the Perspective of the Production Costs of Goods and Services in the Public Domain, Paris: OECD. OECD (2009), Employment Outlook, Paris: OECD. OECD (2010), Economic Survey: Finland, Paris: OECD. Schwartz, H. M. (2001), The Danish ‘Miracle’. Luck, Pluck, or Stuck? Comparative Political Studies, 34: 131–55. Schwartz, H. M. (2010), Iceland’s Financial Iceberg: Why Leveraging up is a Titanic Mistake without a Reserve Currency, unpublished manuscript. Siaroff, A. (1999), Corporatism in 24 industrial democracies: Meaning and Measurement, European Journal of Political Research, 36: 175–205. Siegel, N. A. (2005), Social Pacts Revisited: ‘Competitive Concertation’ and Complex Causality in Negotiated Welfare State Reforms, European Journal of Industrial Relations, 11: 107–26. Sipilä, J. (ed.) (1997), Social Care Services: The Key to the Scandinavian Welfare Model, Aldershot: Ashgate. Visser, J. and Hemerijck, A. C. (1997), A Dutch Miracle. Job Growth, Welfare Reform and Corporatism in the Netherlands, Amsterdam: Amsterdam University Press. World Economic Forum (2010), The Global Competitiveness Raport 2010/2011, Geneva: World Economic Forum, http://www3.weforum.org/docs/WEF_Global CompetitivenessReport_2010-11.pdf (accessed 20 September 2010).
© 2011 Blackwell Publishing Ltd.
145
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00761.x Vol. 45, No. 2, April 2011, pp. 146–160
Has the Nordic Welfare Model Been Transformed?
spol_761
146..160
Jon Kvist and Bent Greve Abstract The Nordic welfare model is undergoing a fundamental transformation. Using Denmark we show how a universal welfare state model is gradually being transformed into an emergent multi-tiered welfare state. Whereas the Danish pension system’s having become multi-tiered in the 1990s, with private schemes – collective and individual – supplementing public schemes is well documented, scant attention has focused on more recent developments in other areas of the welfare state. This article shows how the multi-tiered welfare state spread in the 2000s to policies for families, the unemployed and the sick. Although Denmark still offers universal coverage in core welfare state areas, the increased use of occupational and fiscal welfare as well as changes in public schemes has gradually transformed the nation into a multi-tiered welfare state that is more dualistic and individualistic, with participation in the labour market becoming still more important for entitlement to benefits. These profound changes have taken place in such a way that although core characteristics are still in place, new structures and understandings of the welfare state are also developing. Thus classical typologies need revision, so that they include more focus on this combination of universality and institutional attachment to the labour market. Moreover, measures of what welfare comprises should include not only public but also private elements.
Keywords Universal welfare state; Nordic model; Denmark; Occupational and fiscal welfare; Dualism Introduction The Nordic model is under siege. Ageing populations, immigration, globalisation and, most recently, financial and debt crisis are often framed as pressures that jeopardise the economic sustainability of the Nordic welfare model. Even disregarding these pressures, many Nordic politicians find the model ripe for change. This article examines how recent policy developments call into question whether the Nordic welfare model can maintain its emphasis on equality (including equal access to the labour market for men and for women), generous and universal cash benefits, and comprehensive and high quality Address for correspondence: Jon Kvist, Centre for Welfare State Research, Department of Political Science, University of Southern Denmark, Campusvej 55, DK 5230 Odense, Denmark. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
services (Greve 2007a; Kvist 1999). Do the Nordic countries continue to qualify as having Social Democratic welfare models rather than Liberal or Conservative welfare models (Esping-Andersen 1999)? Existing comparative and historical studies on the Nordic countries may underestimate the scope of recent change. Comparative studies typically use macro-indicators, finding that the Nordic countries continue to be relatively stable in spending on social policy and with a relatively high degree of equality vis-à-vis other countries (e.g. Castles 2004). Historical studies appreciate the complexity of change, but tend to overlook the profound changes in recent years (but see Christiansen et al. 2006). Comparative analysis also looks mainly at direct public intervention while overlooking fiscal and occupational welfare (but see, respectively, Kvist and Sinfield 1996; and Greve 2007b). Assessments within any one policy risk underestimating the scope of change if this also takes place in other policies. In short, what is missing for the Nordic countries are broad investigations of contemporary changes, especially including those incremental changes that may gradually diminish the core in the traditional understanding of the Nordic welfare model. As Mahoney and Thelen (2010) argue, institutional change is often brought about by incremental changes. Fortunately, the value of such studies has been shown for the other two dominant welfare models. In a study of the USA as representing the Liberal welfare model, Hacker (2004) demonstrates how the US welfare system has become more privatised not so much through direct reforms such as eliminating public policies or replacing them with private ones, but rather by a gradual transformation of policy resulting from changing circumstances (‘drift’), internal adaptation of existing policies (‘conversion’) and the creation of new policies without eliminating the old ones (‘layering’). Most recently, Palier and his colleagues revealed how waves of reforms of the Bismarckian model (a nickname for the Conservative welfare model) led in the 1990s and 2000s to a shrinking of social insurance that, in turn, resulted in ‘multiple dualisms’. The first dualisation is made by a private component added to the public system for insiders and the second dualisation is made within the public system with an even sharper distinction between insiders and outsiders (Palier 2010). This article contributes to such investigations by focusing on the Nordic welfare model. We study one country, Denmark, in the 2000s. The 2000s witnessed a polarisation with more people getting rich, more people getting poor and fewer people getting middle incomes (Juul 2010). Whereas changes of inequality can be explained by both policy changes in tax and social policies as well as non-policy changes in the value of, especially, houses and stocks, this article looks only at the former. We investigate policy developments in four areas central to a Nordic welfare regime. Our analysis concludes that the multitude of small changes found across these areas over the ten-year period, when added up and seen together, challenge the notion of the Nordic welfare policies as characterised by equality as the main value inherent in outcomes and policies. What is more, as policy changes unfold they are likely to contribute to further inequality. Some changes – e.g. in the pension system – have a clear impact on future generations. © 2011 Blackwell Publishing Ltd.
147
Social Policy & Administration, Vol. 45, No. 2, April 2011
The article proceeds as follows. In the next section, we describe the method and data of the study. Then follows the analytical section. The analysis of child and family policies finds gradual erosion of benefits and, most recently, cuts in the form of benefit ceilings and lower benefit levels that disproportionately affect families with many children. The analysis of policies for the unemployed indicates a movement towards a more narrow understanding of flexicurity towards emphasizing a ‘work first’ approach over the upgrading of skills, a gradual erosion of benefits and the emergence of a two-track minimum income system. The analysis of pension policies shows a clear movement towards a multi-tiered system in which, universality notwithstanding, occupational welfare combined with the use of fiscal welfare has become increasingly important, making participation in the labour market of outmost importance. The analysis of policies for the ill points out that the widespread existence of sickness benefits through collective agreements has recently been paralleled by a massive growth in private health care insurance schemes stimulated by favourable taxation. The subsequent discussion compares patterns of development across the policy areas. The concluding section assesses the importance of findings on our understanding of the ongoing transformation of the Nordic welfare model.
Theory, Data and Method The literature on the Nordic welfare model identifies a number of constitutive elements. These include comprehensive state responsibility, universal coverage, individualism, high employment, equality of opportunities and results, high quality (public) services, high generosity, a decentralised organisation, strong social partners, a tradition of social dialogue, and some corporatism (Kvist 1999; Greve 2007a). Theoretically, we use this understanding of the Nordic welfare state as a reference point for the analysis. By looking at policy changes and their impact on the welfare state, we use the model as a proxy for the direction of change. The article is thus in line with the social policy tradition of combining institutional changes with how they influence the existing understanding of the model. Our framework is also inspired by Richard M. Titmuss’ social divisions of welfare. In short, public welfare, fiscal welfare and occupational welfare are three distinct routes that provide welfare that are more often than not associated with different equality outcomes (Titmuss 1958; Sinfield 1978). Often, the routes work in tandem, for example when taxes privilege occupational schemes. Normally, we label state welfare and fiscal welfare as public schemes and occupational welfare as private schemes. In the category of private schemes we also find individual schemes. In reality, however, the distinction between the public and private schemes is blurred. Private schemes are regulated and often enjoy favourable (public) tax treatment, for example, most occupational schemes, as compared to other income. The interaction effects between fiscal and occupational welfare have distributional consequences, what Titmuss (1958) aptly named ‘the concealed multipliers of occupational success’. Our contention is that shifts in the particular mix of public and 148
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
private schemes have distributional consequences that may be more or less in accordance with the characteristics of the Nordic welfare model as just described. Because most studies specifically examine one feature or policy area, they give only a partial picture of the transformation across these many distinct elements that together form the Nordic welfare model. In part, an inherent trade-off exists between the number of aspects and the number of cases that a study can cover. To get a fuller picture, we analyse policy change in four areas central to the Nordic welfare model: policies for families, the unemployed, the elderly, and the sick. We examine Denmark, which in the 1990s was seen as the pioneer of the Nordic model, especially given its successful and internationally recognised kick-start of the economy and labour market reforms. Denmark thus gradually took over from Sweden (which suffered a severe economic crisis in the early 1990s) the role of being the representative of the Nordic welfare model. Danish policy developments in the 2000s are also important for comparative research for political reasons. In 2001, the Social Democratic-led coalition government that had ruled since 1993 gave way to a right-wing coalition government of Liberals and Conservatives. They have ruled with the loyal support of the Danish People’s Party, and their policies shed light on how the Nordic welfare model may develop in the other Nordic countries as parties with more focus on migration and nationalistic policies become more powerful. Whilst the Danish People’s Party is an antiimmigration party, it also has a social profile that is reminiscent of old style Social Democratic thinking. Throughout the 2000s the party pressed for social policy reforms disfavouring immigrants, while at the same time trying to preserve or improve the social rights of other groups, especially elderly and vulnerable groups. One of the advantages of a single case study is that data is not constrained to the available comparative data, which can be partial or even misleading. The study uses national institutional data and statistics on public and private schemes that is not yet available in comparative data sets. Furthermore, using the time since 2000 enables both in-depth investigations and an understanding of how many small changes might suggest larger, more fundamental changes in the model. The value of the method used here depends not only on the availability of data but also on the understanding of institutional change. The focus is on whether mainly institutional changes have had an impact on universalism and on equality. In particular, we are interested in change that through the establishment of more tiers in the welfare system leads to a more differential treatment of groups. Are public schemes divided into two or more tiers by introducing new schemes or introducing elements in the existing schemes that leads to a de facto differential treatment of groups who are otherwise in the same category of families, unemployed, elderly or sick? Are new or old private schemes promoted through the tax system, more choice between providers or a hollowing out of the quality or scope of public schemes? © 2011 Blackwell Publishing Ltd.
149
Social Policy & Administration, Vol. 45, No. 2, April 2011
Developments in the Danish Welfare Model This section examines core areas of the Danish model to depict deviation from the Nordic welfare model. We pay attention to changes in both public and private schemes that contribute to a differential treatment of people. In each subsection we, first, describe the conventional understanding of the Nordic welfare model in the given policy area, second, analyse developments of public services and social security schemes, and, third, perform a similar analysis for the private schemes. Policies for families with children The Nordic welfare model combines universal child care with generous leave schemes and family allowances that enable parents to reconcile work and family life thus boosting female labour supply and, arguably, investing in the social and cognitive skills of children (Esping-Andersen 2009). Compared to other countries the Nordic welfare states early developed family policies with an emphasis on high quality day care at an affordable price and relatively high benefit levels (Abrahamson et al. 2005). In the 2000s, Denmark maintained a high coverage rate around 90 per cent for children aged one to six years in day care. User fees have been set at maximum 25 per cent of costs with even smaller fees for low income families. However, user fees have increased in recent years, mostly in line with cost developments, and in 2010 municipalities could charge 30 per cent of costs, if the child care institution served lunch meals. Hence, the costs of having children in day care have risen in recent years. Private and public child care institutions do not differ significantly. They are regulated by the same laws and receive the same subsidies. Free choice between public and private providers and between institutions in and between municipalities has been expanded. Perhaps because there is maximum on user fees that is set by municipalities and not by institutions, Denmark has yet to experience the emergence of large numbers of private day care. That being said, this may change, as the possibility of charging a supplementary payment for specific activities now is possible. Family benefits have, in general, been generous and until recently not subject to any direct cuts. However, the value of benefits has gradually been reduced due to the way indexation takes place, which implies a slight reduction compared to consumer prices of approximately 0.3 per cent per year. The lack of full indexation also goes for social assistance, unemployment and sickness benefits and pension. Not fully indexing benefits is the text book example of drift, i.e. erosion of benefit with no active decisions being made. For the first time since the 1970s benefits were cut in 2010. The government and the Danish People’s Party decided in June 2010 on an austerity package that cut benefits for families. Family benefits are reduced with 5 per cent phased in from 2010 to 2013. A new benefit ceiling of DKK 35,000 annually has also been put in place. The combined loss of income caused by the benefit ceiling and benefit cuts cannot exceed DKK 12,000 per year until 2013, after which the permissible amount of loss increases by DKK 3,000 each year to 150
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
expire completely in 2020. In short, families with many children will receive less and less child family benefits in the coming years. As a result of these cuts, there will be 6,000 more people living in risk of poverty, as defined by having under 50 per cent of the median income (AE 2010). Most of these people live in areas with high concentrations of ethnic minorities (AE 2010). There are few signs of a dualisation at the other end of the income scale through new or expanded private schemes. Collective agreement has for many years given most employees the right to take a day off in connection with the illness of a child. Collective agreements during the 2000s doubled this period to paid leave for the first two days of the child’s sickness. Also, collective agreements for the majority of the labour force introduced care days where employees can take up to four days off for caring for children, the elderly or for other such activities. Because employment–population rates are relatively high, the expanded social rights in collective agreements may not constitute an elitist dualisation trend but rather a broad one between people in and out of work.
Policies for the unemployed Active labour market policies were historically aimed at ensuring full employment and equal access to the labour market for both men and women. The unemployed had relative easy access to generous benefits, either unemployment insurance or social assistance. Private individual unemployment insurance virtually did not exist. In the 1990s the two systems of, respectively, benefits for unemployed and active labour market programmes got linked under the label ‘activation’ and, since then, benefits have become conditional on taking part in activation programmes. Together with a kick-start of the economy in 1993, activation is often said to account for the reduction of unemployment in the 1990s. Emphasis on activation continued in the 2000s with a shift from ‘empowering’ and ‘human capital investment’ to ‘make work pay’ and a ‘work first’ approach. The government wanted to evade the possible lock-in effects that, according to some effect evaluations, were associated with educational programmes, especially in the short-term. The 2003 labour market reform ‘More in work’ distinguished three programmes: guidance and qualification, internships in private or public companies and employment with a wage subsidy to the employer. Making work pay was further sought by lower benefit levels for young people on social assistance and by introducing a new less generous benefit for people coming to the country who would previously have been entitled to social assistance. Another significant change is the establishment of a two-track system for claimants on minimum income schemes, one track for ‘natives’ and one for de facto ‘non-natives’. This has been made possible by changes to the general social assistance scheme and by the introduction of a new benefit. Social assistance now includes the criterion that each partner in a couple must work at least 450 hours over a two-year period to remain entitled to benefits. If both or one partner cannot satisfy this work demand, then the couple will only © 2011 Blackwell Publishing Ltd.
151
Social Policy & Administration, Vol. 45, No. 2, April 2011
receive one social assistance benefit. The work demand obviously affects households where housewifery is more prevalent. Social assistance also got a benefit ceiling, which tends to reduce the level of benefits for families with many children. The new benefit, Start help, pays considerably lower benefits than the ordinary social assistance scheme to people coming to Denmark for the first seven years of their stay. All these changes were said to increase work incentives and do not discriminate on the basis of nationality, but they all disproportionally affect certain groups of ethnic minorities. The size of unemployment insurance benefits has not been cut since the early eighties. Over the last two decades, however, their value eroded. The indexation mechanism described earlier led to smaller increases than a full wage indexation. The unemployed also lost out to those in work when a specific tax-allowance for working people was introduced in 2004. The most important change was a halving of the benefit period. The maximum benefit period was shortened from four to two years as part of the austerity package of June 2010 between the government and the Danish People’s Party. Private unemployment insurance was almost non-existent in the 1990s. Generosity of social assistance and unemployment insurance was high mainly for low income groups, as replacement rates for middle and high income groups were markedly lower than the formula of 90 per cent of previous earnings indicates due to a low benefit ceiling. However, because unemployment was concentrated among low income groups, the low replacement rates for middle and high income groups did not result in pressure for higher benefits or alternative private schemes in the 1990s. Individual private unemployment insurance is now increasing. Following its introduction in 2002 there was a steep increase of insured, peaking at 82,218 people in 2006 (Forsikring & Pension 2010). This was followed by a drop to 70,175 people in 2008 and a boost to 94,504 people being insured in December 2009, due to the financial crisis. Coverage rates varied from around four per cent of the total labour force to five per cent to date. In other words, the share of privately insured individuals is still small, but may grow as other public benefits are eroded due to the lack of full indexation or if adverse economy developments lead to more middle and high income group people becoming unemployed.
Policies for the ill Albeit organised differently, sick employees in all the Nordic countries receive sickness benefits that replace a proportion of their wages, with many receiving full wage replacement. When in need of treatment, people have access to a public health system where the extent and amount of user fees vary across countries and areas within health. Whereas differences may prevail when comparing Nordic countries, their similarities become apparent when comparing them with other countries’ health care systems: their mainly taxfinanced health care policies provide universal and egalitarian health services. The Danish Health Act stipulates that the health care system is based on the principle of easy and equal access for all citizens. Recent years have seen a change in the wording from free and equal to easy and equal access. This 152
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
core principle in the Health Act is further challenged by policy reforms over the last decade, in particular by the shift in governance focusing on guaranteed access within a short time-span and also for planned types of operations, increased free choice among providers stimulated through the tax system, and generous subsidy schemes for private providers. As a consequence, expenditure has risen over the last few years, and an increasing share of patients has been provided for by private hospitals, mainly paid by public means, i.e. regional authorities. An investigation by the Audit of the State Accounts found that private providers were paid too much (Audit of the State Accounts 2009). They do not specify how changes may be made, but indicate that the price regional authorities pay for treatment in private hsopitals should be reduced. Private health insurance has increased dramatically. Actually, there are various types of private health insurance, the most general one is ‘Danmark’ which is a non-profit firm owned by its members that pays out subsidies for medicines and specialist treatment. Today there are about 2 million members of ‘Danmark’ amounting to nearly 40 per cent of the Danish population of 5.5 million people. However, the most notable development has been in private health insurance run by private for-profit companies. In 2002, private health insurance became a non-taxed fringe benefit. As a result more than 1 million people were covered in 2008 compared to 50,000 in 2002 (Ministry of Health and Prevention 2010). Companies finance 90 per cent of the policies, with the insured having privileged access to physiotherapy and psychotherapy as well as to operations linked to physical movement. Policyholders can skip the waiting list for public treatment and go directly to a private specialist for treatment. The rise in private health insurance questions the main principle of health care of easy and equal access to health services. Those covered by these private health insurances are, broadly speaking, mainly working in the private sector and, due to the segregated labour market, this implies that men to a higher degree than women are covered, and that high income earners have better options than low income earners (Kjellberg et al. 2010). Fiscal and occupational welfare have thus been used as instruments to depart from the Nordic welfare model. How important this break is can be questioned, given that in most areas private production of health care is still only of limited importance (Audit of the State Accounts 2009; Economic Council 2009). Inequalities in health policies can be found in the use of both public and private delivery of health services. For example, the Economic Council has studied how the probability of using the extended free choice depends on personal and other characteristics. Most generally, the probability decreases with age, especially with people over the age of 75, who do not make use of the extended choice probably because they feel more secure in a public hospital and because they have more complex diagnoses that can only be treated in public hospitals (Economic Council 2009). Also the Economic Council found that the probability of making use of the extended free choice between hospitals increases for people in employment, of Danish origins, with an education, and a high income (Economic Council 2009). This may, in part, be explained by the fact that these groups face higher costs in terms of lost income caused by waiting times for treatment in public © 2011 Blackwell Publishing Ltd.
153
Social Policy & Administration, Vol. 45, No. 2, April 2011
hospitals. Single people and those without children are also less likely to choose treatment in a private hospital, perhaps because people making use of the extended choice are more dependent on relatives helping them out in various ways than in the public sector where follow up is more institutionalized. More generally, lack of information and inability to process complex information may help explain inequalities in the take up of health policies. Increased choice thus risks increasing inequality (Greve 2009).
Policies for the old The Nordic welfare model for the elderly is characterised by accessible, good quality social services and universal, generous pensions. However, both services and cash benefits are organised differently in the Nordic countries. The pension systems have two goals: the first is to provide a minimum income to raise elderly people out of poverty in old age and to live a life that is normally accepted, especially for those on basic pensions. The second goal is that pensions must ensure a certain degree of income compensation. In Denmark, these goals are met by what can now be seen as a clear and distinct three-pillar pension system (Petersen 1995; Kvist 1997; Green-Pedersen 2007). The national old age pension is thus no longer the sole provision of economic security for the elderly. Occupational-based labour market pensions (enacted on a large scale by the labour market partners in the late eighties and early nineties) are important to achieve a decent standard of living. The increase of occupational pensions implies that the pressure on the public purse due to the demographic change is reduced. Occupational pensions will, in most cases, reduce how much retired people receive in national old age pension, although all will get the basic amount. In this sense, the universal allencompassing pension system remains universal for those without any further means, but at the same time the real value of the pension has been slightly eroded in the last few years as it is not linked to consumer price. However, at the same time a special income allowance to the elderly has almost implied that pensioners have not been those hardest hit by the economic crisis. The value of the national old age pension has been gradually eroded by the lack of full indexation described earlier. At the same time, the Labour Market Supplementary pension (ATP) has expanded in scope to cover people on early retirement pension and social assistance, i.e. coverage for those at the margins of the labour market. For the insiders on the labour market, collective agreements kept their relative high contribution rates to funded schemes, and in some sectors even increased implying an even stronger role for occupational welfare. Individual private saving for old age became less tax privileged when a ceiling on tax-deductible contributions of DKK 100,000 annually was introduced as part of the austerity package in June 2009. The age of retirement will increase in the future. As part of the Welfare Agreement in 2006, the retirement age for the national old age pension will gradually be raised by two years, i.e. half a year in 2019, 2020, 2021 and 2022, making it 67 years of age in 2022. Also, as part of the Welfare Agreement, a demographic adjustment of the retirement age was introduced so that increases in longevity in part translate into a longer working life and do not 154
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
exclusively contribute to a longer time in retirement. Thus, from 2025, increases in longevity will result in higher retirement ages aimed at keeping the period in retirement at 19.5 years as it was in 1995. Again, the push towards later retirement tends to favour the privileged. The tax-deductible contributions to supplementary schemes grew by DKK 3.4 billion from 2007 to 2008 and to a total of DKK 116.3 billion. Contributions to labour market pensions grew by DKK 6.5 billion and to individual schemes fell by DKK 4 billion. Nearly all contributions to labour market contributions are set by collective agreements, whereas they are voluntary in individual schemes. In light of wage development and the financial crisis straining some household budgets, the diverse development is not surprising. The shift towards a pension system with more private pensions that are based on labour market participation obviously favours those with high and stable income over those with smaller and unstable incomes. Overall, women in Denmark have more or less the same participation as men, but not in their fertile years due to maternity leave and disproportionate take up of other leave schemes. The labour market is highly gender segregated with more women working in the public sector, especially in health, social and education areas. Women also live longer than men. Hence, in private pension schemes, fully actuarial pensions would therefore result in women either enjoying lower benefit than men or women would have to pay higher contributions than men to ensure there is no difference to the level of benefits. To counter such practice in the ATP scheme, a parliamentary majority decided in 1998 to adopt a unisex principle which determines that men and women are covered by the ATP in the same way independently of their gender. As a result, there is a significant redistribution of income from men to women (ATP 2010), confirmation of a willingness to ensure gender equality in the Danish model. Pension inequalities exist between pensioners depending on whether they are single or a couple. The average income for single and cohabiting pensioners is the same, i.e. DKK 184,500 annually (ATP 2009). However, the composition differs, with single people having larger public pensions and couples having larger private pensions. Because living costs are higher for single people, they have less available for consumption. Pension inequalities also have an ethnicity dimension. Because people with a minority background, especially people from non-Western countries, save less than native Danes, they will receive less in pensions when they retire. At the moment, only 1.5 per cent of pensioners are immigrants from nonWestern countries, equating to 14,000 people. In 2050, an estimated 7 per cent of pensioners will be immigrants from non-Western countries, equating to more than 100,000 people. They will, to a large extent, be dependant on means-tested benefits like the pension supplement, housing benefits and personal allowances.
Discussion Has the Nordic welfare model transformed fundamentally? The answer to that question is ‘not yet’, because developments in Denmark show this is not the case. Denmark is the Nordic country where the expansion of private © 2011 Blackwell Publishing Ltd.
155
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 1 A qualitative assessment of the importance of tiers across policy areas in Denmark, 2010 Tier/policy area
Child families
Unemployed
Ill
Old
Private individual Private collective Public universal Public selective
+ + +++ +
+ +++ + ++
++ +++ +++ –
+++ +++ +++ +
Notes: + = limited multitier. +++ = high degree of multi-tiered system. The qualitative assessment is based upon the author’s evaluation in the preceding sections.
supplementary schemes and the establishment of two tracks within the public schemes for, respectively, in- and outsider, have gone furthest, and it is safe to say that the transformation of the Nordic model is still a ‘work in progress’. At this early stage of the transformations the emergence of new tiers has not crowded out the public universal tier. There are big differences in the extent to which a multi-tiered system has come into play in the various policy areas. Table 1 shows that the most fully-fledged multi-tier system is for pensions for the elderly. This comes as no surprise because the ‘revolution’ of the Danish pension system started a decade earlier than in the other areas. The least fully-fledged multi-tier system is for families with children. Here the emergence of a multi-tiered system primarily reduces cash benefits for families with many children and potentially lowers the quality of care by freezing budgets. The remaining two areas, however, have seen marked developments towards a multi-tiered system over the last decade, but they operate at different ends of the system. Policies for the unemployed now have a clear two-track character in the public schemes concerning minimum incomes. Policies for the ill are also moving towards a two-track system, but here with the addition of private collective schemes of health care insurance on top of the public universal health care services. There is no welfare policy area in which there is not a development towards a more multi-tiered system. Is there evidence against the emergence of a more multi-tiered system? Although social expenditure is the most commonly used evidence, aggregate measures such as the share of social expenditures in the gross domestic product result from not only policy design but also other factors, such as need and economic developments. When an economy goes awry, the numerator (social expenditure) grows as more people become dependent on social benefits, and the denominator (gross domestic product) declines or even shrinks. Thus the size of social expenditure in the gross domestic product increases. In conventional comparative studies this rise is a sign of ‘more welfare’. By contrast, single case studies avoid such fallacies, because they can draw on finer institutional data, as well as on tax expenditures that may be available 156
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
only for the country in question. In this study, our use of institutional data results in more nuanced findings than usually found in studies using crude proxies. Looking at developments in the number of people either claiming benefits or working in a given area is another way of assessing welfare reforms. This approach is rarely used in comparative studies due to lack of reliable comparative time-series data. As shown, the marked rise of people with private health and unemployment insurance signals a transformation of the Nordic welfare model. Moreover, most areas – with pension as a central exception – have seen a reduction in the generosity of benefits. However, at the same time the Danish system largely ensures low-income earners high coverage. In certain areas (sickness and maternity benefits), such high coverage is also ensured by agreements among the labour market partners. The increased emphasis on free choice is also a challenge as it changes users to consumers. Since not everybody is equally able to exercise choice, the spread of free choice leads to increased inequality. The shift towards more fiscal and occupational welfare has been driven by political parties, and the labour market partners have taken on the role of provider in the areas of pension, sickness and leave benefits. Furthermore, many private companies by now offer health insurance. The Danish People’s Party has been successful in pressing for policy changes. As previously described, a special social assistance scheme now exists for new entrants to the country, and within the general social assistance are accompanying benefit ceilings and work demands. Family allowances are also subject to a benefit ceiling. All in all, these policy changes tend to disproportionately affect claimants in certain ethnic minority groups more than ethnic Danes. At the same time, the Danish People’s Party has on several occasions hindered benefit cuts that would negatively affect the unemployed, the elderly or the sick more generally. These policy changes have led to a dualism in state welfare schemes, where some people are more equal than others. Whereas all may still be covered, i.e. a high degree of universalism, people are no longer receiving the same benefits. This change constitutes a clear break with the equality value inherent in the traditional Nordic welfare model. Will the future bring more or less dualism or multi-tiered welfare states? Integration of ethnic minorities into national labour markets is proceeding faster than predicted only a few years ago. This can be seen by comparing the projections of labour market participation of non-Western immigrants and their descendants made by the Danish Welfare Commission from 2005–06 with those achieved. The completion of tertiary education by the young ethnic minority females now also surpasses that of ethnic Danes. Perhaps the time of ethnicity as the main divider in welfare policies will return to the classic class divider.
Conclusion The data used here show a movement away from the universal welfare state in central welfare state areas through the worker’s increasing need to be in the © 2011 Blackwell Publishing Ltd.
157
Social Policy & Administration, Vol. 45, No. 2, April 2011
labour market if he or she is to acquire the social rights to benefits. Nonetheless, a high degree of gender equality still exists in Denmark, compared to other countries, even with men still earning more than women. Policy changes also indicate that both the emphasis and the long-term perspective of ALMP have changed, with a move towards more work first. However, the ALMP still fulfills the historical ambition of having equal access to the labour market for men and women and a low level of unemployment. Development of private health insurance could imply a farewell to the universal access to health care, although the tipping point has not yet been reached. Free choice and increased use of market mechanisms in the public sector also suggest a risk of less solidarity in the model. Even so, the degree of economic and gender equality is still high in Denmark when compared with other European countries. This may indicate that it is possible to achieve traditional objectives of equality in the Nordic model even when welfare reforms make it more important to be in work to get benefits and introduce new ways of financing and organising the welfare state. However, if these changes continue, then we will be witnessing a gradual erosion of the universal Danish welfare state, with an increase in the focus and use of a more private model. A financial crisis precipitated the introduction of direct benefit cuts, and then only for families with children, and in ways that disproportionally affect socio-economic groups of ethnic minorities. The emergence of more ‘multi-tiered’ or ‘dualised’ welfare states challenges research in more than one way. Policy researchers will be ill-advised to continue looking only at state welfare and using indicators reflecting average situations or households. Private schemes – whether collective or individual – must increasingly be taken into account and stacked on top of those state welfare schemes, if any, for which a particular group may be eligible. The differentiated coverage of the population combined with developments towards more diversity in both labour markets and families means that researchers must use new indicators. These indicators must be more relevant for their research question than indicators on the situation of standard workers and families, or (even less informatively) on aggregate or average indices. The need for better measures is essential, not only for better portraying social reality but also for differentiating between the similar pathways taken by countries that initially belong to different welfare regimes. The USA, Conservative and Social Democratic countries all enforced private schemes and made minimum income schemes and activation measures harsher. These parallel trends may, however, not lead to convergence of welfare models because the starting points in terms of welfare regime were different and because the timing, speed, intensity and extent of the changes to public and private schemes are likely to vary. Carefully crafted comparative analyses may provide useful answers, and choosing countries from different regimes may prove a useful supplement to existing studies comparing developments inside specific regimes (Emmenegger et al. 2011). The extent to which countries end up on the same route of more tiers or dualism calls for further comparative political studies of how this pathway came about for countries with different institutions, political systems and 158
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
parties in government, as well as diverging ideas and normative belief systems. The big question is what dynamics exist between different tiers in the emerging multi-tiered welfare models. In the Nordic context, the question is simple: Will retaining the support of – and tax money from – privileged socio-economic groups remain possible as they become increasingly covered by complementary, supplementary or even alternative schemes to the state welfare schemes that are gradually eroding or at best at a standstill? If not, we may soon experience a tipping point, after which the model may change in ways that can no longer be seen as matching those of the Nordic welfare model we knew from the 1990s. There is thus a need for a rethinking of our worlds of welfare capitalism.
References Abrahamson, P., Boje, T. and Greve, B. (eds) (2005), Welfare and Families in Europe, Aldershot, Ashgate. AE (2010), Ny model for børnecheck er stadig socialt skæv, Copenhagen: Economic Council of the Labour Movement. ATP (2009), Enlige pensionister tjener det samme som gifte – men har mindre til sig selv, faktum: om pension og samfund, nr. 72, September, http://www.atp.dk (accessed 1 August 2010). ATP (2010), Fem milliarder pensionskroner optjent af mænd går til kvinder, faktum: om pension og samfund, nr. 80, May, http://www.atp.dk (accessed 1 August 2010). Audit of the State Accounts (2009), Rates, quality and access to treatment on private hospitals, Copenhagen: Rigsrevisionen. Castles, F. (2004), The Future of the Welfare State: Crisis myths and crisis realities, Oxford: Oxford University Press. Christiansen, N. F., Petersen, K., Edling, N. and Haave, P. (eds) (2006), The nordic model of welfare, Copenhagen: Museum Tusculanum. Economic Council (2009), Dansk Økonomi Efterår 2009, Copenhagen: Det Økonomiske Råd. Emmenegger, P., Häusermann, S., Palier, B. and Selaib-Kaiser, M. (eds) (2011, forthcoming), The Age of Dualization: Structures, Policies, Politics, Oxford: Oxford University Press. Esping-Andersen, G. (1999), Social Foundations of Postindustrial Economies, Oxford: Oxford University Press. Esping-Andersen, G. (2009), The Incomplete Revolution. Adapting welfare states to women’s new roles, Princeton, NJ: Princeton University Press. Forsikring & Pension (2010), Statistik, http://www.forsikringogpension.dk (accessed 1 August 2010). Green-Pedersen, C. (2007), Denmark: A ‘World Bank’ pension system. In E. M. Immergut, K. M. Anderson and I. Schulze (eds), The Handbook of Western European Pension Politics, Oxford: Oxford University Press, pp. 454–95. Greve, B. (2007a), What Characterise the Nordic Welfare State Model, Journal of Social Sciences, 3, 2: 43–51. Greve, B. (2007b), Occupational Welfare, Winners and Losers, Cheltenham: Edward Elgar. Greve, B. (2009), Can Choice in Welfare States Be Equitable?, Social Policy & Administration, 43, 6: 543–56. Hacker, J. S. (2004), Privatizing risk without privatizing the welfare state: The hidden politics of social policy retrenchment in the United States, The American Political Science Review, 98, 2: 243–60. © 2011 Blackwell Publishing Ltd.
159
Social Policy & Administration, Vol. 45, No. 2, April 2011
Juul, J. (2010), Stor stigning i antallet af rige, Copenhagen: Economic Council of the Labour Movement. Kvist, J. (1997), Retrenchment or restructuring? The emergence of a multi-tiered welfare state in Denmark. In J. Clasen (ed.), Social insurance in Europe, Bristol: Policy Press, pp. 14–39. Kvist, J. (1999), Welfare reform in the Nordic countries in the 1990s: Using fuzzy-set theory to assess conformity to ideal types, Journal of European Social Policy, 9, 3: 231–52. Kvist, J. and Sinfield, A. (1996), Comparing Tax Routes to Welfare in Denmark and the United Kingdom, Copenhagen: The Danish National Institute of Social Research. Kjellberg, J., Nyhus, M. and Søgaard, J. (2010), Private sundhedsforsikringer, Copenhagen: Dansk Sundhedsinstitut. Mahoney, J. and Thelen, K. (eds.) (2010), Explaining Institutional Change: Ambiguity, Agency, and Power, Cambridge: Cambridge University Press. Ministry of Health and Prevention (2010), Det danske sundhedsvæsen i nationalt perspektiv, http://www.sum.dk (accessed 1 August 2010). Palier, B. (ed.) (2010), A Long Goodbye to Bismarck? The Politics of Welfare Reform in Continental Europe, Amsterdam: Amsterdam University Press. Petersen, J. H. (1995), Leuven Lectures: Three Essays on Trends Towards Transformation of the Danish Welfare State, Odense: Center for Helsetjenesteforskning og Socialpolitik. Sinfield, A. (1978), Analyses in the Social Division of Welfare, Journal of Social Policy, 7, 2, April: 129–56. Titmuss, R. M. (1958), Essays on ‘the Welfare State’, London: George Allen and Unwin.
160
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00762.x Vol. 45, No. 2, April 2011, pp. 161–179
Gender Equality Revisited – Changes in Nordic Childcare Policies in the 2000s spol_762
161..179
Guðný Björk Eydal and Tine Rostgaard Abstract The Nordic childcare policy model is often reviewed and even recommended internationally for its contribution to gender equality, high female labour force participation and, perhaps more indirectly, to a high fertility rate. Nordic childcare services and parental leave schemes have thus been portrayed in the literature as policies which have managed to facilitate a work–family model of dual earners and dual carers. However, the recent introduction of cash-for-care schemes seems to go against the Nordic dual earner/dual carer model and ideals of gender equality, in supporting parental (maternal) care of the child in the home. At the same time, new upcoming trends of political fatherhood and the perspective of lifelong learning for the child are also changing the Nordic childcare model. This article provides an analysis of how new childcare policy goals have been articulated into policies from the late 1990s to the late 2000s and how these may challenge the traditional goals of the Nordic welfare states.
Keywords Childcare; Policies; Gender; Equality; Nordic Introduction Comprehensive childcare policies have been claimed to be one of the cornerstones of the Social Democratic welfare model of the Nordic countries (e.g. Gornick and Meyers 2003; Kangas and Rostgaard 2007). As Anttonen and Sipilä (1996) note, the Nordic countries are primarily ‘social service states’ in which public social services are locally and universally provided, thus available for all and used without stigma or loss of status. The extensive and generous support for paid parental leaves, individual rights for fathers, together with generous and abundant public day-care for young children is claimed to be one of the main explanatory factors for the development of a Address for correspondence: Guðný Björk Eydal, Professor, Faculty of Social Work, University of Iceland, 101 Reykjavik, Iceland. Email:
[email protected]. Tine Rostgaard, Senior Researcher, SFI, Herluf Trolle’s Gade 11, DK 1052, Copenhagen, Denmark. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
dual earner/dual carer model in the Nordic countries where women and men share paid and unpaid work (Leira 2006). The political argumentations for generous paid parental leave and day-care services in these countries have, accordingly, been based on ensuring both parents have the possibility to work and care as well as to promote children’s well-being and development (Eydal and Kröger 2009). However, traditional policy concerns perhaps stand out today as ‘a bit old and somewhat blue’ as new policy concerns, especially the goals of flexibility and parental choice, are gaining ground. A closer look reveals that recent policy developments are somewhat contradictory to the dual earner/dual carer model. All the Nordic countries have in recent years implemented schemes of cash for childcare which support parental care (in practise often the mother) of the child in the home. This not only works against gender equality but also results in postponed take-up of day-care for the child, thus working against the investment in the child strategy. At the same time, the facilitation of fatherhood through the introduction of special quotas in the parental leave and an increasing interest in the educational investment in the early years of the child, promote gender equality and early take up of day-care. The aim of this article is to take a closer look at what seems to be the somewhat contradictory aims of the Nordic policies of day-care, parental leave and cash-for-care as they have developed from the late 1990s into the 2000s. The full scope of Nordicness will be represented by the inclusion not only of the ‘usual suspects’ of Sweden, Denmark and Norway, but also countries which are more rarely included in comparative studies on Nordic childcare policies: Iceland and Finland. The article concludes with a consideration of what the derived consequences are for the Nordic model as a unique policy regime and for the policy direction of the dual earner/dual carer model. The results show that the Nordic countries are still ‘social service states’ with generous and abundant public day-care and extensive and generous support for paid parental leave. This public support to parents of young children contributes to the continuation of the dual earner/dual carer model in the Nordic countries, which has in recent years been strengthened – in some countries – by the introduction of father’s quotas in the parental leave. However, the cash-for-care schemes exemplify the variations in the promotion of gender equality and the dual earner/dual carer model across the Nordic countries, and – in addition – seem to work against another Nordic ambition, to provide equal opportunities for all children.
Old, Borrowed and Blue? The Traditional Policy Concerns behind Nordic Childcare Policies The Nordic countries are often singled out for their apparent success in increasing women’s labour force participation and for creating a more gender equal society. In this sense, the Nordic way is often a ‘borrowed’ ideal model in the EU political agenda, but a closer look reveals some wearing out of the traditional policy agendas and some important differences between countries. 162
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
As shown in the following sections, traditional policy concerns of gender equality, female labour force participation and the interest of the child have shaped – and continue to – shape childcare policies in the Nordic countries. However, while these are supported by some of the new and upcoming policy ideals, they are challenged by others.
The shared heritage of gender equality and the creation of the dual earner/dual carer model The Nordic family of nations shares both a historical and cultural heritage and has for decades worked explicitly on creating gender equality in workand family-life, both in single country policies as well as in Nordic co-operation, such as the Nordic Council of Ministers (Arter 1999). As Leira (2006) notes, the gender equality model is the model that is predominantly being presented and promoted in Nordic parenthood policies. This model implies the dual earner/dual carer model, calling for men and women’s equal sharing of the responsibility of care for children and paid work. The model may be said to connote a certain gendered citizenship model where the policy discourse is gender neutral – although there is some dispute over this among Nordic feminist scholars (e.g. Lister 2003) – with the explicit policy goal of promoting equality between men and women. Gender equality is in this way perceived to be more than equality of opportunity. Equality of outcome and especially the gendered division of unpaid and paid work is central (Leira 2006). An important part of the Nordic gender equality model is the facilitation, especially, of women’s labour market participation – or what by Lewis (2001) has been named the Adult Worker Model, where men and women are considered equally employable and working full-time – but with the additional Nordic emphasis on equal sharing of care for children between men and women. The comprehensive Nordic childcare policies play a special role especially with regard to the equal sharing of paid labour as it is one of the most necessary prerequisites for a mother’s participation in the labour market (Bradshaw et al. 1996). Despite this importance of day-care for the Nordic dual earner/dual carer model, volumes of day-care differ substantially between the countries. Denmark has by far the highest coverage for children below one year (17 per cent) and for children aged one to two years (90 per cent), as table 1 on volumes of day-care shows. Iceland, Norway and Sweden have quite similar coverage with regard to the one to two year olds (70–79 per cent), while Finland has relatively low figures for this age (42 per cent). The Nordic countries provide day-care for 95 per cent of children aged three to five years old, except in Finland where only 73 per cent of children in this age group are enrolled in public day-care. And while cash-for-care schemes in many cases are mainly used by low income families, as we will show in the following section, day-care services are often a popular choice among other groups in society, e.g. as Kröger et al. (2003) note for Finland, middle and upper class families are typical users of municipal day-care. Despite similar ideology and goals of public day-care, the difference in the coverage of children in day-care services is an indication of a different empha© 2011 Blackwell Publishing Ltd.
163
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 1 Children aged 0–5 years enrolled in public day care in the Nordic countries in 2008, shown as a percentage of all children in the age groups
Denmark Finland Iceland* Norway Sweden
0–1 years
1–2 years
3–5 years
17 1 7 4 –
90 42 79 75 70
97 73 95 96 97
Source: NOSOSCO 2009: 47. Note: * = Figures for Iceland from 2007.
sis on female employment. The political agenda of facilitating parental labour force participation has been especially pronounced in Sweden and Denmark, where the goal of gender equality in work-life has, to a high degree, received broad political backing and has influenced the day-care service policies in the provision of full-time places and, with regard to Sweden, prioritizing provision for children of working parents (Ellingsæter and Gulbrandsen 2007). Gender equality in work-life has to a lesser degree been a concern in, e.g. Norway where the family policy model contributes to what Skrede (2004) terms ‘gender equality light’, in that it supports a highly gender divided work-life, where women more often are outside the labour market.
Increasing fertility Facilitating high female labour force participation is one thing, but achieving this and at the same time ensuring high levels of fertility is another. At the European policy stage, the Nordic countries are often cited as examples of countries that have achieved this combination of high female labour force participation without losing out in fertility (at a Nordic average of 2.0, compared to 1.6 in EU 27, 2007; Eurostat n.d.) (e.g. European Commission 2007: 77). While fertility was previously an important issue in all the Nordic countries, childcare policies have, however, not been designed with that in mind. As Ellingsæter (2009: 16) points out, the ‘Nordic parental leave policies are not intended as pronatalist measures, nor are they the direct cause of change in mothers’ employment and fertility practises’. Although fertility today does not play a significant role within the Nordic countries in policy debate, the collective evidence, however, seems to corroborate the fact that the Nordic welfare policies that enable parents to combine work and family life have a positive effect on the fertility rate (Ellingsæter 2009; Rønsen and Skrede 2010). 164
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Children’s best interest and the perspective of the child A traditionally more explicit motive for Nordic childcare policies has been to ensure children’s best interests. Childcare policies in the Nordic countries have also been influenced by the increase in children’s legal rights that has set a clear mark on childcare policies, e.g. in the emphasis on children’s voices and participation in decision making in day-care services (Eydal and Satka 2006), as well as in the introduction of child care guarantees for day-care provision (Eydal and Rostgaard 2010, in print). New Policy Concerns: Caring Fathers, Flexibility/Choice and Lifelong Learning While the above mentioned policy concerns have shaped and continue to shape the childcare policies in the Nordic countries, new policy concerns have within the last decades unfolded, of which some support the ideals of the Nordic dual earner/dual carer model, whereas others seem to go against these models, as this section will outline. Nordic fatherhood The importance of fathers as carers has been addressed both as a part of the Nordic policies on gender equality and as children’s right to their fathers (Eydal and Gíslason 2008). However, the role of caring fathers and the extent to which the state should take actions, and which actions, to ensure fathers’ participation in care has led to a heated debate in the Nordic countries during the past decades, culminating in the introduction of a father’s quota in four of the five Nordic countries at present. The father’s quota is an individual, non-transferable entitlement to a number of weeks of parental leave, based on the principle of use-it-or-lose-it, in order to stimulate fathers to take up leave (Lammi-Taskula 2007).1 Part of the explanation for the introduction of this quota was the male take-up of parental leave that remained low in the 1970s and 1980s in all Nordic countries. The idea of special quota for fathers was first discussed in Norway in the 1980s as a way to reduce gender inequality and strengthen the father–child relationship. Norway became in 1993 the first Nordic country to establish a father’s quota of one month of parental leave (Brandth and Kvande 2003). The entitlements of fathers in Norway have gradually been increased and from 2011 the father’s quota will be 12 weeks in all (Barne, Likestillings og Inkluderingsdepartementet n.d.). Both Sweden and Iceland later followed suit: in 1995 Swedish fathers became entitled to one month’s father’s quota, to be extended in 2002 to two months, and from 2008 bonus payments to parents who divided their parental leave equally (Duvander et al. 2008). In 2000, Icelandic mothers and fathers gained equal individual rights to paid parental leave of three months each and three months for either parent (Eydal and Gíslason 2008). In Finland, ideas about increased entitlements of fathers to paid parental leave have been debated for years, but no father’s quota has been introduced so far. However, fathers in Finland can have three weeks by © 2011 Blackwell Publishing Ltd.
165
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 2 Parental leave in Nordic countries, percentage of income covered weeks of entitlements and the percentage of total number of benefit days used by fathers in 2007
Percentage of income Total weeks – only mother – only father – father with mother Percentage of total benefit days used by fathers
Denmark
Finland
Iceland
Norway
Sweden
100 50–64 18 0 2 6
70 44 18 (2) 3 5
80 39 13 13 0 33
100/80 42–52* 9* 6** 2 9
80 69 8 8 2 20
Source: NOSOSCO 2009: 30. Note: * = 56 weeks in total and 10 weeks for fathers from 2009. ** = 10 weeks from 2009 and 12 weeks from 2011.
the end of the paid parental leave if they have used two weeks of the joint entitlements of the parents (Lammi-Taskula and Takala 2009). Inspired by the neighbouring countries, Denmark in 1998 introduced a two-week father’s quota, only later to be abolished as part of an extension of the parental leave period in 2002 (Rostgaard 2002; Borchorst 2006), the main political argument being that it is a private and not a state decision as to which parent should take the parental leave. Table 2 outlines parental leave entitlements of Nordic fathers, and their use of leave days, as of 2007. Fathers in Iceland, Norway and Sweden use higher proportion of total days than do fathers in Finland and Denmark, and take-up suggests that Nordic fathers are first and foremost using their individual entitlements (quotas) while mothers use the lion’s share or all of the joint entitlements (Eydal and Gíslason 2008). Not only are there gender differences in take up, there are also differences in take-up according to socio-economic status. Research has shown that fathers with low income tend to take shorter leave than fathers with high education (e.g. Eydal and Gíslason 2008; Lappegård 2008; Olsen 2005). Hence it has been pointed out that policies on the father’s quota fit well with the ideas and lifestyles of middle class families while they, to a lesser extent, accommodate working class families (Stefansen and Farstad 2010).
Facilitating flexibility and choice Along with the political project of the caring father, there has simultaneously been another competing discourse growing – that of free choice of parents to choose the form of care they believe is best for their children (Moss and Kamerman 2009). Schemes of cash-for-care have accordingly been implemented in all the Nordic countries, but at different points in time, providing 166
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
parents with a cash amount if they do not take up day-care. In all countries, we therefore find the paradoxal development of father’s quota policies clearly indicating which parents should take leave (the father), and cash-for-care policies opening up parental choice and user sovereignty, but as we will show, this also results in it mainly being the mother who stays at home. These cash-for-care schemes are different in character, organization and take-up, and their importance vis-à-vis the day-care services also differs in the Nordic countries (Eydal and Rostgaard 2010). Finland had enacted a cashfor-care scheme by 1985 and Norway in 1998 (Salmi 2006). The remaining three Nordic countries have, until recently, to a greater extent promoted the dual earner/dual carer model, implying that mothers also work when their children are young. However, in 2002 Denmark enacted new laws that made it possible for parents to apply to the municipalities for cash-for-care (Dagtilbudsloven nr. 501 06/06/2007) and in 2008 Sweden enacted laws on cash-for-care (Cronholm 2009). By late 2009 about one third of Swedish municipalities had either introduced cash-for-care schemes or planned to do so (Nyberg 2010). Iceland differs in that there are no national laws on cashfor-care, but since 2006 some municipalities have nevertheless implemented local schemes and, by late 2009, 11 out of 77 Icelandic municipalities had enacted cash-for-care schemes (Rannsóknastofnun um barna- og fjölskylduvernd 2010). The political motives behind the schemes differ among the countries, although a common goal is to ensure parental choice in childcare arrangements, as table 3 shows. In Norway and to some extent also in Finland, equality between parents who use day-care services and parents who do not was an additional argument for implementing the cash-for-care, i.e. in these two countries principles about fairness and equal distribution of resources was also at stake. Furthermore, in the case of Norway, an important goal was to provide parents with more time to spend with their children (Ellingsæter 2006; Rantalaiho 2009). In Denmark, the concern has also been to provide municipal flexibility in the provision of day-care services (Dagtilbudsloven nr. 501 06/06/2007). Futhermore, the timing and institutionalization of the cash-forcare schemes differ between the countries, as table 3 shows. The cash-for-care schemes have been proposed and enacted by parties right of the centre and as Ellingsæter (2006) points out with regard to Norway, the preceding political debate clearly exposed the various parties’ family ideals, illustrating that the ‘main dividing line is between different ideal models of early childhood: family based care versus a parent/institutional care mix’ (2006: 41). However, the schemes have received more public support in Finland compared to Norway, where the two discourses of political fatherhood and choice are clearly competing (Rantalaiho 2009). Table 4 shows the rules regarding eligibility to cash-for-care and amounts paid in the Nordic countries, and in the case of Denmark and Iceland, the rules of the capital cities. There are substantial differences in entitlements, e.g. in Finland, Iceland and Norway, all parents are entitled to the cash allowance, while the Danish allowances are only paid to parents outside the labour market. In Sweden, parents receiving certain social benefits cannot receive cash-for-care, and Swedish and the Norwegian schemes are most flexible, © 2011 Blackwell Publishing Ltd.
167
168
Yes 2002 Municipalities Municipalities Increase choice
Source: Eydal and Rostgaard 2010.
Laws on CFC? Year introduced Financed by Implemented by Main goals
Denmark Yes 1985 State/munciaplities State and municipalities Increase choice (originally also: equality)
Finland No (2006) Municipalities Municipalities Equality/to close the care gap
Iceland
Yes 1998 State State Increase choice/ equality/more time for family
Norway
Law in cash-for-care (CFC), goals, financing and the administrational level of the schemes in the Nordic countries, 2009
Table 3
Yes 2008 State Municipalities Increase choice
Sweden
Social Policy & Administration, Vol. 45, No. 2, April 2011
© 2011 Blackwell Publishing Ltd.
© 2011 Blackwell Publishing Ltd.
1–3 years Yes Yes
1–3 years No Yes Yes 10.8
Age of the child? Part-time day care + partial payments of cash-for-care? Can be used to pay others for care?
Universal, for all parents regardless of type of income? Cash-for-care benefits as a percentage of AW*
No 10.7
250 days–3 years Yes Yes
Sweden
6 months–3 years No No, but other schemes for that No 24.8
Copenhagen
Yes 12
6/9 months–2 years No Yes
Reykjavik
Source: Eydal and Rostgaard 2010. Notes: * = AW 2007 is calculated by NOSOSCO (NOSOSCO 2009: 212–13). For further information on NOSOSCO AW calculations, see http://nososco-eng.nom-nos.dk/filer/publikationer/tabeller/descrtyp07.pdf.
Yes 9.4
Norway
Finland
Nordic countries – cash-for-care, rules for eligibility and amounts as a percentage of average wage (AW), 2009
Table 4
Social Policy & Administration, Vol. 45, No. 2, April 2011
169
Social Policy & Administration, Vol. 45, No. 2, April 2011
since parents can combine part-time day-care and partial allowances. Overall, benefits are rather low, representing between ten to 12 per cent of an average wage, only in Denmark is the amount higher than this (25 per cent). The cash-for-care has been popular especially among parents in Finland. In 2008, more than half of the children under three years of age in Finland (56.9 per cent) were in care supported by cash-for-care allowances (Kela n.d.), reflecting public constructions of childhood which has changed in recent years to positively stress homecare as the proper care solution for children under three years of age (Hiilamo and Kangas 2006; Repo 2010). In the case of Norway, the cash-for-care scheme became popular immediately after being implemented. In 1999, 74.3 per cent of children aged one to three years old were cared for by the support of cash-for-care, but by 2008 this percentage was down to 34.9 per cent (Statistics Norway 2010), partly explained by the simultaneous increase in day-care places (e.g. Ellingsæter and Gulbrandsen 2007). However, in both countries, it is mainly mothers who use the cash-forcare scheme and often for longer periods of time (Ellingsæter and Leira 2006; Rantalaiho 2009; Repo 2010; Salmi 2006). Futhermore, in the case of Norway, take-up rates of the cash-for-care benefit are higher among migrant than non-migrant families (Rønsen and Kitterød 2010). At present, there are no studies comparing take-up rates in Swedish and Icelandic municipalities, which renders it difficult to evaluate and project the popularity of the schemes there. In Denmark, the allowance was only paid to 700 children in 2006 (Rantalaiho 2009). Hence, it is too early to declare that the cash-for-care schemes have had impact in all the Nordic countries, but they nevertheless stand out. The (previously) service-intensive Nordic welfare states have now also introduced cash-for-care with regard to care for children, despite the idea that a child should be cared for at home, usually by the mother, until the age of three, which is in sharp contrast to the Nordic tradition of gender equality policies that have aimed at ensuring both parents the opportunity to participate in labour market and care for their children. The evidence of the gendered outcome is clear for the three countries where data is presently available: it is overwhelmingly women who use the scheme. The introduction of cash-forcare schemes has accordingly been criticized as they are feared to contribute to a backlash for mothers in the labour market and for fathers in the families (Sipilä et al. 2010), as they consolidate the traditional division of labour and economic gender inequality. The result for women may be longer absences from the labour market and a subsequent reduction in human capital, employment opportunities and earnings as well as pension contributions (Sipilä et al. 2010). It is a policy which is in line with present tendencies to introduce elements of user-empowerment and consumerism as the cash benefit can also be used for the purchase of care from other side, and a policy which reflects the ethos of individualism which has gained ground especially in the middle class (Sipilä et al. 2010). It is however, also a policy which reflects the trend towards re-familialization of care (Lister 1994; Sipilä et al. 2010), and a policy which risks generating polarization between families, due to the low and not income-related amount which is more favorable for low-income and migrant families. 170
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Furthermore, while the cash-for-care schemes provide an alternative to the norm of full-time day-care services, this policy development may not always be in line with parental preferences. When asked about their preferences for various choices between policy options, parents in Denmark are least likely to support the combination of a choice between public day-care and cash-forcare support for privately organized day-care solutions (Finansministeriet 2005). Likewise, research among Norwegian mothers of young children in 1999 showed that although they did indeed want more time with their children, they would prefer longer paid parental leave and shorter working hours to a cash-for-care option (Hellevik and Koren 2000). Also, the relatively high take-up of the cash benefit in Norway is indeed not only the result of parents exercising their choice, but also due to the fact that public day-care has not met the demand and parents have, accordingly, used cash-for-care to bridge the gap between paid parental leave and day-care (Leira 2002; Lauritzen 2005).
Day-care as a right for and investment in the child As an interesting parallel development to the introduction of more choice for parents, more focus has at the same time been given to the advantages of these services for the individual child, as well as for society. Across the Nordic countries, the provision of day-care has as an important goal to create equal opportunities for children, regardless of their background. The Nordic daycare model is seen as a way to positively shape the life chances of the individual child by ensuring that the child experiences a good, socially developing and educational environment before starting school (Borchorst 2002; Sipilä 1997). The quality of the pre-schools and the outcomes for children have accordingly gained increased attention and the educational as well as the pedagogical role of the day-care services is recognized, thus relating directly to the field of educational and equality policies, not least after the OECD PISA studies indicated that children who participated in kindergarten or pre-school education achieve better results in school (Bennett 2008). The provision of daycare is also believed to be of vital importance for the integration of migrant children and serves the purpose of integrating disabled children (OECD 2001). The best interest of the child is in this way not only a question of whether care in the early years takes place in the family or an institution, but has also become a question of providing the right stimulus and challenges to ensure that the child has the best chances in life. In acknowledging the importance of day-care, all Nordic countries except for Iceland today provide a legal right to day-care for children, following the end of the parental leave, introduced in Denmark in 1998, Finland in 1990, Sweden in 1985, and Norway in 2009 (Eydal and Rostgaard, 2010 in print), and this is rarely found in other comparable countries (Lohman et al. 2009). With the introduction of the cash-for-care schemes, it has accordingly been debated whether such schemes work against the best interest of children, e.g. it has been argued that the importance of day-care is such that it is not acceptable to deny children this public good (Ellingsæter and Leira 2006; © 2011 Blackwell Publishing Ltd.
171
Social Policy & Administration, Vol. 45, No. 2, April 2011
Rantalaiho 2009; Repo 2010). Only in Norway and Sweden is it at present possible to combine part-time day-care with a partial payment of cash-forcare (see table 4).
New Agendas, New Directions? The Nordic Model under Scrutiny What then are the implications of these developments in Nordic childcare policies for the individual child, families and society, and also for the understanding of the Nordic welfare states as a particular cluster with regard to childcare policies? The differences between the countries have been increasing over the past few years, facilitating different childhood and parenthood models. The following overview (table 5) shows how the policies provide care support for care of children according to age. Overall, countries seem to share a concern for the investment in the child and the importance of high quality day-care for children in their early years in order to stimulate early learning, whereas gender equality and the support for the dual earner/dual carer model comes in various versions, and with different emphasis on either the dual earner or dual carer model. Different views on familialism also seem to accommodate different ideals about when it is best for the child to start in day-care. As shown in the article, these ideals are not only related to the national child policy model, but also seem to be related to social class and ethnicity. Denmark stands out as the country that provides most support in line with the dual earner model, even for parents of young children. The organization of the day-care services provision today is set up to match parents’ need arising from their participation in the labour market, providing full-time, full-year provision. Despite the introduction of the cash-for-care scheme in recent years, the Danish model is based on (women’s) early return to the labour market. Last but not least, what distinguishes the Danish policies from the other countries is the fact that from 2000 there has been no father’s quota. Except for two weeks of paternity leave immediately after the birth, fathers and mothers in Denmark can decide themselves which parent should use the parental leave – with the result that it is the mother who takes the majority of leave days. Gender concerns are in this way mainly related to the equal participation of men and women in the labour market and less concerned with equal division of care work in the family. This is still regarded as a private decision for the individual family to make – and may be less of a concern here than in the other Nordic countries, due to the short duration of the parental leave. At best, one may say that Denmark takes a gender neutral stand to the equal division of care work between men and women. Often, the mother is therefore the main carer in the first year of the child’s life in Denmark and thereafter childhood for most children comprises institutional or family daycare outside the family, where one in six children under the age of one are in day-care. In the case of Norway and in particular in Finland, the cash-for-care schemes contribute to long-term absence of mothers from the labour market. 172
© 2011 Blackwell Publishing Ltd.
© 2011 Blackwell Publishing Ltd.
Day care services (90) (Cash-for-care)
Day care services (97)
1–2 years
3–5 years
Source: NOSOSCO 2009. Note: * = in 2010, 42–52 weeks.
Paid parental leave (50–64 weeks) Day care services (17)
0–1 year
Denmark
Day care services (73)
Cash-for-care Day care services (42)
Paid parental leave (44 weeks) Day care services (1)
Finland
Day care services (95)
Care gap – private solutions Municipal schemes of cash-for-care Day care services (79)
Paid parental leave (39 weeks) Day care services (7)
Iceland
Day care services (96)
Cash-for-care Day care services (75)
Paid parental leave (*42–52 weeks) Day care services (4)
Norway
Sweden
Day care services (97)
Paid parental leave Municipal schemes of cash-for-care Day care services (70)
Paid parental leave (69 weeks) Day care services (–)
How are the policy goals translated into childcare policies in 2008 in the Nordic countries? (percentages)
Table 5
Social Policy & Administration, Vol. 45, No. 2, April 2011
173
Social Policy & Administration, Vol. 45, No. 2, April 2011
However, when the day-care volumes for the one to two year olds are examined, it is clear that the consequences of the cash-for-care scheme are dramatically different in these two countries. Only 40 per cent of the one to two year olds are in day-care in Finland, compared to 69 per cent in Norway (table 5). Where the popularity of the cash-for-care scheme in Norway has dropped alongside the increase in the day-care service provision, it is difficult to establish whether the high take-up of cash-for-care in Finland is due to new familialism or due to the relatively low provision of day-care services. The cash-for-care scheme has existed since 1985 in Finland and seems to be an important component in the childcare policies. As Kröger et al. (2003) note, the cash-for-care along with the right to day-care services represent in Finland a new form of universalism, as they in combination provide a right to day-care services and compensate parents who do not make use of day-care services. Nevertheless, the cash-for-care benefits do not reinforce public care service production either and seem to be a choice or option particularly for low income groups. Children of low income families are thus more likely to be cared for at home. The gendered consequence, in addition, is that women and especially low income women – perhaps with precarious labour market positions – take long absences from the labour market. Along with the lack of the father’s quota in parental leaves, gender equality seems less of a concern in Finland than in the other Nordic countries, both with regard to the sharing of paid work and caring. In Norway, the cash-for-care scheme has, as mentioned, signalled less emphasis on gender equality and it does not support the dual carer model. On the other hand, Norway has, like Sweden and Iceland, introduced the father’s quota. As discussed above, take-up ratios of fathers show very clearly that fathers do tend to use their individual rights – the use-it-or-lose-it idea actually works. Thus on this point, the development in Norway – along with Sweden and Iceland – has been more in line with the gender equality goals of the Nordic model than is the case in both Finland and Denmark, with regard to shared caring. In Sweden, the contrast between central and local policies is remarkable. Gender equality and the dual earner/dual carer model has traditionally stood strong in Sweden, resulting in the father’s quota, gender equality bonuses, generous provision of day-care services and a right to day-care, but policy pragmatism may – like in the other countries – have caught up with the ideals as municipalities are now able to offer local schemes of cash-for-care which may work against the equal sharing of work and care between men and women. In addition, the relatively long parental leave may disadvantage women in their return to the labour market. Finally, Iceland has been the newcomer in the Nordic country group of generous and universal childcare policies, but has within a couple of years taken the lead and inspired other countries, also outside the Nordic group, in the progressive approach to equal sharing of parental leave between men and women with the father’s quota of three months. Iceland has, like Denmark, promoted the early return to the labour market, through a relatively short parental leave, however without the similar coverage of day-care services (7 per cent) as in Denmark (17 per cent) for children below one year. Even 174
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 6 Care policies that support or work against the dual earner/dual carer model
Supportive Paid parental leave/entitlements connected to labour market participation Father’s quota Legal claim right to day care Not supportive Long paid parental leave (more then a year) Cash-for-care Care gaps between paid parental leave and day care
Denmark
Finland
Iceland
Norway
Sweden
¥
¥
¥
¥
¥
¥ ¥
¥
¥ ¥
¥ ¥ ¥
(¥)
¥
(¥) ¥
¥
(¥)
though Iceland provides for a relatively high proportion of the children aged one to two years (80 per cent) (table 5), a care gap exists in the period from the end of the parental leave to the age of one year. Summing up, table 6 provides an overview of the various policies that support or work against the dual earner/dual carer model and how the countries place themselves according to these policies. As the table shows, there is no clear direction in the policy line, and a country may have implemented policies that both work against and support the dual earner/dual carer model. All countries offer paid parental leave where entitlements are connected to labour market position, which has a positive effect on the return to the labour market of women. Iceland has installed a day-care guarantee which should smooth the transition from parental leave to labour life. However, all countries have also introduced the cash-for-care, although the schemes seem to be of less significance in Denmark and, so far, are not widespread in Iceland and Sweden. The countries generally also seem to be upholding their course as no reforms or substantial changes have been introduced in the wake of the financial crisis; however, Iceland has recently reduced the level of the benefit in the parental leave and, in Denmark, municipalities have for a number of years cut down the opening hours in day-care in order to cut costs.
Conclusion The Nordic countries have traditionally stood as the proponents of gender equality and the dual earner/dual carer model in their childcare policies. The Nordic countries are still ‘social service states’ with a widely available provision of day-care services to children from the age of one. Services remain © 2011 Blackwell Publishing Ltd.
175
Social Policy & Administration, Vol. 45, No. 2, April 2011
publicly subsidized and parental costs are income-related, also allowing for the participation of children from low-income groups. The extensive and generous support for paid parental leave together with generous and abundant public day-care for young children is likely to contribute to the continuation of the dual earner/dual carer model in the Nordic countries, although the countries have from the beginning had different emphases on dual earning vs. dual caring. New policy agendas of fatherhood have contributed to the traditional policy agendas of gender equality and the dual earner/dual carer model. Fatherhood has been politicized, resulting in the father’s quota in three of the five Nordic countries and this has strengthened the gender equality goal. The traditional focus on the involvement of the child has also been continued and the rights of the child have gained ground in recent years, as has emphasis on the investment in the child. The question is whether the introduction of the cash-for-care schemes has changed the direction of the Nordic childcare policies. The new cash-for-care schemes now introduced in all five countries all provide an alternative to day-care outside the family, but have come into life from different agendas of parental choice/municipal flexibility/fairness, and their importance varies greatly across the Nordic countries. The cash-for-care schemes seem to accentuate the inherent differences between the Nordic countries and in this way emphasize the variations in the promotion of gender equality and the dual earner/dual carer model across the Nordic countries. However, the introduction of the schemes also suggests that the childcare model which has been the trademark of the Nordic countries is ambitious, but perhaps to some degree also unrealistic to uphold, with parents requesting individualized care solutions and municipalities looking for less expensive approaches to providing day-care. Most importantly, however, the new schemes also seem to create a new social divide between children of lower income and/or migrant background who are more often cared for at home, and children of middle and upper income background who are more often in public day-care, a divide which is not easily combined with another Nordic ambition, to provide equal opportunities for all children.
Acknowledgements The article draws on work in the project Föräldraledighet, omsorgspolitik och jämställdhet i Norden, commissioned by the Nordic Council. We would like to thank Ann-Zophie Duvander, Berit Brandt, Johanni Lammi-Taskula and project leader Ingolfur V. Gíslason for stimulating discussions and help with literature. The article also draws on work in REASSESS (Reassessing the Nordic welfare model a Centre of Excellence in welfare research), in which both authors participate. Note 1. It is thus different from the so-called paternity leave, which is also given to the father. The paternity leave is aimed at relieving the mother, rather than giving the
176
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
father individual responsibility for the care of the child, and is usually taken within the first weeks of the birth of the child, while the mother is also on maternity leave. In the Nordic countries, the paternity leave has a duration of two to three weeks (see also table 2).
References Arter, D. (1999), Scandinavian politics today, Manchester and New York, NY: Manchester University Press. Anttonen, A. and Sipilä, J. (1996), European Social Care Services. Is it possible to identify models? Journal of European Social Policy, 6, 2: 87–100. Barne, Likestillings og Inkluderingsdepartementet (n.d.), Styrking av fedres rett til foreldrepenger, http://www.regjeringen.no/nb/dep/bld/pressesenter/pressemeldinger/ 2010/pm-1211.html?id=619700 (accessed 10 October 2010). Bennett, J. (2008), Early Childhood Services in the OECD Countries. Review of the Literature and Current Policy in the Early Childhood Field, Florence: Unicef Innocenti Research Centre. Borchorst, A. (2002), Danish Child Care Policy: Continuity Rather than Radical Change. In S. Michel and R. Mahon (eds), Child Care Policy at the Crossroads: Gender and Welfare State, New York, NY: Routledge, pp. 263–85. Borchorst, A. (2006), The Public-Private Split Rearticulated: Abolishment of the Danish Daddy Leave. In A. L. Ellingsæter and A. Leira (eds), Politicising parenthood in Scandinavia. Gender relations in welfare states, Bristol: Policy Press, pp. 101–20. Bradshaw, J., Kennedy, S., Kilkey, M., Hutton, S., Corden, A., Eardley, T., Holmes, H. and Neale, J. (1996), Policy and the Employment of Lone Parents in 20 Countries, York: Social Policy Research Unit, University of York. Brandth, B. and Kvande, E. (2003), Fleksible fedre. Maskulinitet, arbeid, velferdsstat, Oslo: Universitetsforlaget. Dagtilbudsloven nr. 501 06/06/2007. Duvander, A. Z., Ferrarini, T. and Thalberg, S. (2008), Towards a new family policy model? In J. Palme (ed.), Framtider, Stockholm: Institute for future studies, pp. 18–24. Cronholm, A. (2009), Sweden: individualisation or free choice in parental leave? In P. Moss, and S. Kamerman (eds), The politics of parental leave policies. Children, parenting and the labour market, Bristol: Policy Press, pp. 227–43. Ellingsæter, A. L. (2006), The Norwegian childcare regime and its paradoxes. In A. L. Ellingsæter and A. Leira (eds), Politicising parenthood in Scandinavia. Gender relations in welfare states, Bristol: Policy Press, pp. 121–44. Ellingsæter, A. L. (2009), Leave policy in the Nordic welfare states: a ‘recipe’ for high employment/high fertility? Community, Work & Family, 12, 1: 1–19. Ellingsæter, A. L. and Gulbrandsen, L. (2007), Closing the Childcare Gap: The Interaction of Childcare Supply and Mothers’ Agency in Norway, Journal of Social Policy, 36, 4: 649–69. Ellingsæter, A. L. and Leira, A. (eds) (2006), Politicising parenthood in Scandinavia. Gender relations in welfare states, Bristol: Policy Press. European Commission (2007), Europe’s demographic future: Facts and figures on challenges and opportunities, Luxembourg: Office for Official Publications of the European Communities. Eurostat (n.d.), Population, http://epp.eurostat.ec.europa.eu/portal/page/portal/ population/data/main_tables (accessed 10 October 2010). Eydal, G. B. and Gíslason, I. V. (eds) (2008), Equal Rights to Earn and Care – Parental Leave in Iceland, Reykjavík: Félagsvísindastofnun Háskóla Íslands. Eydal, G. B. and Kröger, T. (2009), Nordic family policies: constructing contexts for social work. In T. Kröger and H. Forsberg (eds), Social Work and Child Welfare Politics. Through Nordic lenses, Bristol: Policy Press, pp. 11–28. © 2011 Blackwell Publishing Ltd.
177
Social Policy & Administration, Vol. 45, No. 2, April 2011
Eydal, G. B. and Rostgaard, T. (2010), Barnomsorg och vardnadsbidrag. In I. V. Gíslason (ed.), Föräldraledighet, omsorgspolitik och jämställdhet i Norden, Köbehavn: Nordisk Ministerrad. Eydal, G. B. and Rostgaard, T. (2010, in print), På vej mod en nordisk børnepasningspolitik – hvad var de politiske processer og dagsordner? In I. V. Gíslason (ed.), Föräldraledighet, omsorgspolitik och jämställdhet i Norden, Köbehavn: Nordisk Ministerrad. Eydal, G. B. and Satka, M. (2006), Social Work and Nordic Welfare Policies for Children Present Challenges in the Light of the Past, European Journal of Social Work, 9: 305–22. Finansministeriet (2005), Det mener brugerne! Brugertilfredshed med dagpasnings-, skole- og ældreområdet, København: Finansministeriet. Gornick, J. C. and Meyers, M. K. (2003), Families that work. Policies for reconciling parenthood and employment, New York, NY: Russel Sage. Hellevik, T. and Koren, C. (2000), Parents Between Work and Care: Effects of the new cash benefit scheme, NOVA Skriftserie 7/00, Oslo: NOVA. Hiilamo, H. and Kangas, O. (2006), Trap for women of freedom to choose? Political frames in the making of child home care allowance in Finland and Sweden A:15/2006, Turku: University of Turku. Kangas, O. and Rostgaard, T. (2007), Preferences or institutions? Work–family life opportunities in seven European countries, Journal of European Social Policy, 17, 3: 240–56. Kela (n.d.), Statitics on number of receipients of Home Care allowances, Helsinki: KELA, http://www.kela.fi/it/kelasto/kelasto.nsf/alias/Pocket_09/$File/Pocket_09.pdf? OpenElement (accessed 10 May 2010). Kröger, T., Anttonen, A. and Sipilä, J. (2003), Social Care in Finland: Stronger and Weaker Forms of Universalism. In A. Anttonen, J. Baldock and J. Sipilä (eds), The Young, the Old and the State: Social Care Systems in Five Industrial Nations, Cheltenham: Edward Elgar, pp. 25–54. Lammi-Taskula, J. (2007), Parental leave for fathers? Gendered conceptions and practices in families with young children in Finland, Helsinki: STAKES. Lammi-Taskula, J. and Takala, P. (2009), Finland: negotiating tripartite compromises. In P. Moss and S. Kamerman (eds), The politics of parental leave policies. Children, parenting and the labour market, Bristol: Policy Press, pp. 87–102. Lappegård, T. (2008), Changing the gender balance in caring: Fatherhood and the Division of Parental Leave in Norway, Population Research and Policy Review, 27: 139–59. Lauritzen, T. (2005), ‘Kontantstøtteelskare’ og ‘kontantstøttehatere’. In B. Brandth, B. Bungum and E. Kvande (eds), Valgfriehtens tid. Omsorgspolitikk for barn møter det fleksible arbeidslivet, Oslo: Gyldendal, pp. 111–23. Leira, A. (2002), Working Parents and the Welfare State, Cambridge: Cambridge University Press. Leira, A. (2006), Parenthood change and policy reform in Scandinavia, 1970s–2000s. In A. L. Ellingsæter and A. Leira (eds), Politicising parenthood in Scandinavia. Gender relations in welfare states, Bristol: Policy Press, pp. 27–52. Lewis, J. (2001), The Decline of the Male Breadwinner Model: Implications for Work and Care, Social Politics, 8, 2: 152–69. Lister, R. (1994), She has other duties: women citizenship and social security. In S. Baldwin and J. Falkingham (eds), Social Security and Social Change, Challenges to the Beveridge Model, Hemel Hempstead: Harvester Wheatsheaf, pp. 31–44. Lister, R. (2003), Investing in the citizen-workers of the future: transformations in citizenship and the state under New Labour, Social Policy & Administration, 37, 5: 427–43.
178
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Lohman, H., Rostgaard, T. and Spiess, C. K. (2009), Towards a framework for assisting family policy in the EU. OECD working papers series, Paris: OECD. Moss, P. and Kamerman, S. (eds) (2009), The politics of parental leave policies. Children, parenting and the labour market, Bristol: Policy Press. NOSOSCO (2009), Social Protection in the Nordic Countries 2007–2008. Scope, expenditure and financing, Copenhagen: Nordic Social-Statistical Committee. Nyberg, A. (2010), Cash-for-childcare schemes in Sweden: History, political contradictions and recent development. In J. Sipilä, K. Repo and T. Rissanen (2010), Cash-for-Childcare: The Consequences for Caring Mothers, Cheltenham and Northampton, MA: Edward Elgar Publishing, pp. 93–139. OECD (2001), Starting Strong: Early Childhood Education and Care, Paris: OECD. Olsen, B. M. (2005), Mænd, orlov og arbejdspladskultur. Fire danske virksomheter, København: Socialforskningsinstituttet. Rantalaiho, M. (2009), Kvoter, valgfrihet, fleksibilitet. Indre spenninger I den nordiske familliepolitikken, Köbenhavn: NIKK. Rannsóknastofnun um barna- og fjölskylduvernd (2010), Umönnun ungra barna í Reykjavík, Reykjavík: Rannsóknarstofnun um barna- og fjölskylduvernd. Repo, K. (2010), Families, work and home care. Assessing the Finnish child home care allowance, Barn, 28, 1: 43–63. Rønsen, M. and Kitterød, R. H. (2010), Cash-for-care in Norway: take-up, impacts and consequences for mothers. In J. Sipilä, K. Repo and T. Rissanen (2010), Cash-for-Childcare: The Consequences for Caring Mothers, Cheltenham and Northampton, MA: Edward Elgar Publishing, pp. 129–55. Rønsen, M. and Skrede, K. (2010), Can public policies sustain fertility in the Nordic countries? Lessons from the past and questions for the future, Demographic research, 22, 13: 321–46. Rostgaard, T. (2002), Setting Time Aside for the Father – Father’s Leave in Scandinavia, Community, Work and Family, 5: 343–64. Salmi, M. (2006), Parental choice and the passion for equality in Finland. In A. L. Ellingsæter and A. Leira (eds), Politicising parenthood in Scandinavia. Gender relations in welfare states, Bristol: Policy Press, pp. 145–70. Sipilä, J. (ed.), (1997), Social Care Services – The Key to the Scandinavian Welfare Model, Aldershot: Avebury. Sipilä, J., Repo, K. and Rissanen, T. (eds) (2010), Cash-for-Childcare: The Consequences for Caring Mothers, Cheltenham and Northampton, MA: Edward Elgar Publishing. Skrede, K. (2004), Familiepolitikkens grense ved ‘likestilling’ light? In A. L. Ellingsæter and A. Leira (eds), Velferdsstaten og familien. Utfordringer og dilemmaer, Oslo: Gyldendal Akademisk, pp. 160–200. Statistics Norway (2010), Statistisk årbok 2009, Oslo: Statistics Norway. Stefansen, K. and Farstad, G. (2010), Classed Parental Practices in a Modern Welfare State: Caring for the under threes in Norway, Critical Social Policy, 30, 1: 120–41.
© 2011 Blackwell Publishing Ltd.
179
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00763.x Vol. 45, No. 2, April 2011, pp. 180–193
Swedish Family Policy – Continuity and Change in the Nordic Welfare State Model spol_763
180..193
Kimberly Earles Abstract The main question addressed in this regional issue is whether or not the Nordic welfare states can still be considered a distinct welfare regime cluster given recent changes, such as the introduction of more private elements into the welfare state. The Nordic welfare states are often described as emphasizing full employment, economic and gender equality, and universal access to cradle-to-grave welfare state benefits and services. In the case of Sweden, often pointed to as the model of a social democratic welfare state, such elements remain intact in most aspects of the welfare state, even given the challenges presented by the global neo-liberal economic paradigm since the 1970s. One way to determine whether or not the Nordic welfare states remain a distinct cluster is to provide an in-depth examination of various welfare state policies in each Nordic country. To contribute to this analysis, an investigation of family policy in the Swedish context will be provided. Even given recent challenges, such as the introduction of private for-profit childcare providers and a home care allowance, I argue that Swedish family policy has remained largely social democratic in its underlying goals, and thus acts to support the case for a distinct Nordic welfare regime cluster.
Keywords Sweden; Welfare states; Family policy Introduction The Swedish welfare state, as one of the Nordic welfare states, is often viewed as the model of a social democratic welfare state in the postwar period. The Nordic welfare states have traditionally been linked together as a distinct social democratic welfare regime cluster, which emphasizes full employment, economic and gender equality, and universal access to cradle-to-grave welfare state benefits and services (Esping-Andersen 1990). In recent years, many scholars have begun to question whether or not the Nordic welfare states remain distinct in the wake of the shift from Keynesianism to neo-liberalism Address for correspondence: Kimberly Earles, University of Guelph, 50 Stone Road East, Guelph, Ontario, N1G 2W1, Canada. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
as the dominant global economic paradigm (Ervik and Kuhnle 1996; Kautto et al. 2001; Greve 2004). When considering Swedish family policy, despite the introduction of private for-profit childcare in the early-1990s, as well as more recent reforms such as the 2008 home care allowance for parents who stay at home with young children, the underlying goals of economic and, particularly, gender equality, as well as full employment remain crucial. As such, I argue that Swedish family policy, while experiencing some challenges, has remained largely social democratic in its underlying goals, and acts to support the case for the continued existence of a distinct Nordic welfare regime cluster. During the 1960s–70s, declining fertility rates as well as women’s increasing labour force participation rates, were both important factors in the development of Sweden’s public childcare system and the introduction of parental leave. As women’s labour force participation increased, policymakers grew concerned about the corresponding decrease in fertility levels. These concerns were met with the development of policies that encouraged women to be both mothers and workers, and since the 1970s, Sweden has been at the forefront in developing programmes and services that help both women and men to balance family and paid employment. Family policy in Sweden has largely had a gender equality focus, particularly around women’s labour force participation and men’s role in childrearing. The childcare and parental leave systems in particular seek to encourage women’s labour force participation while, at the same time, aiming for a more equal distribution of childrearing responsibilities between mothers and fathers. As such, these programmes address issues of gender equality, not only in redefining women’s roles, but also men’s roles.
Women’s Employment The history of social democracy in Sweden is rooted in a model of full employment, which has supported the development of a comprehensive social democratic welfare state. In postwar Sweden, open unemployment remained under three per cent (Esping-Andersen 1990), exceeding this benchmark only three times from 1950–91 (Olsen 2002). However, the model of full employment in Sweden has differed from other nations since the 1970s in that full employment applies to both women and men, as women’s labour force participation is just as important and just as needed in the Swedish system. Until the 1960s, women were still largely viewed as housewives and mothers, and treated as such by the state (Jenson and Mahon 1993); however, beginning in the 1960s, Swedish women were encouraged into the labour force in order to fill Sweden’s continuing postwar labour shortage. This marked the end of the male breadwinner model and the beginning of the dual earner model in Sweden. During the 1960s, women’s labour force participation rose to more than 50 per cent, increasing further to 63 per cent in 1970, and 81 per cent in 1980, where it has remained fairly constant ever since. The increase in women’s labour force participation also included married women and women with children aged under seven, who saw their labour force participation rates rise from below 50 per cent in the 1970s to over 85 per cent in 1980 (Gustafsson and Jacobsson 1985; Hoem 1995). By the early-1990s, women constituted 48 per cent of the Swedish © 2011 Blackwell Publishing Ltd.
181
Social Policy & Administration, Vol. 45, No. 2, April 2011
labour force (Olsen 2002), the highest percentage anywhere in the world. In Sweden, as in the other Nordic countries, the welfare state became the most important female labour market, providing women with good pay, job security and flexibility, including quality part-time jobs (Esping-Andersen 2002).1 As a result, women currently make up 74 per cent of public sector workers (Statistics Sweden 2010). As such, the welfare state not only provides family-friendly programmes, such as childcare and parental leave, but it also acts as an important source of employment for women. Taken together, the data demonstrate that women, even married women and women with young children, began entering the labour market in greater numbers prior to the development of childcare and parental leave policies in Sweden. However, it was then these women who were engaged in paid labour, as well as the trade unions that represented them, who pushed for the introduction and development of such family policies, which would help to strike a better work–life balance.
The Development of Swedish Family Policy Childcare In the 1960s, political parties were divided on the issue of childcare, with the centre-right advocating a care allowance and the Social Democrats advocating publicly provided childcare (Bergqvist et al. 1999). From the mid-1960s, the blue-collar Swedish Trade Union Confederation (LO) became the leading advocate for the expansion of public childcare (Daune-Richard and Mahon 2001), as women workers continued to demand access to childcare services. Thanks to LO’s strong ties with the Social Democratic Party (SAP), childcare was given a high priority and the number of publicly provided municipal childcare spaces increased rapidly from the late 1960s onward, from under 12,000 in 1965 to over 136,000 by 1980 to nearly 730,000 by 2002 (Curtin 1999; Daune-Richard and Mahon 2001; Mahon 1999; Swedish National Agency for Education 2003). This was partly the result of lobbying by key women in the trade union and political spheres, as well as the state’s own agenda to foster an environment where women could work outside the home while still having children, thus achieving full employment and relatively high fertility rates simultaneously. The Swedish social democratic welfare state, like the other Nordic welfare states, recognized access to reliable childcare as a basic requirement of a society in which the majority of both parents work outside the home. The state was thus prepared to provide working mothers and fathers with the services needed to allow them both to work in the labour market. This constitutes a crucial gender equality element that distinguishes the Nordic social democratic welfare states from other European welfare states. While women’s employment has become increasingly important in many European countries since the 1970s, the liberal and conservative-corporatist welfare states have not been willing to provide universal access to public childcare or generous parental leave programmes to encourage the growth of women’s employment. Instead, families are generally left to fend for themselves when it comes to 182
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
childcare, often relying on informal networks or on the market. While many European countries could today be described as ‘dual earner’ societies, or at least ‘one and a half earner’ societies (e.g. the Netherlands), it is the Nordic welfare states that first took active measures to achieve a dual earner society, and that have taken additional steps towards achieving an earner–carer society, where both mothers and fathers are expected to engage in paid labour and in unpaid domestic caring responsibilities. In 1963, only three per cent of all pre-school aged children in Sweden were enrolled in public childcare, while over 35 per cent of all mothers of pre-school children were in the labour market (Nyberg 2004), which demonstrates the absolute need for expanded childcare at this time. Since the 1960s the state has exercised a great deal of power in its quest for a national model of childcare with minimum standards (Alvestad and Pramling Samuelsson 1999). For example, in 1966 state grants to full-time childcare centres doubled in order to encourage expansion and improve the quality of care, and in 1968 the government appointed the National Commission on Childcare, which laid the foundation for the pre-school model that exists in Sweden today (Nyberg 2004; Swedish Institute 2004a). After more than two decades of long queues for childcare due to demand outpacing availability, the Social Democratic government introduced the Act on Child Care in 1995 under which municipalities became obliged to provide childcare without ‘undue delay’ (Bergqvist and Nyberg 2002; Swedish Institute 2004a), which generally translates into within three months of asking for a space. The result has been not only a high quality, accessible and affordable public childcare system, but also high rates of women’s labour force participation, and a higher fertility rate than the European average. As access to childcare increased, the fertility rate in Sweden also increased from just over 1.5 in the late 1970s and early 1980s, to 2.14 in 1990 (Hoem and Hoem 1996). However, Sweden faced another decline in fertility rates during the 1990s, with the economic crisis, which brought with it increasing unemployment and uncertainty. Today, Sweden’s fertility rate stands at 1.91, lower than the necessary replacement rate of 2, but higher than in most European countries (Statistics Sweden 2009). Research in industrialized countries has shown that such high fertility rates are linked with high female employment such as in Sweden, which goes against previous thinking that women who did not participate in the paid labour force had more children. Statistics and research suggest that today women have more children when they know they can balance being a mother with having paid employment (Alfredsson 2005). In Sweden, this is clearly linked to both the availability of childcare and the generosity of the parental leave system.
Parental leave Parental leave was introduced in Sweden in 1974, replacing maternity leave, which had focused solely on mothers. Women’s increasing level of employment, particularly married women and women with small children, was the impetus for the creation of the parental leave system. As mentioned above, at the time, there were differing opinions on how to best deal with the increasing © 2011 Blackwell Publishing Ltd.
183
Social Policy & Administration, Vol. 45, No. 2, April 2011
proportion of women and mothers in employment. The SAP was somewhat divided on the issue, but the unions advocated a comprehensive childcare and parental leave solution. The work of the LO’s Family Council on this issue and its influence within the SAP led to the development of childcare and parental leave as the favoured option (Mahon 1999; Lindberg 2005). In the end, there has been a great deal of agreement among the political parties around the necessity of parental leave, but there remain differences of opinion when it comes to the details. When parental leave was first introduced it comprised six months of leave per child, which parents could divide in any way, at a 90 per cent income replacement rate (Nyberg 2004). That this leave was now open to both mothers and fathers was at the time unique and very forward thinking (Wennemo 2005; Ferrarini 2005). The fact that the system was adopted with a great deal of consensus across party lines indicates the strength of the belief in the underlying objective to move Sweden from a male breadwinner to a dual earner society (Bygren and Duvander 2005), and even to an earner–carer model. The parental leave system was viewed as a complement to the development and expansion of the childcare system, as both were aimed at helping mothers and fathers better balance work and family life (Cohen et al. 2004). And further amendments to the parental insurance system were made to accommodate this goal. For example, in 1978 parental leave was extended to nine months, with the last three months paid at a flat rate; alternatively, the additional three months could be used to reduce the workday for one parent (Daune-Richard and Mahon 2001). The main objectives behind the parental insurance system in Sweden are the well-being of the child, women’s economic independence, and the involvement of fathers in family and home life (Drew 2005). The last two objectives are related directly to gender equality, with an emphasis on an earner–carer model, which includes a more equal division of household responsibilities, particularly around childrearing. For even in the late 1960s and early 1970s, the Family Policy Committee that proposed the parental leave system had discussed adopting an additional month reserved solely for fathers. In the end, the majority of the members of the committee voted against such an idea (Lindberg 2005), but the discussion had begun. This posed a serious challenge, since in 1974 only three per cent of those accessing parental leave were fathers, and fathers took less than one per cent of the total days available (Nyberg 2004). Over time this number has increased, but not as quickly as the architects of the system had hoped (Lindberg 2005). For example, by 1992, 48.3 per cent of fathers took an average of 63 days leave (Daune-Richard and Mahon 2001). While this was an improvement, further reforms were taken in the 1990s and 2000s, which will be discussed in detail below, in an attempt to encourage mothers and fathers to share parental leave more equally.
Recent Family Policy Reforms While the 1990s were, for the most part, a decade marked by welfare state contraction in Sweden due to a severe economic crisis, the childcare system actually expanded during this time, and is now more comprehensive than 184
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
ever. In addition, while benefit levels were decreased somewhat in the 1990s, the length of leave within the parental insurance system was extended, and increased individualization within the system is the focus of current debate in Sweden. However, there have also been elements introduced into Swedish family policy which challenge the social democratic goals of the system, such as private for-profit childcare and a home care allowance. Yet, the overall impact of these reforms has not been great as private for-profit childcare accounts for just a fraction of all available childcare, and only about one-third of all Swedish municipalities have adopted the home care allowance. As will be demonstrated below, the majority of the family policy reforms introduced in the past two decades are related to furthering the underlying social democratic goals of gender equality and full employment.
Childcare Beginning in the 1980s, the Swedish state had to deal with initiatives for private for-profit childcare, mainly coming from the Swedish Employer’s Association’s (SAF) and the Swedish Federation of Industries’ proposal for a private day-care company. In response to this proposal, Pysslingen, an Electrolux subsidiary company was created to open up childcare to the for-profit sector (Daune-Richard and Mahon 2001; Strandbrink and Pestoff 2006). The challenge was launched by Sollentuna, a suburb of Stockholm, which attempted to contract out the provision of childcare to Pysslingen. In response, the Social Democratic government passed the Lex Pysslingen legislation in 1982, which banned all private for-profit childcare, and regulated the growth of other non-public forms such as parental cooperatives (DauneRichard and Mahon 2001). The SAP’s concern regarding private for-profit childcare was that it would turn municipal institutions into ghettos for children of the low-paid or children with special needs, thus breaking with the universal principle of the Swedish welfare state. However, when the SAP was defeated in the 1991 election, the ruling centre-right coalition abolished Lex Pysslingen, and the Social Democrats did not re-instate it when they returned to office in 1994, as their primary concern remained Sweden’s economic crisis. Because several suburbs of Stockholm governed by centre-right parties continued to challenge the law, Pysslingen was ready to expand once it was legally able to do so in 1992 (Mahon 2005). Although private for-profit childcare centres are now allowed in Sweden, they must follow the same fee rules and curriculum standards as public centres in order to receive municipal subsidies (Daune-Richard and Mahon 2001). Since the early 1990s, the popularity of private childcare has increased only somewhat in Sweden. In 1994, there were 27 Pysslingen childcare centres in and around Stockholm, by 1999 there were 40, and in 2009 there were 77 (Strandbrink and Pestoff 2006; Mahon 2005; Pysslingen n.d.). Yet public provision has remained the norm in the rest of Sweden. In 2008, while 34 per cent of all children in Stockholm county were enrolled in some form of private pre-school, including parental co-operatives and private for-profit childcare, the percentage was as low as five per cent in other counties, making the © 2011 Blackwell Publishing Ltd.
185
Social Policy & Administration, Vol. 45, No. 2, April 2011
national average 18 per cent (Statistics Sweden 2010). The numbers continue to indicate that private for-profit childcare has not really taken off in Sweden. There are many theories as to why, from citizens’ existing satisfaction with the public childcare system to the fact that it is difficult to make a profit when following state regulations in order to receive funding. In addition, the Swedish Schools Inspectorate has the ability to issue injunctions and withdraw operating permits if private childcare providers do not meet national standards (Ministry of Education and Research 2010). The childcare system that exists in Sweden today also contains many recent developments that reinforce the social democratic nature of Swedish family policy. First, as mentioned above, every parent in Sweden is now guaranteed a childcare space without undue delay, making the system extremely accessible. Second, in 2002 the state introduced a system of maximum fees that the municipalities could charge for childcare, which made the system even more affordable, as parents are now charged no more than one to three per cent of their income in childcare fees, depending on the number of children in the family who are enrolled in childcare (Swedish National Agency for Education 2007). Parental fees currently make up approximately ten per cent of the gross costs of childcare, with the state and the municipalities covering the remainder (Swedish National Agency for Education 2009a). Third, the system is now more inclusive than ever, incorporating more categories of children, such as children of the unemployed (since 2001) and children of parents who are on parental leave looking after a sibling (since 2002). The childcare system has also taken on an increasingly educational focus since 1996 when the responsibility for childcare was moved from the Ministry of Health and Social Affairs to the Ministry of Education and Science. As a result, in 1998 a new pre-school class for six-year-olds was introduced, which is technically voluntary but 95 per cent of all six-year-olds who have not yet started compulsory school attend (Swedish National Agency for Education 2009b). And, in 2003 all children aged four to five became entitled to 525 hours of free attendance in childcare per year, something which has recently been expanded to include three-year-olds as well. These free pre-school hours are a reflection of the state’s attempt to encourage more children to attend pre-school to start to engage in the process of lifelong learning from an early age. These changes have resulted in a more universal childcare system, and one that the OECD (2001) supports as a model for other countries. Today in Sweden, more than 85 per cent of children aged one to five attend pre-school, and 74 per cent of children aged six to nine attend leisure-time centres, which provide care for school-aged children before and after school (Government Offices of Sweden n.d; Ministry of Education and Research 2008). This illustrates how a public childcare system marked by low fees and a high level of legitimacy, quality and accessibility is now an integral feature of Sweden’s welfare state (Swedish Institute 2004a). The system also offers parents, and particularly mothers, more opportunities to engage in paid labour, whilst also having children, the result being high labour market participation rates for women and relatively high fertility levels. 186
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Care allowance An ongoing challenge to the economic and gender equality aspects of the Swedish social democratic welfare state is reflected in recent attempts to introduce a care allowance in Sweden. The Christian Democratic Party (KD) was one of the four parties that formed the governing centre-right coalition in Sweden from 1991 to 1994. One of the most important beliefs for the KD is that ‘The family is the most important social unit’, as it promotes ‘concepts based on the Christian ethic’ (Christian Democratic Party n.d.). The party is in favour of more ‘choice’ for parents to decide how to raise their children, which is a principal argument in favour of care allowances (Bergqvist et al. 1999). As a result, the KD was in support of a care allowance, which would make it possible for parents of young children to stay at home longer than the time allotted by the parental leave system, or to employ a domestic nanny rather than to place their child in public childcare. Once elected in 1991, the centre-right coalition was divided on the issue of a care allowance. Through several rounds of negotiations with its governing partners, the Christian Democrats were eventually successful in implementing a care allowance in July 1994 (Bergqvist et al. 1999). However, when the Social Democrats returned to power just a few months later they immediately abolished the allowance (Bergman 2004), which meant that it existed only for a few months in 1994. The Social Democrats disagreed with the allowance because of its potential to encourage women to stay at home to take care of their children, thus reinforcing stereotypical notions about gender and care. In addition, the allowance allows parents to hire working class women to take care of their children as nannies, thus reinforcing gender stereotypes, and exacerbating class tensions among Swedish women. While the care allowance existed in Sweden for only a short time in 1994, it is an issue to which the Christian Democrats have remained committed. When the centre-right coalition returned to power in 2006 the care allowance re-emerged on the political agenda. In 2008 the government, again, introduced such an allowance, granting municipalities the ability to provide an untaxed benefit of SEK 3,000 per month for a ‘parent’2 to take leave with a child between the ages of one and three rather than to utilize the public childcare system. There is no previous work requirement for the allowance, although it is to be used after parental leave has been exhausted, and requires that the other adult in the house be engaged in either employment or education. The experience of other countries with similar allowances, such as Norway, indicates that it: (1) is mainly used by mothers, not fathers; and (2) can lead to further labour market marginalization of those lower educated women who tend to take advantage of the leave (Ferrarini and Duvander 2010; Westlund 2007), thus demonstrating both the gender and class implications. The introduction of such a leave appears to go against the dominant gender paradigm in Sweden, which encourages full employment throughout one’s adult life, with the exception of periods of parental leave, for both women and men. By encouraging (particularly) women to detach from the labour force for several years, the home care allowance threatens women’s economic inde© 2011 Blackwell Publishing Ltd.
187
Social Policy & Administration, Vol. 45, No. 2, April 2011
pendence while, at the same time, reinforces the notion that child-rearing is predominantly a woman’s responsibility. In fact, the Left Party argues that the home care allowance is a ‘direct anti-feminist reform that encourages women to stay at home and be supported by their husbands’ (The Local 2009). Thus, it does nothing to encourage men’s role in child-rearing, which, again, goes against the direction of previous family policy initiatives (Westlund 2007). In addition, the allowance has not proven to be popular with Swedish-born parents, but is becoming more popular with immigrant women who face greater labour market marginalization. This was the fear of the Swedish Teachers Union, which issued early warnings about the allowance’s potential for segregation and isolation among immigrant communities in Sweden. The fact that the allowance has only been adopted by 100 of Sweden’s 290 municipalities (Landes 2010a) supports this fear of segregation, while, at the same time, demonstrates the limited impact of the reform to date. As such, the home care allowance could prove to be detrimental to the gender equality project in Sweden, which has, for more than 40 years, tried to move away from traditional gender roles and norms. It may also have broader implications for gender equality among immigrant communities, as well as employment levels among immigrant women.
Parental leave Over time, the Swedish state has sought to increase men’s role in childrearing by encouraging a more gender equal sharing of parental leave between mothers and fathers. In 1995, the first ‘father’s month’ was introduced, which meant that 30 days of parental leave was now reserved for each parent (Nyberg 2004). In 2002, this was extended to two months per parent (Berg 2005). As a result, the system today offers parents 13 months of leave per child at an 80 per cent income replacement rate, with an additional three months available at a flat rate. The total of 16 months currently available to parents includes two months reserved for mothers and two months reserved for fathers, leaving 12 months to be divided as the parents wish. The parental leave system is very flexible, allowing parents to use parental leave until the child’s eighth birthday in a variety of ways, from full-time leave to part-time work and part-time leave (Swedish Institute 2004b). Due to the generosity and flexibility of the parental leave system, fathers have been increasingly taking up more leave; by 2008 men were using an all-time high of 21 per cent of parental leave, an increase of 11 per cent in just 11 years (Swedish Social Insurance Agency 2010). While this is not anywhere near equal, there has certainly been movement in a positive direction since the introduction of both reserved months. In April 2004, the government appointed the Thorwaldsson Commission to examine the parental leave system, to determine whether the system works in the best way for children and contributes to greater equality between women and men. On 15 September 2005 the Thorwaldsson Commission presented its proposals, which included a recommendation to increase paid parental leave to 15 months which would be divided into five months for the mother, five months for the father, and five months for the parents to split as they wished; 188
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
in addition, all mothers would have the right to 30 days leave before the birth, and the parents could take 30 days leave together after the birth (Berg 2005). These recommendations are very controversial, supported by some, but lacking support in the general population. One of the major issues behind the current debate over parental leave is the fact that the system has actually had some unintended negative side effects for women’s labour force participation. While parental leave has been successful in ensuring women’s labour force attachment before and after having a child, because women take the majority of parental leave, employers seeking to hire new employees or to promote employees see women as more of a risk, leading to statistical discrimination. The Ministry of Finance in Sweden found that employers exclude women from occupations where absence is costly, resulting in a gender-segregated labour market (Nyberg 2004). This is cause for concern, as gender equality and full employment are two underlying goals of the Swedish welfare state. Anna Thoursie, an economist at the LO and Thorwaldsson Commission member, suggests that if both mothers and fathers had a fixed number of months under the parental leave system, then employers would be forced to adjust their expectations (Thoursie 2005). The LO, the Left Party, the Green Party, as well as the Social Democratic Women’s Federation and the Social Democratic Youth League all believe that parental leave should be divided equally between parents in order to remedy this situation (Berg 2005), as does the ombudsman for gender equality, Anne-Marie Bergström (The Local 2008). The SAP, however, was more hesitant about the commission’s recommendations, as it feared the Swedish people were not behind such a reform (Berg 2005). In the case of parental leave, popular opinion has constantly lagged behind more progressive politics. For example, before the first father’s month was introduced, surveys showed a majority were not in favour; similar results were found before the second father’s month was introduced, although a majority was now happy with the one month that had already been introduced (Ferrarini 2005). As such, politicians have had to walk a fine line when it comes to how far to push their citizens in the direction of gender equality, although the citizens do tend to support such changes after they are introduced. A recent report by the Swedish Confederation for Professional Employees indicates that, at the current pace of change, it would take another 51 years before Swedish parents would achieve an equal sharing of parental leave (Vinthagen Simpson 2010). While the SAP did nothing to implement the proposals of the Thorwaldsson Commission prior to losing the 2006 election, the party has, more recently, spoken out in favour of increasing individualization within the parental leave system. For example, in 2008 SAP leader Mona Sahlin stated: The question is not whether we are going to share parental leave, but how . . . I don’t think that we can get at the statistical discrimination currently experienced by all women of a fertile age in Sweden unless dads also have their foreheads tattooed so that employers can see and realize that this man could also be away, this man could also become a father. (O’Mahony 2008) © 2011 Blackwell Publishing Ltd.
189
Social Policy & Administration, Vol. 45, No. 2, April 2011
The solution proposed by the Social Democrats is to increase the individual quotas within the system in the short-term, leading to an outright individualization in the long-term (Vinthagen Simpson 2008). Upon winning the 2006 election, the current centre-right government proposed and then introduced an equality bonus for couples that share parental leave equally, which came into effect in July 2008. The bonus encourages the lower income earner (usually the woman) to return to work earlier and the higher income earner (usually the man) to take their share of parental leave (Ministry of Finance 2007; The Local 2007). For low and medium income families, the bonus meant that it became more financially advantageous to share leave more equally than not (Ferrarini and Duvander 2010). The gender equality bonus in parental leave continues the tradition of encouraging female employment and fathers’ involvement in child-rearing, while also fitting neatly with the neo-liberal ‘freedom of choice’ concept of the centre-right parties. As such, the policy has few detractors. However, as Ferrarini and Duvander (2010) point out, the bonus is technically complicated and it can take a year before couples receive payment. As such, to date the reform has not had the desired result of a more equal sharing of parental leave, and the Moderate Party has promised to revise the bonus, as part of its 2010 election platform (Landes 2010b).
Conclusions While Swedish family policy has been challenged by neo-liberalism in the past two decades, and has even incorporated certain neo-liberal elements during this time, it largely remains social democratic, in that the underlying goals remain centred on equality and full employment. In addition to some neoliberal reforms, such as the introduction of private for-profit childcare and a home care allowance, there have been many changes in the past two decades that are more in line with social democratic values. Such changes include the introduction of the first and second ‘fathers’ months’, as well as the recent introduction of the gender equality bonus in the parental leave system. During this time, the childcare system has also expanded greatly, to include all children in need of a space, even those of parents who are unemployed or on parental leave with another child. The childcare system has also taken on an increasingly educational focus during this time, and remains a model for other countries. It is clear that the neo-liberal elements that have been introduced into Swedish family policy in the past two decades are minor and their impact has been limited. The fact that private for-profit childcare has not expanded much beyond the wealthy suburbs of Stockholm, and that only about one-third of Sweden’s municipalities have even implemented the home care allowance demonstrate the limited impact that these reforms have had on Swedish society and the social democratic welfare state. When comparing such reforms to those more in line with the equality goals of social democracy, such as the gender equality bonus introduced into the parental leave system, as well as the expansion of childcare to increase the system’s accessibility and affordability, it becomes clear that the underlying social democratic goals of Swedish family 190
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
policy remain intact and exemplify the continued relevance of the Nordic welfare regime cluster in terms of its focus on full employment, equality and universalism in welfare state services.
Notes 1. The vast majority of women who work part-time in Sweden work what is called ‘long part-time’, or 20–34 hours per week, and only a small percentage work ‘short part-time’, or 1–19 hours per week (Statistics Sweden 2010). 2. The home care allowance is framed in gender-neutral terms as though men and women will be equally enticed to use the allowance.
References Alfredsson, K. (2005), First, a few facts, Swedish Institute, http://www. sweden.se/templates/cs/Print_CommonPage_12527.aspx (accessed 13 May 2006). Alvestad, M. and Pramling Samuelsson, I. (1999), A Comparison of the National Preschool Curricula in Norway and Sweden, Early Childhood Research and Practice, 1, 2 [electronic version], http://ecrp.uiuc.edu/v1n2/alvestad.html (accessed 17 March 2007). Berg, A. (2005), Commission proposes controversial parental leave changes, National Institute for Working Life, The European Foundation for the Improvement of Living and Working Conditions, http://www.eurofound.europa.eu/eiro/2005/10/feature/ se0510103f.htm (accessed 9 November 2005). Bergman, S. (2004), Collective Organizing and Claim Making on Child Care in Norden: Blurring the Boundaries between the Inside and the Outside, Social Politics: International Studies in Gender, State and Society, 11, 2: 217–46. Bergqvist, C., Kuusipalo, J. and Styrkarsdóttir, A. (1999), The Debate on Childcare Policies. In C. Bergqvist et al. (eds), Equal Democracies? Gender and Politics in the Nordic Countries, Oslo: Scandinavian University Press, pp. 137–57. Bergqvist, C. and Nyberg, A. (2002), Welfare State Restructuring and Child Care in Sweden, In S. Michel and R. Mahon (eds), Child Care Policy at the Crossroads: Gender and Welfare State Restructuring, New York, NY: Routledge, pp. 287–307. Bygren, M. and Duvander, A.-Z. (2005), Who takes care of the children? Couples’ workplace situation and their division of parental leave, Working Papers in Social Insurance: 2, http://www.forsakringskassan.se/sprak/eng/publications/dokument/wp0502.pdf (accessed 13 July 2006). Christian Democratic Party (n.d.), Our core beliefs, http://www.kristdemokraterna.se/ otherlanguages/English.aspx (accessed on 17 December 2008). Cohen, B., Moss, P., Petrie, P. and Wallace, J. (2004), A New Deal for Children? Re-forming education and care in England, Scotland and Sweden, Bristol: Policy Press. Curtin, J. (1999), Women and Trade Unions: A Comparative Perspective, Aldershot: Ashgate Publishing Limited. Daune-Richard, A.-M. and Mahon, R. (2001), Sweden: Model in Crisis. In J. Jenson and M. Sineau (eds), Who Cares? Women’s Work, Childcare, and Welfare State Redesign, Toronto: University of Toronto Press, pp. 146–76. Drew, E. (2005), Parental Leave in Council of Europe member States, Directorate General of Human Rights, Strasbourg, http://www.coe.int/equality (accessed 9 November 2005). Ervik, R. and Kuhnle, S. (1996), The Nordic Welfare Model and the European Union. In B. Greve (ed.), Comparative Welfare Systems: The Scandinavian Model in a Period of Change, London: Macmillan Press Ltd, pp. 87–107. © 2011 Blackwell Publishing Ltd.
191
Social Policy & Administration, Vol. 45, No. 2, April 2011
Esping-Andersen, G. (1990), The Three Worlds of Welfare Capitalism, Cambridge: Polity Press and Princeton, NJ: Princeton University Press. Esping-Andersen, G. (2002), A New Gender Contract. In G. Esping-Andersen (ed.), Why We Need a New Welfare State, Oxford: Oxford University Press, pp. 68–95. Ferrarini, T. (2005), Interview (Associate Professor, Sociology, Swedish Institute for Social Research, Stockholm University, Stockholm). Ferrarini, T. and Duvander, A.-Z. (2010), Conflicting Directions? Outcomes and New Orientations of Sweden’s Family Policy, Stockholm University Linnaeus Center on Social Policy and Family Dynamics in Europe (SPaDE) Working Paper: 4, http://www.su.se/pub/jsp/polopoly.jsp?d=12097&a-53788 (accessed 4 May 2010). Government Offices of Sweden (n.d), The pre-school system, http://www.sweden. gov.se/sb/d/7172 (accessed 17 May 2010). Greve, B. (2004), Editorial Introduction: Are the Nordic Welfare States Still Distinct?, Social Policy & Administration, 38, 2: 115–18. Gustafsson, S. and Jacobsson, R. (1985), Trends in Female Labor Force Participation in Sweden, Journal of Labor Economics, 3, 1, pt. 2: S256–74. Hoem, B. (1995), The Way to the Gender-Segregated Swedish Labour Market. In K. Oppenheim Mason and A.-M. Jenson (eds), Gender and Family Change in Industrialized Countries, Oxford: Oxford University Press, pp. 279–96. Hoem, B. and Hoem, J. M. (1996), Sweden’s family policies and roller-coaster fertility, Journal of Population Problems, 52, 3/4: 1–22. Jenson, J. and Mahon, R. (1993), Representing Solidarity: Class, Gender and the Crisis in Social-Democratic Sweden, New Left Review, 201: 76–100. Kautto, M., Fritzell, J., Hvinden, B., Kvist, J. and Uusitalo, H. (2001), Introduction: how distinct are the Nordic welfare states? In M. Kautto, J. Fritzell, B. Hvinden, J. Kvist and H. Uusitalo (eds), Nordic Welfare States In the European Context, London: Routledge, pp. 1–17. Landes, D. (2010a), Home childcare trend fuels segregation fears, The Local, http://www.thelocal.se/24222/20100105/ (accessed 5 January 2010). Landes, D. (2010b), Moderates add feminist tough to policy platform, The Local, http://www.thelocal.se/24572/20100125/ (accessed 27 April 2010). Lindberg, I. (2005), Interview (Senior Researcher, Agora (think tank), and Member of the Family Policy Committee, Stockholm). Mahon, R. (1999), ‘Both Wage Earner and Mother’: Women’s Organizing and Childcare Policy in Sweden and Canada. In L. Briskin and M. Eliasson (eds), Women’s Organizing and Public Policy in Canada and Sweden, Montreal/Kingston: McGillQueen’s University Press, pp. 238–79. Mahon, R. (2005), Rescaling Social Reproduction: Childcare in Toronto/Canada and Stockholm/Sweden, International Journal of Urban and Regional Research, 29, 2: 341–57. Ministry of Education and Research (2008), School-age childcare, http://www.sweden.gov.se/sb/d/2098/a/72443 (accessed 17 May 2010). Ministry of Education and Research (2010), Responsibility for schools in Sweden, http://www.sweden.gov.se/sb/d/8009/a/72375 (accessed 6 October 2010). Ministry of Finance (2007), Budget Bill for 2008: Putting Sweden to work – increasing opportunities, http://www.sweden.gov.se/sb/d/8186/a/88569 (accessed 24 September 2007). Nyberg, A. (2004), Parental Leave, Public Childcare and the Dual Earner/Dual Carer-Model in Sweden, Swedish National Institute for Working Life Discussion Paper. OECD (2001), Starting Strong: Early Childhood Education and Care, Paris: OECD, http://www1.oecd.org/publications/e-book/9101011e.pdf (accessed 17 May 2006).
192
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Olsen, G. (2002), The Politics of the Welfare State: Canada, Sweden and the US, Oxford: Oxford University Press. O’Mahony, P. (2008), Sahlin pushes for more paternity leave, The Local, http://www.thelocal.se/10414/20080311 (accessed 24 March 2008). Pysslingen (n.d.), http://www.pysslingen.se/ (accessed 9 September 2009). Statistics Sweden (2009), Sweden’s population 2008: Highest fertility in 15 years, Stockholm: Statistics Sweden, http://www.scb.se/Pages/PressRelease_264551.aspx (accessed 21 October 2009). Statistics Sweden (2010), Statistical Yearbook of Sweden 2010, Stockholm: Statistics Sweden, http://www.scb.se/Pages/Product_30937.aspx (accessed 6 October 2010). Strandbrink, P. and Pestoff, V. (2006), Small-scale Welfare on a Large Scale: Social cohesion and the politics of Swedish childcare, Research Reports, 5, Huddinge: Södertörns högskola, http://www.diva-portal.org/diva/getDocument?urn_nbn_se_sh_diva-774-1_ fulltext.pdf (accessed 11 July 2006). Swedish Institute (2004a), Childcare in Sweden, http://www.sweden.se/templates/ BasicFactsheet_4132.aspx (accessed 9 November 2005). Swedish Institute (2004b), Equality between women and men, http://www.sweden.se/ templates/BasicFactsheet_4123.aspx (accessed 9 November 2005). Swedish National Agency for Education (2003), Descriptive data on childcare, schools and adult education in Sweden 2003, Swedish National Agency for Education Report no. 236, Stockholm: Swedish National Agency for Education. Swedish National Agency for Education (2007), Five years with maximum fee [English summary of Report no. 294], http://www.skolverket.se/sb/d/1926/a/10038 (accessed 17 May 2010). Swedish National Agency for Education (2009a), Costs: national level, Report no. 333, http://www.skolverket.se/sb/d/190 (accessed 6 October 2010). Swedish National Agency for Education (2009b), Children, pupils and staff: National level, Part 2, Report no. 331, http://www.skolverket.se/sb/d/190 (accessed 6 October 2010). Swedish Social Insurance Agency (2010), Social Insurance in Figures 2009, Stockholm: Swedish Social Insurance Agency, http://www.forsakringskassan.se/nav/ a286f7e4ab7756b46e7dc7ac22854a88 (accessed 6 October 2010). The Local (2007), Government offers more money for stay-at-home parents, The Local, http://www.thelocal.se/8382/20070904 (accessed 25 March 2008). The Local (2008), Swedish dads steer clear of paternity leave, The Local, http://www.thelocal.se/10420/20080312 (accessed 24 March 2008). The Local (2009), Legislate right to full-time employment: Left Party, The Local, http://www.thelocal.se/21772/20090830/ (accessed 3 March 2010). Thoursie, A. (2005), Interview (Senior economist, Swedish Trade Union Confederation, and Member of the Thorwaldsson Commission, Stockholm). Vinthagen Simpson, P. (2008), Social democrats demand shared parental leave, The Local, http://www.thelocal.se/14028/20080830 (accessed 9 March 2009). Vinthagen Simpson, P. (2010), ‘51 years to shared parental leave’: report, The Local, http://www.thelocal.se/25314/20100303/ (accessed 7 March 2010). Wennemo, I. (2005), Interview (Head of Economic Policy, Swedish trade Union Confederation, Stockholm). Westlund, J. (2007), Increased Parental Choice Can Lead to Reduced Gender Equality, NIKK magasin (Nordic Gender Institute), http://www.nikk.uio.no/?module= Articles;action=Article.publicShow;ID=595 (accessed 17 December 2008).
© 2011 Blackwell Publishing Ltd.
193
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00764.x Vol. 45, No. 2, April 2011, pp. 194–205
Narrowing of Public Responsibility in Finland, 1990–2010 spol_764
194..205
Merja Jutila Abstract Public responsibility in Finland has narrowed in the last 20 years while the sphere of the private sector has been increased. The economic crisis of the early 1990s was not the cause, but an accelerator of public sector/welfare state retrenchment in Finland. Based on which, it was easy for the advocates of neo-liberal reforms to argue that the changes were a must. The welfare state programmes however, are popular among the Finnish population and therefore large one-time cutbacks have not been possible beyond the immediate aftermath of the economic crisis. This article looks into three different methods through which the Finnish welfare state has been gradually cut since then: (1) by not raising income transfers along with the rising cost of living and wages; (2) by reducing funding of public services; and, on the other side of the coin (3) through regular tax cuts contracting the revenue side. Welfare state retrenchment in Finland has therefore been achieved in a subtle fashion through slow gradual weakening of social programmes on one hand, and through cuts in revenue on the other that have left proportionally more in the hands of the wealthier. These combined movements have resulted in a drastic reversal in the trend in income inequality in Finland.
Keywords Welfare states; Retrenchment; Public sector; Neo-liberalism; Inequality; Finland Introduction The Finnish welfare state has undergone a profound transformation in the 20-year period since 1990 until now. The emphasis has reversed from expanding and improving the welfare state to tightening social spending, privatizing or making public services more cost-effective, and moving policy more towards activation and individual responsibility, away from public responsibility. While some austerity measures were necessary following the drastic early 1990s economic crisis, the focus since has remained on tax cuts and paying off Address for correspondence: Merja Jutila, Ph.D., The City University of New York, 365 Fifth Avenue, New York, NY 10016, USA. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011 Table 1 Economic growth in Finland as percentage of the GDP, GDP per capita in Finland in euros, inflation in Finland, social spending as percentage of the GDP in EU25 and in Finland, and year-on-year percentage change in social expenditure in Finland in real terms Year
GDP growth
GDP/capita in euros
Inflation
1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
7.6 8.0 13.3 12.0 5.9 -4.5 -3.1 1.1 5.0 8.9 3.5 8.4 8.8 4.8 7.8 5.7 2.9 1.4 4.4 3.2 6.3 7.6 3.6
12,579 13,547 15,308 17,079 18,000 17,092 16,470 16,566 17,312 18,777 19,368 20,929 22,716 23,753 25,541 26,945 27,650 27,968 29,107 29,946 31,713 33,970 35,041
4.8 4.3 7.8 6.2 5.8 1.8 0.7 2.0 1.4 4.8 –0.2 2.1 3.4 0.9 2.6 3.0 1.3 -0.4 0.6 0.5 1.3 3.2 2.7
Social spending/ GDP – EU25
Social spending/GDP
Change
26.5 26.7 27.0 27.4 27.2 27.3 26.9 26.4
24.6 29.2 33.1 34.2 33.7 31.5 31.4 29.1 27.0 26.3 25.1 24.9 25.7 26.6 26.7 26.8 26.2 25.4
6.9 8.8 7.1 2.0 2.5 0.7 2.6 –0.8 –0.2 0.5 –0.4 2.5 4.3 4.0 4.6 2.6 2.5 1.6 2.3
Source: Statistics Finland 2008a; The National Institute for Health and Welfare n.d.a.
national debt. Income transfers and social services have been left unrepaired and unimproved and, instead, their funding has been tightened even further. The country experienced fast economic growth from 1994 onwards (see table 1), at a rate of almost nine per cent in 1995, 1997 and 1998 (Statistics Finland 2008a), but none of the money was spent on improving the impoverished transfers and services as the economy grew. Further, there were budget surpluses from 1998 until 2009 (Statistics Finland 2008a), but the surplus money was spent on additional tax cuts and further debt payments rather than on improvements in social transfers or services. The neo-liberal politics that aim at the reduction of the size of the public sector through public sector spending cuts and efficiency requirements, tax cuts, privatization of public enterprises and outsourcing of public services, etc. had already gained ground elsewhere in the world in the 1980s, pushed by international organizations like the OECD. The idea is to give more room for © 2011 Blackwell Publishing Ltd.
195
Social Policy & Administration, Vol. 45, No. 2, April 2011
the operation of the private sector at the expense of the public. Following the dramatic economic crisis, this doctrine also became the framework for restructuring the Finnish welfare state. Reducing costs and paying back debt became the priority in Finland, while the function of social policy has increasingly become the creation of conditions for a well-functioning economy. As the neo-liberal ideas have gained ground, the welfare state has become an outdated idea to many. As the country’s economy has experienced fast growth, the fruits of the growth have not been distributed equitably. In just a couple of years in the mid-1990s the income gap in Finland widened back to what it had been at the beginning of the 1970s (Statistics Finland 2007a). In a country that consistently over decades had worked toward reducing inequalities, this marks a profound transformation. While inequality has become accepted among the political decision-makers as a necessity for the competitiveness of the Finnish economy, of all the OECD countries, income inequality in the last ten years has risen the fastest in Finland and Sweden. The OECD report agrees that this is because social benefits in Finland have not been increased in line with wages, and because income taxes have been lowered (OECD 2008). Where I disagree is that the report seems to assume that this is a natural process of the anomaly of high equality correcting itself as countries converge, not accounting for the man-made decisions by political decision-makers that have created this outcome.
The Change in Paradigm of the Finnish Welfare State Until the 1980s, the focus in Finland had been on controlling societal divisions and reducing inequalities (Saari 2006: 11). As a result, poverty in Finland fell from 18 per cent in 1966 and 12.2 per cent in 1981, to only 6.4 per cent in 1993 (Statistics Finland 2006a). This development, however, was disrupted by the early 1990s economic crisis in Finland. The economic crisis started with the banks and was the worst recession of any of the OECD countries since the depression of the 1930s. The growth rate in Finland remained at zero in 1990 with the following three years witnessing negative growth with the GDP declining approximately 13 per cent in total (Uusitalo 1996: 1). At the same time, because of all the bankruptcies (exceeding 7000 businesses in 1992, Statistics Finland 2006b) and public sector layoffs, unemployment grew, reaching 18 per cent of the Finnish working population at its worst point in the winter of 1993–94, up from four per cent just three years earlier (Uusitalo 1996: 2). At the same time, the Finnish national foreign debt increased from 14 per cent of GDP in 1990 to 57.8 per cent of GDP in 1994 (Statistics Finland 2008b). The economic crisis resulted in the severest reform process of any of the social democratic welfare states with a government savings programme that amounted to 12 per cent of GDP in three years from 1994 to 1996. Although the economy started to again grow quickly in 1994, and the following year the Social Democrats won a landslide victory in the parliamentary elections (holding on to power in coalition governments for the next 12 years), the emphasis permanently stayed on trimming expenses. 196
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
According to Juho Saari (2006), the battle of ideas in Finland lasted for a couple of years, but the ones that won out, with the help of the economic crisis, have since been rather unquestioned in Finnish politics. The goal since has been a welfare state that, instead of creating equality and lowering poverty, ‘continuously renews itself by creating institutional structures that enable competitiveness’ (Saari 2006: 105–6). This has meant making labour markets more flexible and decentralizing income negotiations, because work has to be ‘encouraged’. Therefore income inequalities have to be increased, taxes lowered, public services privatized and social benefits cut as well as made more conditional. And just as in many other countries, neo-liberalism in Finland has also meant the increasing power of the Ministry of Finance (Patomäki 2007: 77). The Ministry of Finance specifically has acted as a jealous guardian against any increases in spending – paying back the national debt and ‘preparing for the future’ having been more important. The austerity politics then legitimized have continued long after the budget had been balanced in 1998. Any discussions about actually improving social policy have been out of the question since (Saari 2006: 255). Social politics have become a tool for achieving goals such as employment and competitiveness of business (Saari 2006: 40–1), and instead of the goal of equality, reducing the cost and the number of recipients is now what constitutes successful social policy (Julkunen 2006: 69). Increasingly, then, social policy is there not to compensate for income loss or to provide care, but rather to use carrots, and sticks, to encourage individuals to participate in the private sector to provide for themselves and their families. The result is that while the welfare state and taxes have been cut, inequality and poverty in Finland have increased. By the end of the 1990s the income gap had deteriorated to what it had been at the beginning of the 1970s (Statistics Finland 2007a). This is because from 1995 (when the Finnish economy started to grow) to 2006, while the average income of the bottom ten per cent in Finland rose by ten per cent in real terms, the incomes of the top ten per cent rose by 65.9 per cent and of the top one per cent by an astonishing 125.6 per cent (Statistics Finland 2007b). At the same time, the number of low income people (60 per cent of the median income) has doubled since 1993. In 2005, 12 per cent of Finns (up from 6.4 per cent during the worst year of the recession) were considered low income, making less than €12,500 a year (Statistics Finland 2007b). And even during fast economic growth, before the global recession, the number of those considered low income rose to 13.6 per cent in 2008 (Sippola 2009). Finland could have afforded to return the welfare state to the level that it was before the economic crisis. Instead, an ideological shift occurred among the political decision-makers that prioritizes economic growth, and believes in the accumulation of wealth, as well as inequality, to be beneficial for it. The new ideology places the welfare state under austerity measures diminishing redistribution and reducing funding for public services, while relaxing taxation on the wealthiest. This is a full reversal of the earlier welfare state goal of equality, and instead knowingly making policy changes that result in an increase in inequality. © 2011 Blackwell Publishing Ltd.
197
Social Policy & Administration, Vol. 45, No. 2, April 2011
The economic crisis in essence helped to usher in the era of neo-liberal austerity politics in Finland. It was not the cause of the changes but rather a facilitator that could be used to justify them. Some of the ideas were already there, ready to be activated when the political opportunity arose (Heiskala and Luhtakallio 2006). Esko Aho, Prime Minister of the Centre Party, 1991–95, explained that since the building of the welfare state had been completed in the 1980s, the depression facilitated the necessary transformation towards a more modern, enabling approach to well-being, away from the role of the state (Interview, 9 January 2009), admitting that without the depression the changes that some political and economic elites had wanted to bring about would have been difficult to accomplish. Yet being the head of a centre-right coalition, the cuts and the reforms made were not surprising. But what is interesting is that the following coalition headed by the Social Democrats continued on the same path for the next eight years (and a further four as a coalition partner in the 2003–07 coalition headed by the Centre Party) well into the period of economic growth and years of budget surpluses.
Retrenchment Politics Following the initial rounds of cutbacks in the mid-1990s, scholars (Kantola and Kautto 2002; Heikkilä and Uusitalo 1997) argued that the character of the Finnish welfare state had been maintained. I argue that a fundamental change in character as a matter of fact has happened. The change, however, has happened gradually over many years, out of sight, after those initial cutbacks. The process has not been one of obvious assaults on the welfare state, but rather a slow deterioration of social transfers and starving of the public services ever since. Reversal in the trend of inequality is the obvious indicator of this change. Although social expenditure as a percentage of GDP is generally used as the means for cross-national comparisons of the welfare states, it can be misleading because when a country’s GDP grows, the social expenditure percentage appears to be lower even if the amount spent had stayed the same, and while during hard economic times a country’s GDP may fall, and therefore social expenditures will appear to rise even without the increase in unemployment expenses. Further, the percentage of GDP does not indicate what kinds of programmes the money is spent on or what percentage of the population benefits from it (e.g. American health care being the costliest of all OECD countries but failing to cover one in six Americans). Also, added efficiency in social service provision will work to lower social expenditure in comparison to the other countries. Therefore, the aggregate percentages do not reflect the type of welfare state or the changes in actual spending. One will also be misled by simply looking at aggregate social spending. In the case of Finland, for example, the aggregate social expenditure has continued to increase slightly during the past two decades, despite the various ‘cost-containment’ and ‘recalibration’ measures (Pierson 2001). Therefore, just looking at the total amount spent, one could conclude that cuts were not made. Yet, social programmes have been weakened across the board and the slight overall increase is produced by rising demand – first, due to the higher 198
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
number of unemployed in the 1990s and especially due to the increasing number of elderly, adding to pension and health care costs. Therefore, without the efforts to restrict spending, the aggregate social expenditure would have continued to grow at a much faster rate than it has. Only the slight increase in aggregate social spending, to account for the increasing demand, has instead been achieved by thinning out the safety net. This becomes especially evident when looking into different policy areas in more detail. I will turn to the deterioration of income transfers first.
Income transfers In the aftermath of the depression, social transfers in Finland were cut across the board. Cutbacks in social spending (which in 1993 was 34.8 per cent of GDP, see table 1) seemed urgent and unavoidable. First, the centre-right coalition of 1991–95 cut back benefit levels and set stricter time limitations particularly in areas of sickness and unemployment (Kosonen 1998: 347–8) while delaying service improvements, and then starting in early 1996 ‘the Rainbow Coalition’, comprising parties of the political left and the right, made cuts especially in the benefits of those relying on minimum income, namely the worst off (Kosonen 1998: 349). Overall, due to the cuts made, it is estimated close to ten per cent of the social expenditure was cut in the 1990s (Kosunen 1997: 41). After that, a more drastic yet concealed change in the welfare state has been achieved in real terms by not raising benefits along with the rising cost of living and wages, and by tightening eligibility, as well as by adding user fees to public services, gradually worsening the purchasing power of the transfers. The effect of not raising benefit levels has been the severest on the groups that tend to have the lowest incomes. The people living in poverty in Finland tend to be students, young families, especially single parents with children, and pensioners. The benefits for these groups have also been the ones that have been left to lag as wages and prices have risen. For instance, the student benefits for university students have not been raised since 1992 (until August 2008). Rather, they were lowered in 1995. Child allowances were last raised in 1994 (other than the raise of €10 for a third child starting in 2009), but were cut in 1995. As a result, student benefits for university students in relation to wages fell 35.1 per cent from 1994 to January 2009 (and 19.2 per cent since 2000), child allowances for the first child have fallen 40 per cent since 1994 (and 22 per cent from 2000), as the child care benefit for a single parent fell by 53.2 per cent since 1994 and 19.4 per cent since 2000. At the same time, the full basic pension, although increasing most years by a few percentage points due to being tied to the basic pension index (which is the average of the third quarter cost of living index divided by 1.16, see The Social Insurance Institution of Finland (KELA) n.d.), fell by 28.2 per cent for a single person since 1994 and 16 per cent since 2000. And the supplementary means tested income support in relation to wages has fallen 28.2 per cent since 1994 and 16 per cent since 2000 (Aaltonen 2009: 13). Clearly, the incomes of those found to be the poorest in society have been left to lag, resulting in poverty and huge income differences. The public sector is providing less toward the well-being of the weakest © 2011 Blackwell Publishing Ltd.
199
Social Policy & Administration, Vol. 45, No. 2, April 2011
and instead increasingly leaving them to fend for themselves for their livelihood. This same phenomenon can be found in the provision of public services in Finland.
Tightening the public service provision The municipalities in Finland are in charge of providing services to their residents from day-care and health care to education. As the state’s share of funding to the municipalities has been constantly cut, the municipalities have been forced to try to provide the services without raising municipal taxes. The idea has been to encourage increased efficiency, innovation, privatization and outsourcing, as well as to encourage municipalities to band together into bigger entities. In the budget for 2008, for instance, Ministry of Health and Social Affairs state transfers to municipalities in 2007–11 were reduced by yet another €60 million (Kaarto 2007). The amount was to be made up by increasing user fees in health and social services in the municipalities beyond what could not be accomplished with added efficiency to compensate for the lesser resources. Other than the increased user fees, the result of the efficiency attempts instead, has often been fewer staff and increased waiting times, while those who can afford are driven to the private sector. As a result of these trends, while the state’s share of the total social expenditure (including the income transfers) in 2008 is the same as before the economic crisis in 1990 (approximately 25 per cent), the municipalities’ share of the total social expenditure has increased from 15.6 per cent to 18.6 per cent from 1990 to 2008 and the share of the insured has climbed from 8 per cent to 11.2 per cent. The winning party to these developments have been the employers, whose share has dropped from 44.1 per cent to 38.4 per cent of the total social expenditure (The National Institute for Health and Welfare n.d.b). In essence, the trend has been to reduce the private sector’s financial burden of paying for social expenditures by channeling responsibility from the employers to the municipalities and the insured through and by the state. This trend continued in 2009 when the current centre-right government abolished the national pension payment from most employers in April 2009. While the financial burden on employers has been reduced, there has been a simultaneous aspiration to move provision of public services to the private sector through outsourcing and privatization. The assumption is that the private sector can provide services cheaper and more efficiently. Therefore, many formerly public institutions have been incorporated or sold to private investors, and an increasing amount of public sector services is bought from private providers. In the public sector in turn, the application of management models of the private sector has been encouraged (Suleiman 2003: 124). By 2007, over a quarter of social services, such as housing services, day-care and domestic services, were provided by private service providers, non-profit organizations and private companies, the share of the latter especially increasing since the mid-1990s (The National Institute for Health and Welfare n.d.c). Yet, the benefits are dubious from the savings’ point of view. In March 2010, for instance, municipal medical providers were complaining that since they had started outsourcing and privatizing, medical expenditures had actually 200
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
risen despite the promises made (Iltalehti 2010). It seems that privatization and outsourcing in many cases have specifically been an ideological project rather than one based on well thought-out efficiency or savings plans. From an institutional point of view, what has happened in Finland is a move from government to governance, a shift from state resource guidance to guidance by markets (Alasuutari 2006: 43), reducing the sphere of the public sector. Instead of budget allocations for a particular purpose performed by the public sector employees, now demand from customers is critical, reinforced by fees, vouchers, etc. for determining what services are provided. Often the providers of the services are third parties, organizations and private service providers (Alasuutari 2006: 43–4). Increasingly, citizens are left to fend for themselves, putting them in differing bargaining positions according to their purchasing power in the private sector. All this may result not just in deeper socio-economic divisions, while draining the public sector of resources, but driving those that can afford to the private sector where they will pay for part of their services out-of-pocket. In the long run, the result will be reducing support for public services (Blomberg and Kroll 1999: 57) and likely further reductions in the funding of public services.
Tax cuts The other side of the coin of public sector retrenchment in Finland have been changes in fiscal policy. Tax cuts are an indirect way of reducing the abilities for welfare state improvements by cutting state revenue, and another ideological decision going along with the reduction of the size of the public sector. There are four main categories of taxes in which this change has happened in Finland: capital gains, state income and municipal taxes and the tax on wealth. First in 1993 capital gains tax was made a flat tax that is taxed at a lower level than earned income tax. This is generally held as the turning point in the growth of income inequality in Finland, enabling the take-home incomes of the wealthiest to skyrocket. This is because the wealthiest tend to make their money from investments and not from salaries, and even then because of the lower tax, there has been a tendency of converting salaries into bonuses that are taxed at the lower rate. Since then, capital gains tax has fallen from 32 per cent to 29 per cent (Pietiläinen 2008). Progressive state income tax, on the other hand, has been reduced every year, with the exception of two years, by 7 per cent since 1995, also leaving proportionally more in the hands of the wealthiest and resulting in income inequality to widen. The progressive state taxes have fallen on average from 37.5 per cent to 30.2 per cent in 12 years (Raeste 2007). At the same time, the emphasis in taxation has been shifted toward municipal flat tax, which again benefits the wealthiest (being lower than the highest progressive tax), and also varies according to the municipality. Municipal tax is what pays for most of the services from health care to education in a municipality. The poorer the population of the municipality (rural, elderly, ill, etc.), the higher the tax for accumulating the necessary funds for the higher© 2011 Blackwell Publishing Ltd.
201
Social Policy & Administration, Vol. 45, No. 2, April 2011
than-average need for social services; and the wealthier the population, the lower the flat tax, such as in the small affluent town of Kauniainen right next to Helsinki (16.5 per cent in 2009, with the most common municipal tax being 19 per cent). Also, starting in 2006, the government abolished the tax on wealth. And in 2007, inheritance taxation was relaxed. As a result, when in 1999 the 1000 with the highest incomes paid 56 per cent in taxes, in 2007 they paid 47 per cent. ‘Fiscal policy not only has failed to level income equalities, it has exaggerated them’ (Taimio 2007: 13, my translation) by lowering the progressiveness of taxation, affecting those at the opposite ends of the income scale the most. Poverty may have increased because of the falling actual levels of social transfers, but the increasing inequality has been created by fiscal policy that specifically allows the wealthiest to get wealthier and, through that, widening the income gap and creating a society of unequals.
Role of the Social Democratic Party Since the early 1990s, well into the period of fast economic growth fueled by Nokia and the hi-tech boom, the Finnish political decision-makers have enacted a series of policies weakening the Finnish welfare state that, at the same time, continues to be very popular among the population. In a 2009 study by Finnish Business and Policy Forum EVA, it was found that 82 per cent of Finns surveyed agreed with the statement ‘although maintaining a high level of social security and other public services costs a lot, the Finnish welfare state is always worth the cost’ (38 per cent answered ‘very much’ and 44 per cent ‘agree’) (see Haavisto and Kiljunen 2009: 33). A total of 79 per cent further thought that the current market economy does not sufficiently take the weakest in society into consideration (see Haavisto and Kiljunen 2009: 21). The Association of Finnish Local and Regional Authorities study in the same year likewise found that half those surveyed prefered raising municipal taxes over cutting municipal services, while 20 per cent were opposed to it (Helsingin Sanomat 2009). The Social Democratic Party that was in power from 1995 to 2003 ended up practicing politics that were in direct conflict with what it had stood for in the past. Instead of being the defender of the welfare state, it became the party to launch the welfare state retrenchment programme in Finland. What happened was that even though the Rainbow Coalition was led by the Social Democrats (and included also the centre-right National Coalition (NCP), as well as the smaller Left Alliance, The Greens, and The Swedish People’s Party, named ‘Rainbow Coalition’ for comprising parties covering the whole political spectrum) it ended up practicing politics that are most identified with right-wing parties. The phrase ‘there is no other choice’ was used frequently by the political decision-makers to communicate to the public that the austerity measures were not political value choices, but rather necessary and unavoidable decisions. Tax cuts were regular, social spending was cut and benefits were not raised, not even to make up for inflation, despite the budget surpluses and fast economic growth. At the same time, incentives were added, publicly owned 202
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
companies were privatized, and public services outsourced to private companies. On countless occasions, the governing coalition decided that additional debt payments and additional tax cuts were more important than increasing funds to the starving public services or raising income transfers that were horribly lacking and subjecting people to live in poverty. The severity of the depression provides an explanation for the austerity politics in the first half of the 1990s, but since then it is not clear that these were the only alternatives and that debt had to be lowered as fast, for example, since it was nowhere near threatening the EMU criteria (not to mention that most of the other EMU countries were constantly breaking the rules), or that all those numerous tax cuts were a must. The recovery politics in Sweden, in contrast, did not result in inequality to grow as much, although the recession in Sweden was not quite as severe. Sweden instead, not being in the Eurozone, for instance, opted to keep its national debt for most of the 1990s substantially above the Stability and Growth Pact requirements (Kiander 2005: 63). It is clear that an ideological shift happened in Finland when even the political decision-makers of the political left started advancing the neo-liberal politics. This is especially interesting, given the Finnish welfare state has continued to be very popular among the Finnish population at the same time that it has endured significant scaling back. It seems that the Finns as a matter of fact are growing increasingly opposed to the growing income gap. Taloustutkimus survey commissioned by Aamulehti found that over 70 per cent of Finns in 2006 were opposed to the increasing income gap compared to 60 per cent in 2001 (Helsingin Sanomat 2009), for example. Therefore, it would be extremely hard to make the case that the political decision-makers were simply carrying out the wishes of the voters. The ideological shift was not discussed or decided on in Finnish politics, rather it crept in piecemeal, decision by decision. To this day, it is hard to find an open proponent of neo-liberalism in Finland. Yet privatization, tax cuts, efficiency requirements in the public sector, etc. have a great number of proponents, particularly among the political and economic elites, and are actively practiced at every level of government. Executed by top civil servants, Anna Kontula (2008: 76) writes that it was never decided – neo-liberalism crept in quietly. Heikki Patomäki further points out that neo-liberalism has been imitated rather uncritically and without imagination in Finland. Anything unique is due to the strength of previously created historical institutions, and the vastness of resistance (Patomäki 2007: 107).
Conclusion Welfare state retrenchment in Finland has occurred in a subtle fashion through a slow gradual weakening of social programmes on one hand, and regular cuts in fiscal policy that favour the wealthier on the other. While the depression ended in 1994, cutbacks continued throughout the decade. Those making the cuts argued that it was not necessarily a value shift, but that cutbacks had to be made. Yet, when it came time to raise the benefits back up, there was always a reason why that could not be done. As the Finnish economy started performing well, the problems facing the welfare state © 2011 Blackwell Publishing Ltd.
203
Social Policy & Administration, Vol. 45, No. 2, April 2011
became ones that look into the future and anticipate potential problems that may emerge therefore necessitating austerity measures in case of a rainy day. The result has been a continuous welfare state retrenchment project over a period of approximately 20 years, although the reasons and justifications for it have varied. While the public sector is reduced in favour of the private, a much more unequal society is being created.
References Aaltonen, E. (2009), Sosiaaliturvauudistus – linjaukset, presentation in Jyväskylä, 5 May 2009, http://www.stkl.fi/2009_SATA_Aaltonen.pdf (accessed 3 March 2010). Alasuutari, P. (2006), Suunnittelutaloudesta kilpailutalouteen: miten muutos oli ideologisesti mahdollinen? In R. Heiskala and E. Luhtakallio (eds), Uusi jako, Miten Suomesta tuli kilpailykyky-yhteiskunta? Helsinki: Gaudeamus. Blomberg, H. and Kroll, C. (1999), Who Wants to Preserve the ‘Scandinavian Service State’? Attitudes to Welfare Services among Citizens and Local Government Elites in Finland, 1992–6. In S. Svallfors and P. Taylor-Gooby (eds), The End of the Welfare State?, New York, NY: Routledge. Haavisto, I. and Kiljunen, P. (2009), EVAn arvo- ja asennetutkimus 2009: Kapitalismikansan käräjillä, Finnish Business and Policy Forum EVA, Yliopistopaino. Heikkilä, M. and Uusitalo, H. (eds) (1997), Leikkausten hinta: Tutkimuksia sosiaaliturvan leikkauksista ja niiden vaikutuksista 1990-luvun Suomessa, Stakes: Raportteja 208. Heiskala, R. and Luhtakallio, E. (eds) (2006), Uusi jako, Miten Suomesta tuli kilpailykykyyhteiskunta?, Helsinki: Gaudeamus. Helsingin Sanomat (2009), Puolet kannattaa kuntaveron korotusta palveluiden turvaamiseksi, Helsingin Sanomat, 18 February 2009, http://www.hs.fi/tulosta/ 1135243625849 (accessed 18 February 2009). Iltalehti (2010), Editorial: Ulkoistaminen lisäsi kuluja, Iltalehti, 12 March 2010, http://www.iltalehti.fi/paakirjoitus/2010031211280284_pk.shtml (accessed 12 March 2010). Julkunen, R. (2006), Kuka vastaa? Hyvinvointivaltion rajat ja julkinen vastuu, Vaajakoski: Gummerus Kirjapaino Oy. Kaarto, H. (2007), Valtio vie terveyskeskusmaksujen korotushyödyn kunnilta, Helsingin Sanomat, 31 July 2007, http://www.hs.fi/kotimaa (accessed 31 July 2007). Kantola, A. and M. Kautto (2002), Hyvinvoinnin valinnat: Suomen malli 2000-luvulla, Helsinki: Edita Publishing Oy. Kiander, J. (2005), Julkisen talouden kehitys 1995–2004 ja sopeutuminen verokilpailuun. In Kymmenvuotiskatsaus 2005, Statistics Finland, http://www.stat.fi/tup/ julkaisut/isbn-952-467-461-0_kiander.pdf (accessed 23 March 2010). Kontula, A. (2008), Onko vasemmistosta haastajaksi? In H. Kuusela and M. Rönkkö (eds), Puolueiden kriisi, Helsinki: Like Kustannus. Kosonen, P. (1998), Pohjoismaiset mallit murroksessa, Tampere: Vastapaino. Kosunen, V. (1997), Laskusuhdanteesta leikkauksiin. In M. Heikkilä and H. Uusitalo (eds), Leikkausten hinta: Tutkimuksia sosiaaliturvan leikkauksista ja niiden vaikutuksista 1990luvun Suomessa, Stakes: Raportteja 208, pp. 33–44. Kosunen, V. (1997), Lama ja sosiaaliturvan muutoksent 1990-luvulla. In M. Heikkilä and H. Uusitalo (eds), Leikkausten hinta: Tutkimuksia sosiaaliturvan leikkauksista ja niiden vaikutuksista 1990-luvun Suomessa, Stakes: Raportteja 208, pp. 45–101. The National Institute for Health and Welfare (n.d.a), Table 8. Social expenditure as a share of GDP in EU and EEA countries, 1990–2007,%, Table 7. Social expenditure as a share of GDP by function 1980–2008,% and Table 4. Social expenditure, year-on-year percentage
204
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
change in real terms, %, http://www.stakes.fi/tilastot/tilastotiedotteet/2010/ liitetaulukot/Tr02_10_liitetaulukot.pdf (accessed 18 September 2010). The National Institute for Health and Welfare (n.d.b), Table 6. Financing of social expenditure, 1980–2008, %, http://www.stakes.fi/tilastot/tilastotiedotteet/2010/ liitetaulukot/Tr02_10_liitetaulukot.pdf (accessed 29 September 2010). The National Institute for Health and Welfare (n.d.c), Yksityiset sosiaalipalvelut, http://www.stakes.fi/FI/Tilastot/Aiheittain/Sosiaalipalvelut/yksityisetsosiaalipal velut.htm (accessed 29 April 2010). OECD (2008), Growing Unequal? Income Distribution and Poverty in OECD Countries, http://www.oecd.org/els/social/inequality/GU (accessed 28 May 2010). Patomäki, H. (2007), Uusliberalismi Suomessa, Lyhyt historia ja tulevaisuuden Vaihtoehdot, Vantaa: Dark Oy. Pierson, P. (2001), Coping with permanent austerity: welfare state restructuring in affluent democracies. In P. Pierson (ed.), The New Politics of the Welfare State, New York, NY: Oxford University Press. Pietiläinen, T. (2008), Tuloeliitille vuosituhannen isoimmat pääomavoitot, Helsingin Sanomat, 4 November 2008, http://www.hs.fi/tulosta/1135240780933 (accessed 4 November 2008). Raeste, J.-P. (2007), Tuloverotus laskenut seitsemän prosenttiyksikköä vuodesta 1995, Helsingin Sanomat, 24 January 2007, quoting the Taxpayers’ Association of Finland president Teemu Lehtinen (Vero 2007-event), http://www.hs.fi/talous/artikkeli/ Tuloverotus + laskenut + seitsem% C3 %A 4n+prosenttiyksikk%C3%B6%C3%A4+ vuodesta+1995/1135224547506 (accessed 24 January 2007). Saari, J. (2006), Suomen malli – Murroksesta menestykseen?, Helsinki: Yliopistopaino. Sippola, A.-R. (2009), Pienituloisia jo yli 700 000, Helsingin Sanomat, 19 January 2009, http://www.hs.fi/tulosta/1135242856569 (accessed 19 January 2009). The Social Insurance Institution of Finland KELA (n.d.), National Pensions Index, http://www.kela.fi/in/internet/english.nsf/NET/100908142337HS?OpenDocument (accessed 18 September 2010). Statistics Finland (2006a), Pienituloisten lukumäärät ja osuudet 1966–2006, http://www. stat.fi/til/tjt/2006/tjt_2006_2008-05-16_tau_003.html (accessed 9 December 2008). Statistics Finland (2006b), Vireillepannut konkurssit velallisen toimialan mukaan, 1986–2006, http://www.stat.fi/til/konk/tau.html (accessed 5 April 2010). Statistics Finland (2007a), Tuloerojen kehitys Suomessa, http://www.stat.fi/til/tjt/2005/ tjt_2005_2007-05-16_tau_001.html (accessed 21 March 2008). Statistics Finland (2007b), Yleinen tuloerojen kehitys, http://www.stat.fi/til/tjkt/2006/ tjkt_2006_2007-12-21_kat_001_fi.html (accessed 9 December 2008). Statistics Finland (2008a), Gross domestic product (GDP) at market prices 1975–2008, http://www.stat.fi/til/vtp/2008/vtp_2008_2009-02-27_tau_001_en.html (accessed 18 September 2010). Statistics Finland (2008b), Julkisyhteisöjen EMU-velka 1975–2008, http://www.stat.fi/til/ jali/2008/jali_2008_2009-09-30_tau_002_fi.html (accessed 10 March 2010). Suleiman, E. (2003), Dismantling Democratic States, Princeton, NJ: Princeton University Press. Taimio, H. (2007), Talouskasvun hedelmät – kuka sai ja kuka jäi ilman?, Helsinki: Tammerpaino. Uusitalo, H. (1996), Economic crisis and social policy in Finland in the 1990s, Social Policy Research Centre Discussion paper No. 70: 1.
© 2011 Blackwell Publishing Ltd.
205
Social Policy & Administration issn 0144–5596 DOI: 10.1111/j.1467-9515.2010.00765.x Vol. 45, No. 2, April 2011, pp. 206–219
Leaving the Nordic Path? The Changing Role of Danish Trade Unions in the Welfare Reform Process spol_765
206..220
Henning Jørgensen and Michaela Schulze Abstract For a long period, Denmark has been labeled a ‘model country’ with a comprehensive welfare state and a successful model of corporatist policy-making. Danish unions are considered amongst the strongest in the world, and they have for a long time been a distinct part of the political system, and as social partners, they were strongly integrated into decision-making processes. The analysis of the Danish welfare and labour market policy during the last two decades documents a profound change in the arrangement and in the status of the social partners (especially unions) in the Danish political system. The results show that two important pillars of the Danish model – the social partner basis and the collective trust in partnership – are eroding. Unions are no longer part of the law-making process and, since 2007, they are formally excluded from the organization of the decision-making process. Recent developments point at weaker unions that operate more as lobbyists instead of being strong corporatist institutions or part of the decision-making process. The results of the study are thought-provoking and the basis for a revised thinking of the Danish and the Nordic model.
Keywords Denmark; Welfare state; Social partners; Activation; Labour market policy; Nordic model Introduction: Collective Action and Loss of Power Resources The loss of influence and power resources of unions during recent decades is not to be denied. The explanation for this is, however, not easy and straightforward. Is it because of growing economic globalization, new forms of production, search for ‘competitiveness’, and deregulation that unions have lost power resources, or is it because of changing attitudes, a fragmented membership and loss of members? Or is it perhaps because of other reasons? Structural changes might have embedded unions in new kinds of relationships, declining unionization, and to this changing attitudes and voting behaviour amongst employAddress for correspondence: Henning Jørgensen, Centre for Labour Market Research, Aalborg University, Fibigerstæde 1, 9220, Aalborg Ø, Denmark. Email:
[email protected]. Michaela Schulze, University of Siegen, Department of Sociology, Adolf-Reichwein-Straße 2, 57068 Siegen, Germany. Email:
[email protected] © 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ , UK and 350 Main Street, Malden, MA 02148, USA
Social Policy & Administration, Vol. 45, No. 2, April 2011
ees might have brought a stronger political fragmentation; but none of these developments determines the use of power resources and the distribution of influence in a system. Here relationships between the labour market and the welfare state system are important and – we will argue – political decisionmaking and shifting ways of organizing such relationships are decisive. Institutional pillars of support for private interest representation and representational legitimacy are amongst the important factors. But structures and resources are not the only decisive elements for decisions and priorities. These cannot be derived from contexts or taken for granted. Unions are not only economic agents but also actors on the political scene. Regulating the labour market and gaining influence is a question of power resources and the use of these resources in intelligent ways. This also includes political decision-making processes and arrangements. Nationally, special institutional arrangements and opportunity structures have been established. The loss of power of unions – the collective ability to mobilize and act as a united actor on behalf of its members’ interest vis-à-vis employers’ representatives and the political-administrative system (Korpi and Palme 2003) – is fundamentally linked to the political system and the way welfare state arrangements are functioning. The ability to organize might be weakened (Visser 2006), but the ability to act in a political way is another question. The creation or recalibration of political and administrative institutions is most important as to the ability of unions to solve collective action problems (Rothstein 1992; 2003), and external legitimacy and concrete opportunity structures count as well. Political decisions and strategies might contribute to the loss of power resources by changing policies and their institutional implementation structures and by shaping capabilities of unions. In this way, governmental strategies and ways of including or excluding central organizations and groups in institutional arrangements and decision-making processes are central to union power resources. This is to be documented, analyzing the developments in the Danish welfare and labour market system during the last two decades, concentrating on labour market policy, the politics of industrial relations and the repercussions that changes in the content and organization of decisionmaking power have had on union power resources. The basis for the following analysis and conclusions are threefold: existing literature on recent Danish labour market history and welfare state reform (desk research), a Ph.D. thesis from 2010 by one of the authors (Michaela Schulze), involving interviews with Danish decision-makers and experts, as well as document studies, and studies carried out by the CARMA group in Aalborg during the last 30 years (amongst these, see Jørgensen 2006/2007 and Jørgensen 2009/2010). Methodologically, more kinds of data collection and data analysis are involved and, in this way, triangulation strengthens the validation of statements.
Danish Labour Market and Welfare Arrangements and Power Resources Danish unions are considered amongst the strongest in the world. Besides strong membership resources (unionization of more than 80 per cent), trade © 2011 Blackwell Publishing Ltd.
207
Social Policy & Administration, Vol. 45, No. 2, April 2011
unions have for a long time been a distinct part of the political system, and as social partners, they are strongly integrated into decision-making processes. Welfare state policy was one of the major fields of union influence besides the private regulation through collective agreements. Social partners have strongly influenced both public and private regulation. Corporatism has had a long history in the Nordic countries, including Denmark. And the collective mobilizing of unions was hugely important for securing social democratic governments up to the 1980s (Allern et al. 2007; Magnusson et al. 2008). The influence of Danish trade unions is widely recognized. A strong union movement and corporatist structures were all part of the development of flexible growth-oriented economies, high employment, compressed wage structures, and welfare state policies in the Nordic countries (Korpi 1978; Korpi 1983; Cameron 1984; Esping-Andersen 1990; Jørgensen 2002; Minnich 2003). Since the late 1990s, new patterns of decision-making processes and political priorities are, however, visible in Denmark. A new liberal-conservative government took office in Denmark in 2001 and political decisions as to include or exclude unions from decision-making bodies are to be analyzed alongside other political initiatives influencing union capabilities and strategies. Unionization was actually growing until 2003. However, political developments were signaling a new future of corporatism in Denmark and a repositioning of the unions in relation to the political arena. The social partners have a common interest in institutionalized influence in the political system. Unions are more dependent on such kinds of arrangements than employers, as they have few other political resources to veto political decisions and public policies. Especially in Denmark, where both economic and formal ties between the Social Democratic Party and the Danish Trade Union Confederation (LO) were abolished in 2003. The employers have power over the means of production, employment decisions and arguments for competitiveness, and they have exit opportunities that are not the case with trade unions. Unions are only achieving power to influence decisions and political agendas by voice and loyalty options. Besides, they have to use their strategic resources and capabilities in combined efforts in order to defend rights and arrangements within the professional field with a political project including egalitarian values and interests unknown to employers. That is why this article concentrates on trade unions and their power resources and not on social partners in general. The welfare state and the labour movement are closely interrelated and the same goes for the relationship between the Scandinavian welfare states and corporatism. But the very idea of the labour movement implies economic, political and cultural goals, and a vision transcending the present defensive character of trade union action. Special opportunity structures have been created for the unions to influence public policies and to find ways of developing social dialogues with employers – and it is the recent Danish labour market and welfare state history we concentrate on in accounting for the weakening of the power resources of the unions. Until the end of the 1990s, trade union influence on social policy remained strong and stable (Jørgensen 2003). This is because trade union membership is linked to the unemployment 208
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
insurance system and to the fact that unions were integrated into the political processes. At that point, the government wished to make its own reforms and this put pressure on the unions. With the coming of the liberal-conservative government in 2001, unions were put on the side-line and new reforms were planned, totally abandoning political corporatism in this policy area in 2007. The social partners were no longer part of public administration but put into advisory roles only. From 1 August 2009, the municipalities took over all formal responsibility for labour market policy, giving the social partner fewer chances to influence the policies adopted. Three periods can be distinguished in the relationship between the state and the unions since 1990. The political developments have changed the influence of the LO and the other peak organizations rather profoundly. The period from the early 1990s up to 1998 is marked by reforms and the changing but strong influence of the LO. Politicians and the social partners prepared the first labour market reform in 1993. Major changes occurred from the second labour market reform in 1998: civil servants prepared the reforms, and the LO was not part of the policy-making process (Jørgensen 2003). In this reform period, the LO lost part of its traditional influence in the political system. While political corporatism changed rather profoundly, administrative corporatism was even expanded. The second period – between 1999 and 2006 – is characterized by major activation reforms and further changes in the corporatist system. The influence of unions in the political system diminished. The period from 2007 onwards shows the most significant changes concerning union influence in politics. The focus will be on the last two periods because the changes form evidence of an end to the Danish model of high influence of unions in the policy process (e.g. corporatism, agenda setting activities, political influence and external legitimacy). The main argument of the article is twofold: on the one hand, labour market policy and activation reforms at the beginning of the 1990s strengthened corporatism in Denmark, but political developments since have reduced its role; and now corporatism seems to be threatened or abandoned in Denmark. Is a path-breaking political development to be recorded? An imbalance between rights and obligations in the Danish system can already be observed, special problem groups have felt reductions of benefits and benefit periods, and stronger punishment has been introduced in the system. The central question is whether the unions have been able to prevent this from happening and to protect the institutions and goods, which are of utmost importance to their collective capability to mobilize and act. The relative balance of power amongst actors is at stake. In addition, the unions are now losing members, as in most other European countries. The article will be organized into four main sections: the following section will give an overview on the main theoretical approaches that are important for the understanding of unions in the welfare reform process. Following that, we will discuss the influence of Danish unions in activation, and welfare reforms and recent political developments under crisis conditions. Finally, in the concluding section, we will answer the questions raised above whether the reforms indicate a backsliding from the Danish ‘model’. © 2011 Blackwell Publishing Ltd.
209
Social Policy & Administration, Vol. 45, No. 2, April 2011
Theoretical Approach and Problems as to Research Questions According to our twofold argumentation, the theoretical approach will be twofold as well. On the one hand, research on welfare state retrenchment has to be acknowledged here. This discussion goes hand in hand with the research on the Scandinavian universal welfare state model. Danish labour market and social policy reforms that started in the 1990s indicate a special path to labour market policy reform by the help of ‘activation’. On the other hand, research on union influence on welfare policy development must also be taken into account, and the relationships between social policy developments and labour market regulations have to be scrutinized. Many researchers have referred to the Danish welfare state as a part of the universal welfare state regime and a ‘model country’ in this respect. Denmark is characterized by a high level of social rights and universalism (EspingAndersen 1990: 27–8). However, liberal elements and market orientation have always been important parts of the Danish economy. In this context, several researchers have pointed at the significance of the Danish and the Swedish welfare state as representatives of the Nordic model too (Kautto et al. 1999). Denmark has developed a special mixture of Scandinavian welfare state arrangements and liberal traditions as to labour market regulations. Denmark is one of the countries with the most developed social security systems, and the country with the strongest labour market policy in the world at the moment. But at the same time there exists a long tradition of liberal practices within the labour market and in some policy fields as well. The politics of industrial relations is most important: the private employer and employee organizations can, on a voluntary basis, make collective agreements without the intervention of politicians. The agreements regulate many issues normally dealt with in Parliament – examples are working time, minimum wages, holiday arrangements and co-determination. The liberal tradition also manifests itself in the preferred use of soft policy instruments to hard ones. Moreover, employers are generally free to hire and fire and they do not have to pay high social contributions – the tax-burden is on the employees. Since 2000, employers do not even contribute to the financing of the labour market policy – but they have been placed in pivotal decision-making positions within the system ever since the start of the labour market policy in the 1960s. One major characteristic of the Danish model is the high degree of unionization and the important role unions played in the policy-making process (Composton 1995). On the one hand, labour market partners influenced social policy reforms, they were part of labour market commissions that prepared reforms and, besides which, they always played a significant role in the implementation process of these reforms (Jørgensen 2003). One central element of explanation is the Ghent system, the special unemployment scheme that combines unemployment insurance and trade union membership (Lind and Møller 2006: 10). Consequently, unions played an important role in employment politics. In times of welfare state expansion, unions played a major role in the political process by fighting for the expansion of social rights and benefits. 210
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Denmark and Sweden are prominent examples of high union influence in welfare state development and expansion. In recent literature, Denmark is still seen as a successful welfare state. Politicians and researchers speak highly of the Danish system in general and the labour market regulations in particular (Jørgensen and Madsen 2007). Recent reforms, however, indicate that Denmark might be leaving the Nordic path of generous welfare benefits and high trade union involvement in social and labour market policy. Our results imply a revised picture of the Danish system.
From a Corporatist Model Country to a Deposed Conqueror: Trade Union Influence in Activation Reforms Trade union influence from the early 1990s to 1998 The 1990s became a strong reform period. The Parliament ‘Folketing’ passed the first activation reform in 1993 with a sweeping majority (Bogedan 2005: 17). The new social democratic led government wanted to reduce unemployment and increase the willingness of people to take up employment and the possibilities to do that (Goul Andersen and Pedersen 2007: 12; Larsen 2005: 123–4; Lind and Møller 2006: 7). The law limited unemployment benefit to seven years and divided it into two periods: a passive time (four years) and an active time (three years). Thus, every recipient had the right and obligation to activation after four years. Besides which, individual action plans to improve the job placement with the addition of education and job rotation as important measures were introduced (Larsen 2005: 124). ‘Activation’ was to have active measures, qualitative good offers to unemployed people in order to give people ‘new chances’. In addition, the law determined a decentralization and flexibilization of labour market policy. The government established 14 regional labour market boards, which are tripartite bodies. Unions, employers and municipalities decided which activation programmes were to be implemented in the regions (Etherington and Jones 2004: 29; Jensen 1999: 7), and the first reform indicated an important strengthening of regional corporatism (Jørgensen 2003; 2009/2010). Trade unions did not only play a vital role in the implementation process but also in the preparation of the law. In the Zeuthen Commission, which prepared this reform, unions had been strongly represented, besides which, employers, politicians, civil servants and experts participated (Madsen 2005b: 284; Martin and Thelen 2007: 26; Mailand 2005: 139), and the government adopted the requirements of the commission (Larsen and Goul Andersen 2008: 9–10). The work of the Zeuthen Commission is seen as a successful example of corporatist policy-making (Mailand 2005: 139; Martin and Thelen 2007: 26). The government passed the second labour market reform in 1996. This reform marked a continuation of the activation path (Goul Andersen and Pedersen 2007: 13). The duration of unemployment benefits was further reduced to five years: two years passive and three years active ( Jensen 1999: 5). Unemployed people were obliged to take up every job after a certain time and would have to accept a commute time of four hours per day. © 2011 Blackwell Publishing Ltd.
211
Social Policy & Administration, Vol. 45, No. 2, April 2011
The preparation of the next labour market reform differed from the first. Civil servants prepared the second reform and unions took no part in the policy-making process. This period was mainly characterized by an increasing influence on administrative corporatism. However, the second reform, especially, signalled the beginning of the change of elements in the Danish system.
Continuing loss of influence: union and activation reforms between 1999 and 2006 The activation reforms between 1999 and 2006 indicated the first loss of trade union influence in the policy-making process. The first revisions shortened the benefit period to five years and changed leave programmes. The third and fourth activation reforms became effective in 1999. The third labour market reform specified a reduction of the unemployment benefit duration to four years and strengthened incentives for early retirement (Lind and Møller 2006: 8; Goul Andersen and Pedersen 2007: 13). The unemployment figures were falling rapidly and new jobs were being created. As opportunities within the labour market increased, the benefit period could be reduced without any serious threat to most wage earners. The fourth activation reform differed from the other reforms insofar as social assistance law (and not unemployment) was reformed. Everybody was obliged to take up a job or participate in activation programmes. But unemployed people still had clear influence on their own action plan. The social democratic government, which was in office until 2001, started the change to the Danish way of policy-making. Trade unions did not play a pivotal role in these later processes, as civil servants and policy experts prepared the reforms. As the unions did not have significant influence on the reforms, it is difficult to label the reforms further ‘revival of corporatism’ (Goul Andersen 2001: 20). Instead, it is more appropriate to speak of the first change to the Danish system. In November 2001, the new liberal-conservative government under the lead of Anders Fogh Rasmussen took office. The change of government immediately threatened the power resources of the unions. The proposal of a state-run unemployment insurance fund was not supported at the parliamentary level, but another of establishing cross-occupational unemployment insurance funds was. This broke the dominance of the Ghent system in Denmark and it soon weakened the collective capabilities of the unions. Next, the right-wing administration implemented the reform initiative More People to Work just one year later ( Jørgensen 2009/2010; Lind and Møller 2006). Since then, everybody has had to accept every job offered from the first day on which they were to be activated. Individual action plans were substituted by narrow ‘job plans’, alongside which stricter controls for unemployed and social assistance recipients were implemented. In addition, the government reformed the unemployment service insofar that it also affects the role of unions. Private providers are allowed to offer activation programmes and, actually, they have taken over most of the activation ‘industry’. The qualitative aspects of activation – especially educational measures – have been reduced strongly; and as private producers are to cover all expenditures themselves, cost commitment is a preferred strategy. This goes hand in hand 212
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
with the reduced quality of the activation offers (Socialpolitik 2010) and increased constraint for self-help in terms of finding a job (before the unemployed are activated). The More People to Work reform has sweeping implications for the role of unions: first, it results in outsourcing and private providers are out of reach for unions. Second, it could also result in the loss of union membership, due to private providers, the lack of connection to unions and the new pressures put on unemployed people, which may decrease the incentive to be a union member (Lind 2007: 57). However, unions remained responsible for part of the unemployed, which partly explains why unions did not antagonize the reform (Goul Andersen and Pedersen 2007: 21; Lind and Møller 2006: 9). In 2006, another activation reform was passed (Goul Andersen and Pedersen 2007: 19; Jørgensen 2009/2010: 351) – the Welfare Compromise. This reform introduced more systematic controls of unemployed people: frequent contacts were installed; each unemployed person had to verify that he or she took the job search seriously; until 2010, the unemployed person had to write four job applications per week – even if no job openings were to be found; if unemployment lasted longer than nine months, the unemployed person was obliged to participate in an activation programme – or lose his or her benefits. The government aim was for ongoing commitment and faster job placement of the unemployed and social assistance recipients (Goul Andersen and Pedersen 2007: 21). In this second reform period, administrative corporatism had become less important, too, which we will comment on in the next section. The inclusion of the unions in all kinds of law-making via corporatist commissions has been reduced strongly over the years: during a ten-year period in the 1960s/1970s, the unions were represented in 53 commissions, while during the 2000s in only 18 (Klitgaard and Nørgaard 2010: 23). This is another example of the fall of political corporatism in Denmark. And the unions were threatened by new governmental interventions in industrial relations and by efforts to weaken the Ghent system.
A new era: union influence and activation reform since 2007 The year 2007 marked a new era in public policy-making in Denmark. On the one hand, the activation path proceeded with further restrictions and stricter eligibility criteria. It was only job placement that counted – unemployed people should now fear having activation started. More economists were talking of strengthening the ‘motivation effect’, e.g. sticks instead of carrots. On the other hand, reforms go hand in hand with a quitting of the Danish way of corporatist policy-making. Three reform initiatives must be highlighted here: the Welfare Compromise (2006/2007), the reform in 2009, which was only the preliminary endpoint of the decreasing union influence and, finally, a crisis package in 2010. The Welfare Compromise of 2006/2007 was developed by the Welfare Commission which worked between 2004 and 2006 without any significant influence from the unions; social partners had the chance to take a stand after the commission finished its work (Velfærdskommissionen 2005; 2006). The LO dealt intensively with the claims of the commission and intensively criti© 2011 Blackwell Publishing Ltd.
213
Social Policy & Administration, Vol. 45, No. 2, April 2011
cized some of them but unions had become lobbyists only. The recommendation to reduce unemployment benefits to 2.5 years was one of the main points of union criticism. From the union point of view, this recommendation was incompatible with the Danish welfare model, and the government had to revise some of the claims after the loud criticism of the unions (Kvist 2009: 39). Finally, the reform was passed with a vast majority. Some of the major components are summarized by Jørgensen (2009/2010). Stronger ‘work first’ elements were decided on which were reflected on the content side by the introduction of stricter activation demands, new programmes and clear time schedules to be followed by the employment authorities. On the administrative side, decision-making arrangements were changed again: 91 new job centres were created in 2007, which were to be responsible for the unemployed and social assistance recipients; 77 of the job centres had a shared leadership of state and municipal representatives – clearly a provisional ‘solution’; 14 had municipalities responsible for all kinds of unemployed people. Most important at that time for the question of union influence was the abolition of regional labour market policy and regional corporatism. Social partners were placed in new advisory roles only – they were no longer part of public administration and no longer responsible for the activation programming. So it is easy to see that unions have lost influence in political as well as administrative corporatism. Governance structures were changed in ways excluding the social partners. Furthermore: from 1 August 2009, municipalities took over formal responsibility for labour market policy, along with the content of activation measures, the decision-making competences, and the financing of measures as well – even if the state had installed a complicated system of refunding most expenditure, but also controlling the municipalities. In case the municipalities activate insured unemployed people, they would get a refund of 75 per cent of expenditure; otherwise, they would have to pay 50 per cent themselves. In case of people on social assistance, the figures were 65 per cent and 35 per cent. The government believed in economic incentives – but also in the fruitfulness of having the unions outside the institutional set-up for decisionmaking and implementation of labour market policy. The unions were having trouble using the advisory role to influence policy, and the abolishing of regional labour market policy was a loss to the social partners. Also, the employer side was strongly against this change of organization of labour market policy. The 28 private unemployment insurance funds were being strongly controlled and benchmarked by the state. They had new responsibilities in employment policy – but they did not get money from the state to carry out these tasks. New tensions between the unemployment insurance funds and the municipalities also arose because the municipalities would have to bear economic consequences of decisions taken in the unemployment insurance funds as to the question of unemployed people being ready to take up a job in the open labour market. Finally, in June 2010, the right-wing government – now headed by the liberal Lars Løkke Rasmussen – succeeded in having a plan for ‘restoring’ the economy realized. The economic crisis and the deficit in the public budget 214
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
were used as a reason for cuts in budgets and direct attacks on interests of the unions. Members of the unions were hit directly, and the position of the unions was weakened once more: now tax reductions for trade union membership fees have almost been abolished – having the low fees of ‘yellow’ unions as a ceiling. Second, the requirements for re-establishing the right to unemployment benefits are doubled, and, third, the length of the unemployment benefit period is being reduced from four to two years – without any kind of compensation as for example higher benefits. This shortening of the benefit period and the loss of rights for welfare state support caused strong protests. It was clear that the decisions contributed to undermining the power resources and action capabilities of the unions.
Leaving the Nordic Path? Trade Union Influence in Retrospect Denmark has for a long period been considered as a ‘model country’ with a comprehensive welfare state and a successful model of corporatist policymaking. During recent years the ‘flexicurity’ system of labour market performance has further contributed to this status (Madsen 2005a; 2005b; Jørgensen and Madsen 2007). However, our analysis documents rather profound changes in the arrangements and in the status of the social partners in general and the unions specifically within the system. Recalibration of existing welfare state arrangements and changes as to the content of labour market policy during recent years witness that the position of the unions in the Danish system has been altered – we think in a systematic and path-breaking way. The unions’ collective capabilities of mobilizing and acting have seriously been weakened. This can be measured along a number of operational points. Today, the trade unions are no longer included in preparations of law-making as this happens at the parliamentary level – often in closed finance bill negotiations – and inclusion in commission work cannot be taken for granted (Klitgaard and Nørgaard 2009; 2010; Christiansen and Klitgaard 2008; 2010). This indicates a profound change in the policy style ( Jørgensen 2002; 2009/ 2010). Corporatist policy-making was a major characteristic of the Danish system. Social partners were always involved in the policy-making processes up until the end of the 1990s. Now things have changed, especially since the right-wing government took office in 2001. It is not only political corporatism that has been fading; but also administrative corporatism. Regarding the organization of decision-making in the labour market policy area, trade unions have lost influence and they have been formally excluded since 2007. Since 2009, they no longer have any administrative function to perform and no political corporatist structures exist as the municipalities have taken over formal responsibility for labour market policy – but in a schizophrenic way. The municipalities are fully responsible for activities, but at the same time the state is strengthening its control arrangements and its power potentials. Now it is a question of steering relations between the state and the municipalities. It is no © 2011 Blackwell Publishing Ltd.
215
Social Policy & Administration, Vol. 45, No. 2, April 2011
longer the state vis-a-vis the social partners in pivotal positions within the system. Recalibration of governance structures has made unions more powerless. And, finally, as to the content of the policy, profound changes have taken place giving the Danish labour market policy a different profile than was the case during the 1990s. The offensive and ‘active’ labour market policy which became famous as part of the Danish flexicurity system is no longer in use as activation measures have been qualitatively and quantitatively reduced and the policy has been turned into a more ‘work first’-based system. Sticks have been preferred to carrots. At the same time, the target groups have been broadened, and including people on social assistance has put these on an equal footing with insured unemployed people. This again has reduced the importance of trade unions and of the Ghent unemployment insurance system. The duration of the unemployment benefit period has been drastically reduced – from seven to two years – and the requirements of re-entering the benefit systems have been strengthened. Besides, full tax reductions for ‘normal’ trade union fees have been abandoned. The pattern is of of trade union unfriendly governmental policies. Unions now have to operate more as lobbyists instead of being part of strong corporatist institutions and decision-making arrangements. They used to be both policy-makers and policy-takers, which created responsibility and commitment. Besides which, an anti-collective and non-egalitarian ideology has been mushrooming during the last decade, contributing to the de-legitimization of collective action. During recent years it has also been quite unclear which kinds of issues are to be handled by Parliament or by the private collective bargaining system. This is in relation to questions like equal pay, equal treatment of men and women, maternity leave and caring days. These are often questions of the highest relevance for women and the work– family balance. Political and professional regulations now interact and supplement each other. This gives the political system good opportunities to influence the regulation of the labour market and the vital interest representation of wage earners. From our analysis we must draw the conclusion that the Nordic approach to growth and welfare seems to have one of its pillars eroded in Denmark: the social partner basis and the collective trust in partnership. Moreover, the unions suffer the most. The present right-wing government openly enacts policies hostile to unionization and trade union action. Traditionally, the political colour of the government is a very important factor as to power and influence in a system. Right-wing governments want to weaken the strong position of the unions in ways such as to exclude trade unions from the preparation of important political decisions, to weaken the position of trade unions in administrative bodies, to recalibrate policy arrangements and to change the content of policies. The opportunity structure of the unions has been changed in Denmark, despite which the relationship between the colour of the government and labour market policy decisions are not that clear and undisputed. Especially not in systems in which political compromises have been cultivated, and which historically include both sides of the political spectrum as well as the social partners ( Jørgensen 2002). In a Nordic ‘flexi216
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
curity’ system, interests and institutional arrangements are interrelated in a much more complex way, and the ‘institutional complementarities’ produced are traditionally supplying the government with incentives not to exclude actors from decision-making arrangements or to favour one party only. However, this is exactly what has happened in Denmark. Competences and decision-making power have been changed quite fundamentally: the abilities of the unions to ensure resources have been weakened, opportunity structures have been recalibrated, and legitimacy of collective action have been openly questioned. Union resources and capabilities have been strongly shaped by governmental action – even if they have not been totally damaged by the developments. This must also be remembered. But in this respect the Nordic path is no longer followed in Denmark. Social dialogue and partnership yield diminishing returns; now they function mostly as a kind of ‘damage control’ mechanism. No direct or perfect ‘fit’ between the colour of the government and the outcome of decision-making is to be expected in any system. Now the Danish trade unions have suffered a severe loss of power resources; and in the short run, they have fewer chances of securing members’ interests as to political power. Political influence and power are functions of union vitality and collective intelligence. It might take time for the trade unions to learn how to be architects of ‘modernization’ processes and constructive veto-players at the same time – no matter what kind of government is in place. A broad and internationally oriented political project supplementary to the professional collective agreement nexus in ‘the Danish Model’ is still missing, we think. But one thing is certain: social scientists will have to think in a more structured and realistic way about the character and future of the Danish welfare and labour market ‘model’.
References Allern, Elin H., Aylott, N. and Christiansen, F. J. (2007), Social Democrats and Trade Unions in Scandinavia: The Decline and Persistence of Institutional Relationships, European Journal of Political Research, 46, 5: 607–35. Bogedan, C. (2005), Mit Sicherheit besser? Aktivierung und Flexicurity in Dänemark, ZeS-Arbeitspapier Nr. 6/2005, Bremen. Cameron, D. R. (1984), Social Democracy, Corporatism, Labor Quiescence and the Representation of Economic Interest in Advanced Capitalist Society. In J. H. Goldthorpe (ed.), Order and Conflict in Contemporary Capitalism, Oxford: Clarendon Press, pp. 143–78. Christiansen, P. M. and Klitgaard, M. B. (2008), Den utænkelige reform. Strukturreformens tilblivelse 2002–2005, Odense: Syddansk Universitetsforlag. Christiansen, P. M. and Klitgaard, M. B. (2010), Afmagtens mekanismer: Den danske fagbevægelse og arbejdsmarkedspolitikken siden 1960’erne, Politica, 42, 1: 5–26. Composton, H. (1995), Union Participation in economic policy making in Scandinavia, 1970–1993, West European Politics, 18, 1: 98–115. Esping-Andersen, G. (1990), The Three Worlds of Welfare Capitalism, Cambridge: Polity Press and Princeton, NJ: Princeton University Press. Etherington, D. and Jones, M. (2004), Welfare-through-work and the re-regulation of labour markets in Denmark, Capital & Class, 28, 2: 19–45. © 2011 Blackwell Publishing Ltd.
217
Social Policy & Administration, Vol. 45, No. 2, April 2011
Goul Andersen J. (2001), From the Edge of the Abyss to A Sustainable Third Way Model? Unemployment and Unemployment Policies in Denmark 1975–2000, Working Paper COST A 13, Working Group Unemployment, Aalborg. Goul Andersen, J. and Pedersen, J. J. (2007), Continuity and Change in Danish Active Labour Market Policy 1990–2007: The Battlefield between Activation and Workfare, CCWS Working Paper, Aalborg. Jensen, P. H. (1999), Activation of the Unemployed in Denmark since the early 1990s. Welfare or Workfare?, CCWS Working Paper, Aalborg. Jørgensen, H. (2002), Consensus, Cooperation and Conflict: The Policy Making process in Denmark, Cheltenham: Edward Elgar Publishing. Jørgensen, H. (2003), The Role of the Trade Unions in Social Restructuring in Scandinavia in the 1990s, Revue Française des Affaires Sociales, 57, 4: 151–76. Jørgensen, H. (2006/2007), Arbejdsmarkedspolitikkens fornyelse – innovation eller trussel mod dansk ‘Flexicurity’?, Brussels and Copenhagen: ETUI-REHS. Jørgensen, H. (2009/2010), From A Beautiful Swan to An Ugly Duckling: Changes in Danish Activation Policies since (2003), European Journal of Social Security, 11, 4: 337–68. Jørgensen, H. and Madsen, M. P. (eds) (2007), Flexicurity and Beyond – Finding a new agenda for the European Social Model, Copenhagen: DFØF Publishing. Kautto, M. et al. (eds) (1999), Nordic Social Policy: Changing Welfare States, London and New York, NY: Routledge. Klitgaard, M. B. and Nørgaard, A. S. (2009), Arbejdsmarkedsreformer med og uden kommissioner. In J. Grønnegaard Christensen, P. E. Mouritzen and A. S. Nørgaard (eds), De store kommissioner. Vise mænd, smagsdommere eller nyttige idioter, Syddansk Universitetsforlag, Odense, pp. 91–116. Klitgaard, M. B. and Nørgaard, A. S. (2010), Afmagtens mekanismer: Den danske fagbevægelse og arbejdsmarkedspolitikken siden 1960’erne, Politica, 1: 5–26. Korpi, W. (1978), The Working Class in Welfare Capitalism, London: Routledge. Korpi, W. (1983), The Democratic Class Struggle, London: Routledge (Swedish original 1981). Korpi, W. and Palme, J. (2003), Klassenpolitik und Wohlfahrtsstaatsabbau: Kürzungen von Rechten der sozialen Sicherung in 18 Ländern 1975–1995. In W. Müller and S. Scherer (eds), Mehr Risiken – Mehr Ungleichheit? Abbau von Wohlfahrstsstaat. Flexibilisiering von Arbeit und die Folgen, Frankfurt am Main: Campus, pp. 221–55 (for English version, see (2003) American Political Science Review, 97, 3: 425–46). Kvist, J. (2009), De nordiske modellene etter 2000 – udviklingen i Danmark, Fafo notat 10/2009, Oslo: Fafo. Larsen, C. A. and Goul Andersen J. (2008), How Ideas Can Have an Independent Causal Effect on Policy Change: The Case of New Economic Ideas that Changed the Danish Welfare State, Paper for the 4th International Conference on Welfare State Change: Policy Feedback, the Role of Ideas, and Incrementalism, Store Restrup Herregaard, 30 January–1 February 2008, Aalborg. Larsen, F. (2005), Active Labour Market Policy in Denmark as an Example of Transitional Labour Markets and Flexicurity Arrangements – What can be learned? In T. Bredgaard and F. Larsen (eds), Employment Policy from Different Angels, Copenhagen: DJØF Forlag, pp. 115–33. Lind, J. (2007), A Nordic Saga? The Ghent System and Trade Unions, International Journal of Employment Studies, 15, 1: 49–68. Lind, J. and Møller, I. H. (2006), Activation for what purpose? Lessons from Denmark, International Journal of Sociology and Social Policy, 26, 1/2: 5–19. Madsen, P. K. (2005a), How Can It Possibly Fly? The Paradox of A Dynamic Labour Market in Scandinavian Welfare State, CARMA Research Paper 2/2005, Aalborg.
218
© 2011 Blackwell Publishing Ltd.
Social Policy & Administration, Vol. 45, No. 2, April 2011
Madsen, P. K. (2005b), The Danish Road to ‘Flexicurity’: Where are we? And How Did We Get there? In T. Bredgaard and F. Larsen (eds), Employment Policy from Different Angels, Copenhagen: DJØF Forlag, pp. 269–89. Magnusson, L., Jørgensen, H. and Dølvik, J. E. (2008), The Nordic Approach to Growth and Welfare – European Lessons to be Learned?, Brussels: ETUI. Mailand, M. (2005), The Involvement of Social Partners in Active Labour Market Policy – Do the Patterns fit Exceptions from Regime Theories? In T. Bredgaard and F. Larsen (eds), Employment Policy from Different Angels, Copenhagen: DJØF Forlag, pp. 135–51. Martin, C. J. and Thelen, K. (2007), The State and Coordinated Capitalism: Contributions of the Public Sector to Social Solidarity in Postindustrial Societies, World Politics, 60, 1: 1–36. Minnich, D. J. (2003), Corporatism and Income Equality in the Global Economy: A Panel Study of 17 OECD Countries, European Journal of Political Research, 42, 1: 23–53. Rothstein, B. (1992), Labour Market Institutions and Working Class Strength. In S. Steinmo et al. (eds), Structuring Politics: Historical Institutionalism in Comparative Politics, Cambridge: Cambridge University Press, pp. 33–56. Rothstein, B. (2003), Sociala fällor och tillitens problem, Stockholm: SNS Förlag. Socialpolitik nr. 1 (2010), Tema: Vejen til arbejde?, København: Socialpolitisk Forening. Velfærdskommissionen (2005), Fremtidens velfærd – vores valg, København. Velfærdskommissionen (2006), Høring om Fremtidens velfærd – vores valg, København. Visser, J. (2006), Union Membership Statistics in 24 Countries, Monthly Labour Review, 129, 1: 38–49.
© 2011 Blackwell Publishing Ltd.
219
