Prepared by Joakim Palme, Institute for Future Studies, Stockholm
Discussion paper prepared for the UK Presidency, October 2005
The issue (1)
In the wake of the present political crisis of European integration, the future of the European Social Model has attracted new attention. This is partly following the increased understanding that the pursuit of a Single European Market can only be successful if the European populations support it. This will only happen if there is trust in governments to handle the social consequences of production and service markets being exposed to competition. The ‘destructive’ forces of market competition have to be met by ‘constructive’ social policies. Here a number of studies have identified that Scandinavian countries are best performers in combining excellent poverty and inequality reductions with high employment rates and growth records that have beaten most European countries over the past decade. When it comes to core issue, in the what the Commission has called the ‘modernisation’ of the European social dimension, the Scandinavian countries also appear to have something to contribute:
(i) they have developed and reformed their pension systems and elderly care programs,
(ii) the gender balance is addressed in a much more elaborated way than elsewhere, and
(iii) working life is distinguished by high quality jobs and flexible labour markets. This boils down to an urgent need to design policies that can resolve the presumed trade-offs between equality and efficiency.
Two questions
Can we identify the policies that have worked? Can such policies be used in other countries and therefore in context that are different? In what follows, I will argue that we should answer both questions with a yes. This ‘yes’ is however not unconditional. We need, firstly, to sort out some of the misconceptions about the nature of the Scandinavian model, and, then, to be specific about how institutions have been designed and - this is important - combined. The two steps of the analysis are vital if we want to avoid the temptation resort to prejudice and/or wishful thinking. Instead, we should base our policy recommendations on a knowledge-based assessment of how the world works. Another critical factor is not to confuse the goals with the means. We should be faithful to our goals but ruthless in examining if the means we use are adequate.
Which are the misconceptions?
One important misunderstanding is about costs. The standard comparisons are based on gross expenditures, which exaggerate the costs in the Scandinavian welfare states. This has to do with fact that in the Scandinavian countries, benefits are taxable and social services expenditures are made up of wages to employees in the social service sector, which are also subject to taxation. This means that a substantial part of the gross expenditures will flow back to the public purse. Other European countries not only spend less on social services, they also tend to have stronger reliance on wage continuation schemes (not included in public social expenditures) and have various tax exemptions for social benefits as well as a stronger use of tax expenditures. OECD will soon publish new data on net expenditures (where most of these accounting differences have been adjusted for), which are likely show that Sweden have lower expenditures than France and Germany and Denmark are on British levels.
Few observers identify the crucial importance of market incomes for explaining the relatively good positions of vulnerable groups such as families with children and, in particular, single mothers and their children. This suggests that we can make politics work with markets in a way that is also socially equitable.
The importance of a good incentive structure is of great importance for equality and efficiency objectives alike, which is rarely recognised. But the reliance on universal benefits rather than means-tested ones is an excellent way to avoid poverty traps that threaten to lock recipients into ‘benefit dependency’. For Scandinavian women, the early introduction separate taxation of spouse income in combination with heavily subsidised social services gave good incentives for them to become taxpayers. Comparatively generous social insurance benefits are balanced by demands on ‘activation’ on the behalf of benefit recipients.
Yet another myth about the ‘big’ Scandinavian welfare states is that cost control is bad. With exception of the deep employment crisis in Sweden and Finland during the early 1990s, the Scandinavian countries have run surpluses in their budgets. We can also see that the some of the new reforms score very high on cost control. The Swedish pension reform with its reliance on a defined contribution formula is perhaps the most extreme case, where contribution rates are fixed [to the next ice age if we are to believe some of the reformers].
The design of each program is critical when it comes to the equality/efficiency criteria. But we should acknowledge that the combination of policies might be critical for the outcome. I hinted to one such instance above in connection to incentives and services as beeing important for triggering increased female labour supply. But we should also understand that this had been preceded by the entry of young women at all levels of education and coincided with labour shortage in the private sector as well increased public employment. Here there are obvious parallels with the current British development.
There is a long history of free-trade policy, exposure to foreign competition and successful exploration of international markets. This has led to dramatic structural transformations of economic structures. These changes were politically accepted because of the strong commitment to take care of the victims of the restructuring. What Schumpeter labelled the destructive forces of capitalism have been balanced by socially constructive forces in terms of unemployment benefits and retraining. Interestingly, the trade strong trade unions have in various ways been playing an active role in reinforcing this restructuring. The so-called ’solidaristic wage policy’ of the Swedish LO is an excellent example. But the gains of these policies have been obvious in terms of the news jobs having better quality and higher pay.
What could be learnt about policy design?
Scandinavian countries have established a universal model of social protection, where benefits and services based on residence are combined with earnings-related social insurance programmes. Social polices are combined with strong emphasis on education and active labour market policies as well as flexible labour markets.
History
History shows that the reforms of social policy programmes in the Scandinavian countries are responses to changing economic and social structures, as well as to political mobilisation. The important phases in this development have involved clear elements of learning and diffusion. The universalism emerged as a response to the different needs of the rural and urban populations, as well as to the political mobilisation of these interests. The post-war period expansion of earnings-related social insurance was a strategy to include workers and salaried employees, and both public and private sectors, within the same system of protection. The expansion of services has partly been a response to ageing populations but is also intrinsically connected to the growth in female labour force participation as well as by political participation by women.
The strategy of redistribution
The various social policy models follow different redistributive strategies (Korpi and Palme 1998). The targeted model follows the same principles as Robin Hood applied by following the means-testing principle, and only taking from the rich and giving to the poor by financing benefit payments from general taxation. The basic security model follows a simple egalitarian strategy by paying flat-rate benefits, i.e. providing the same benefit levels to both rich and poor. The corporatist model, in its classical form, redistributes resources primarily within different corporations. The comprehensive model, by relying on universal earnings-related social insurance benefits gives in fact more to those who already have, following the preaching of Matthew rather than that of Robin Hood. However, in reality, most countries apply combinations of different kinds of programme and the relevance of different principles varies between different sectors. Even if the earnings-related principle might be central to social insurance it is less applicable in, for example, the provision of social services.
The core issue is whether there are trade-offs between the provision of different kinds of benefits. In reality, and contrary to the expectations of many, it seems to be the case that the more the middle class is involved in the welfare state, the better the situation will be for vulnerable groups and the more social inequalities will actually be reduced by tax/transfer programmes. The reason might be found in the way interests are organised in different kinds of social policy models. Here, a vital distinction has to be made between the distributive profile of benefits (and taxes) and the size of the sums that become subject to redistribution. Moreover, there appears to be a correlation between the distributive profile and the size of sums for redistribution - the more benefits are targeted, the smaller the sums will become. This gives rise to a strongly positive correlation between the size of sums and the size of inequality reduction. Hence, paradoxically, the more benefits are targeted to the poor in a country, the smaller is the reduction in inequality achieved by the welfare state (Korpi and Palme 1998).
But the success of the Scandinavian model cannot be explained without understanding how improved incentives, human resources, social services and job opportunities have promoted employment and participation, particularly among women. It is also somewhat paradoxical, that the importance of market incomes also for vulnerable groups is explaining the good social policy performance.
The life-cycle perspective
The Scandinavian success in combating poverty and social inequalities is perhaps best exemplified by declining poverty rates among families with children and among the elderly. These groups represent the phases in the life cycle that historically have been the most poverty-stricken and where we still find large variations from country to country, even among the most advanced industrial nations. The results from a study by Kangas and Palme (2000), based on the Luxembourg Income Study, give some indications. The analysis was based on data from different points in time, the first around 1970 and the second around 1990, and different countries representing different social policy strategies. Major differences were found both over time and among countries. A central tendency was that poverty rates showed a more cyclical pattern at the earlier point in time. The cyclical pattern was similar to that Rowntree had observed, i.e. poverty was higher among families with children and among the elderly. In the 1990s, this pattern had vanished in the Scandinavian countries included in the study. In Canada, where old-age poverty had become very low, families with children still faced clearly higher poverty risks. The cyclical component was still strong in the United States. Even if old-age poverty had gone down a little compared to the situation in the 1970s, poverty had gone up among families with children.
Family policy for children
The analysis family policy institutions in the EU-countries show that they are important factors for explaining cross-national patterns of gendered distributions of paid and unpaid work, socio-economic outcomes, such as poverty patterns among children, and childbearing Institute for Futures Studies 2004). One important linkage seems to be the way family policy institutions simultaneously support or discourage female labour force and childbearing.
The most common type of family policy model, predominant not only in Continental European countries but also in new member states supports a male breadwinner family type, where paid and unpaid work is highly divided so that fathers have the main responsibilities for carrying out paid work on the labour market whereas mothers carries out the lion’s share of reproductive care work in the home. Benefits in support of such a division of work include for example childcare leave benefits paid in low flat-rate amounts as well as various mainly flat-rate cash and tax benefits in support of children and the dependent spouse (wife). Earnings-related parental leave benefits to both parents hardly exist and childcare for the youngest children is less developed. In countries with male breadwinner models female labour force participation and fertility rates are generally at a comparatively low level, with the exception of France. Poverty levels among children are typically at medium levels, possibly because male breadwinner support is generous enough to stimulate mothers to stay at home to care for the children at the same time as labour market participation and women’s earnings are held at a lower level than what would have been possible otherwise.
The other main type of family policy is the dual earner model developed in the Scandinavian countries. This family policy model supports female labour force participation and to a large extent also male participation in care of children. The main characteristics of this model are parental leave benefits directed to both parents, which are generous both regarding duration and earnings replacement, and publicly subsidised childcare after the termination of parental leave. This model of family policy has enabled these countries to combine high levels of female labour force participation with relatively high levels of fertility. Countries with dual earner models of family policy also have the smallest gender differences in care work, which probably is partly due to family policy measures supporting male involvement in care of children. Due to the combination of generous earnings-related benefits, publicly subsidised services and high female labour force participation, poverty levels among households with children are kept at low levels in these welfare states.
A third model of family policy has neither support to the traditional family nor to the dual earner family is well developed; instead parents have to rely on the market to provide such arrangements. In countries with market-oriented models of family policy poverty levels are highest while fertility levels and female labour force participation rates are at medium levels.
Thus, whereas increased family policy generosity overall seems to decrease poverty and facilitate childbearing it appears to have a more ambiguous relationship to female labour force participation. High female labour force participation is in turn related to both higher fertility levels and lower child poverty. Comparative analysis evidence indicating first that child poverty has become associated with only one earner in the family. Thus the traditional family model seems less and less able to provide for child welfare. This must raise concern about future human capital formation in several different dimensions. The potential to achieve the desired number of children for families that rely on a single breadwinner are at the very least not favoured by this trend. Nor can we expect the cognitive development of children living below poverty levels to be competitive.
Evidence on educational performance indicates that educational attainment of mothers is important and possibly increases in importance over time. Nor can we find any detrimental effects from maternal labour force participation in the data from the PISA study. Combined, those two pieces of evidence give reason to believe that the need to enrol more women in the labour force in the future need not have negative effects on the cognitive development of children. Rather it is likely to yield positive effects on educational attainment by reducing the risks of child poverty.
Insuring the labour force: life long learning
The analysis Eurobarometer data suggests that a rather clear distinction can be made between some Northern European countries, above all the Scandinavian countries and countries in southern Europe, as regards attitudes related to participation in, experiences of and motivations for learning (Institute for Futures Studies 2004). The Scandinavian countries show:
- A small ’skills gap’; a low proportion of individuals stating that they do not possess the skills that are important for them in working life.
- A small ‘credentials gap’; a low proportion of individuals stating that they do possess the skills that are important for them in working life, but that they can not provide concrete evidence to support this.
- A high proportion of individuals have recently taken part in studies or training.
- A high proportion of individuals stating that there would be no obstacles for them to take part in training or studies.
- A small proportion of individuals stating that there is nothing that could encourage them to take up studies or training.
Social citizenship at old age
The analysis of the Luxembourg Income Study, give evidence of cross-national variation in old-age poverty (Kangas and Palme 2000). A central tendency was that poverty rates showed a more cyclical pattern at the earlier point in time. In the 19070s, the cyclical pattern was similar to that Rowntree had observed, i.e. poverty was higher among the elderly. In the 1990s, this pattern had vanished in the Scandinavian countries included in the study. In Canada, where old-age poverty had become very low, families with children still faced clearly higher poverty risks. The cyclical component was still strong in the United States.
The observed variation from country to country can be linked to the design of the social policy. The generosity of the public pension programmes correlates with poverty rates among the elderly. Here, of course, the universal basic pension in the Scandinavian countries is the most important factor. The relationship between pensions and income inequality involves two kinds of paradoxes. Firstly, the highest degree of inequality in public pensions among the countries we have data for is found in Finland and Sweden. The most equal public pensions are found in Australia and US. Yet, if we examine the distribution of gross income (with all sources of income taken into consideration), the ranking of the countries is largely reversed. This is due to the fact that private pensions and capital income are much more important in countries like Australia and other countries where benefits only provide for basic security. Since private pensions tend to be much more unequally distributed, suggesting that the more private pensions are ‘crowded out’ by earnings-related public programmes, the less inequality there will be. Finland, a country with no income ceiling in its earnings-related programme, is a good case in point.
Secondly, countries that have basic pensions that also benefit the rich are more successful in combating poverty among the elderly than countries that target their basic pensions to the poor. Two different factors are probably contributing to that. The first is that universal provisions have a better take-up, the second that universal design has been successful in creating broad coalitions for maintaining high benefit levels.
The major Swedish pensions reform of the 1990s represents a formidable challenge to anyone wishing to characterise social policy reforms in what are described as modern or mature welfare states. When it comes to the social policy objectives - the provision of both basic security and security of income within the framework of a public, compulsory system - remain unchanged. The means, however, have altered: It is reshaping both the income and basic security components of the system, as well as the role and forms of pre-funding. In short, the reform is, firstly, replacing the old earnings-related defined benefit system with a defined contribution system where basically 18.5 per cent of earnings is the financial basis of the income pension. Secondly, basic security is ensured by a universal guarantee pension. The size of this guarantee is graduated in relation income pension. The defined contribution approach gives both excellent cost control and implies that the life time earnings provides an excellent incentive structure. More work will pay in terms of earned pension entitlements.
A framework for reform
The Scandinavian performance in terms of employment rates and economic growth over the pas decade shows that equality does not necessarily have to be in a major conflict with efficiency (Shapir 2005; Ailinger and Guger 2005). From an egalitarian perspective, the real challenge is to question traditional views about how the world works, and may work in the future. There is also increasing evidence to suggest is that inequality is harmful to efficiency, growth and social cohesion. Good intentions and political commitments are however not enough to make the reform work successful, the design of the existing programmes needs to be critically evaluated. In this context, the Scandinavian tradition of welfare research can make a contribution, not only to study whether the social policy goals are actually being achieved, but also to examine unintended consequences of the programmes.
A framework for reform inspired by the Scandinavian experience starts from a notion that in order to be successful in meeting new needs with restricted resources, as well as in avoiding trade-offs between equality and efficiency, we must improve incentives, human resources, social services and employment opportunities. Social citizenship is about the establishment of a proper balance between rights and responsibilities.
The central questions when it comes to incentives are: How can poverty traps be avoided? How can marginal effects be reduced? How can welfare state programmes be designed so that it pays more to work while entitlements are protected? The Scandinavian rule of thumb is to use universal benefits and services rather than means-tested ones. The reason is that as soon as we start means testing, it will affect the profitability of, particularly, low-income people - often women - to engage in paid employment. Another strategy is to make social insurance provisions earnings-related, making it profitable for people to work and pay social security contributions. The more they earn and pay, the better the benefit entitlements will be. If we are interested in improving the efficiency of welfare state programmes, empirical research suggests that we should worry less about the aggregate social spending and level of taxation, and more about the actual design of both programmes and methods of financing.
The Scandinavian strategy when it comes to the human resources is education and training, as well as other forms of active labour market policy, such as public relief work and forms of subsidised employment. The aim of these measures is to improve, or at least maintain, the employability of unemployed persons. Public expenditures can promote growth and equality simultaneously by affecting the distribution of two aspects of human resources; skills and health, in a favourable direction.
Social services can be seen as another form of resource, making it possible also for adults in families with small children, or frail elderly relatives, to participate on the labour market. The approach here is to give heavy subsidies to public services, such as day-care facilities and the care of frail elderly people. This has undoubtedly contributed to the overall high employment rate among women. Social services may be seen as investments that in a dynamic way provide people with the opportunity to become tax payers and thus to contribute to balancing state finances. Social services are also needed to ensure the full participation of all citizens in society in general and not only on the labour market. The access to services is therefore a democratic problem.
Social policies cannot make up for failures in economic policy to provide employment opportunities. This means that a successful strategy has to be based on successful macro-economic policy making. The fundamental problem of unemployment in Scandinavia is that there are too few jobs. But successful macro-economic policies are not likely to be enough, either, if the skills of the unemployed persons do not match what the demands on the labour market.
The human capital response to the ageing of societies - Children must come first in the ESM
"The European Social Model (ESM) is about social inclusion and equality of opportunity." (Barrosso, July 12, 2005).
There is no ESM in the sense of a unified strategy for organising social protection and labour markets. Instead, there are different institutional strategies among the European countries, often labelled ‘models’. There is however a common set of indicators/benchmarks of policy performance emerging from the process of establishing an Open Method of Coordination (OMC). The OMC is about the goals and hence about common European values that deserve to be taken seriously.
There are good reasons to point to the common concerns generated by our common futures of ageing populations for the macroeconomic development. Weak growth of the European Single Market has repercussions for all countries. There is hence good evidence to suggest that a more balanced population development in Europe should be a common concern. Moreover, within the OMC framework, the Member States of the EU have agreed on common objectives when it comes to employment, pensions, health care and social inclusion. There is very little to suggest that these objectives can be reached without a deeper, and at the same time explicit, concern with human capital formation along the lines outlined in the present report. This is in congruence with the work in the Commission on Modernising Social Protection but would also benefit from being operationalised within the OMC framework. Human capital formation ought to be an integral part of the thinking also around other common objectives of the Member States.
The ageing of our societies is putting more pressure on redistribution, not less. The bad news is that the ageing of European societies is putting increased pressures on redistribution, whether performed by the state, market or family. This warrants us to rethink our social and economic policies. The debate has been unduly focussed on pension reforms and capital accumulation (i.e. saving) to ensure future living standards for the elderly. We should instead focus on how social policy interacts with education, fertility and other fundamental determinants of the future tax base. This strong policy recommendation is based on our observation a causal structure where education is the central driving variable for GDP increases in Europe (Institute for Futures Studies 2004). In order to design sustainable social policies for the future we need to put our children and youth first. This is contrary to what has been done so far within the framework of the Open Method of Coordination. The British Presidency could and should take initiatives to change that.
The reason is simply that without massive attention to the future tax base in terms of both the number of tax payers and their productivity, there will be little left worth labelling sustainable pensions and health care systems, not to speak about social inclusion. At the heart of this is finding new ways to reconcile production and reproduction. This is what the new gender balance should be about. Working life needs to be made more flexible. This demands new public-private partnerships. The gap in the OMC process needs to be filled; the situation of children, youth and their families should be a new domain. But this is not enough. The number of indicators needs to be reduced. The production of the statistics needs to be speeded up. There needs to be sanctions or at least incentives to encourage better performance. The social deficits of the Member States need to be addressed as promptly as those in their public finances. This is not done overnight but Europe has to try harder and act wiser.
The way in which the British presidency has addressed the future of the ESM has created a momentum for the politics of European integration. The British approach of setting quantifiable policy goals and designing credible policies that are subsequently evaluated fits very well with the OMC process. For us who believe that change bears the potential for betterment this is hopeful. Improvement is badly needed when it comes to ’social-Europe’ as we know it. European leaders should join forces make the massive but necessary investments in human capital. This is also true about the fact that even if the ESM is about our self-interests as European citizens, in the end it is about our moral obligations towards our children. The efforts will however have to be very strong if Europeans are to put enough trust in the further and future European integration. This may be coming late but it should not be too little.
Biography
Joakim Palme is the director of the Institute for Futures Studies. He is also associated to the Swedish Institute for Social Research, Stockholm University, as a professor. From 1999 to 2001, he chaired the Welfare Commission appointed by the Swedish Government.
Footnotes
(1) This is an abbreviated version of a contribution commissioned by Policy Network for a larger project on the European Social Model.
delicious
digg
facebook
