Creating a competitive EU labour market for the 21st centur
LABOUR MARKET (INCLUDING LABOUR FORCE SURVEY)
Welfare to Work
Replacing or supplementing labor market incomes?
Social policy evens out the distribution of income. It prevents social unrest, it satisfies the taxpayers’ sense of justice and it insures against random variations in people’s lifetime careers. Ideally, it insures risks that are not privately insurable, either because risk markets suffer from adverse selection or because private insurance comes too late in a person’s life, when the veil of ignorance has already been lifted.
Welfare programmes “… serve to even out differences in life chances, to achieve greater equality between generations and to redress inequality by race, gender, or health status. More generally, these programmes are intended to help people reallocate income over the lifecycle, to insure against events which cause income loss, and to provide a sense of security to all citizens”.
This is the theory. In practice poverty is dominated by joblessness, and large quantities of money are spent on the condition that the beneficiaries do not work and do not earn. The replacement of labor income with public transfers becomes the dominant form of assistance.
As employment is the most important source of income, the replacement of labor income is a plausible implication of the social-policy motives. The problem, however, is that the fraction of people not having a job is not exogenously given but depends heavily on policy itself. Benefits under the condition of not working operate like a wage paid for idleness which the market wage has to exceed. Since no one is willing to work at a market wage below the social benefit attainable without working – with the exception of those who fear being stigmatized as “welfare scroungers” – this benefit is a lower bound on market wages.
However, in a market economy, an upper bound on an individual’s market wage is given by his (or her) productivity, i.e. the value added he or she is capable of creating. Thus there is a fundamental problem with people whose productivity is below the benefit that the welfare state is willing to provide. These people, in principle, cannot find a job in a market economy under traditional policies. The wage has to be above their benefit to make them offer their labor, and the wage has to be below productivity to make firms demand this labor. The two conditions are mutually exclusive. Although driven by good intentions, the wage replacement policy turns out to be a policy of increasing the reservation wage – the wage below which a worker will refuse a job – and of preventing the creation of jobs which otherwise would have been available.
This problem used to be minor when benefits were low relative to average incomes. However, the gradual expansion of the welfare state (expressed as the proportion of gross domestic product being spent on unemployment compensation and social assistance) has increased the number of people who are affected and has therefore increased the number of unemployed, in particular among the less educated, whose productivity is low relative to the minimum income which the state provides them. Unfortunately, this situation seems unlikely to change in the future.
The productivity effects of the New Economy are likely to stimulate aggregate income growth and with it the growth of social standards. However, the number of people who just cannot keep up with the New Economy and who are unable to cope with modern work requirements may be increasing. The digital divide may not only be a problem among nations but also among the people within a nation.
European integration may increase the desire for harmonization of social standards. If traditional welfare benefits are harmonized, many people in the less-developed regions of Europe may find themselves in a situation where their labor productivity is below common European benefit standards. In Europe, there are regions where labor productivity is only a quarter of that elsewhere. Harmonizing social standards without changing the conditions under which social benefits are paid would undoubtedly create mass unemployment in many of the less-developed regions if the benefits were sufficient for the more productive regions. The problem of the Mezzogiorno would spread. Thus it is opportune to search for alternative ways of designing the welfare state, ways that make it possible to help the needy without driving many of them into unemployment. Basically, these ways involve redefining poverty and the conditions under which the welfare state delivers its benefits. To satisfy the definition of welfare programmes, it is not necessary to make benefits conditional on people being jobless. They could also be made conditional on people being employed and not earning enough. A new definition of poverty would capture that. It is not a person who does not work who is poor but someone who works to his physical and mental capacity and is nevertheless unable to earn a sufficient income.
With this definition of poverty, the welfare state would not replace labour income but supplement it when it is inadequate. Supplementing income to reach a social target level has very different implications for the functioning of the labor market than the current system because it circumvents the problem described above. Even people with very low labor productivity would be able to find jobs because social benefits would no longer establish a lower bound to wages. People would be willing to work at very low wages, because they know that this would make them eligible for social benefits, and, for the same welfare state expenditures, they could even have higher incomes than in the current system.
A number of mainly Anglo-Saxon countries have followed this line of reform and have moved from a wage replacing to a work complementing welfare system.
In the OECD countries, an important part of social protection against unemployment is unemployment insurance. In order to be eligible for compensation, claimants must have worked and contributed to the insurance fund for a given period of time, they must be involuntarily unemployed and they must be actively looking for work.
Financial assistance for those no longer eligible for unemployment insurance takes two forms: unemployment assistance and social aid. Unemployment assistance is designed as a follow-up benefit to unemployment compensation, paying a lower benefit than unemployment insurance.
Social aid is given to those who qualify for neither unemployment insurance nor unemployment assistance. The government acts as a provider of last resort to secure a minimum standard of living. Social assistance in the EU member countries normally has an unlimited duration (see European Commission, MISSOC 2000).
Replacement policy
The traditional social security systems of most OECD countries can be characterized as passive. Benefits are provided to secure a minimum standard of living, and recipients receive the benefits without a strong obligation to look for work. This is especially true for social aid, which is provided without any significant obligation imposed on the recipient. Such a social security system leads to welfare dependency. It encourages inactivity, does not provide sufficient incentives to look for work and increases the opportunity cost of working in the market economy. In short, by following a wage replacement policy, the traditional social security system pushes the reservation wage up and thus destroys part of the employment opportunities which otherwise would have been available.
The extent to which the required wages are artificially pushed up is influenced by the level of unemployment benefits and social assistance, the duration of entitlement, the coverage of the system and the strictness with which the system is operated – as well as social attitudes. The influence of the welfare system on reservation wages can be represented and quantified by the net replacement rate (NRR) defined as:
NRR= Benefit income when unemployed – tax on benefit income/ Earned income + benefit income when employed – tax on earnings and benefits
The net replacement rate is the fraction of current or potential income which the social system provides to a person if he or she does not work. It varies according to the type of household, employee, sector of industry, wage and salary group and the reasons for not working.
A simple table can show the net replacement rates for an average production worker receiving unemployment benefits (at the beginning of receipt of benefits) or social assistance (long-term benefit recipient). It demonstrates that the net replacement rate at the beginning of unemployment is relatively high for a couple with two children but lower for someone who is single. Hence, the breadwinner has little incentive to seek regular work. This is all the more true if the (participating) spouse is long-term unemployed. There are, of course, differences in the net replacement rate from one country to another.
The net replacement rates for long-term benefit recipients are lowest in the United States and Spain and highest in the Scandinavian countries (except Norway), Switzerland and the Netherlands.
The replacement rate can be explained by the intended insurance function. However, a replacement rate also defines a minimum reservation wage, below which no one is willing to accept a job. In fact, for most people the minimum reservation wage may be even higher than that because when they decide to work they not only require a compensation for the lost special benefits but also for the time lost for leisure and for working at home or even for the loss of black market income. The higher the replacement rate, the better is the insurance protection, but the lower is the number of jobs which employers are willing to provide, given the skill distribution of the unemployed.
High unemployment of low-skilled workers
The destruction of jobs and output resulting from the traditional policy is particularly severe at the lower end of the income distribution. Workers with an income below the average will have a higher replacement rate than that reported, and what is more: the replacement rate would be above one for people who do not work because their productivity and potential wage is below the level of social aid. However, since they do not work, the wage at which they would find employment is not known. Thus, no statistical information is available on the replacement rates of this important group.
Social aid (and to a lesser extent unemployment assistance) is particularly problematic for the functioning of the labor market because, unlike unemployment insurance, it is a lower bound on the feasible wage distribution very much like a legally prescribed minimum wage. This lower bound is of limited importance for average production workers, but it destroys jobs for the less well qualified whose labor productivity is below the social aid level or not sufficiently above it to compensate for the work effort.
The consequence of social aid is that it compresses the wage distribution and concentrates unemployment on the lower qualification (or productivity) levels. As even less educated people are normally able to fulfill some useful functions in the economy, a wage is conceivable at which these people could find employment. The problem is that society considers this wage to be too low and it is therefore replaced by a higher level of social aid; but the good intentions turn out to have adverse employment consequences for those people who seemingly benefit. Indication of the compression of the wage distribution through the social system show it regresses the earnings dispersion amongst relatively low earners as measured by the ratio of the median decile to the lowest decile of the wage distribution with the ratio of social aid and the average wage income across the 18 OECD countries for which the data were available. There is a significant negative correlation between these variables indicating that in countries like the United States and Canada, which have low levels of ordinary welfare payments, the distribution is indeed much wider than in countries like Denmark, Austria, Finland and Sweden where welfare payments are rather high.
An overview of the employment situation among the EU countries shows the majority of European countries are reporting high rates of unemployment. Many of the low-skilled unemployed are unemployed for over a year. The standardized unemployment rate for EU members is nearly 11,5 per cent. In every country, and for both sexes, the less skilled have the higher rate of unemployment. Average unemployment rates in excess of 10 per cent occur in several countries Spain, Greece, Portugal and even there they are experienced only by women. For the less skilled such rates prevail in Finland, France, Germany, Ireland, Italy (for women), Spain, the UK (for men) and Canada.
The unweighted average unemployment rate of workers with a “below upper-secondary education” is roughly 50 per cent higher than the general unemployment rate of the 28 EU member countries. Again, it should be noted that the statistical information is incomplete and misleading since those whose productivity is below social aid do not work enough to qualify for benefits and often do not look for work and, therefore, do not count as unemployed.
Black market activities
The policy of providing social assistance through replacing labor income has not only destroyed jobs by increasing reservation wages; it has also worked as a policy of subsidizing black market activities. It is true, of course, that this was not intended, but as informal labour is the natural alternative to formal labor and as the payment of benefits stops when formal labor income is obtained, it is clear how the incentives have worked.
Simple study reveals that the underground economy has reached a high level in most OECD countries. Black market activities are, of course, only partially brought about by the generous provision of social assistance. Other factors like high tax rates also play a significant role. The replacement policy, however, not only provides incentives to work in the informal market, but also to use informal labor for home improvements, especially where home ownership is widespread. The spread of do-it-yourself stores not only reflects a fashion but also a rational reaction to economic incentives which undermine the division of labor and prevent the productivity gains that it entails.
Attempts to keep disincentive effects under control
The disincentive effects have not been overlooked by policy-makers, but in most countries policy reactions have not called into question the traditional policy as such. There have been a number of attempts, however, to keep the disincentive effects under control. One obvious provision in this regard has been the limitation of the periods during which people are eligible for benefits. Simple study gives an overview for EU countries. In some countries, however, eligibility is of unlimited duration. Belgium is an extreme example, but even a country like Germany, which at first glance seems to have restrictive rules, pays unemployment assistance (Arbeitslosenhilfe) at a replacement rate of slightly more than 50% for an indefinite period of time.1 Entering the formal labor market as unemployed and then moonlighting is an extremely attractive option under these conditions.
Another attempt to keep misuse under control is the imposition of sanctions if a job is left voluntarily or if a job offered an unemployed person is not accepted. Simple study give an overview of the situation in some European countries. At first glance, the table signals a rather harsh approach in some countries, which even exclude the unemployed from benefits if they refuse a job twice. The actual policies are, however, much milder than suggested by the study, since the jobs rejected must have been appropriate, and the definition of what is appropriate is always a matter of ambiguity. Also, of course, the provisions do not alter the fact that public money flows if people do not work and stops flowing if they begin to work. This is a challenge to the ingenuity of beneficiaries to invent reasons why an appropriate job cannot be found or why one offered is not appropriate.
The problem may not be avoidable with unemployment insurance because benefits have to be provided if someone does not work. However, the provision of social aid and social assistance to the long-term unemployed or people who have never entered the labour force are clearly another matter. Here, in particular, the traditional policy should be reconsidered.
