The Welfare State and the Employment Problem

by Assar Lindbeck
The Welfare State and the Employment Problem
Assar Lindbeck
The American Economic Review
Start Page: 
End Page: 
Select license: 
Select License

The Welfare State and the Employment Problem

Social security, provision of public-sector services, job-security legislation, minimum wages, and centrally regulated (bargained) relative wages are important elements of the modern welfare state-each designed, often quite successfully, to enhance economic security and to redistribute income. The consequences of these policies for the unemployment issue are the topic of this paper.

I. Welfare-State Spending

It is a commonplace that very generous unemployment benefits with long, or even unlimited, duration and with lax work tests contribute to unemployment persistence. This is both because such benefits reduce the incentives for job search and because they boost real wages by raising the reserva- tion wage of workers.

Public-sector provision of subsidized social services, such as child care, education, health care, and old-age care, also has con- sequences for aggregate employment and unemployment, as it tends to shift the pro- duction of such services away not only from households, but also from the market. Hence, while such policies expand public- sector employment, they are detrimental to production and employment in the private service sector. The net effect is, most likely, to expand aggregate employment. Later on, if the government faces a financial crisis, and public service production is therefore cut rapidly and substantially, it may, how- ever, be difficult for the private service sec- tor to pick up the slack fast enough to prevent a rise in aggregate unemployment. This is, indeed, what is happening today in the Nordic countries.

*Institute of International Economic Studies, Uni- versity of Stockholm, Stockholm S-106 91, Sweden.


It is also a commonplace that marginal tax wedges, which are an unavoidable con- sequence of high welfare-state spending, favor not only leisure but also household service production (do-it-yourself work) -at the expense of buying such services in the market. Obvious examples are cook- ing, repairing homes and durable consumer goods, gardening, and child care. In several high-tax countries, it often pays the household to produce such services itself, even if it is only 25-50 percent as productive as craftsmen in the market. The high elasticity of substitution between home production and purchases of services in the open market makes high marginal tax wedges, including those created by payroll taxes and the VAT, much more damaging for the service sector than for the goods- producing sector of the economy. Indeed, tax wedges have become a serious obstacle to expansion of production and employment in the private service sector in several coun- tries in Western Europe.

11. Job-Security Legislation

Job-security legislation is sometimes a substitute, but more often a complement to social security. It is rather generally agreed that job-security legislation which raises the costs of hiring and firing labor, "labor turnover costs" for short, tends to stabilize aggregate employment at the initial level, whatever this happens to be. Thus, it con- tributes to making employment and unem- ployment persistent. One can, however, en- rich the analysis if some additional aspects are taken into account.

The Character of the Macroeconomic Dis- turbance.-When aggregate employment is high initially, and the business cycle is mild and fairly regular, most observers probably regard employment persistence, which may be brought about by job-security legislation,


as socially advantageous. Moreover, as high labor turnover costs make variations in the number of employees expensive, firms are induced to vary the number of working hours per employed instead (Katharine G. Abraham and Susan N. Houseman, 1993). Such work-sharing may also be seen as so- cially advantageous, in the sense that the burden of macroeconomic disturbances is then shared more evenly among individual employees during the course of the business cycle.

A more negative evaluation of the employment effects of legislated labor turnover costs is natural if the economy, as in much of Europe today, instead happens to be in a deep and prolonged recession. It is then high unemployment rather than high employment that is stabilized. High labor-turnover costs would be expected to be particularly damaging if firms are very uncertain as to whether they will need newly hired workers also in the future. Such un- certainty tends to be especially large after deep recessions and during periods of rapid structural change.

Firm Size.-The restraints on employment expansion due to job-security legisla- tion may be particularly pronounced for small and medium-sized firms. One reason is that many such firms have a rather un- even distribution of age groups in the work force; as a consequence, some of these firms, in particular fairly new ones, may have no workers close to retirement age. As a result, overstaffing due to overoptimistic earlier hiring may not disappear quickly by way of natural attrition, such as retirement. More- over, small and medium-sized firms often have only limited access to liquid assets and credit to finance labor hoarding during peri- ods of low sales. Such firms would also be expected to suffer more than large firms from the misconduct of individual workers, because of the close interpersonal relations between employees in small firms. High leg- islated costs of individual dismissal, and not just collective dismissal, may therefore dis- courage the hiring of labor, especially in small and medium-sized firms. Such reluc- tance on the part of small firms to expand employment has serious macroeconomic im- plications, as the entry and growth of such firms is potentially highly conducive to in- creased aggregate employment.

Consequences for Wages.-Labor turnover costs also help those initially employed in the formal sector of the economy, the "insiders," to push up wages above the reserva- tion wage of those who are not employed there, the "outsiders," without the latter being able to get jobs. Moreover, in busi- ness upswings, insiders may use their mar- ket power to raise wages, rather than to make it easier for outsiders to get jobs by accepting unchanged wage rates. Such legis- lation, therefore discriminates against out- siders in the labor market (Lindbeck and Dennis Snower, 1988). For these reasons, legislated labor turnover costs are also more likely to reduce the aaerage level of aggre- gate employment over the cycle in the con- text of models where the consequences for wage formation are taken into account than in models where such consequences are neglected.

Seniority Rules.-Job-security legislation, or more frequently bargaining agreements, may also include rules for the order in which employees can be fired-usually, in fact, according to the principle of "last-in, first out." Such seniority rules are often asserted to be advantageous from a distri- butional point of view. The argument is that they help low-productivity workers with high seniority (i.e., low-productivity insiders) to keep their jobs. Moreover, low-productivity workers who are fired can avoid being stig- matized as "inferior labor" when firing is based on seniority rules rather than on poor work performance (Robert Gibbons and Lawrence Katz, 1991). These gains for low- productivity workers in terms of the dis- tribution of jobs occur, however, at the expense of low-seniority workers and out- siders. The latter group includes entrants to the labor market, such as young workers and those who often move either between jobs or in and out of the labor force. Job- security legislation is, therefore, a mixed blessing from a distributional point of view.

Moreover, avoidance of stigmatization implies, in fact, that information about rela- tive productivities of workers is concealed from employers, thereby accentuating the asymmetry of information. As a consequence it becomes more difficult to get the right person in the right place. Economists usually regard such efficiency costs due to asymmetric information as a market failure, which should not be encouraged. Thus, du- bious distributional advantages may be achieved at rather high costs.

Last-in first-out rules may also make it expensive, perhaps even prohibitive, for contracting firms to retain key employees who are important for the future success of the firm, perhaps even for its survival. Such success often requires young, newly trained, hardworking and enthusiastic employees. It is also important to take into account con- temporary tendencies among firms to over- haul their internal organization, by way of administrative decentralization and in- creased reliance on individual initiatives. These developments tend to make jobs more heterogeneous and workers poorer substi- tutes for each other, while much job-secur- ity legislation is based on the notion that individual workers are good substitutes. This is bound to create inefficiencies. Set against these negative consequences for productiv- ity, however, is the fact that firms are, in principle, encouraged to provide training for their work forces when legislated labor turnover costs make individual workers re- main on the job longer than otherwise.

Regulations of Types of Contracts.-When job-security legislation makes new hiring ex- pensive for firms, they are encouraged to shift to types of contracts that are not cov- ered by legislation, such as temporary and fixed-term employment contracts, which tend to reduce job security and incentives for training. In response, legislators may be tempted to restrict the use of such contracts; indeed, such legislation exists today in most countries in Western Europe, even though there has recently been a tendency to liberalize these rules.

Moreover, job-security legislation often requires complex, long, and expensive bar- gaining between the firm and its employees, or unions, which raises the costs of negotiat- ing within firms. This tends to make intrafirm transactions less attractive as com- pared to transactions in the market. In the context of Coase's theory of the firm, with the demarcation line between the firm and the market defined as a situation where the marginal transaction costs are the same within firms and in the market, higher trans- action costs within firms tend to increase the incentives for firms to use subcontrac- tors, rather than to employ workers. Such adjustments by firms partially undermine the ambitions behind job-security legislation. As a consequence, political and union pressure would be expected to emerge to restrict the freedom of firms to use subcontractors. In- deed, such restrictions have existed in some countries, such as in Sweden from 1973 to 1993. This may have further negative con- sequences not only for efficiency in pro- duction, but also for the possibility of increasing aggregate employment after deep recessions.

Thus, an overall evaluation of the conse- quences of job-security legislation has to be based on judgments about the relative im- portance of contradictory effects. It would seem, however, that existing legislation was better adopted to conditions in the past than to today's world with large macroeco- nomic disturbances, high and persistent un- employment, rapid structural change, great uncertainty, and increasing heterogeneity of jobs and workers. If prevailing legislation was adequate when it was introduced dur- ing the 1950's and 1960's, it is probably less adequate today (Lindbeck et al., 1994).

111. Wage Formation

Attempts by the government or unions, or both, to equalize relative wages against pre- vailing market forces is another important aspect of welfare-state policies. Such poli- cies may certainly be an additional explana- tion for unemployment. The natural recom- mendations are not only that minimum-wage legislation be eased (in countries where it is binding), but also that bargaining be more


decentralized. The latter is likely to contribute to a more market-conforming wage structure, including wages that facilitate employment for low-productivity workers and that create sufficient encouragement for investment in human capital. Such recom- mendations of larger wage differentials make less sense in countries where these differentials are already substantial, such as in the United States, where the working poor may pose a more serious problem than unemployment.

Special problems arise in countries with high average tax rates for low-income groups, and with high benefit levels for both the unemployed and people living on social welfare. The reason is, of course, that the reservation wage of low-productivity work- ers may be too high to make them accept jobs at the after-tax wages offered by firms. Such workers will therefore prefer to remain unemployed. (As pointed out in the literature, however, job searchers may accept job offers earlier when the distribution of income is less dispersed, as they then cannot have much hope of high-wage offers later on [Christopher A. Pissarides, 19831.1

IV. Solutions?

Wage subsidies might be a way out of these various problems. The argument for such subsidies is based on the reasonable principle that it is better if the government pays people for working rather than for not working. An obvious technique, in particu- lar at the end of recessions, is temporary marginal employment subsidies, for instance in the form of reduced payroll taxes for additions to the work force of a firm. While such subsidies reduce the marginal costs of increasing the number of employees, the costs of more hours of work per person, including overtime, would not be subsidized. Firms are, therefore, stimulated to hire more workers rather than to increase the number of hours worked per employee. As a result, the position of outsiders would improve.

On the other hand, an obvious problem with marginal employment subsidies is that they may prevent downward adjustments of wages. The problem is accentuated if employees and unions expect similar subsidies to be reintroduced the next time unemploy- ment rises. There is thus a risk that the government, after a while, will feel compelled to prolong the subsidies, and perhaps even make them permanent, which would result in various undesired side-effects. One is that, later on, total wage costs would vary considerably among firms, depending on their pre~~iouspaths of hiring; another is that some firms may choose to split into contracting and expanding parts to get sub- sidies, not to mention outright cheating with the subsidies.

A more selective version of employment subsidies involves reserving them for low- productivity workers, possibly for the stock of such workers rather than just for new hirings. This makes sense particularly in the case of handicapped citizens. Such a policy is more problematic if low productivity is the result of insufficient training, for then the subsidies, like progressive taxation, un- avoidably imply an implicit tax on investment in human capital, as the subsidy is reduced if a worker acquires higher produc- tivity, regardless of whether it is measured in terms of the level of training or the wage rate. This type of policy, then, is in direct conflict with the common, and solid, idea that the best way, in the long run, to im- prove the incomes and job opportunities of low-productivity workers is to improve their training. (Reduced payroll taxes confined to low wages are, however, much less of a problem in countries where these taxes are regressive.)

Another, and perhaps better, policy op- tion is to reserve wage subsidies for the long-term unemployed, for instance people with more than 6 or 12 months of unemployment behind them. A variation of this theme is that the long-term unemployed be allowed to turn over their unemployment benefits to firms in exchange for work (Dennis Snower, 1993). One (possibly seri- ous) problem with this approach, though, is that people who have been unemployed for fewer months than the period required for a subsidy may then reduce their search activi- ties, as they expect to be assisted in getting a job anyway after a while.

Thus, there is a menu of conceivable poli- cies to improve the employment situation, and to limit the unfavorable effects of vari- ous welfare-state programs on aggregate employment and unemployment. To this menu may be added, of course, various ele- ments of active labor-market policies, including training programs, improved labor- market exchange systems, and better counseling and job-search assistance. Vast empirical research warns, however, against excessive hopes of the efficacy of each of these policies, as isolated actions, as a way of reducing unemployment.

In several West European countries, a well-functioning labor market also requires reforms of the unemployment benefit system, reduced marginal tax wedges (in par- ticular for services), liberalized job-security legislation, and more market-conforming systems of wage formation, as well as the removal of discrimination against private production of social services and other ob- stacles to the entry of firms. There is little risk that such reforms would undermine the automatic stabilizers of the government budget to any large extent; moreover, in some advanced welfare states, these auto- matic stabilizers may even be too strong today in the sense that exploding govern- ment deficits in deep recessions tend to result in instability in financial markets and higher interest rates.


Abraham, Katharine G. and Houseman, Susan

N. "Does Employment Protection Inhibit Labor Market Flexibility?" Staff working papers, W. E. Upjohn Institute for Em- ployed Research, Kalamazoo, MI, May 1993.

Gibbons, Robert and Katz, Lawrence. ''Layoffs and Lemons." Journal of Labor Economics, October 1991, 9(4), pp. 351-380.

Lindbeck, Assar and Snower, Dennis. The in- sider-outsider theory of employment and unemployment. Cambridge, MA: MIT Press, 1988.

Lindbeck, Assar; Molander, Per; Persson, Torsten; Petersson, Olof; Sandmo, Agnar; Swedenborg, Birgitta and Thygesen, Nils. Turning Sweden around. Cambridge, MA: MIT Press, 1994.

Pissarides, Christopher A. "Efficiency Aspects of the Financing of Unemployment Insur- ance and other Government Expenditures." Review of Economic Studies, January 1983, 50(1), pp. 57-69.

Snower, Dennis. "Getting the Benefit Out of a Job." Financial Times, 23 February 1993,

p. 12.

  • Recommend Us