This Selected Issues paper reviews some features of the labor market in Finland during the 1990s. In particular, it reviews patterns of employment and wage distribution across sectors. The paper describes the main regulatory features and formal structure of labor market arrangements: the wage negotiation process, employment legislation, income replacement regulations for the unemployed, and the recent changes in these areas introduced by the government. The paper also describes the structure of labor taxation, and reviews the theory and existing empirical evidence concerning the effects of taxation on labor market outcomes.


This Selected Issues paper reviews some features of the labor market in Finland during the 1990s. In particular, it reviews patterns of employment and wage distribution across sectors. The paper describes the main regulatory features and formal structure of labor market arrangements: the wage negotiation process, employment legislation, income replacement regulations for the unemployed, and the recent changes in these areas introduced by the government. The paper also describes the structure of labor taxation, and reviews the theory and existing empirical evidence concerning the effects of taxation on labor market outcomes.

I. Labor market issues1

A. Introduction

1. Labor market conditions in Finland have undergone dramatic changes during the 1990’s. At the beginning of the decade, Finland was hit by a severe recession, reflecting the unwinding of a domestic asset price bubble, the collapse of trade with the former Soviet Union and a severe shock in the terms of trade. Between 1990 and 1993 GDP declined by 11 percent, while the unemployment rate rose from 3.4 percent to 18.4 percent. Economic conditions improved substantially after 1994. However, while quarterly GDP has almost reached at end-1996 its level prior to 1990, the unemployment rate fell only slightly, and was still 15.0 percent, s.a., in December 1996. The persistence of high levels of unemployment, has been at the center of policy discussions in Finland. In October 1995, the government adopted an “Employment Progamme” that aimed at cutting unemployment substantially. In 1995 the OECD published a “Jobs Strategy” which contained a series of recommendations for future policies, and some country specific notes for Finland (Appendix I).

2. The aim of this paper is to review recent labor market trends, report changes in labor market regulations and assess the effect on unemployment of the increase in labor taxes and unemployment benefits that took place in the last 15 years.2 The paper is organized as follows: Section B reviews some features of the labor market in Finland, in particular, recent patterns of employment and wage distribution across sectors; Section C describes the main regulatory features and formal structure of labor market arrangements: the wage negotiation process, employment legislation, income replacement regulations for the unemployed, and the recent changes in these areas introduced by the Government; Section D describes the structure of labor taxation, in particular employers’ non wage labor costs, and reviews the theory and existing empirical evidence concerning the effects of taxation on labor market outcomes; Section E briefly draws some conclusions.

B. Aspects of the Labor Market

Unemployment by age, gender and geographical distribution

3. The economic recovery has had a limited effect on the labor market and unemployment rates have remained much above their pre-recession levels. During the recession (1990–93), unemployment increased at an extraordinarily rapid pace, unprecedented in OECD economies (Figure 1), in spite of a 2 percent reduction in labor force participation. Although employment started recovering in 1995, especially in the export-related sectors, at end-1996 it remained some 15 percent below its 1990 level.

Figure 1.
Figure 1.

Finland: Unemployment, 1988–96

Citation: IMF Staff Country Reports 1997, 060; 10.5089/9781451813135.002.A001

Sources: Statistics Finland, Statistical Yearbook of Finland; and Finnish Labour Review.

4. Unemployment rates declined in 1995–96 across all age groups, except for workers above 50 years of age, where it increased (Figure 1, middle panel). This pattern is partially explained by labor market regulations. Persistently high unemployment rates in the older segments of the labor force reflect, to a large extent, some of the features of the unemployment benefits system which has allowed workers to effectively retire at age 53.3 4 Moreover, the sharp decline in the unemployment rate in the cohort below age 25 reflects recent changes in the unemployment benefit regulations (see Section C) that have made young people without completed vocational training no longer eligible for unemployment benefits, and have instead broadened the scope of training programs and student allowances. Female’s unemployment rates, lower than those of men during the recession, have proved more persistent in recent years and the gap, that had widened through 1994, has now virtually disappeared (Figure 1, top panel).5

5. The average duration of unemployment doubled from 24 weeks in 1988 to 48 weeks in 1996, still on an upward trend with respect to 1995. However, the average duration of completed unemployment spells slightly declined to 24 weeks in 1996 from 25 weeks in 1995 while, the share of long-term unemployment in unemployed job-seekers which had increased to 33 percent in 1994 (SM/95/208, 8/21/95), fell to 31 percent in 1995, and remained unchanged in 1996.

6. Historically, unemployment rates have varied widely across the country, in particular between the southern and the northern part of the country (Table 1). This pattern remained broadly unchanged during the recession. However, 1996 data suggest that new employment is concentrating mostly in the southern region, while unemployment seems to be more persistent in the northern parts of the country.

Table 1.

Finland: Employment by Administrative District

article image
Sources: Statistics Finland, Statistical Yearbook of Finland, and staff calculations.

Shifts in output composition and employment trends

7. Trends in total employment in the 1990s were dominated by the rapid changes which affected the production structure of the economy. While output and employment declined rapidly, the subsequent recovery was until recently, (EBS/97/99, 6/11/97) lop-sided.

8. The recovery was early and particularly strong in the export oriented industries (wood and paper, electronics and engineering, and by sector, manufacturing). This explains why employment developments varied markedly in 1993. It also explains why total employment performance was lackluster. The sectors which grew fasted are also sectors where the production technology is particularly capital intensive. Assuming the sectoral output share had not changed with respect to 1987, employment in 1995 can be estimated to have been approximately 3.5 percentage points higher than its actual level.6 The shift of production towards more capital intensive sectors alone explains an increase of about 3 percentage points in the unemployment rate.7

9. The change in the composition of output is one of the factors explaining the limited growth of employment. Two features are particularly relevant: (i) substantial gains in productivity, evidenced by a decline of the wage share across all industries and an increase in total turnover per worker,8 and (ii) a reorganization of industrial production, evidenced by an increase in the share of workers employed by medium sized enterprises and a decline in the average number of employees per enterprise.

10. Between 1988 and 1994, the wage share in turnover declined across all industries by 16 percent on average, and by a larger amount in the textile, wood and construction sectors (Table 2). The average real turnover per worker increased by about 33 percent during 1988–94 (Table 3).

Table 2.

Finland: Wage Share in Turnover Across Industrial Sectors, 1988–94 1/

article image
Sources: Statistics Finland, Statistical Yearbook of Finland, and staff calculations.

Ratio of total expenditure in wages to total turnover.

Table 3.

Finland: Real Turnover Per Worker Across Industrial Sectors, 1988–94 1/

(Percentage change) 2/

article image
Sources: Statistics Finland, Statistical Yearbook of Finland, and staff calculations.

Average turnover per worker deflated by the producer price index (1990=100) of the respective sector.

Percentage change over a two year period.

11. This productivity increase was particularly strong for large enterprises. During 1988–94 the percentage of labor force employed by enterprises with 50–999 employees has increased. However, the opposite occurred in the production shares: the share in total turnover of large enterprises (over 1,000 employees) increased from 54 percent to 56 percent (Table 4).

Table 4.

Finland: Employment and Turnover by Enterprise Size in the Industrial Sector, 1989–94

article image
Sources: Statistical Yearbook of Finland, and staff calculations.

Thousands markka.

12. In the public sector, employment has followed closely the overall trend in the rest of the economy. The public sector continues to represent about 30 percent of total employment, which, although still among the highest of the OECD, has remained almost unchanged since the recession. This implies that the public sector has effectively restructured employment together with the private sector. The process appears to have affected particularly female employment, traditionally concentrated in the public services. Employment in these activities is also unlikely to pick up in the short term, even if the output share of the public services is still below its pre-recession levels, because of the ongoing process of fiscal consolidation. This, to some extent, explains why female unemployment rates have failed to recover in the same way and have come to lie above those of men. In summary, the poor employment performance during the 1990s cannot be explained solely by sectoral shifts in output. Productivity in industrial sectors increased rapidly, particularly in large enterprises, possibly reflecting the shedding of labor in service like activities provided by industrial enterprises. What brought about this generalized drop in the demand for labor per unit of output? The answer has to be found, at least in part, in the significant changes in labor regulation and taxation that took place over the last 10–15 years. These are discussed in Sections C and D.

Trends in wages

13. Wage differentiation across sectors has historically been low in Finland. This appears to have been an outcome of centralized wage agreements, which covered 90 percent of employees, and left little room for wage negotiations at the sectoral and plant level (see Section C, SM/94/219, 8/15/94).

14. In recent times, there appears to have been an increase, though with a delay, in wage dispersion, which could be viewed as a positive response to the widely different unemployment rates across sectors. The standard deviation of wages across sectors has increased over the whole period 1988-94 (Table 5). In particular, it is possible to infer a negative relationship between sectoral annual changes in wages and the sectoral unemployment levels during 1987-95 (Figure 2, top panel). Wages increased least in the sectors where unemployment rates were higher, thus evidencing a general ability of the system to react to economic conditions. The elasticity of sectoral wages to sectoral unemployment was quite low in the short run. The figures show that wage growth in sectors where the unemployment rate is respectively 5 percent and 10 percent would differ only by 2 percent per year. Thus, it would take a long time before sizeable wage gaps are opened. Indeed it took a long time, around four years, before a significant gap was opened in relative sectoral wages as an effect of unemployment differentials during the 1990s (Figure 2 middle panel).

Table 5.

Finland: Real Producer Wages Per Worker Across Industrial Sectors, 1988-94 1/

(Percentage change) 2/

article image
Sources: Statistics Finland, Statistical Yearbook of Finland, and staff calculations.

Real producer wages were calculated using sectoral nominal wages and deflated by the producer price index of the respective good (1990=100).

The percentage changes are calculated over the two year period.

Across all industries.

Standard deviation across levels.

Figure 2.
Figure 2.

Finland: Trends in Wages

Citation: IMF Staff Country Reports 1997, 060; 10.5089/9781451813135.002.A001

Source: Statistics Finland, Statistical Yearbook of Finland and staff calculations.1/ Each point represents the annual combination of sectoral unemployment rates and corresponding sectoral wage growth in the period 1987-1995. The sectors included are: agriculture, manufacturing, construction, trade, financing and insurance, and public services.2/ Per capita sectoral wages relative to per capita wages in the manufacturing sector (1990 = 100).3/ Average increase of the index of wages and salaries over the negotiated contractual increases.

15. The rigidities in the wage determination process have been identified as an important factor behind the delay in the recovery of private sector employment, in particular, in the non-tradeable, labor-intensive, sectors. Indeed, during 1990-92, real producer wages across the economy increased by 7 percent (Table 5), despite an overall loss of 10 percent in GDP in the corresponding period and a sluggish increase of 1.7 percent in productivity.9 Only after a period of two years of strong recession did real wages begin to decline.

C. Labor Market Organization and Regulations in Finland

Wage bargaining system

16. During the 1980s, wage agreements have been centralized, and negotiated between the employers’ and the employees’ unions; the central government was also involved in the process though not to the same extent as it used to during the 1970s. The agreements have generally covered some 90 percent of employees. Decentralized upward adjustments have been allowed through agreements at the plant level, which are evidenced by the a “drift” in wages above the centralized negotiated agreements (Figure 2 lower panel). In 1993, in an effort to promote decentralization of wage setting agreements, wage negotiations were conducted entirely at the sectoral level and the government disengaged itself completely from the negotiation process. However, this resulted in high wage increases across all sectors, as imitation led to higher wage increases even in sectors where productivity growth was lower. A centralized agreement with government involvement was again concluded in 1995.

17. There are no legal minimum wages in Finland. However, according to the contract of the Employment Act, the employer must comply at least with the pay and other terms laid down for the work in question in a generally applicable collective agreement for the sector. This has effectively enforced contractual minimum wages at the sectoral level within the framework of the centralized negotiated agreements.10

Job protection legislation

18. The legislation protecting labor is not particularly strict in Finland. Dismissal requires written notice and statement of reason only if the employee so requests. Then the employer must outline to him the reasons for the termination of the contract in writing. Before the termination for an individual reason the employer must give the employee his/her view concerning the reasons for the termination. When the reason is financially related, it must be explained to the employee or the employees’ representative. There is no severance pay.11 12 According to an OECD study (OECD 1994) these features place the Finnish system among the least rigid in the OECD (Table 6).

Table 6.

Employment Protection Legislation in Europe 1/

article image
Source: OECD, 1994 “The Jobs Study”.

Ranking of OECD countries with respect to strictness of job protection legislation. Rankings increase with the strictness of job protection.

19. Fixed-term contracts were allowed only if justified by the temporary nature of the job, but a wider use of this kind of contract has been allowed since 1996, in line with the relaxation of the regulations which has been taking place since 1994 (SM/95/208, 8/21/95). The share of fixed term contracts in 1994 was 13.5 percent, well above the OECD average of 10.5 percent (OECD 1995) and subsequently has risen to some 17 percent in April 1997. According to estimates by the Ministry of Labor,13 the use of these kind of contracts has recently increased rapidly: indeed, the share of such contracts in new jobs created within the last 12 months reached 60 percent in April 1997. In early 1997, the regulations on the use of fixed term contracts were further relaxed. This could be one of the factors behind the increase in the flow in and out of employment recently observed (see Section B). In this respect, further efforts are necessary to create incentive to extend use of fixed-term contracts within the private sector.

20. Part-time jobs are allowed but have not been widely used, because, in general, they have been associated with lower skills and lower pay (SM/95/208, 8/21/95). The share of part-time employed in total employment has not changed appreciably in the last decade, from 8.2 percent in 1983 to 8.0 percent in 1996. This lies well below the average share for OECD countries, which has increased from 14.4 percent in 1983 to 18 percent in 1995. To some extent, the small share of part-time jobs appears to be related to the generous unemployment benefit system which does not provide an incentive to take on low paid jobs (OECD 1995).14

21. Working hours have been fixed at 40 per week, with the total amount of overtime hours restricted to 330 hours per year. Over the past two years the government has continued to push for increased flexibility in working hours arrangements. In the 1994-95 negotiation round, unions agreed to contracts in which suitable working hours could be negotiated at the local level, particularly for the export industry and the service sector. In 1996, the flexibility in the determination of working hours was further increased with the introduction of the so-called “compressed working week”, a new Working Time Act, which allows more flexibility in hours of work at the workplace level came into force in November 1996.15 16

Unemployment benefits

22. In the tradition of the Nordic countries, Finland has a well developed and generous unemployment benefit system. A job-seeker between 17 and 64 who has registered for full time employment and for whom a suitable job, a training program or another labor market measure cannot be found, is eligible for unemployment benefits. Persons receiving a pension or maternity allowance or who have turned down job training programs do not qualify.

23. The Finnish system has historically comprised a basic allowance (Unemployment Allowance) of unlimited duration, and an earnings-related allowance (Unemployment Insurance) for members of the unemployment funds for a specified period (see below paragraph 25). In the 1980s the system become progressively more generous. A reform in 1985 significantly rose before and after tax unemployment insurance benefits (OECD 1995) and relaxed some of the eligibility criteria; subsequently the duration of the earnings-related entitlement was increased (OECD 1995, Economic Survey 1988).17 As a result the after tax replacement rates rose to levels among the highest in the OECD area (OECD 1995 and 1996, see below paragraph 28).18

24. The regulations for the financing of the unemployment benefit system (see Section D), have implied that employers’ contributions had to rise when unemployment rates surged at the beginning of the 1990s, thus increasing gross labor costs. The OECD “Jobs Strategy” recommendations suggested a change in some of the regulations governing entitlement to benefits, to encourage job search, and to allow a reduction in private sector unemployment insurance contributions (SM/95/208, 8/21/95) and therefore of gross labor costs.19

25. The system was reformed in 1994 to comprise: (a) a basic daily allowance for 500 workdays for which there is no means testing, if there has been six months of employment in the two years preceding unemployment; previously, the basic daily allowance was strictly means tested; (b) a labor market support allowance, which is paid, upon means testing, to those with no previous employment, and without means testing for 180 days and upon means testing after that, for those or those no longer eligible under other schemes; this has unlimited duration; and, (c) an earnings-related allowance for 500 workdays for those who qualify for the basic daily allowance and are members of unemployment funds. Until 1992, the earnings related allowance used to be calculated as 45 percent of the difference between the previous daily wage and the daily unemployment allowance; this share was then reduced to 42 percent.

26. In effect, the result of this reform was to broaden the base for the entitlement to the basic allowance and, once again, extend the duration of benefits. Subsequently, the persistence of high unemployment rates and the surge in expenditure of the unemployment funds, increased the pressure to reform the unemployment insurance system.

27. In the beginning of 1997, the unemployment benefit legislation was changed to entail a better targeting of the benefits, stricter requirements for the entitlement and some decline in the duration and extent of the benefits. Specifically: (a) the period of previous employment required to be eligible for the earnings-related allowance was raised from 6 to 10 months; (b) the age limit to qualify for additional days of the earnings-related benefit for the long term unemployed, was raised from 55 to 57,20 c) the age limit to qualify for labor market support was increased from 20 to 24 for young people with no training;21 and (d) the waiting period for receiving labor market support was raised from 5 to 7 days. It was also decided that if the earnings-related allowance is renewed after a period of subsidized employment, the allowance will be calculated only on 80 percent of the previous earnings (rather than 100 percent as before).22

28. A series of OECD studies contain international comparisons of the income replacement rates for workers earning average salaries.23 24 Finnish gross replacement rates do not appear to be particularly generous compared to those of other OECD countries (OECD 1994): in 1994, the replacement rate for a single earner was about 53 percent of average earnings, against 52 percent of the OECD average. However, net replacement rates, after taxes and including other benefits such as child and housing allowances, particularly for longer durations of unemployment, are among the most generous in OECD countries (OECD 1995, SM/95/208, 8/21/95) (Tables 7 and 8).

Table 7.

Replacement Rates in the OECD 1/2/

article image
Source: OECD (1995) “The Jobs Study”

The replacement rate is the unemployment benefits payments as a percentage of average earnings. Average earnings are defined in terms of Average Product Worker wages (APW), which are the earnings equivalent to those of a full time worker in the country.

All rates reported refer to a single earner household formed of couple with no children.

Table 8.

Finland: Unemployment Benefit Statistics

article image
Source: ETLA database and staff calculations.

1990 prices.

Average makkaa; paid per day to unemployed workers.

Total benefits paid in the year, on average, to an unemployed worker.

29. The 1996 reform reduced somewhat the generosity of the system, but overall, the cuts in the extent and duration of the benefits have been quite minor. Thus unemployment trap situations remain quite possible.25

Labor market programs

30. Government operated active labor market programs comprise subsidized employment and job training programs. The percentage of the labor force in subsidized employment has doubled since pre-recession years to reach in 1996 around 3 percent of the labor force. Training programs have increasingly been directed to the younger segments of the population, offsetting the loss in the entitlement to labor market support. The benefits of the labor market training programs have been controversial (SM/95/208, 8/21/95). In particular, there is some evidence (OECD 1996) that the effectiveness of the programs has been declining, as evidenced by an increasing percentage of program participants who have remained unemployed.

D. Labor Taxation

31. Total taxes paid on labor income comprise (i) taxes paid by the employee, typically income tax and social security contributions; and (ii) taxes paid by the employer, usually social security contributions and payroll taxes, referred to as non wage labor costs. Total labor taxation, or the tax wedge, can be defined as the ratio of total labor taxes to gross labor costs (Figure 3).26

Figure 3.
Figure 3.

Finland: Tax Wedge

Citation: IMF Staff Country Reports 1997, 060; 10.5089/9781451813135.002.A001

Sources: OECD Revenue Statistics, Statistics Finland and staff calculations.1/ The tax wedge is defined as the ratio of sum of income taxes and social security contributions to gross labour costs, defined by total wages (take home pay plus income taxes and employees’ social security contributions) plus employers social security contributions.

32. In Finland employees pay income taxes and contributions to the National Pension and National Sickness Insurance schemes in percentage of their wages. Since 1992 they are required to contribute to the expenditure of the Unemployment Insurance funds in the measure of 5.5 percent of total expenditure of the funds; this percentage was raised to 18 percent in 1995. In 1993 employees were also required to contribute to the funding of employment pensions schemes, also in percentage of their wages (SM/95/208, 8/21/95). Employers contribute to the social security funds which administrate National Pension and National Sickness Insurance and Unemployment Insurance schemes,27 with contribution rates determined each year in the budget. The contribution rate for the Unemployment Insurance funds is determined as a percentage of the total expenditure of the funds, and was initially fixed at 47 percent28 In 1995, this percentage was lowered from 47 to 32. Finally in the spring of 1996, the government fixed the nominal rate of employers’ contributions to the unemployment funds at their current level. This would imply that the share of employers contributions would return to 47 percent from its current 32, since expenditures are projected to decrease. Historically, labor income taxation in Finland has been among the highest in the OECD, as can evidenced by comparing labor tax revenue as a percentage of GDP with other countries (Table 9): in 1992 labor taxes represented around 30 percent of GDP against the OECD average of 20 percent. Average and marginal income tax rates have been high in Finland with respect to other OECD countries and have not undergone much change until very recently29 (OECD 1995) (Table 10). Moreover, social security contributions, have increased substantially, both for employers and employees, after the surge in expenditures of the unemployment funds that followed the recession. In 1995 employers’ social security contributions amounted to roughly 25 percent of total paid wages and have represented the main source of increase in real labor costs in the past four years (Table 11).

Table 9.

Total Taxes on Labour as a Percentage of GDP

article image
Source: OECD (1995) “The Jobs Study”.
Table 10.

Finland: Income Tax Rates 1/

article image
Source: OECD (1995) “The Jobs Study”.

Average income tax rates.

Earnings are defined in terms of the average production worker wage (APW), which is calculated on the basis of the average earnings of a full time worker in the country.

Table 11.

Finland. Labor Income Taxation

article image
Source: OECD Revenue statistics, Statistical Yearbook of Finland, and staff calculation.

Defined as average paid wages plus social security contributions deflated by the producer price index and averaged over total workers.

Defined as average earned wages after taxes and social security contributions defeated by the consumer price index and averaged over total workers.

In 1993, employees were required to contribute to the funding of employment pension schemes

Defined as total paid wages plus social security contributions.

Defined as total earned wages net of taxes and social security contributions.

33. The overall level of labor taxation can be measured by the tax wedge on labor income, defined as the difference between employers’ gross wages and employees’ take home income. The average tax wedge has increased by around 20 percent during 1989-95, thus representing an important factor adversely affecting the labor market.

The impact of taxation on unemployment: empirical evidence in Finland and OECD countries

34. The taxation of labor income opens a wedge between the net after tax take-home pay of the workers and the gross labor costs to the employers. The overall effect of taxation on the labor market will depend on the relative elasticities of the supply and demand curves and on the assumptions on labor market behavior.30 A survey of the effects of taxation on unemployment can be found in Zee (1996) and OECD (1995).

35. Taxation will reduce real wages received by workers and thus lower the benefit of working with respect to enjoying leisure, this will tend to decrease the participation rate (Killingsworth 1983) and employment. An increase in taxation which reduces the net after tax real wage with respect to the unemployment benefits will reduce the incentive to search for jobs (Mortensen 1986), and thus employment. In general, taxation will reduce labor demand if it implies an increase in real labor costs. This will occur when it is not possible for the employer to pass through increases of payroll taxes into lower nominal wages, or when increases in income taxes result in higher wages demanded. The theoretical outcome on the labor market of increased taxation will then depend on the elasticity of labor demand and labor supply, whether perfect or imperfect competition is assumed in the goods market, and on how wages are determined. Taxation will have no long run effect on unemployment levels (Elmeskov et al. 1993) only if (i) workers bear the burden of taxation in the long run through a decline of real consumption wages; (ii) taxation applies equally to wages and unemployment benefits; and (iii) the labor demand curve does not change in the long run.31 However, in the short run taxation may lead to increased unemployment.

36. There is a considerable amount of empirical investigation seeking to evaluate the impact of taxation on labor market decisions (OECD 1995, Zee 1996). In cross country studies, the evidence of a negative relationship between the level of taxation and unemployment rates is generally weak (OECD 1994). However, many time series studies support the hypothesis that labor taxes increase real product wages and therefore unemployment (Bean, Layard and Nickell, 1986).

37. The evidence for Finland is mixed. Eriksson, Suvanto and Vartia (1990) found evidence that the elasticity of wages to the wedge was 1 in Finland during 1970-85, thus suggesting that tax increases were shifted back to real consumption wages leaving real product wages, employment and unemployment unaffected.32 Calmfors and Nymoens (1990), provide further evidence of this effect, while estimating a wage determination equation over the period 1962-85 for Sweden, Norway, Denmark and Finland. Only in the cases of Norway and Sweden do they find evidence for some long term impact of taxation on labor costs while they find no evidence of an effect of taxation on labor costs in the long term for Finland. Using more recent Finnish data, Tyrväinen (1995 a) reached different conclusions. His estimates led him to reject the hypothesis that the overall effect of the wedge on wages was 1. Real producer wages would be resilient. Indeed he found a negative long run elasticity of wages to changes in employers’ payroll taxes of around 0.3 while the elasticity was of employment of around 0.2. In another study, where the labor demand and supply equations are formulated in a slightly different manner (Tyrväinen 1995 b), and the model is estimated for 10 OECD countries, the elasticity of wages to increases in the wedge is found to be 0.5 in Finland, a coefficient which is similar to that of Japan and Australia, lower than that of Canada and Germany, but higher than that of the U.K., Italy, and Sweden.

38. Using a model similar to that specified by Tyrväinen (1995 b), the impact of the wedge on wages and employment was estimated on Finnish data (econometric procedure and results are reported in Appendix III) extended to 1995. The estimated elasticity of wages to the wedge was around 0.4, not very different from that reported by Tyrväinen (1995 b). However, the impact on employment of changes in the wedge, which are transmitted through changes in wages, appears to be greater than found in previous work. This suggests that the elasticity of employment to the wedge has increased after the recession.

E. Conclusions

39. In recent years, Finland has experienced a dramatic increase in unemployment which, to a great extent, reflected the severe economic recession that hit the country in the early nineties. The problem has remained severe on account of the rigidities in the system that did not allow for a timely reaction to the shock.

40. At the beginning of the recession, labor market regulations and the wage negotiation process resulted in real wage rigidity. Moreover, the increase in unemployment was exacerbated by rising taxes on labor and the effect of the increase in the level of unemployment benefits since 1985.

41. To overcome the negative developments in the labor market, the Government changed labor market regulations to allow increased flexibility in work arrangements with regard to both the nature of employment contracts and the number of the working hours. These changes have had positive effects, including the recent increases in the flows into and out of employment, but did not contribute much to the decline in unemployment rates.

42. The unemployment benefit system remains very generous, especially in terms of duration, and eased the pressure on job-seekers to take on part-time or low paid jobs. High benefits involve a high reservation wage for low skilled workers in low productivity sectors and thus limit downward the movement of wages.

43. Taxation has been increasing, not least as result of the perverse mechanism that ties social security contributions to the expenditure of the unemployment funds. This indicates that a vicious circle may emerge in which high unemployment leads to higher taxes which lead to higher labor costs and higher unemployment rates. Econometric analysis has confirmed that changes in the tax wedge have a powerful effect on employment.


article image
Source: OECD (1996).


  • Beenstock, Michael, (1988), Modelling the Labor Market Chapman and Hall, London.

  • Bean, Charles R., Richard C., Layard and Stephen J., Nickell (1986), “The Rise in Unemployment a Multi-Country Study,Economica, 53: S1-S22.

    • Search Google Scholar
    • Export Citation
  • Calmfors, Lars, and Ragner Nymoens (1990), “Real Wage Adjustment and Unemployment Policies in the Nordic CountriesInstitute for International Economic Studies WP No. 461.

    • Search Google Scholar
    • Export Citation
  • Calmfors, Lars, Ed., (1990), Wage Formation and Macroeconomic Policy in the Nordic Countries. Oxford University Press.

  • Jackman, R., Richard, Layard and Stephen J., Nickell (1996), “Combating Unemployment: Is Flexibility Enough?CEPR working papers 293.

    • Search Google Scholar
    • Export Citation
  • Elmeskov, Jørgen, and Karl, Prichelnam (1993), “Unemployment and Labor Force Participation, Trends and CyclesOECD Economic’s Department Working Papers No. 130.

    • Search Google Scholar
    • Export Citation
  • Elmeskov, Jørgen, (1993), “High and Persistent Unemployment: Assessment of the Problem and its CausesOECD Economic’s Department Working Papers No. 132.

    • Search Google Scholar
    • Export Citation
  • Eriksson, Tors, Antti Suvanto and Penti, Vartia (1990), “Wage Formation in Finlandin Wage Formation and Macroeconomic Policy in the Nordic Countries: a Summary. Calmfors L. Ed. Oxford University Press.

    • Search Google Scholar
    • Export Citation
  • Henry, S.G.B., and Dennis, Snower (1996), “Economic Policies and Unemployment Dynamics in EuropeIMF publications.

  • Ministry of Labour, (September 1996), “Follow-up Report for Finland’s Employment Programme.”

  • Nickell, J. Stephen, (1990), “Unemployment: A SurveyEconomic Journal, Vol. 100 pp. 391439.

  • Nickell, J. Stephen, (1995), “Labor Market Dynamics in OECD CountriesCEPR Discussion Paper No. 255.

  • OECD, (1994), The Jobs Study: Evidence and Explanations. Part II.

  • OECD, (1995), The Jobs Study: Taxation, Employment and Unemployment.

  • OECD, various years, Country studies: Finland.

  • OECD, (1996), Employment Outlook.

  • Tyrväinen, Timo, (1995a) “Wage Determination, Taxes and Employment: Evidence from FinlandBank of Finland Studies, E:3.

  • Tyrväinen Timo (1995 b), “Real Wage Resistance and Unemployment: Multivariate Analysis of Cointegrating Relations in 10 OECD CountriesOECD, Jobs Study working paper series No. 10.

    • Search Google Scholar
    • Export Citation
  • Tyrväinen, Timo, (1995 c), “Wage Determination in the Long Run, Real Wage Resistance and Unemployment: Multivariate Analysis of Cointegrating Relations in 10 OECD EconomiesBank of Finland Discussion Papers 12/95.

    • Search Google Scholar
    • Export Citation
  • Zee, Howell, (1996), “Taxation and UnemploymentIMF working paper 96/45.


Prepared by Luisa Zanforlin; Tarhan Feyzioğlu contributed to Appendix III.


These issues are also discussed in SM/94/219 (8/15/94) and SM/95/208 (8/21/95).


See footnote 20 and Section C paragraph 27.


In the beginning of 1997 the legislation was changed to increase by two years the minimum age to for entitlement to additional days of unemployment benefits.


In spite of the overall decline in the labor force, female participation rates have remained among the highest in OECD economies.


This calculation controls for the increase in productivity which occurred in all sectors during 1987-95.


At the actual labor force participation rate of 1995.


Turnover, or total sales, was analyzed rather that output per worker because more disaggregated data were available.


Real producer wages were calculated by using sectoral nominal wages and deflating them with the producer price index of the respective good.


In 1993, in an attempt to favor youth employment unions agreed that a wage 10-50 percent below the minimum could be offered to unemployed young people. Young people were entitled to refuse such jobs without loosing the entitlement to the unemployment benefit. This kind of contract has had little success, and has been virtually abandoned.


Unless the dismissal is ruled as unfair by a Court of Justice.


The advance notice to be given to the worker is also short: two months notice for those employed more than four months and six month notice for those employed more than fifteen years. The regulations were changed in 1996, as part of the implementation of the jobs that have been employed for less than a year.


“Follow-up Report for Finland’s Employment Programme” Ministry of Labor, September 1996.


The definition of part-time jobs was changed in 1997. According to the new definition the share of part-time jobs in the first quarter of 1997 was 11 percent.


The “compressed week” allows for extended shifts and week-end only shifts.


Since 1996, employer and employee can agree on a job alternation leave from 3 months to a year, where the employee would receive 60 percent of the unemployment allowance he/she would otherwise have received, and the employer is obliged to hire an unemployed person during the employee’s leave, albeit not necessarily in the same task.


In 1987 the Employment Act required the state to provide either subsidized employment or training for the long term unemployed. The effect of this regulation, in particular the requirement to provide short periods of employment, generates successive renewals of the earnings related allowance and therefore implicitly extends the duration of the benefits.


The replacement rates is the unemployment benefit payment as a percentage of previous earnings.


See also Appendix I.


Workers above age 55 can claim the unemployment insurance (UI) benefit until they reach age 57, then claim extension of the benefits until they are 60 years of age, then claim an unemployment pension until they reach age 64, and finally qualify for workers pension. This structure of benefits effectively lowers the pension age to 55 and explains the high unemployment rates registered among this segment of the labor force (Section I).


This explains, at least in part, the decline in the labor force for the relevant age group (Section I).


A calculation of the unemployment allowance before and after the 1996 reform is presented in Appendix II.


In the OECD studies the concept of average production worker (APW) level of earnings is developed to allow for cross country comparisons.


An “unemployment trap” situation occurs if the worker faces an income loss when shifting from unemployment to employment. For example, if a worker shifts from unemployment to a job where he earns % of the APW, this may increase his gross income by 22 percent. However, at the increased level of income he will face higher marginal tax rates and loose the entitlement to housing and child care cash benefits. According to recent OECD calculations, this will imply an income loss of 51 percent (OECD 1996); only if tax deductions for work-related expenses are allowed for, there is a positive net income gain of around 2 percent.


Gross labor costs are defined as total wages plus social security contributions paid by the employer.


The Unemployment Insurance scheme finances the earnings-related allowance system; the Labor Market Support is entirely publicly funded.


Employers contributed 47 percent and the state budget financed the remainder. From 1992, employees were also required to contribute (see text).


In 1997 the main changes introduced were: (i) a 1 percent reduction of all earned income marginal tax rates at the central government level; the range is thus from 6 percent to 38 percent; (ii) a decline in the average tax rate from 17.53 percent to 17.43 percent at the local government level; (iii) a 4 percent increase of the brackets for the earned income tax at the central government level; (iv) an increase in the earned income deduction at the local level. Together these changes are estimated to reduce revenue by Fmk 5.5 million, roughly equivalent to 1 percent of GDP.


For a survey of analytical models of unemployment see Nickell (1990), Snower (1995).


This implies that in the long run the elasticity of the labor demand curve does not change.


The implication is that labor supply is completely rigid in the long run, as assumed in Elmeskov (1993).