Chapter 5 Trends in between- and within-group
5.2.1 Trends in aggregate time series
The period from 1977 to 1990 was characterized by strong economic growth. In 1977, the Finnish economy was in recession. The recovery started in 1978 and stable growth continued during the 1980's. An average annual growth rate of the real earnings of the median worker was almost 2.5% between 1977 and 1990. Turnaround in the stable growth came with the economic crisis in 1990. Average real earnings declined for several years and were in 1995 almost at the level of 1990. Figure 1 shows the development of real annual earnings of wage earners by deciles based on the published figures of the Income Distribution Survey. For comparability over time, the figures include all wage earners, therefore combining part-time and full-time workers as well as workers who have been employed only part of the year.
Obviously this causes a problem when comparing periods with high and low unemployment rates.
Figure 1. Annual real earnings by decile in 1990 marks.
Note: Data includes all wage earners.
Source: The Income Distribution Survey. The Survey was not conducted in 1981 nor 1985.
Aggregate measures of inequality
Dispersion of earnings among full-time workers has increased for both men and women from 1980 to 1989. Since the gap between men and women simultaneously decreased, the rise in inequality among all employees was less than among men and women separately. Among women the top decile gained compared to the median, whereas among men the widening of earnings distribution occurred both in the 90/50 ratio and in the 50/10 ratio. After 1989 the earnings dispersion declined sharply. Since figures are based on data on full-time year-round workers, this decline was probably partly caused by the relative increase in unemployment among low earners.
Figure 2. Trends in earnings dispersion.
Note: Data includes only full-time year-round workers. Missing data from the years 1978, -79, -81, - 82, -84 and -85 interpolated linearly using the previous and following year with non-missing data.
Source: OECD, Employment Outlook 1996 Table 3.1. Year 1977 author’s calculations from IDS microdata.
Trends in the return to education
Earnings differences between different educational categories continued their decrease that had started earlier in 1970's (Eriksson and Jọntti, 1997). Some leveling of the trend, and even a partial rebound in returns to education can be seen in between 1985 and 1990 but the decline of relative wages of more educated employees seems to have continued during the 1990's recession. The 1990's development is again problematic: an increase of unemployment was more severe for workers with less education. However, it is not clear whether this increased or decreased the wage differences among the employed workers.
Besides the trends, also the levels of relative earnings are of interest. Due to the classification used by Statistics Finland, a college - high school wage differential, often used in international comparisons, cannot be calculated from these data.2 Here a more natural comparison is the earnings premium of various education categories compared to employees with only the compulsory nine years of education. On average, during the period under the study, men with university education have earned approximately twice as much as men with only compulsory education. For women the difference has been approximately 80%.
Interestingly, lower secondary education (two years of vocational schooling) seems to bring no wage gain. In several years, the average earnings of employees with lower secondary schooling have been less than the average earnings of employees with only compulsory schooling3.
2 High school graduates are classified as having ''upper secondary education''. However, in the 1995 census only slightly over 40% of all persons classified into this group had a high school diploma as the highest degree. The rest had completed some vocational upper secondary level education (SVT 1997).
3 This is partly explained by differences in work experience. The share of workers with only compulsory schooling is higher in the older cohorts.
Figure 3. Annual earnings of employees by the level of education compared to earnings of employees with compulsory level education.
Source: The Income Distribution Survey. Data includes both part time and full time workers.
Trends in the supply of skills into labor market
Educational attainment of new labor market entrants has risen quickly. As the new entrants replace the older, less educated workers, the average education level of the labor force increases, though more slowly than the average education level of the new entrants.
Therefore, education is rather unequally distributed across the age groups. In 1995, 25-34 year old Finns had a higher average education level than those in the same age group in other Northern-European countries, but 55-64 old Finns had much less education than their Northern-European counterparts. (OECD Education at Glance 1995).
Figures 4a and 4b show the trends in the levels of education among wage earners between 1977 and 1995. Clearly, for both men and women the number of workers with only compulsory education has steadily declined. It appears that the share of lower secondary education first increased but started to decline as the share of those with upper secondary schooling increased. Although the number of employees with higher education has grown more than 50% for men and almost 90% for women, their share of all employees, even at the end of the period, was only about 20% .
Figure 4. Wage earners by the level of education.
Source: The Income Distribution Survey. Missing data from years 1981 and -85, when the Survey was not conducted, replaced by the average of previous and following years.
Trends in industry composition of the labor force
The composition of employment by industry has also changed rapidly in Finland. In 1970, a quarter of the workforce was still involved in primary production. By 1995, the share of primary production had declined to 8%. Employment in the service sector has increased fairly steadily since 1970 reaching 65% of all workers in 1995. Since the service sector employs women and employees with higher education in larger proportion, the increase of demand for services benefits these groups. As a matter of fact, the service sector has largely absorbed the increase in the labor force, caused by an increase in women's labor force participation rate.
Interestingly, manufacturing and construction maintained their share of employment until the economic downturn in the beginning of 1990's. In fact, the absolute levels of employment in both manufacturing and construction were almost identical in 1970 and in 1990. The largest change in the industrial structure occurred in the 1990's when the economic crisis wiped away some 200,000 manufacturing and construction jobs.
Figure 5 Employment by industry.
Note: Classification based on SIC 1979, after 1990 on SIC 1988. Change in classification corrected by adjusting the figures after 1990 using the ratio of workers in each category according to SIC 1988 and SIC 1978 in the year 1989 when both figures were available.
Source: Labor Force Surveys 1970-1995.