EUROPEAN PAPERS ON THE NEW WELFARE

Part-Time Pensions and Part-Time Work in Sweden

5. Part-time work and labour force participation

The critical question is if a part-time pension system leads to a higher or lower number of hours worked in the economy. To answer that, certain questions have to be posed. The first question is if the part-time pension system leads to more people being employed. The second question is how the number of hours worked per person is influenced. The combined answer to those two questions tells us how the total number of hours in the economy is influenced by the introduction of the scheme.
Part-time work is common in our economies not only among older people. We can learn something from the experiences of part-time work among other demographic groups. I will give some examples on experiences from part-time work below.
Women often work part-time in many countries, more so than men in the same age groups. Is part-time work among women combined with high or low labour force participation? This question has been dealt with by Jelle Visser (2002), who studies the pattern in the EU countries, and especially the development in the Netherlands11. He finds a strong positive correlation between the share of women who work part-time and the number of women who are employed. In countries like Greece, Italy and Spain few women are employed and a low share of them are working part-time, while in countries like Denmark, Sweden and the United Kingdom many women are working part-time and the female employment rate is high. In the Netherlands a rapid expansion of part-time work has been accompanied by a rapid rise in the female employment rate. One interpretation of the development is that in countries in which women previously had a low employment rate, they now enter the labour market by part-time work, and that the possibilities of getting part-time work facilitate that process. Sweden had such a development in the 1970s and the 1980s, when the female employment rate rose at the same time as the part-time rate among employed women increased. During the last two decades more women have started to work full-time (or long part-time) in Sweden, and the female employment rate has continued to be high. According to Visser it is not self-evident that the development will go the same way in the Netherlands as in Sweden. The Netherlands may continue to be a part-time country on a permanent basis.
Young people are another group who often work part-time. In many cases part-time work is combined with (full-time) studies in high school or university. As mentioned earlier the transition from education to permanent jobs is taking place at a gradually higher age. However, there are large differences in labour force participation between different countries. Countries that have early entry into the labour force are mainly those who have an apprenticeship system (not only during the years the apprenticeship lasts) and those where it is common to work alongside studying. Both cases indicate that an early introduction of part-time work with part-time (or full-time) studies may be a way of increasing labour supply12.

A third group of part-time workers are those who combine part-time work with a part-time disability pension. In Sweden it is rather common to combine work and a disability pension (granted on medical indications) and the share of the disability pensions which are part-time pensions has increased. An important question is how the existence of this option influences the labour supply. Cross-country comparisons indicate that labour force participation is relatively high among disabled people in Sweden.
The experiences accounted for here do not show that a high rate of part-time work necessarily implies that the rate of full-time work is reduced to the same extent. The total share of employed people may be higher. If this means that the total number of working hours is higher in countries with a high incidence of part-time work is another question. There are good arguments for carrying out a study of the same type as Visser’s (2002), and this time compare the share of part-time employment and the employment rate among people of older active age. Is it so that countries that have a high part-time employment rate among older people also have a high employment rate in the same age group?
We have done such a study based on observations for the EU countries for the period 1990-2003. The results are given in Tables 3 and 4 below. In both tables the part-time share is the independent variable and the employment rate the dependent variable. Table 3 shows OLS estimates with controls for country and year of observation. In Table 4 fixed effect models are estimated (fixed effects for countries). The main results are that the sign of the coefficient for the part-time share is positive, as expected, in all estimations and that the estimates are significantly different from zero. The value of the coefficient for the part-time share is higher for those aged 55-59, especially for men, than for those aged 60-64.

Table: 3: The employment rate in percentage of people aged 55-64 in the EU countries 1990-2003. OLS regressions with employment rate as the dependent variable and part-time share as independent variable
wande-tabella3_eng.gif
source: Eurostat.
Notes: ***= significant on the 1% level; reference categories are Austria (country) and 1990 (year); observations are missing for Austria, Finland and Sweden for 1990-1994, and for Austria, Germany and Luxembourg, for 2003.

Table 4: The employment rate in percentage of people aged 55-64 in the EU countries 1990-2003. Fixed effect regressions (for country) with employment rate as the dependent variable and part-time share as independent variable
wande-tabella4_eng.gif
source: Eurostat.
Note: ***= significant on the 1% level; **= significant on the 5% level; observations are missing for Austria, Finland and Sweden for 1990-1994, and for Austria, Germany and Luxembourg for 2003.

11 See also Rasmussen et al. (2004) for the development of part-time work among different groups in New Zealand, the Netherlands and Denmark.
12 See for example Ryan et al. (1991).


Pages: 1 2 3 4 5 6 7 8 9 10 11


Tags: , ,