A central objective of the Lisbon Process was full employment defined as an employment rate of at least 70 percent until 2010. Although most of the EU member states moved ahead, the Lisbon process failed to reach this target. Furthermore, the major part of employment gains was related to non-standard employment, especially in the form of part-time work, fixed-term contracts, temp-agency work and self -employment. Whereas many welcomed this development as a blessing for flexible labour markets, demanding even more of this kind of employment relationships in favour of the Lisbon benchmark, others were highly critical hinting to disastrous intended or unintended side-effects such as low or volatile income, dead-end jobs instead of stepping stones, high job insecurity, and poverty in old-age. In 2003, the European Employment Task Force stepped in as a kind of broker of these two visions by recommending to direct the European Employment Strategy and the related open method of coordination (OMC) towards a proper balance of flexibility and security (Kok et el. 2004). Dubbed already early by ingenious Dutch researchers as 'flexicurity' (Wilthagen 1998), the European Commission took over these recommendations and after long debates eventually succeeded in reaching some kind of consensus about the common elements of the flexicurity strategy (European Commission 2007).
Despite many conceptual drawbacks of the flexicurity strategy (Keune/Jepsen 2007; Schmid 2010a), its central objective of increasing employment and labour force participation is still valid. Even taking into account the fact that the current crisis led to a drastic increase of unemployment in most of the EU -member states, the long-term perspective of most EU member states is still one of labour shortage for two reasons: one quantitative related to the ageing society, one qualitative related to the rapid change of technology and global competition. Whereas migration might fill this gap to some extent, policies raising labour force participation and life-long -learning are generally seen as the more sustainable solution. Furthermore, changing work preferences, especially among women traditionally tied to unpaid work in the private households, hint to unexploited potentials of endogenous factors driving labour force participation. Preferences for labour market participation might still be blocked by institutional barriers of various sorts: employment protection, tax incentives, lack of child care or elderly care infrastructure, and wage discrimination.
Whether one likes the flexicurity-oxymoron or not, a further increase of labour force participation therefore seems inevitably be connected with a greater variety of employment relationships. The aim of the following essay, therefore, is to test this assumption in a preliminary way through systematic descriptive work and conceptual reflections: first by comparing the development of non-standard employment in EU member states from 1998 to 2008; second by relating this development to the dynamics of economic welfare and labour force participation; third by exploring the main determinants of this development; and fourth by discussing the policy consequences to overcome the weaknesses of the current flexicurity strategy and to provide guidelines for advancing the Post-Lisbon employment strategy.
The Change of the Employment Relationship in the European Union
The following view on the dynamics of employment relationships is based on the European Labour Force Survey using the following definitions for labour force participation and non-standard employment:
(1) Activity rate or labour force participation rate = (employed + unemployed) as per cent of working age population (age 15 to 64)1
(2) Part-time employment rate = employed in part-time work and in open-ended contracts or in own account work (2) as per cent of working age population; or part-time employment share as a proportion of total employment
(3) Fixed-term employment rate = employed in fixed-term contracts (including temp-agency work with fixed-term contracts and part-timers in fixed-term contracts) as per cent of working age population; or fixed-term employment share as a proportion of total employment
(4) Self-employment rate = own account workers (self-employed without dependent employees) in full-time as per cent of working age population; or self-employment share as a proportion of total employment
(5) Aggregate non-standard employment rate = sum of (2, 3 and as per cent of working age population. (3)
Figure 1 shows the development of the aggregate non-standard employment rate for 24 EU member states (4). The first pattern that hits in the eyes is that countries belonging to the 'social-democratic' regime (including Netherlands as a 'hybrid') rank highest in terms of the combined indicator for non-standard employment. (5)
[FIGURE 1 OMITTED]
However, with around one quarter of the working-age population non -standard employment is also fairly well developed in the 'liberal' system of UK, and even in family centred or so-called conservative employment systems like Austria, Belgium, France, Germany, Italy, Spain and Portugal. (6)
On the other hand, it is remarkable that most of the new member states cluster together in the left corner of the figure, which means displaying low non-standard employment rates of around 10 percent, and some countries showing even declining rates.
This leads to the second pattern that immediately can be observed from Figure 1. Most countries are situated above the diagonal line that serves as an implicit time axis. If all countries would lie on this diagonal, nothing would have changed from 1998 to 2008; this is true for some countries, e.g. for UK, Greece, and Hungary. Some countries, especially Lithuania and Latvia, experienced even a decline in the aggregate non-standard employment rate. In most other countries, however, especially in Italy, Poland, Spain, Germany and Netherlands, the non-standard employment rate increased by about five to ten percentage points. The decomposition of non-standard employment into its three components of part-time work, fixed-term employment and self-employment, confirms our expectation: part-time work is the most prominent element in non-standard employment of most countries. As already hinted at the beginning by pondering about the definition of "standard" employment from a life -course perspective, there are good reasons to argue that at least open- ended part-time work in the range of 20 to 35 hours deserves to be counted as standard, and not "atypical" anymore. Part-time work is common especially in well developed knowledge and service economies. Part-time employment rates - including the nontrivial number of self-employed people working in part-time - however display great variation between the EU member states, ranging from one percent in Romania to 27 percent for the "champion" Netherlands. Fixed -term employment rates (including part-timers with fixed-term contracts) vary "only" between (roughly) one percent in Romania again and 16 percent in Spain; whereas the self-employment rate (excluding part-time) displays a minimum of two percent (Luxembourg) and a maximum of 12 percent (Greece).
Behind any variation of figures there are possibly hidden patterns. Are these three components of "flexible" employment (part-time work, temporary work, own-account work) complementary or substitutive? A first answer to this question can be found by simply correlating the various forms of non-standard employment across the 24 country observations in 2008. (7)The strong positive correlation between open-ended and fixed- term part-time employment (r=0.71) is intuitively clear since both contractual forms are complementary. One plausibly can assume that a majority of open-ended part-time employment is the continuation of fixed- term part -time work. The same explanation can be given for the positive correlation between fixed-term part-time work and fixed-term full-time work (r=0.34): a substantial part of fixed-term part-time contracts might lead to fixed-term full-time contracts, although such interpretations cannot directly be derived from such correlations. A bit more difficult to explain is the strong correlation between fixed-term part-time employment and part-time self- employment (r=0.61). Common underlying causal factors of this correlation probably are supply constraints, in particular of single or married women (or of the few single men) having children who can devote only part of their time to gainful employment. This interpretation is corroborated by the significant correlation between open-ended part-time work and part-time self-employment (r=0.52). (8)
Finally, the strong negative correlation between full-time self-employment and open-ended part- time work (r= -0.46) indicates a substitutive relationship between these forms of non-standard employment. It seems that not all forms of non-standard employment are driving labour force participation - at least not for all target groups. This substitutive pattern forecasts the decline of full-time self-employment in favour of part-time employment especially for countries that need to catch up with the 'developed' countries in terms of non-standard employment and labour force participation. Furthermore, it can be assumed that formerly self-employed people in agriculture, retailing or sweat-shops transit into dependent part-time work and combine this small but regular income with volatile income from various kinds of informal work on the side (especially in small- sized agricultural production), moonlighting or even illegal work.
The differentiation of these observations by gender provides further hints to the reasons of rising non-standard employment. Figures 2 and 3 clearly show that the variation of non-standard employment among women in the EU is much higher than among men. The minimum and maximum non-standard...