Temporary employment: The trade-off between efficiency and equity

Elke Jahn, Regina T. Riphahn, Claus Schnabel, 10 October 2012



Over the last three decades, the use of flexible forms of employment such as fixed-term and temporary agency work contracts has increased substantially throughout much of Europe. This development has been driven by government efforts to ease restrictions on temporary employment, whereas the regulation of permanent contracts has been left essentially unaltered. The reforms of temporary employment have intended to increase overall employment by lowering dismissal and adjustment costs for flexible jobs and thereby providing firms with new opportunities. Generally, two-tier labour markets can increase labour market flexibility when it seems to be politically infeasible to reduce employment protection for workers with permanent contracts. Moreover, if a considerable share of flexible jobs is ultimately transformed into regular jobs, aggregate unemployment might decline.

Some pros and cons of increased labour market flexibility

Increasing labour market flexibility is widely expected to increase firms’ productivity and competitiveness, for instance by enabling firms to screen workers, by avoiding firing costs and by reducing labour costs. In some cases employees may benefit from flexible work arrangements as well: additional job opportunities can make it easier for workers to enter the labour market or to escape unemployment, and they generate the possibility to accumulate human capital on the job. However, increased labour market flexibility comes at a price: as flexible workers are less protected against job loss than workers with permanent jobs, they face higher unemployment risks and lower job tenure. Moreover, working conditions in flexible jobs are often poor.

A crucial question for both academics and politicians is how the deregulation of employment protection legislation has affected the labour market and society as a whole. A number of articles have addressed the effects of such reforms on labour market outcomes and have surveyed the extant evidence (e.g., Boeri 2011). However, we know relatively little about the connections between employment protection regulation and aggregate efficiency and equity. In recent research (Jahn et al. 2012), we identify some aggregate patterns suggesting that there appears to be a trade-off between efficiency and equity when deregulating employment protection.

Flexible employment and macroeconomic outcomes

We study the correlation between labour market regulation and the incidence of flexible employment on the one hand, and overall employment, productivity, and inequality of income distribution on the other. We look at the EU15 countries in the period 1985-2008 and use the OECD indicators of employment protection for regular and temporary employment (see Venn 2009) as well as shares of flexible employment (fixed-term and temporary agency work). Note that these four indicators are not independent. In particular, there is a negative correlation between the level of employment protection of temporary jobs and the shares of fixed-term and agency work contracts.

Using these four indicators and aggregate data from OECD statistics (OECD 2011), we estimate pooled least squares regression models with controls for country fixed effects. We analyse the contemporary correlations between employment protection legislation and the share of employment in fixed-term and temporary agency contracts with several indicators of macroeconomic outcomes. The coefficient estimates and their standard errors are shown in Table 1. Each entry is based on a separate regression model.

Table 1. Employment protection, incidence of flexible employment and macroeconomic outcomes; EU15 countries, 1985-2008

Notes: Linear regression results of macroeconomic outcomes on the strictness of employment protection legislation (EPL) and the incidence of flexible employment forms; each cell represents the results of one OLS regression with country fixed effects; data source: OECD (2011); the Gini coefficient is based on equivalised household disposable income, before and after taxes and transfers, respectively (five years averages); GDP is measured in US $, constant prices, constant PPPs, OECD base year and divided by the total number of employed; the share of agency work contracts is taken from CIETT (2011) and refers to the period 1996-2008; standard errors in parentheses, ** p<0.01, * p<0.05

Source: Jahn et al. (2012).

Efficiency: Employment and productivity

Looking at overall employment first, our results yield broadly intuitive patterns: deregulation (i.e., lower values for the OECD indicators) is associated with higher employment. This confirms the overall thrust of the policy initiative. Higher shares of fixed-term contracts and temporary agency employment go hand in hand with higher employment levels. In the next column of the table, we measure the association of labour market regulations and the share of temporary employment with productivity, i.e., GDP per employed person in a given country and year. The results confirm expectations and show that productivity is negatively associated with higher levels of labour market regulation, even conditional on country fixed effects. Although not all results are robust to various time trend controls, they suggest that lower regulation is associated with higher employment and productivity. Thus, by and large, our evidence seems to support deregulation from an efficiency point of view.

Equity: Income distribution

Efficiency is not all that counts from a wider societal perspective. In the last columns of Table 1 we therefore investigate the association of regulation and the incidence of flexible employment with the equality of the income distribution. We consider Gini coefficients of the distribution of equivalised household incomes, both before and after taxes and transfers. The results are startling: in countries with strictly regulated labour markets, the distribution of household incomes is significantly more equal than in countries with flexible labour markets. These outcomes are robust to controls for time trends and to alternative indicators of inequality, such as the mean log deviation of incomes. We find no strong connection between the shares of temporary employment and inequality, but the associations reported suggest that equality is higher where temporary employment shares are lower.


Although our investigation at the macroeconomic level is largely descriptive and needs to be taken with a pinch of salt, it points to some relationships that have not been given much attention in prior research and by policymakers. Our cross-country analysis indicates that in countries with flexible labour markets, overall employment and productivity tend to be higher but the distribution of household incomes is less equal than in countries with strictly regulated labour markets. This suggests that there may be a trade-off between equity and efficiency when dual labour markets are supported. Temporary employment can thus be both a boon and a bane to the labour market and to society as a whole. It remains an open question whether the gains from enhanced labour market flexibility outweigh the costs.


Boeri, T (2011), “Institutional reforms and dualism in European labor markets”, in O Ashenfelter and D Card (eds.), Handbook of Labor Economics, 4B:1173-1236, Amsterdam: Elsevier BV.

CIETT (2011), CIETT database, http://www.ciett.org.

Jahn, EJ, RT Riphahn, and C Schnabel (2012), “Feature: Flexible Forms of Employment: Boon and Bane”, The Economic Journal, 122: F115-F124.

OECD (2011), OECD STAT database.

Venn, D (2009), “Legislation, collective bargaining and enforcement: updating the OECD employment protection indicators”, OECD Social, Employment and Migration Working Papers No. 89, Paris: OECD.

Topics: Labour markets
Tags: efficiency, equity, temporary contracts, unemployment

Senior Researcher at the Institute for Employment Research, Nuremberg

Professor of Economics at the University of Erlangen-Nuremberg

Professor of Economics at the University of Erlangen-Nuremberg