European Commissioner Lazlo Andor – ETUI Conference “Labour Markets in the Crisis”
Product Markets: The Evidence
Labor market: the evidence
The Data
In order to investigate the determinants of labor market regulation, we consider two time- varying measures for twenty-one OECD countries for the period from 1985 to 2003.
These two measures capture the degree of employment protection related to the firing decisions and the level of insurance provided to the unemployed, respectively. Data on the former measures are coded and collected by the OECD and described in the OECD Employment Outlook (2004). The latter data are also collected at the OECD and are described in the OECD Benefits and Wages (several issues); because original data are available only for odd years, data for even years have been obtained by linear interpolation. The indicator on employment protection ranges from 0 to 6 (from least to most restrictive) and measures the restrictions placed on the firing processes by both labor legislation and collective- bargaining agreements. This index includes an assessment of the legislative provisions, as well as the enforcement dimension, as they provide a measure of the judicial practices and court interpretations of legislative and contractual rules. This indicator is also provided separately for regular and temporary workers.
For the regular workers, the indicator on the employment protectory regulation has three main components:
(a) Difficulty of dismissal-that is, legislative provisions setting conditions under which a dismissal is justified or fair;
(b) Procedural inconveniences that the employer may face when starting the dismissal process; and
(c) Notice and severance pay provisions.
The index also provides a measure of the regulation of fixed-term contracts and temporary work agencies. This is intended to measure the restrictions on the use of temporary employment by firms with respect to the type of work for which these contracts are allowed and their duration. The employment legislation for regular contracts constitutes the core component of the overall summary index of employment protective legislation (EPI) strictness that we use.
The indicator on the level of insurance provided to the unemployed represents the unemployment benefit replacement rate for low- income workers in their fi rst year of unemployment. This is measured by the average replacement rate-that is, the ratio of the unemployment benefit to the last wage, and for a worker that earns 66 percent of average worker earnings.
The Euro and Labor Market Reforms
As for the product market, all our regressions are estimated with generalized least squares, allowing for heteroschedasticity of the error term, and include the lagged value of the left- hand side variable and country and time dummies.
We consider the generosity of the unemployment benefits, as defined earlier, to be a measure of labor market regulation. In one column, we start from the basic specification, with tests only for the effects of the European single market and of the euro. We then add the interaction of EMU with the lagged value of the dependent variable (one column, our measures of competition (another column), and additional possible explanatory variables encountered in the literature, such as economic crisis and fiscal and political variables (another column). Finally, in more two columns report the results of the regressions that include the effects of the lagged variable of regulation in the product market, the alternative variable of regulation in the labor market (EPL), and the level of unemployment benefits in the trading partners.
The results show that while the ESM had no impact on this measure of labor market regulation, the introduction of the euro led to an increase in the generosity of the unemployment benefit. No other variable shows any explanatory power, with the exception of the level of unemployment benefits in the trading partners, which presents a puzzling result, however, as more unemployment benefits in trading partners is associated with less unemployment benefits in the home country. When using the degree of EPL as a measure of labor market regulation we do not find any effect of EMU-or any other plausible explanatory variable-on labor market reforms. More generally, we found that this index of labor market reform moved much less than that of product market.
Additional Evidence
The indicator of labor market reform used may give an overly narrow view of the evolution of labor markets in Europe. These indicators of flexibility refer only to the primary labor market. But two other factors, related to each other, have changed. One has been the development of a vast labor market in several countries based on temporary contracts with very few, if any, of the rigidities of the primary labor market.
For instance, much of the increase in employment reported in France, Italy, and Spain has occurred in this secondary market. The second change is that in the last ten or fifteen years, several European countries seem to have experienced a substantial amount of wage moderation. We investigated whether the adoption of the euro has contributed to achieving wage moderation in these seemingly unreformed labor markets. This is of course important as an indicator of second-round effects: that is, whether inflationary shocks get a second- round boost from wage increases. This shows that the countries that joined the EMU in 1999 have experienced a significant increase in wage moderation in the period leading up to the common currency: that is, between 1993 and 1999. After this period, there is no evidence of an additional effect of euro adoption on the degree of wage moderation. These results are consistent with the fact that in preparation for EMU membership, many countries had to put their houses in order.
This meant inflation reduction and fiscal rigor (in areas including public salaries).
More specifically, the dependent variable is the growth of nominal wages. On the right- hand side, in addition to the lagged dependent variable, we have lagged inflation and our variables capturing simple market membership and EMU membership. The former (but not the latter) has a negative and statistically significant coefficient, indicating, at least at first sight, an effect of simple market membership on wage moderation. However, we show that this result is driven by the countries’ membership of the simple market and their preparation to join the EMU and attempts to achieve convergence criteria. In fact, we added a dummy for EMU countries in the run- up to the euro (1993 to 1998) and another one after they adopted the single currency. As this shows, the pre- euro dummy variable has a significant negative coefficient. Meanwhile, the coefficient on the post euro period is insignificant. We also investigated possible differential effects between EMU and non- EMU countries relative to the effects of (lagged) inflation, but we found no differences.
