We note that retroactivity can play an important role in explaining the negative effects of expansion on the employment of unrelated firms (Table 5b). The degree of retroactivity is measured by the number of days between the entry into force of the collective agreement between related companies and the date of publication of the extension to non-related companies. The negative average effect of salary reflects the effect of extensions on the variation in employment growth for the typical administrative delay (180 days). The interaction between the processing effect and the administrative delay results in a one-day increase in administrative delay to the change in the rate of post-extension employment growth. This situation is negative for unrelated companies, while for related companies it is insignificant or even positive. The difference between related and unrelated businesses is consistent with the discussion before us above, as retroactivity is expected to directly affect unrelated firms, while it may indirectly benefit related firms by reducing competition between unrelated firms. Another option is to use a fuzzy DDR that, based on the two data points mentioned above, explicitly takes into account the small decrease in the likelihood of an extension between January 24, 2011 and February 28, 2011. Formally, DSD fuzzy can be described by a result equation and a treatment equation. The processing equation models the probability of renewal of a collective agreement based on relative time based on a constant (α), a decoy (Ti) equal to a date from which the probability of an extension is zero, and a function that controls the effect independent of relative time on both sides of the threshold (f.): Overall, the results do not allow for strong conclusions about the effectiveness of representativeness criteria in mitigating the negative effects of extensions. However, even if the representativeness criteria do not effectively ensure that the interests of unaffiliated firms are taken into account, representative criteria can play a useful role in the long term in encouraging the degree of organization of employers, particularly if they are gradually implemented over time. This can be useful in itself, as it can help improve the quality of labour relations and the degree of trust between social partners (see also Box 3.2 at the IMF, 2016). We are now looking at the role of representativeness and retroactivity in the impact of extensions on the employment performance of unrelated and related businesses.
The level of representativeness of employers` organisations is measured by the share of employees of related companies in the total employment of the sector concerned. This definition corresponds to the criterion of representativeness introduced in the context of the 2012 labour market reform. In the current context, controlling the relative effects of time is essential for two main reasons. First, the portion of the year in which companies are subject to extensions depends on the date of extensions, as dependent variables will be measured in October 2010 and October 2011. A subsequent extended agreement necessarily has a shorter period of time to produce effects than an agreement that was extended earlier. Second, economic conditions can affect both the timing of the agreements and their actual content. For these reasons, the results probably depend on the relative time. Our analysis assumes that the relative effects of time are linear or square and may differ between the two sides of the threshold.9 Regressions are weighted (based on the number of people employed in 2010) in order to obtain the average effect of extensions on total employment (not the average effect of extensions on collective agreements). Panel B: Regressions are weighted by the number of employees in 2010.