The results are robust for a variety of different specifications (Table 3). By focusing on specification with linear relative time effects, we now compare the basic results based on fuzzy RDD and net RDD with alternative specifications or samples (panels A and B) in the following order. “Controls” include only membership status as control and not average company size and export intensity, as provided for in the basic specification. “Bandwidth” expands the observation window to include agreements signed after the new government took office in June 2011 until the end of August 2011. “Falsification” assesses whether there is discontinuity when a placebo date is used to evenly distribute the number of chords on each side of a fictitious threshold (around mid-May) after reform (March-August). The results indicate that the exclusion of controls and the extension of bandwidth do not change the quality of the results relative to the base value. The falsification test does not indicate an interruption of employment growth around the fictitious reform date. The rest of this document is structured as follows. Section 2 presents the economic and institutional context at the time of the experiment. Section 3 describes the experience, explains how it is used using a DDR, and explains the validity of this approach in this context. In Section 4, our coordinated employer-employee dataset is supplemented with information on collective agreements and renewals (including their schedule). Section 5 presents evidence of the impact of the extension of management in all areas and, separately, for companies affiliated with an employer organization and companies that are not and are at the occupational level. It also analyses how the impact of enlargements depends on the degree of representativeness of employers` organisations and the role of retroactivity linked to administrative delay in the treatment of enlargements.
Section 6 presents some additional results with respect to wage inequality. Finally, Section 7 is concluded. In the current context, controlling the relative effects of time is essential for two main reasons.