In order to compare the dispersion of wages between workers in collective bargaining and those not covered by collective agreements, it is important to take into account a possible sampling: for example, if collective agreements mainly concern men or certain sectors, the dispersion of wages may be lower in collective bargaining, wages tend to be more similar than for men or in certain sectors. than in the labour force as a whole. Different empirical techniques can be used to account for these differences in composition between the negotiating groups. The regression used in this section dates back to Juhn, Murphy and Pierce (1993) and has since been widespread.19 For each country and each level of bargaining, a regression of the standard hourly wage is performed separately on a large number of explanatory variables: age, sex, education, size of enterprise, type of contract, years of employment in the enterprise, the sector and the profession. Differences in composition are then corrected by replacing the coefficients and residues of each level of collective agreements with those that are not accounted for. Box 3.1 describes the empirical approach in detail. Collective bargaining is a two-party process. Both parties – employers and workers – are taking action together. There is no intervention by third parties. It is a method of data and reciprocal taking and not a take-it-or-leave-it method to reach the settlement of a dispute. For differences in composition, the empirical approach to the adaptation of wage statistics, i.e. the .
B the average salary or the ratio D9/D1, is as follows. Workers whose wages are not governed by collective bargaining, b 1, are used as a reference. In Belgium, France and Spain, where no data are available for unsersered workers, collective bargaining at company level is used as a reference. The counterfactual wage of employees covered by collective bargaining b2 is then calculated, because even if the reduced returns of age, education and seniority explain in some way the lower dispersion of wages through collective bargaining, it is essentially non-observer factors that reduce the dispersion of wages (Figure 3.6). The work of collective bargaining takes many forms. First, negotiations can take place between the single employer and the single union, which is called sole proprietorship bargaining. This form prevails in both the United States and India. In France, Italy and Portugal, no change in the level of bargaining and the degree of effective centralisation has been observed, as collective bargaining on wages in these countries remains fairly centralised, as stated in Chapter 2. This may seem surprising, given that in recent decades the principle of proper functioning has been reversed in France and other reforms have favoured negotiations at company level.
However, as noted in Chapter 2, decentralisation in France concerned only non-wage working conditions, while wage negotiations remained exclusively in the hands of sectoral negotiations. .