The outcome of different bargaining models: the effects on wages, employment and the employment mix

The paper analyses data on wages, employment and labour composition in the Uruguayan manufacturing sector during 1985-1999 in order to get some evidence on the effects of union action on these variables. The whole period is first studied using a model in which no assumptions on the underlying bargaining model are made. The results support the hypothesis of two different bargaining frameworks in the 80s and 90s. Therefore, a right-tomanage bargaining model is specified for the 80s and a recursive contracts model for the 90s. Union effects are such that while in the 80s the effect of trade unions were to increase wages and hence decrease employment, in the nineties they moderated wage demands in exchange of more job stability. They not only managed to have a positive direct effect on employment but also to buffer the negative effects of increased openness and demand fluctuations on employment. The existence of unions also had an impact on labour composition, favouring a higher share of non-production workers in total employment. The result can be linked to the fact that firms moved to more capital intensive – or at least more skilled labour intensivetechnologies to avoid union costs. A final finding is related to the fact that the change in the Uruguayan bargaining regime at the beginning of the 90s – by which the mandatory extension of contracts vanished – favoured a more decentralised negotiation scheme and thus ended with the homogenous impact found in the 80s, since coordination in bargaining was lost.