Unemployment rate and working- hour constraints 449 Unemployment rate and working-hour constraints Empirical evidence from the Swiss labour force José V. Ramirez University of Geneva, Switzerland 1. Introduction The surge in unemployment in Switzerland since 1991 has prompted calls for a reduction of standard working time and, above all, a limitation on the use of overtime, particularly in the more affected sectors. These calls are principally founded on the apparent uncoupling trend that has characterised aggregate wages and productivity during this period. In industry, apart from the building sector, the productivity per worker, in equivalent full-time and in real terms, increased dramatically between 1990 and 1995; in the machine-building sector, for example, productivity rose by 17 per cent, in the textile sector by more than 40 per cent, and in the chemical industry the increase reached 88 per cent! A t the same time, the aggregate real wages continued to follow the downward trend which had characterised them since the middle of the 1980s. Hence, with this picture in mind, we shall not discuss whether mandated reductions of working hours are a viable solution to reduce unemployment in Switzerland, but we can reasonably wonder if workers who face a relatively higher risk of unemployment, also face relatively tougher constraints on their working hours, particularly those who work overtime. The belief that the unemployment rate may increase the probability of a worker being constrained to put in more than his or her desired number of working hours is implicitly put forward by the literature on the efficiency wage (Bowles, 1985; Shapiro and Stiglitz, 1984). T he idea of an effect of the “reserve army of unemployed” on the effort level of the worker in his job is an old one in the economic analysis of the labour market and it is worth noticing that, 150 years after Das Kapital , the positive trade-off between the overwork and the unemployment rate is empirically confirmed by Stewart and Swaffield (1997), with regard to the most recent survey on the English male labour force. The relatively new interest in the empirical analysis of working-hour constraints is likely to be connected with the weak explanatory power of some empirical facts by the classical model of working-hour determination (Robbins, 1930). Surveys in the European Community and the USA on whether workers would like to work more, fewer or the same number of hours reveal a clear trend: between 40-60 per cent of respondents would like to change their working hours (Dickens and Lundberg, 1993; European Economy, 1991; Kahn and Lang, International Journal of Manpower, Vol. 19 No. 6, 1998, pp. 449-460, © MCB University Press, 0143-7720