Population ageing is often viewed as a source of major economic problems after the turn of the next century, calling for strong policy responses if we wish to avoid growing disequilibrium in a large number of fields (Clark et al., 1978; Bös and von Weizsäcker, 1989). In fact, close inspection of predictable effects of population change shows that they are not uniformly important. At one extreme, there is little doubt that it will be hard to maintain current performances of pension systems with age structures like those expected for the next century: large adaptations will be necessary either in the relative level of pensions, in contribution rates by workers, or in retirement age. On the other hand, the impact of ageing on a wide range of economic phenomenona is surprisingly low. This is for instance the case with consumption patterns, even for goods and services whose demand is strongly related to age at the individual level (e.g. health expenditures).

This paper will try to examine where labour force problems fall between these two extreme situations, between high or almost nonexistent sensitivity to ageing. We shall do so by considering two basic aspects of labour force problems and labour force management:

• The age-productivity relationship and the question of optimal human capital accumulation and on-the-job training in an ageing population. We shall first examine how changing age distributions may affect average productivity, relying both on simulations and simple analytical relations. This impact will appear to be rather weak and its sign may be positive or negative. This does not exclude that changes in the age-productivity profiles could be profitable in an ageing population: we shall examine how educational policies could be used for that purpose (section 1).

• […]