Are firms willing to employ a greying and feminizing workforce?

Abstract

Are employers willing to employ more older individuals, in particular older women? Higher employment among the older segments of the population will only materialize if firms are willing to employ them. Although several economists have started considering the demand side of the labour market for older individuals, few have considered its gender dimension properly; despite evidence that lifting the overall senior employment rate in the EU requires significantly raising that of women older than 50. In this paper, we posit that labour demand and employability depend to a large extent on how the age/gender composition of the workforce affects firm's profits. Using unique firm-level panel data we produce robust evidence on the causal effect of age/gender on productivity (value added per worker), total labour costs and gross profits. We take advantage of the panel structure of data and resort to first differences to deal with a potential time-invariant heterogeneity bias. Moreover, inspired by recent developments in the production function estimation literature, we also address the risk of simultaneity bias (endogeneity of firm's age-gender mix choices in the short run) by combining first differences with i) the structural approach suggested by Ackerberg, Caves and Frazer (2006), ii) alongside more traditional IV-GMM methods (Blundell and Bond, 1998) where lagged values of labour inputs are used as instruments. Results suggest no negative impact of rising shares of older men on firm's gross profits, but a large negative effect of larger shares of older women. Another interesting result is that the vast and highly feminized services industry does not seem to offer working conditions that mitigate older women's productivity and employability disadvantage, on the contrary. This is not good news for older women's employability and calls for policy interventions in the Belgian private economy aimed at combating women's decline of productivity with age and/or better adapting labour costs to age-gender productivity profiles.