From the beginning of the 1990s to the 2008-2009 crisis, labour productivity in France increased steadily, by around 1.3% per year. It then experienced an unprecedented fall during the crisis, losing 2.6% compared with its previous trend. From 2010, productivity started to grow once again, although at a slower pace than pre-crisis.

Labour productivity is linked to many factors. Traditionally, economic analysis has focused on the role of capital, which is complementary to the labour factor in the production process, and that of “total factor productivity”, which describes how capital and labour contribute to production.

Many explanations have been put forward in the theoretical literature to attempt to account for the slowdown in labour productivity observed recently: a slump in investment and hence in the amount of capital due to the crisis, less efficient allocation of factors of production (capital and labour) between branches of activity or between enterprises, change in labour quality, etc.

An accounting breakdown of labour productivity gains, differentiating inter-branch and intra-branch contributions linked with capital and employment, identifies the determining factors of the slowdown in productivity post-crisis.

Average growth in labour productivity dropped from 1.3% per year for the 1990-2007 period to 0.9% during the 2010-2016 period. A less efficient capital allocation between branches post-crisis would probably be responsible for 0.2 points in this slowdown in labour productivity. A lower capital intensity would make a more moderate contribution to the slowdown, of around 0.1 points. The allocation of labour, on the other hand, appears not to play any part. Finally, total factor productivity appears to contribute 0.2 points to the slowdown, i.e. a significant contribution to the decline in pace.

Until now, the model used to forecast employment in the short term was based on a linear productivity trend, but this has now been challenged as a result of post-crisis developments. A new model has been built, using as a basis the previous accounting breakdown. The aim is to explain the slowdown in labour productivity in the long term by capital destruction both during and after the crisis, and the decline in efficiency of the reallocation of factors of production. The short-term employment forecast that is possible using this new model appears to be more accurate than with the tools used until now.

This new tool will now be used for forecasting employment in Conjoncture in France. According to the model, employment should increase at a slower pace in 2018 than in 2017 (+203,000 jobs over the year at the end of 2018 against +257,000 in 2017). Apart from employment policies, this slowdown is probably due essentially to the slowdown in value added in the non-agricultural market branches.