Trade liberalisation and the labour market in Morocco

2 Preface In the last two decades, trade liberalisation has mainly been a characteristic of developing countries. The consequences have been substantial and widespread in all the branches of these societies. In particular, we would investigate how trade policy affects household's welfare through the labour market channel. The existing literature suggests that in developing countries tariff reductions impact more on wages than on the employment level. Recently, some authors have shown evidence of the importance of examining the determinant of the low net employment change and they show the importance of the churning effect, i.e. the workers' reallocation. Following the methodology suggested by Davis and Haltiwanger (1990), we analyse the churning effect in the Moroccan economy. The data derives from the Annual Industrial Census. The sample covers 1,300 enterprises and contains data for five macro-sectors for 1990 and 2002 in Morocco. The five sectors covered are: clothing and textiles, food processing, chemicals and plastics, metallurgy, and electrical machines. One of the substantial advantages of this survey is that it contains extremely detailed information at the firm level. For each firm we have information on the sales, production, exports, and start-up data. In particular we have detailed information on labour supply for each firm, with employment divided by gender, skills and employment period. After a general overview of the sample firms' characteristics, we compute the indexes of job creation and job destruction at sectoral level. Gross job creation (POS) is defined as the sum of the new places available through expansion of existing firms and creation of new establishments within the sector. Similarly, gross job destruction (NEG) is computed by adding up employment losses over shrinking and dying establishments within a sector. Adding up POS st and NEG st produces SUM st , a measure of the gross job reallocation rate in sector s between t-1 and t. After classifying firms on the base of their trade orientation, size and sector of activity, the results show a significant simultaneous job creation and destruction in all the cases. Following Davis and Haltiwanger (1992), WE decompose excess job reallocation in two components. One component represents the contribution of reshuffling employment among sectors, and the other component represents the contribution of excess job reallocation within sectors. The job reallocation decomposition suggests that the churning effect is mostly explained by movement of workers within sectors. Consequently, the swing across sectors is negligible and firms' heterogeneity …