Khalid ElFayoumi, Gregory Auclair
20 S.: Anh.
(Working Paper Series / Economic Research Forum ; 1282)
Against what theory predicts, large productivity gaps across sectors persist and the process of structural transformation is stagnant in many developing economies. This wedge between observed and optimal labor allocations suggests the presence of institutional and market frictions, which impose costs on the reallocation of labor from low to high productivity sectors, thus leading to sub-optimal allocations and a loss in aggregate labor productivity. Using a panel of crosscountry sector-level data, we estimate a dynamic panel error correction model that captures the dynamic adjustment of labor flows across sectors. We find that, on average, labor flows from low to high productivity sectors, closing around 15.4 percent of labor productivity gaps each year. The pace of this flow varies across country income groups and geographical regions, with high-income countries enjoying a more fluid structural transformation process than lower income countries. Heterogeneity also arises across sectors, suggesting a positive role for sectoral policies. With regard to labor market regulations, we find a significant positive association between the pace of labor reallocation across sectors and the freedom level of labor market institutions. However, in contrast to neo-classical intuition, we find that lowering firing costs slows the structural transformation process. Results suggest that the discouraging effect of having lower job security on the labor supply side is stronger than the benefits that firms gain from more flexible labor market conditions. Hence, policy reforms need to steer between the goal of easing job creation and destruction, while supporting labor supply incentives to reallocate and shift industries through strong social nets, labor protection, and risk sharing.
Keywords: Labor misallocation, labor reforms; structural transformation; multi-sectors growth models; Error Correction Model