Essays on misallocation and productivity

Living standards are closely linked to an economy’s productive capacity, but productivity in the rich world has stagnated in the past decade. Economists developed a wide range of explanations for the productivity slowdown and the aim of this dissertation is to provide novel insights on the dynamics of productivity and its drivers. It does so by adopting two complementary perspectives: acknowledging that economic agents are heterogenous and that, as a consequence, aggregate TFP depends both on the productivity of each unit and on how efficiently inputs are allocated across f irms, I investigate both the first and second moment of the productivity distribution, as well as their determinants. The first two papers of the dissertation deal with the second moment, which has been proven to be extremely relevant by the large literature on resources misallocation started by Restuccia and Rogerson (2008) and Hsieh and Klenow (2009). In the first, I study to what extent the credit cycle influences the efficient allocation of resources across firms, relying on an empirical framework similar in spirit to Rajan and Zingales (1998) and exploiting a balanced firm level dataset covering manufacturing and services industries of 18 European countries from 2006 to 2014. I document that a higher availability of credit leads to a significant decline in both capital and labor misallocation, suggesting that, when credit is above its trend, the additional funds f lowing in the economy reduce financial frictions, allowing financially constrained but productive f irms to implement their projects. Furthermore, the richness of the data allows to reconcile the different findings in the literature, as the estimated effect is larger when evaluating exclusively manufacturing sectors or countries with relatively small domestic credit markets, while tends to disappear for financially developed economies. In the second paper, instead, I investigate the dynamic evolution of productivity misallocation, using the same dataset and three different methodologies: β-convergence, σ-convergence, and the distribution dynamics approach proposed by Danny Quah (1993, 1996). Interestingly, there is no evidence of full distribution convergence in capital misallocation and only slow distribution convergence in labor misallocation, but β-convergence in both inputs allocative efficiency holds conditionally as well as unconditionally. This points to the existence of a mean reversion process in factors misallocation, that is contrasted by exogenous shocks that hinder the convergence of the overall distribution. Understanding what are the determinants of the interplay between these two opposing forces opens interesting paths for future research. The third paper, which is co-authored with Lilas Demmou and Irina Stefanescu, deals again with the productivity and finance nexus, but looking at the first moment of the productivity distribution. Using a cross-country firm level dataset from 1995 to 2015 and a panel fixed effects econometric approach, we investigate how the impact of financial constraints on productivity growth is mediated by sectoral intangible intensity. The paper argues that, despite their aggregate rise, intangible assets often fall short of desired levels, because financing the acquisition of intangibles is more difficult than that of tangibles (e.g., uncertain valuation and lower pledgeability). As a result, financial frictions are expected to be even more binding for productivity growth in sectors where intangibles have become a pivotal component in firms production function. The empirical analysis confirms our conjectures, showing that financing frictions act as a drag on productivity growth and especially so with respect to firms operating in intangible intensive sectors.