• Citer

Financial constraints and productivity growth across the size spectrum: microeconomic evidence from Morocco

par Ferragina, Anna Maria;Mazzotta, Fernanda;Sekkat, Khalid
Référence Eurasian Business Review, 6, 3, page (361-381)
Publication Publié, 2016-12

Article révisé par les pairs

• Détails
• Contenu

Résumé :

To what extent do financial constraints act as conditioning factors on firm productivity? Is this impact different across firm size classes and across sectors and in what way is it different? We explore these issues for Morocco. We use firm level data, taken from two surveys of the World Bank Enterprise Surveys (WES) for Morocco (2004, 2007), to investigate the relationship between two proxies of a firm’s financial constraint (perceived financial constraint as the most serious obstacle and credit constraint) and its productivity growth. The analytical framework assumes a Cobb–Douglas production function for firms where technical progress is not exogenous but depends on financial constraint. Two different econometric estimates are adopted: within estimators and GMM system estimates to address endogeneity and to correct for potential selection, simultaneity and omitted-variable bias (OVB). The estimations using the GMM give results that are consistent with the expectations: they show that each of the two financial constraint proxies has a significant and negative impact on small- and medium-sized firms (i.e. firms with less than 100 workers) but not on large firms (i.e. firms with more than 100 workers). To see whether the main results of the analysis persist (i.e. credit constraints harm small-sized firms more than large-sized firms) several robustness checks are performed. We change the threshold of the separation between small- and large-sized firms from 100 to 50 workers. We keep the threshold at 100 but exclude alternatively small-sized firms that are too small (less than 20 workers) and the large-sized firms that are too large (more than 500 workers). We keep the threshold to 100 and the whole sample but add a control variable which gives the legal status of the firm. Finally, we examine the sensitivity of the results to the sectoral composition of the sample. The main result holds for all sectors except “Wearing apparel”.