Sindacati e performance economica dell'impresa: Un'analisi empirica

The present work represents an attempt at investigating the controversial relationship between labor unions and firm economic performance, with particular regard to innovation and to what is known in economics as the hold-up problem. The study is articulated in three core sections. First, the main theoretical models for unions behavior are presented, ranging from the standard microeconomic models such as the Monopoly union model and the Efficient contract model to more recent models that emphasize the institutional nature of unions, like the Collective Voice/Institutional Response model. Second, the empirical evidence on the effect of unions on the major indicators of firm performance is collected and laid out. Productivity, profitability and investment are addressed separately from one another in this respect. Third, after an institutional contextualization of the role of labor unions in Italy, the actual econometric analysis is carried out. Data from a panel of Italian firms, drawn from the RIL-AIDA matched dataset, are used for the years 2005, 2007 and 2010. By resorting to different econometric techniques and models, such as a standard OLS Linear Probability Model, a Probit Model, and a Fixed Effects Model, I try to estimate the effect of unions on innovation. It is shown that the initial positive correlation between unions and innovation, attributable to omitted variable bias and opposed to the evidence from the literature, disappears after a detailed control for factors like size of the firm, age, region and industry sector, eventually becoming negative. Hence, the final estimates obtained, which are fairly robust to the different specifications of the model and to appropriate checks, support the literature in finding a modest presence of hold-up for Italian firms. However, the limitations of the data currently available, linked to possible noise and to scarce variability, especially within observations, impose some caution in drawing secure conclusions. The weak evidence found certainly opens the way to alternative interpretations, but the need for further research remains paramount.