Machines and Skills: Technology, Employment, and Labour in Italy, 1871-2011

In the last few years the technology question became one of the most important argument debated both inside and outside academia. For example, employment has been reshaped by dramatic events like the Great Recession and the Covid-19 but also by a quieter ongoing evolution in the mix and location of jobs. In the last fifty years the diffusion of Information and Communication Technologies (ICT) has led to a transformation of production and employment structures. In the decade ahead, the next wave of artificial intelligence (AI) technology may accelerate the pace of change. Millions of jobs could be phased out even as new ones are created. More broadly, the nature of work could change for nearly everyone as new machines become fixtures in the workplace throughout the world. As intellectual exercise, social scientists develop different ways to interpret such changes, and broadly speaking there are two main views of technological change, one optimistic and one pessimistic. Among the optimists we find, for example, Levy and Murnane (2013) who argued that we can ‘dance’ with robots because as in previous waves of technical change technological change has if anything a galvanizing effect on jobs, increasing demand for labor and raising wages. Contrarily, pessimists propose a link between job polarization and technological change, arguing that decreases are caused by concentration among middle-skilled jobs of “routine” tasks, which are easier to automate due to defined set of standard instruction, as documented by Autor (2019). Historically, the relationships between technology and employment have long been at the heart of economic theory and history. Classical economists such as Adam Smith and Karl Marx, and social scientists as Joseph Schumpeter, had already identified technological change as a fundamental factor explaining changes in terms of division of labor, quality and quantity of labor demand, as well as income structures. The question of whether technology creates or destroys jobs is also the starting point for many recent studies that have examined the effects of different types of technologies and the structural factors that influence the amount of employment change. Qualitative aspects have received increasing attention, with the analysis of the type of jobs that are created or destroyed by new technologies, of how change the composition of skills and of wages. But, in this way, what is the difference between Charlot of Modern Times (1936) and today’s workers? Can we reconstruct the line of evolution from then to present day? This thesis aims to contribution to the production and discussion of quantitative evidence on the relationship between technology and work in Italy in historical perspective (1871-2011). The reason for this choice derives from the observation of the persistence of the lack of this data. Therefore, the adoption of a historical-quantitative approach seemed to be inevitable in such a circumstance to delve into the following main four issues: First, how does technology redefine the skills of workers? Second, does all new technology lead to de-skilling or does certain technology improve or even create skill? Third, how are skills distributed? Fourth, does all technology have the same impact on work skills or do the specific and idiosyncratic features of those skills have an effect? The research is divided as follows: the first chapter investigates the historical and economic literature over the last two centuries, exploring the multiple dimensions and the direction of theories and studies on technology and labor. The central questions across two centuries of historical literature may be classified according to two of main effects. Among technology’s effects can be to enable skills, to complement them (enabling technologies), or to replace them altogether (replacing technologies). The second chapter presents the new dataset of Italian occupations, the “Unified Italian historical Database of Occupations” (UIHDO) based on the professional classification of Italian demographic censuses (1871-2011) according to the HISCO-HISCLASS framework. This section describes how the Italian database was constructed and considers its shortcomings. The third chapter analyzes the effects of technologies on the quantity and quality of Italian workers in terms of skills, the dynamics of the arrival of new jobs and the exit of others and the rise and fall of middle-skilled workers. I use the technological change framework because that capture the evolution of concrete types of technology and their different power of change when applied to employment, actual work functions and the skills of the people doing the jobs. Finally, appendices present supplemented materials that clarify the methodologies and expand the information for each census.