An agent-based simulation of the Swiss labour market : an alternative for the labour market policy evaluation

Active labour market policies (ALMPs), which are primarily addressed to integrate (or reintegrate) into the labour market those who are on the edge of unemployment, are generally analysed in the context of a macroeconomic approach or within a micro-econometric model. A critique of the usual economic analysis of labour markets states, however, that a standard economic approach does not take into account the importance of the social environment in the demand-supply matching process. In order to address this problem Agent-based Computational Economics (ACE) can present both a generic model in order to explain a complex phenomenon at a very abstract level and can be based on empirical data to describe a particular case study. The purpose of this paper, that follows (ACE) models approach, is therefore to link micro-level and macro-level evaluations with an integrated (micro/macro) simulation model, trying to further diversify the battery of instruments used for ALMPs evaluation. The implemented ACE model defines a virtual regional labour market with firms and worker agents. Firms are assigned different sectors and have sector-specific skill requirements. The policy maker agent defines active labour market policies, thus financing the unemployed workers' human capital investment. During the simulation periods each firm opens vacant positions and receives applications; it then randomly chooses one applicant which fit with the firm's skills requirement. Firms are also hit by shocks and dismiss their employees that are 'too costly' (the worker's 'cost' depending on two factors, his productivity and his nationality). Unemployed workers have to invest in their human capital to qualify for vacancies opened in different sectors. The model has been structured in modules (taking in consideration Subsidized training and Job displacement effects), which let the user to enable or disable some features to explore different policy options. Following the exposed perspective, we have developed a case study to test and validate the application of the proposed model and framework, exploring the regional labour market in the southern Switzerland (Ticino and some districts of the Grisons), where the particular geographical position of the Italian-speaking part of Switzerland was an incentive to build strong relationships between the cantons and the neighbouring Italian provinces. The model describes an environment where individual agents, their micro-level decisions and the macro level policies interacts in a systemic way. Specifically, this study seeks to identify traceable connections between micro and macroeconomic scales exploring the regional labour market. At this point, the model is still experimental and calibration is very preliminary but it already allows to approximate a quite a number of stylized features of the southern Swiss regional labour market. The model is programmed into NetLogo, a program specifically designed to accommodate agent-based modelling.