Statistically Assisted Programme Selection - International Experiences and Potential Benefits for Switzerland

Allocating people into social programmes on the basis of statistical tools has gained increasing acceptance in recent years. The need for better targeting of active labour market programmes recently has arisen in Switzerland, too. Statistically assisted programme allocation could be an effective tool to increase the effectiveness of labour market policies. It is a software tool that assists the caseworkers in public employment centres to find adequate measures for unemployed people to get them back to work. This paper gives a short overview of the methodologies applied in various countries and assesses their capabilities to serve the purpose of improving overall programme outcomes. Based on these experiences, an approach suitable for Switzerland is proposed and its potential benefits are investigated. On the basis of very rich data about programme participants in 1998 and 1999, a reallocation into programmes is simulated retrospectively. The reallocation is done according to individual-specific estimates of the usefulness of programmes for each person. The results of the simulations show that the reemployment share could have been improved substantially. Hence, a statistical allocation system could be very fruitful for Switzerland.
(http://www.sjes.ch/papers/2003-III-4.pdf)
Download Internet Appendix (pdf, 80 kb)