Download this Paper Open PDF in Browser

The Dependent Variable Problem in Quantitative Studies of Active Labour Market Programmes: Uncovering Hidden Dynamics?

REC-WP Working Paper No. 03/11

32 Pages Posted: 15 Jul 2011  

Tim Vlandas

University of Reading

Date Written: 2011

Abstract

The question of what explains variation in expenditures on Active Labour Market Programs (ALMPs) has attracted significant scholarship in recent years. Significant insights have been gained with respect to the role of employers, unions and dual labour markets, openness, and partisanship. However, there remain significant disagreements with respects to key explanatory variables such the role of unions or the impact of partisanship.

Qualitative studies have shown that there are both good conceptual reasons as well as historical evidence that different ALMPs are driven by different dynamics. There is little reason to believe that vastly different programs such as training and employment subsidies are driven by similar structural, interest group or indeed partisan dynamics. The question is therefore whether different ALMPs have the same correlation with different key explanatory variables identified in the literature? Using regression analysis, this paper shows that the explanatory variables identified by the literature have different relation to distinct ALMPs. This refinement adds significant analytical value and shows that disagreements are at least partly due to a dependent variable problem of ‘over-aggregation’.

Keywords: Labour market policy, active labour market policies, ALMP, political economy

Suggested Citation

Vlandas, Tim, The Dependent Variable Problem in Quantitative Studies of Active Labour Market Programmes: Uncovering Hidden Dynamics? (2011). REC-WP Working Paper No. 03/11 . Available at SSRN: https://ssrn.com/abstract=1884931 or http://dx.doi.org/10.2139/ssrn.1884931

Tim Vlandas (Contact Author)

University of Reading ( email )

Humanities and Social Sciences Building
Whiteknight campus
Reading, RG6 6AH
United Kingdom

Paper statistics

Downloads
256
Rank
99,521
Abstract Views
1,144