Asymmetric Effects of National-based Active Labour Market Policies

Abstract

Altavilla C. and Caroleo F. E. Asymmetric effects of national-based active labour market policies, Regional Studies. Labour market policies settled at the national level imply a ‘one-size-fits-all’ labour market strategy. This strategy might not sufficiently take into account region-specific economic structures. Whether active labour market programmes might asymmetrically affect labour markets at the regional level is evaluated. The results for Italy suggest that while in the South employment is mainly driven by social and economic context variables, in the North the employment dynamics are significantly explained by policy interventions. Two policy implications are suggested. First, the success of active policies depends on the regional labour market conditions. Second, policy-makers should adjust labour policy strategy to regional economic structures.

Altavilla C. and Caroleo F. E. 国家层面劳动力市场政策的不对称效应，区域研究。国家层面的劳动力市场政策是一种“一刀切”的劳动力策略。这一策略可能并未充分考虑特定区域的经济结构。本研究对积极的劳动力市场项目是否会对区域层面的劳动力市场产生不对称影响进行了评估。意大利的结果表明南部就业主要由社会与经济背景变化所驱动，而北部的就业活力则可以由政策介入进行解释。政策影响表现在两方面。首先，积极政策的成功取决于区域劳动力市场状况。其次，政策制定者应当调整劳动力政策策略以与区域经济结构相协调。

Altavilla C. et Caroleo F. E. Les effets asymétriques des politiques actives en faveur du marché du travail au niveau national, Regional Studies. Les politiques en faveur du marché du travail établies au niveau national laissent supposer une stratégie pour l'emploi ‘taille unique’. Il se peut que cette stratégie ne tienne pas assez compte des structures économiques spécifiques à une region. On évalue si, oui ou non, les programmes actifs en faveur du marché du travail puissent toucher asymétriquement les marchés du travail sur le plan regional. Les résultats pour l'Italie laissent supposer que dans le Sud l'emploi est piloté plutôt par des variables de contextes sociaux et économiques, tandis que dans le Nord la dynamique de l'emploi s'expliquent dans une large mesure par des interventions politiques. On propose deux conséquences pour la politique. Primo, la réussite des politiques actives dépend des conditions du marché du travail régional. Secundo, les décideurs devraient adapter leur stratégie politique aux structures économqiues régionales.

Altavilla C. und Caroleo F. E. Asymmetrische Auswirkungen nationaler aktiver Arbeitsmarktpolitik, Regional Studies. Bei der auf nationaler Ebene festgelegten Arbeitsmarktpolitik wird eine verallgemeinerte Arbeitsmarktstrategie verfolgt. Bei dieser Strategie werden die spezifischen Wirtschaftsstrukturen der einzelnen Regionen unter Umständen nicht ausreichend berücksichtigt. Wir untersuchen, ob sich aktive Arbeitsmarktprogramme auf regionaler Ebene asymmetrisch auf die Arbeitsmärkte auswirken können. Aus den Ergebnissen für Italien geht hervor, dass das Beschäftigungsniveau im Süden in erster Linie von sozialen und wirtschaftlichen Kontextvariablen bestimmt wird, während sich die Beschäftigungsdynamik im Norden in signifikantem Umfang durch politische Interventionen erklären lässt. Wir erörtern zwei Konsequenzen für die Politik. Erstens hängt der Erfolg aktiver Politiken von den Bedingungen des regionalen Arbeitsmarkts ab. Zweitens sollten die Politiker ihre arbeitspolitische Strategie an die regionalen Wirtschaftsstrukturen anpassen.

Altavilla C. y Caroleo F. E. Efectos asimétricos de las políticas activas del mercado laboral en un ámbito nacional, Regional Studies. En las políticas del mercado laboral establecidas a nivel nacional se sigue una estrategia del mercado laboral con un enfoque único. Es posible que con esta estrategia no se tengan lo suficientemente en cuenta las estructuras económicas específicas para las regiones. Aquí evaluamos si los programas activos del mercado laboral podrían afectar de modo asimétrico a los mercados laborales en el ámbito regional. Los resultados para Italia indican que si bien el empleo en el sur está principalmente impulsado por las variables de contexto social y económico, en el norte las dinámicas de empleo se explican significativamente mediante las intervenciones políticas. Sugerimos dos consecuencias políticas. Primero, el éxito de las políticas activas depende de las condiciones del mercado laboral en un ámbito regional. Segundo, los políticos deberían ajustar la estrategia de políticas laborales a las estructuras de la economía regional.

• Active labour market policies
• Factor-augmented vector autoregression (FAVAR)

• 积极的劳动力市场政策
• 因子增强向量自回归

• Politiques actives en faveur du marché du travail
• Modèle de vecteurs autorégressifs augmenté par des facteurs (Modèle FAVAR)

• Aktive Arbeitsmarktpolitik
• Faktoraugmentierte Vektorautoregression (FAVAR)

• Políticas activas del mercado laboral
• Autorregresión vectorial con factores aumentados

JEL classifications:

• C33
• J64

Notes

1. The theoretical foundations of the Beveridge curve are substantially twofold: the first, starting from the Hansen model (Hansen, 1970; Holt and David, 1966), derives the matching function from an aggregation over distinct markets in the presence of frictions and of limited mobility of labour. More recently, an alternative approach was developed that arises from a matching function combined with job search (Mortensen and Pissarides, 1994; Pissarides, 2000; Petrongolo and Pissarides, 2001).

2. For a review of the debate on the evaluation of the effects of the ALMPs on labour markets, see Kluve et al. (2006), Kluve (2010), Calmfors et al. (2001), Caliendo et al. (2008), and Büttner and Prey (1998).

3. For a comprehensive survey of the evaluation studies carried out in Italy, see Trivellato et al. (2003) and Trivellato (2009).

4. For a detailed overview of the existing knowledge of ALMPs in Europe, see European Commission (2006).

5. For each year the variation index is calculated as the ratio between the standard deviation and the mean of a given variable.

6. These two policies are only considered because they account, on average, for more than 85% of the total ALMP participants.

7. For a detailed description of the variables used in the analysis and their sources, see Appendix A.

8. The exact relationship between expenses and participants in ALMPs is derived in Altavilla and Caroleo (2006a) and Calmfors et al. (2001).

9. ALMP participation in Campania and Basilicata is high because the denominator of the policy measure (labour force) is relatively low.

10. The recent literature on factor models has solved the curse of dimensionality, allowing very large panels of data to be decomposed into a small number of common factors (Stock and Watson, 1998, 2002).

11. The Southern regions are Campania, Abruzzo, Molise, Apulia, Basilicata, Calabria, Sicily and Sardinia.

12. The twelve regions are Piedmont, Aosta Valley, Liguria, Lombardy, Trentino-Alto Adige, Veneto, Friuli-Venezia Giulia, Emilia Romagna, Marche, Tuscany, Umbria and Lazio.

13. All variables considered in the factor analysis and their transformations are reported in Appendix A. When computing the common factors, the last two variables, employment and participation rate, which are included in Y_it, are not considered.

14. The principal components are orthogonal linear combinations of the data that explain the maximal variances of the data contained in X_it.

15. In fact, given the rotational indeterminacy problem, unless identification assumptions are imposed on the factor loadings, it will always be possible to find some rotation of the factors that explains the same amount of total variation in the data but which implies a different set of factor loadings.

16. Robustness of the main results is examined by adopting alternative identification schemes. The results remain substantially unchanged.

17. In complete principal component analysis, with no factors dropped, the communality is equal to 1.0, or 100% of the variance of the given variable. As the model does not extract all the variance, the proportion of variance of a particular variable that is due to common factors is called the communality. The proportion of variance that is unique to each variable is then the respective variable's total variance minus the communality.

18. Both Italy and Germany have major regional divides, with Southern Italy and the eastern portion of Germany recording higher unemployment rates than the rest of the respective countries.