Unemployment and Firm Entry and Exit: An Update on a Controversial Relationship

Abstract

Santarelli E., Carree M. and Verheul I. Unemployment and firm entry and exit: an update on a controversial relationship, Regional Studies. The present study explores the relationship between unemployment and subsequent firm entry and exit for 103 Italian provinces for the period 1997–2003. Two models are estimated. The first model concentrates on unemployed individuals starting or closing a business in the province in which they live. The second model incorporate cross-border effects by taking into account possible start-ups by unemployed individuals from adjacent provinces. Findings show that a positive effect of unemployment on net entry is not due to a positive ‘push’ effect on entry, but rather to a negative effect of unemployment on firm exit. This indicates a lack of dynamics in the Italian labour market.

Santarelli E., Carree M. et Verheul I. Le chômage et l'entrée et la sortie des entreprises: une mise à jour d'un rapport controversé, Regional Studies. La présente étude cherche à examiner le rapport entre le chômage et l'entrée et la sortie ultérieures des entreprises pour 103 provinces italiennes pendant la période allant de 1997 jusqu'à 2003. On estime deux modèles. Quant au premier modèle, on se concentre sur les chômeurs qui créent ou qui ferment une entreprise dans la province où ils habitent. Pour ce qui est du deuxième modèle, on y inclut des effets transfrontaliers, en tenant compte des créations d'entreprise éventuelles par les chômeurs qui habitent les provinces limitrophes. Il s'avère que l'effet positif du chômage sur l'entrée nette ne s'explique pas par un effet ‘moteur’ positif au moment de l'entrée mais plutôt par l'effet négatif du chômage sur la sortie des entreprises. Ceci indique un manque de dynamique sur le marché du travail italien.

Chômage Entrée et sortie des entreprises Provinces italiennes

Santarelli E., Carree M. und Verheul I. Arbeitslosigkeit und Betriebsgründungen bzw.-schließungen: aktuelle Informationen zu einer kontroversen Beziehung, Regional Studies. In dieser Studie wird die Beziehung zwischen Arbeitslosigkeit und anschließenden Betriebsgründungen bzw.-schließungen in 103 italienischen Provinzen im Zeitraum von 1997 bis 2003 untersucht. Wir stellen Schätzungen anhand von zwei Modellen auf. Im ersten Modell konzentrieren wir uns auf arbeitslose Einzelpersonen, die in der Provinz, in der sie leben, einen Betrieb gründen oder schließen. Im zweiten Modell integrieren wir grenzübergreifende Auswirkungen, indem wir mögliche Neugründungen durch arbeitslose Einzelpersonen aus angrenzenden Provinzen berücksichtigen. Aus den Ergebnissen geht hervor, dass eine positive Auswirkung der Arbeitslosigkeit auf die Nettozahl der Firmengründungen nicht auf einen positiven ‘Push’-Effekt auf die Betriebsgründungen zurückzuführen ist, sondern vielmehr auf einen negativen Effekt der Arbeitslosigkeit auf die Betriebsschließungen. Dies weist auf eine mangelnde Dynamik des italienischen Arbeitsmarkts hin.

Arbeitslosigkeit Betriebsgründungen und –schließungen Italienische Provinzen

Santarelli E., Carree M. y Verheul I. Desempleo y entrada y salida de empresas: actualización de una relación controvertida, Regional Studies. En el presente estudio analizamos la relación entre el desempleo y la posterior entrada y salida de empresas para 103 provincias italianas durante el periodo de 1997 a 2003. Evaluamos dos modelos. En el primer modelo nos centramos en personas desempleadas, que empiezan o cierran un negocio en la provincia en la que viven. En el segundo modelo incorporamos los efectos transfronterizos teniendo en cuenta posibles empresas emergentes por parte de personas en paro de provincias adyacentes. Los resultados indican que un efecto positivo del desempleo en la entrada neta no se debe a un efecto positivo de ‘empuje’ sino a un efecto negativo del desempleo al cerrar las empresas. Esto indica una falta de dinámica en el mercado laboral italiano.

Desempleo Entrada y salida de empresas Provincias italianas

Keywords:

• Unemployment
• Firm entry and exit
• Italian provinces

JEL classifications:

• J60
• R11
• R23

Acknowledgements

This paper is the result of a series of visits by Martin Carree and Ingrid Verheul to the Department of Economics, University of Bologna, in 2005, and by Enrico Santarelli to the Department of Applied Economics, Erasmus University Rotterdam, and to the Department of Organization and Strategy, Maastricht University, in 2006 and 2007. Comments from seminar participants at the Max Planck Institute of Economics (Jena, Germany, April 2006); the 3rd Hellenic Workshop on Efficiency and Productivity Measurement (Patras, Greece, June 2006), and the 33rd Annual Conference of the European Association for Research in Industrial Economics (Amsterdam, the Netherlands, August 2006) are gratefully acknowledged. The authors also want to thank three anonymous referees for their useful suggestions.

Notes

Earlier work on Italy includes studies by Audretsch and Vivarelli (1995, 1996), Garofoli (1994), and Santarelli and Piergiovanni (1995).

An example is the 1998 Bersani Law deregulating the retail trade sector (Schivardi and Viviano, 2007).

Unemployment may also affect those who are not yet officially part of the labour force. Di Pietro (2006) finds a negative relation between youth unemployment rates and university dropout rates in Italian regions. Low unemployment rates apparently are an incentive to drop out of university education which is very low cost and without selection mechanisms in Italy.

One way of coping with the negative effect of unemployment on firm formation is by including business cycle and average income level variables in the model.

Audretsch and Jin (1994) propose to reconcile these seemingly contradicting relationships.

For an overview of recent developments from the new economic ‘geography’ approach, see Armington and Acs (2002).

Although the authors are aware that analysis of inactivity rates by region (cf. Faggio and Nickell, 2005) might also prove useful to understand the relationship between employment dynamics and rates of new firm formation, this paper focuses only on the unemployment issue.

The results including entry and exit rates relative to the number of incumbent firms (in the previous period) were also examined. These results are very similar to those including entry and exit rates relative to the total labour force.

Before the Bersani Law retail establishments were required to have a permit from the town council. The Bersani Law abolished this permit for smaller firms, which now only have to give notice of their activity. For the estimated effects of a similar deregulation on entry and exit rates in Dutch retailing, see Carree and Nijkamp (2001).

The authors are aware that studies by Garofoli (1994) and Santarelli and Piergiovanni (1995) found contrasting evidence.

Provinces with at least one important ‘traditional’ (according to the definition used by Unioncamere) industrial district are: Ascoli Piceno (shoes), Arezzo (golden jewellery), Avellino (leather), Bari (footwear), Biella (textiles – wool), Brescia (metal household artefacts and machinery for the textile industry), Como (silk), Ferrara (mechanical engineering), Macerata (leather products), Mantova (stockings), Modena (knitwear, biomedical industry and ceramics), Pisa (leather), Pordenone (cutlery), Prato (textiles), Parma (ham), Pesaro–Urbino (furniture), Pavia (machinery for the footwear industry), Siena (furniture), Treviso (sporting footwear), Vicenza (leather), Verona (furniture). and Viterbo (ceramics). Note that the definition of industrial district used here excludes local systems dominated by ‘focal’ or leading firms occupying strategic and central positions due to their extensive network of customers and suppliers (for a further specification, cf. Lazerson and Lorenzoni, 1999).

The number of adjacent provinces in Italy varies between one for Sassari and Trieste to eight for Mantova, Milano and Perugia, and even nine for Firenze (Florence). The average number of adjacent provinces is 4.46.

Johnson and Parker (1996, p. 684) recommend constructing a vector autoregressive (VAR) system including lags of firm entries and exits. The short time-series dimension of the present panel does not allow for several lags. However, ran each of the regressions was run, including the one-period lagged entry and exit rates in both the entry and exit equations. The results leave the general conclusions unaltered.

Whereas interpretation of the impact of entry subsidies on entry rates is straightforward (also Del Monte and Scalera, 2001), the role of regional capital incentives in attracting plants to low-income areas is more complex. For a discussion of the latter effect, see, for example, Faini and Schiantarelli (1987).