http://orcid.org/0000-0002-6325-7351
![Sample our Built Environment journals, sign in here to start your access, Latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5926/TOC-Subject-Sample-2021-Built-Environment.jpg"})
ABSTRACT
Do subsidized new firms survive longer? Evidence from a counterfactual approach. Regional Studies. This study provides a statistically robust evaluation of the impact of capital subsidies on the survival of start-ups in Italy. The analysis considers the subsidies provided by the main Italian regional policy, Law 488/1992, during 1996–2009. By exploiting an unusual characteristic of the L. 488 selection process, a quasi-experimental design with high internal validity – the regression discontinuity design – is applied for the first time to the survival analysis. The results show a lower default risk in subsidized start-ups. These findings contradict the view that start-ups receiving L. 488 assistance simply used the grant funding to delay exit.
摘要
受资助的企业是否能生存更久？来自反事实方法的证据，区域研究。本研究对资本赞助对于意大利新创企业的生存之影响，提供统计上可信的评估。本分析考量1996年至2009年间，由主要的意大利区域政策488/1992法案所提供的资助。本文透过探讨488号法案选择过程的不寻常特徵，第一次将具有高度内部效度的准实验设计 – 断点迴归设计 – 应用至生存分析。研究结果显示，受资助的新创企业拥有较低的违约风险。这些研究发现，与接受488号法案协助的新创企业仅只是运用该资金来延迟出场的看法相左。
RÉSUMÉ
Les nouvelles entreprises subventionnées, jouissent-elles d’une plus forte pérennité? Les résultats d’une approche contrefactuelle. Regional Studies. Cette étude fournit une évaluation statistiquement robuste de l’impact des subventions en capital sur la survie des nouvelles entreprises en Italie. L’analyse porte sur les subventions octroyées conformément à la principale politique régionale en vigueur en Italie, à savoir la loi 488/1992, pendant la période allant de 1996 jusqu’à 2009. En exploitant une particularité du processus de sélection de la loi 488, une approche quasi-expérimentale et d’une validité interne élevée – la notion de discontinuité par régression – est appliquée pour la première fois à l’analyse de survie. Les résultats laissent voir des risques de défaillance moins élevées pour ce qui est des nouvelles entreprises subventionnées. Ces résultats vont à l’encontre du point de vue selon lequel les nouvelles entreprises qui bénéficient de l’aide sous l’égide de la loi 488 n’utilisent la subvention que de reporter à plus tard la fermeture.
ZUSAMMENFASSUNG
Überleben subventionierte neue Firmen länger? Evidenz eines kontrafaktischen Ansatzes. Regional Studies. Diese Studie enthält eine statistisch robuste Bewertung der Auswirkung von Kapitalsubventionen auf das Überleben von neu gegründeten Firmen in Italien. Bei der Analyse werden die Subventionen untersucht, die aufgrund der wichtigsten italienischen regionalpolitischen Maßnahme, dem Gesetz 488/1992, im Zeitraum von 1996 bis 2009 gewährt wurden. Durch die Nutzung eines ungewöhnlichen Merkmals des Auswahlprozesses von Gesetz 488 wird für die Überlebensanalyse erstmals ein quasi-experimenteller Aufbau mit hoher interner Validität – der Regressions-Diskontinuitäts-Aufbau – angewandt. Die Ergebnisse lassen auf ein niedrigeres Konkursrisiko von subventionierten neu gegründeten Firmen schließen. Diese Resultate widersprechen der Auffassung, dass Neufirmen, die Subventionen nach dem Gesetz 488 erhielten, diese Mittel lediglich zu einer Verzögerung des Konkurses nutzten.
RESUMEN
¿Sobreviven más las nuevas empresas subvencionadas? Evidencia de un enfoque contrafactual. Regional Studies. Este estudio contiene una sólida evaluación estadística del impacto de las subvenciones de capital en la supervivencia de empresas emergentes en Italia. En el análisis se examinan las subvenciones que se concedieron entre 1996 y 2009 gracias a la principal política regional italiana, Ley 488/1992. Al examinar la particularidad poco usual del proceso de selección de la Ley 488, se aplica por primera vez al análisis de supervivencia un diseño casi experimental con un alto nivel de validez interna: el diseño de regresión descontinuada. Los resultados muestran un riesgo más bajo de impago en las empresas emergentes subvencionadas. Estos resultados contradicen la perspectiva de que las empresas emergentes que recibieron ayuda por la Ley 488 simplemente utilizaban la financiación subvencionada para retrasar el cierre.
ACKNOWLEDGEMENTS
The authors are grateful to participants at seminars in Rome (Sapienza University and Tor Vergata University) and Turin (AISRE), and to Augusto Cerqua, Marusca De Castris and the anonymous referees for the useful comments and suggestions. They thank Salvatore Mignano, Ministry of Economic Development, who made the administrative dataset of Law 488/1996 available. The views expressed in this article are those of the authors and do not necessarily reflect those of the Istituto Nazionale di Statistica (ISTAT).
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.
SUPPLEMENTAL DATA
Supplemental data for this article can be accessed at https://doi.org10.1080/00343404.2016.1190814
ORCiD
Guido Pellegrini http://orcid.org/0000-0002-6325-7351
Notes
1. See, inter alia, Blundell and Costa Dias (Citation2009) for descriptions of the difference-in-difference, matching and difference-in-differences matching estimators.
2. Objective 1 corresponds to all regions in the South of Italy, except for Abruzzi, which lost its Objective 1 status at the end of 1996; while the Objective 2 and 5b areas are all located in the Centre-North of the country or Abruzzi. For more details on the features of L. 488, see Bronzini and De Blasio (Citation2006).
3. Since 2001, the L. 488 scheme has been extended with separate auctions to the tourism, trade and transport sectors. This paper only considers subsidies to manufacturing sectors.
4. However, in the sample, 82.7% of the firms were founded not more than one year before the call for tender. See the fourth section for more details.
5. Firms that apply for the incentives renounce any other public subsidies, even without any guarantee of receiving the L. 488 funds.
6. The lower this ratio, the more likely is the firm to obtain the subsidy. The highest subsidy applicable is determined by the European Union and varies with the dimension of the firm (favouring small firms) and with the location of the production unit (favouring the most disadvantaged areas).
7. These new indicators, signalling only the absence/presence of certain requirements, have played a minor role in determining the rankings. Therefore, the heterogeneity in the number of indicators should not affect the quality of the ranking, and scores calculated with a different number of indicators can safely be pooled together. For a detailed description of the rationale of each indicator, see Bernini and Pellegrini (Citation2011), section 3.
8. The intensity of the subsidy is far higher in the southern regions: they are the only eight Objective 1 Italian regions in the 1994–99 European Union regional policy cycle.
9. This is a standard approach in the presence of multiple cut-off but similar samples. The design is converted into a two-dimensional RD: for each cut-off, the assignment variable is centred at 0 and all observations are pooled into a single dataset (e.g., Jackson et al., Citation2007). However, this procedure relies on the assumption of the random allocation of subsidies to the firms ranked around the unique cut-off point.
10. Moreover, firms interested in merger, acquisition, conversions or transfers in the period (less than the 2% of the total) were dropped. For the fuzzy RDD, see the fifth section.
11. There are several problems related to the active/inactive status in the Infocamere Archive. The Infocamere Glossary states that ‘in some cases there is a considerable presence of companies that are inactive for the registry, but they actually are not, especially those registered in the Rome Archive’. Therefore, it was decided to apply the more conservative hypothesis, i.e., that inactive firms are alive. Actually, the adoption of such an assumption strengthens the results: the survival probability of treated firms is higher than the plausibly overestimated survival probability of non-treated firms, and the difference is statistically significant.
12. One reason is that many of the registered firms never started the activity because they never received the grant. This explains why there are more inactive firms in the control group with respect to the treatment group.
13. The incentives of Law 488 are only confirmed if the new plants reach at least five years of age.
14. On the contrary, the decision to revoke the Law 488 subsidy depends only on the failure to respect some parameters, and therefore it is independent from regional authorities.