Skip to main content

Which projects are selected for an innovation subsidy? The Portuguese case


Several empirical studies have analyzed which firm characteristics influence government evaluators in the decision to select specific firms to participate in Research and Development and Innovation subsidy programs. However, few authors have provided a precise analysis about the selection process of applications submitted for public support. The aim of the present article is to assess differences in investment project characteristics (expected impact) between firms with approved and non-approved applications and to understand which kinds of projects are selected for a subsidy. The analysis is focused on the case study of applications submitted to the Portuguese Innovation Incentive System (SI Innovation) between 2007 and 2013. The impact variables under study are those used in the selection procedure to grant the firm a subsidy, namely the expected impact on exports, value creation, productivity, patent application and qualified employment. Using a counterfactual analysis and Propensity Score Matching estimators, the results show that firms with approved applications are those that expect to invest more and forecast a higher increase in exports and productivity as the result of the investment project. However, these firms in comparison with the control group (those with non-approved applications) have investment projects with a lower contribution to growth and lower economic efficiency (return on investment in terms of productivity). The conclusions of this study could be useful for policy-makers since it provides evidence about firms’ strategic choice concerning investment projects submitted for an Innovation subsidy.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4


  1. An ex-ante analysis refers to an evaluation done before investment project implementation in order to assess which applications will be granted.

  2. Public support for RDI can be direct or indirect. Direct public support refers to direct public expenditure on RDI, such as subsidies, grants and R&D infrastructure, whereas indirect public support is linked to fiscal incentives, public procurement, technology transfer and legal framework (Conte et al. 2009:13).

  3. Net Present Value (NPV) corresponds to the sum of the present value of the project’s expected cash flows (positive and negative) occurring over the life of the project; Internal Rate of Return (IRR) is the rate of return earned by the project based on discounted cash flows; Pay Back Period is the required period of time for nominal cash flow from the project to cover the initial amount of investment (Damodaran 2006: 199–211).

  4. “Total factor productivity (TFP) is the portion of output not explained by the amount of inputs used in production. As such, its level is determined by how efficiently and intensely the inputs are utilized in production” Comin (2008: 6685).

  5. Credit rating refers to evaluation of the firm’s ability to meet its financial obligations. The higher the score the lower the risk and the greater the firm’s ability to attract debt capital.

  6. The transversal characteristic of SI Innovation, SI IDT and SI Qualification was their integration in both the Thematic Operational Program, through the Operational Competitiveness Program (COMPETE), and the Regional Operational Program (excluding the Azores and Madeira).

  7. For more details see Resolução de Conselho de Ministros n° 86/2007 of 3 July.

  8. Portaria n° 353-C/2009 of 3 April and Portaria n° 1103/2010 of 25 October.

  9. Funding of R&D activities, which comes before innovation, was provided by the SI IDT.

  10. Also delimited by Decreto-Lei n° 287/2007 of 17 August (line l of article 3), where innovation corresponds to the “implementation of a new or significantly improved solution for the company, new product, process, organizational method or marketing, with the aim of strengthening its competitive position, increasing performance, or knowledge”.

  11. For more details see Appendix 2 of SI Innovation regulation (Portaria n° 1464/2007 of 15 November and its amendments).

  12. Productivity was measured by: i) gross added value per employee; ii) gross operating surplus by assets ratio; iii) gross value of production by intermediate consumption ratio.

  13. For more details see the following Portuguese legislation: Decreto-Lei n° 287/2007 of 17 August, Portaria n° 1464/2007 of 15 November, Portaria n° 353-C/2009 of 3 April, Portaria n° 1103/2010 of 25 October and Portaria n° 274/2012 of 6 September.

  14. This distinction is only made in the call for applications from 2009.

  15. For more details see for example calls for application n° 04/SI/2010, 05/SI/2011, 02/SI/2012 and 12/SI/2012.

  16. For more details see article 5 of SI Innovation regulation (Portaria n° 1464/2007 of 15 November and its amendments).

  17. For example: i) crowding out effect of private R&D expenditure – when all or part of public expenditure replaces the firm’s own investment (see e.g. Busom 2000; Erden and Holcombe 2005; Aerts and Thorwarth 2008; Cavallo and Daude 2011); ii) deadweight effect - firms would have carried out their strategic investment project even in the absence of subsidies (see e.g. Bronzini and de Blasio 2006; Skuras et al. 2006; Tokila et al. 2008); iii) subsidized firms’ inefficiency – in the post-intervention period, subsidized firms show a lower increase of economic performance than non-subsidized firms (see e.g. Bergström 2000; Bernini and Pelligrini 2011; Jorge and Suárez 2011); iv) ineffectiveness of public support in achieving targets, e.g. no effects on internationalization (Silva 2011), no effects on employment (Wallsten 2000; Sissoko 2011) or in alleviating financing constraints (Sissoko 2011; Silva and Carreira 2012).

  18. With the Stata command “teffects psmatch”.

  19. With the Stata command “psmatch2”.

  20. In 2010 new conditions of access to SI Innovation were introduced (see e.g. call for application n° 04/SI/2010, 05/SI/2011, 02/SI/2012 and 12/SI/2012). Reporting intensity of investment above a threshold was a mandatory requirement for access to finance through SI Innovation.

  21. Criterion C3 - Creation of highly-skilled jobs.

  22. Dynamics of value creation index = ΔGAV / ΔTurnover, where GAV is the gross added value.

  23. Labour productivity corresponds to gross added value per employee.

  24. Crépon et al. (1998) used in their study patent per employee as a measure of innovation output.

  25. The original sample has 5880 observations, but for some applications relevant information is missing. We also excluded all multi-region projects (destined to be implemented in more than one region at NUTS 2 level), due to their small representativeness (34 applications) and the difficulty of matching these observations.

  26. Probit regression model was also performed (results available on request), but compared to the Logit regression model it shows weaker results in terms of Log likelihood and Pseudo R2.

  27. We consider that a firm is financially constrained if it has difficulty in accessing external finance (e.g. due to high cost or collateral requirement) or because external financing is not available (e.g. due to high risk or a lack of collateral), and for one of these reasons firms will not be able to make the investment or part of it.


Download references


The authors are grateful to the Comissão de Coordenação e Desenvolvimento Regional do Alentejo (CCDRA) for the information provided in the framework of the activities developed by UMPP – Unidade de Monitorização de Políticas Públicas/Public Policy Monitoring Unit at the University of Évora (Portugal). Special thanks go to Peter Berkowitz (DG-Regio), Wolfgang Petzold (Committee of the Regions), Nicola Francesco Dotti (Vrije Universiteit Brussel), Frank Crowley (University College Cork), Oto Potluka (University of Basel), Prof. Maria Luisa Mancusi (Università Cattolica del Sacro Cuore- Milan), Prof. Pierre Mohnen (Maastricht University), Prof. Julien Ravet (Université libre de Bruxelles), Prof. Bruno van Pottelsberghe (Université libre de Bruxelles) and the two anonymous referees for valuable comments and suggestions on the earlier version of the article.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Anabela Santos.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


Appendix 1: Benchmark review of literature

Table 7 Benchmark studies focused on the probability of receiving a public subsidy

Appendix 2: Descriptive statistics

Table 8 Mean-comparison tests: approved versus non-approved applications

Appendix 3: Collinearity Diagnostics

Table 9 Variance Inflation Factors (VIF) and correlation matrix

Appendix 4: Assessing balancing quality

Fig. 5
figure 5

Kernel density plots for raw and balanced data: all Portuguese regions. Source: Authors’ own elaboration

Fig. 6
figure 6

Kernel density plots for raw and balanced data: Norte region. Source: Authors’ own elaboration

Fig. 7
figure 7

Kernel density plots for raw and balanced data: Centro region. Source: Authors’ own elaboration

Fig. 8
figure 8

Kernel density plots for raw and balanced data: Alentejo region. Source: Authors’ own elaboration

Table 10 Covariate balance summary statistics

Appendix 5: Differences between approved and non-approved applications by region

Table 11 ATET: Foreseen impact of investment project in SI Innovation, by NUTS 2 level region

About this article

Verify currency and authenticity via CrossMark

Cite this article

Santos, A., Cincera, M., Neto, P. et al. Which projects are selected for an innovation subsidy? The Portuguese case. Port Econ J 18, 165–202 (2019).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Subsidy
  • Innovation
  • Selection procedure
  • Propensity score matching

JEL classification

  • O38
  • O31
  • C14