Evaluating the impact of innovation policy instrument mix on firm-level research and development intensity

Firm-level innovation is a well-established determinant of national competitiveness and economic growth. Many studies highlight that private firms may under invest in innovation due to market and system failures. This has motivated a large body of research on the role of public funding to support firms’ innovation activities. At the level of the firm, a suite of innovation policy instruments operationalise innovation policy. While impact evaluations traditionally focus on individual innovation policy instruments, such analyses have limitations. Firms often apply to public funding agencies for different innovation policy instruments. Many firms accrue more than one instrument, resulting in an innovation policy instrument mix. Different instruments can have a complementary or substitutive relationship. It is essential to consider how interactions between instruments in the mix influence firms’ innovation outcomes. Despite this imperative, there is lack of empirical research evaluating the impact of innovation policy instrument mix on firm-level innovation. To address this gap, this thesis develops a novel conceptual framework for evaluating static and dynamic complementarity and substitution between innovation policy instruments. To apply this framework empirically, the analysis creates three unique panel datasets through a series of data merges. Each dataset captures detailed information on the type and source of innovation policy instruments firms in Ireland receive, and contains 16,084 observations of 3,098 firms from 2007 to 2014. This data facilitates pairwise tests for complementarity and substitution between Research and Development (R&D) tax credits and R&D/innovation support from Ireland’s three key national funding agencies: Enterprise Ireland (EI), Industrial Development Agency Ireland (IDA) and Science Foundation Ireland (SFI). The panel nature of the data enables estimating a number of lag-structured models to test for both contemporaneous and longer-term impacts. To control for potential selection bias and endogeneity associated with the allocation and receipt of public funding, the econometric analysis employs an instrumental variable method. Using R&D intensity as a measure of firm-level innovation, results indicate that firms receiving a combination of R&D tax credits and EI support benefit more than firms that receive the same instruments separately. This static complementary relationship materialises in the same year firms receive this instrument mix, and persists for two years following receipt of the mix. Static complementarity is also identified between a combination of R&D tax credits and IDA or SFI support up to two years after the mix is initially received. The tests for dynamic complementarity reveal that a complex relationship between instruments unfolds over time. The transition from receiving an R&D tax credit in one year to an instrument mix composed of an R&D tax credit and EI support in the next year produces a substitutive relationship. In contrast, dynamic complementarity is identified when firms transition from receiving an R&D tax credit to a mix of the R&D tax credit and IDA support, or when firms transition from receiving SFI support to a combination of the R&D tax credit and SFI support. These results highlight that the sequence in which firms receive different instruments through time plays an important role in driving impact. While this study makes an important academic contribution to the field of innovation policy evaluation, the results also have potential policy implications in the area of public funding for innovation.