Researchers and policymakers have recently argued that industrial policy should focus more on stimulating growth in innovative small and medium enterprises (SMEs). However, few studies in Sweden and internationally have evaluated whether this kind of selective firm support is effective. This can to a high extent be explained by limited availability of data on firm support.

Here the effects of two selective government support schemes in Sweden are analysed that target innovative SMEs, namely VINN NU and Forska & Väx. Both programs are developed and administered by Vinnova, and the goal of the analysis is to provide recommendations that can facilitate future evaluations of selective firm support schemes.

This report is made possible by access to a unique micro database compiled by Growth Analysis, which is the Swedish Government Agency for Growth Policy analysis. The database includes information on all firms that have received different kinds of government firm support, which makes it possible to evaluate whether these support programs have any growth effects. The information from the database is used to match firms that received support from Vinnova against similar firms that did not receive support. Similar firms (twin firms) were identified using Coarsened Exact Matching (CEM), which has several advantages compared to the more frequently used Propensity Score Matching (PSM).

The effects of VINN NU and Forska & Väx are analysed on the number of employees, labour productivity, sales, sales growth, and the share of workers with higher education or research positions. The results do not indicate any statistically significant effects of the support programs during the period studied on: the number of employees, labour productivity, the share of highly skilled workers, or the proportion of researchers in the firm. The only positive effect found is limited to firms with at most six employees that after the support programs were ended increased sales with 20 percent compared to similar firms that did not receive any support. For larger firms, no positive and statistically significant effect is found.