Productivity, financial performance, and corporate governance: evidence from Romanian R&D firms

ABSTRACT

We study the impact of financial performance and corporate governance on the productivity of the Romanian Research and Development (R&D) sector. We draw on a dataset consisting of 116 Romanian R&D companies covering the time span from 2007 to 2016. Firm productivity is computed using several metrics of TFP (total factor productivity). We show, based on bootstrap panel quantile regressions, that size, financial profitability, and foreign ownership are key drivers of R&D companies’ productivity. Intangible assets and taxation have no significant effect on R&D firms’ productivity, while the degree of independence in decision-making and owners’ presence in firms’ management negatively impact TFP. With women on the board, state-owned applied research institutes benefit from higher productivity compared with R&D private firms, if they are in lower quantiles.

KEYWORDS:

• Productivity
• R&D firms
• corporate finance and governance
• panel quantile regression
• Romania

JEL CLASSIFICATION:

• D24
• O30
• G32
• C33

Disclosure statement

No potential conflict of interest was reported by the author(s).

Acknowledgements

We are grateful to Scott W. Hegerty, Xavier Galiègue, Marcel Voia and the participants at the INFER Annual Conference (Brussels, 2019) for constructive comments and suggestions. All remaining errors are ours.

Notes

1 As argued, in this category, one can include private firms as well as public companies that are independent research institutes. These public companies are self-financing. In other words, their activity is not supported by the government budget.

2 Romanian Government Decision no. 929 - October 21, 2014 regarding the “National strategy for research, development and innovation 2014–2020.” (Romanian Government, 2014). The priority fields in terms of public R&D investment outlined in this document are defined along with EU strategy. They are the following: (i) bio-economy, (ii) ICT, (iii) energy, environment, and climate change, (iv) eco, nanotechnology, and advanced materials, (v) health, (vi) patrimony and cultural identity, and (vii) new and emerging technologies.

3 Romania joined the European Union (EU) in 2007. In this paper, we analyze the period 2007–2016 (the choice of the time span is also constrained by data availability). Hence, over the investigated time span, there is no change of Romania’s EU membership status.

4 Another competing measure of firms’ efficiency is Tobin’s q. However, compared with this ratio, the TFP measure of efficiency is relatively unaffected by the choice of accounting methods (Hill and Snell 1989). Further, the TFP approach allows for different factor intensities in different firms and industries (Schoar 2002).

5 Only the firms for which observations are available for at least six out of the eleven years of the time span are retained in the analysis. None of the companies defaulted during the analyzed period. Out of 116 companies, about 65% are located in the Bucharest area. Overall, the companies in our sample perform R&D activities in various domains. Nevertheless, among the 47 public research institutes, those located in the capital area carry out very specialized activities.

6 The methodology is heavily drowned from Albulescu, Miclea, and Grecu (2021).:

7 $K_{it}=Kinitia{l}_{it}+MI{N}_K$. Another way to overcome the issue of negative values for the stock of capital is to exclude them from the sample. However, in this case our panel analysis may suffer from the broken panel bias. While in this paper we present the results using the first option for the capital stock series, we have also run all the regressions without negative values (these additional findings can be provided upon request). There are not important differences in terms of empirical results among the two series of findings. Hence, we further on present the coefficients of the estimations that avoid the broken panel problem.:

8 A level of indicator equal to B means direct or total ownership less than 50% but one stakeholder own over 25% (medium-low ownership concentration) and a level equal to C means a direct or total ownership over 50% and one stakeholder own over 50% with an unknown share (medium-high ownership concentration).

9 $D2\mathrm{\_}FDI$ variable remains almost the same. However, in order to get closer to the definition of associates according to which 10% to 50% of the voting power is held by the FDI enterprise (OECD 2008), we have kept the 25% threshold.

10 $D1\mathrm{\_}Indep$ variables is positive, but very small).

Additional information

Funding

This work was supported by a grant of the Romanian National Authority for Scientific Research and Innovation, CNCS – UEFISCDI, project number PN-III-P1-1.1-TE-2019-0436