Firms' R&D dilemma: to undertake or not to undertake R&D

Abstract

It is well known that in most industries a significant proportion of firms do not perform innovative activities. Although empirical studies on the determinants of R&D often have taken this fact into account by considering the dependent variable as a censured one, there is not an explicit theoretical model to explain the zeros. The concern of this letter is to discuss a simple theoretical model where firms simultaneously decide whether to undertake or not R&D activities jointly with the level of the R&D investment. It is shown that a firm performs R&D activities only when its optimal level of R&D expenditure is higher than a threshold. Additionally, it is shown that both the probability of undertaking R&D activities and the R&D expenditure increase with market power, with the elasticity of demand with respect to quality and with the elasticity of quality with respect to R&D. Finally, from this simple theoretical framework we discuss a suitable econometric model that threats these decisions simultaneously.

Acknowledgements

We are very grateful to Jordi Jaumandreu for all his helpful comments and suggestions. We acknowledge financial support from CICYT, projects BEC2002-02527 and SEC2000-0268, and from Xunta de Galicia PGIDT00PXI30005PN.

Notes

¹ For example Cohen and Klepper (1992) find, in a sample of American manufacturers that overrepresents the biggest firms, that 16 percent of the total observations are zero R&D expenditures. Moreover, the proportion of business units that report zero expenditures range, for the two digit industries, from 5 to 50%. And according to a representative sample of Spanish manufacturing firms, almost 15% of the firms with more than 200 workers, and 70% of the firms under this size, do not report to perform formal R&D (see González, Jaumandreu and Pazó (2001).

² For a recent theoretical analysis see, for example, the paper of Lee (2002) in which the R&D intensity is determined jointly by consumer characteristics and the R&D elasticity of quality, but there is no explanation for the zero expenditure.

³ Some pioneering experiments to introduce Tobit analysis in the examination of the determinants of innovative activities are Cohen, Levin and Mowery (1987) and Cohen and Klepper (1996). Crépon, Duguet and Mairesse (1998) analyses the links between productivity, innovation and research at the firm level taking into account explicitly that only a small proportion of firms engage in research activities. More recently Parisi and Sembenelli (2001) provide econometric evidence on the elasticity of private R&D on its price applying a censored panel data regression model to a panel of Italian firms.

⁴ In fact, several authors have expressed their worries about the effects of sample selection biases in most of the samples used in the analysis of innovative activities. See, for example, Cohen and Levin (1989) and Griliches (1990).

⁵ Most theoretical and empirical work has focused on process R&D activities, but empirical evidence point to the dominance of product or mixed R&D expenditures. For example, Scherer (1984) reports that most industrial R&D is product-oriented (75.4% of R&D expenditures of American firms are allocated to product innovation activities).

⁶ The utility specification follows Dixit and Stiglitz (1977), and introduces quality in a way similar to Levin and Reiss (1988), Motta (1992) and Sutton (1998), among others.

⁷ See, for example, Arrow (1962) and for a more recent discussion, Metcalfe (1996).

⁸ This is a rather usual assumption, that considers R&D a non-rival input of production. Klette and Griliches (2000) describe this assumption as ‘once a new product improvement is developed and introduced, no additional resources are needed to produce this improved product as compared to the older version’.

⁹ This a Dorfman-Steiner type of condition (Dorfman and Steiner (1954)) as that obtained by Lee (2002).

¹⁰ These authors modelled the housewives decision to enter the labour market depending on unobserved reservation wages.