آزادسازی، آشفتگی صنعت و توازن فعالیت های R & D

In this paper we investigate the relationship between market structure and the composition of R&D activities. We present an empirical model to demonstrate that when market structure is shuffled by an institutional discontinuity, such as liberalisation, basic and applied activities respond in opposite ways: the former decrease whereas the latter increase. As a consequence, market turmoil is likely to provide firms with short-term incentives, shifting the allocation of resources towards applied and development activities. The model is tested on an original data set including innovation measures from the incumbent Public Telecommunications Operators of 17 European countries. We argue that the dynamics outlined in the model are likely to prevent firms (industries) sustaining an appropriate rate of innovative activity.

Over the last decade, European telecom markets have eventually started their processes of liberalisation. The old monopolies of Public Telecommunications Operators (PTOs) have systematically been replaced by liberalised markets, in which domestic and foreign investors are striving hard to establish and consolidate their market positions. The vast majority of OECD countries are now well ahead in the process of market liberalisation and the debate about the effects of competition has reached its peak. Very many of the arguments tend to support the idea that the net effect of the institutional change will be positive, both socially and for the incumbent firm. Nevertheless, there also seems to be a widespread awareness about the existence of potential sources of market failures, as witnessed by the ever-increasing regulatory activity performed by national authorities. Among the controversial issues, the debate about the actual suitability of market incentives to generate and sustain an appropriate rate of technical progress within the country has until now occupied a relatively marginal position. The crucial question is how, and how much, firms’ R&D expenditure profiles will be affected by institutional transition generating industry turmoil, both in terms of total amount and in terms of composition. This paper focuses on the latter issue, that is, on the effects of liberalisation on the balance between basic and applied R&D.1 In this section we will report some of the prominent concerns that have been expressed by many policy experts all over the world on the fate of industrial basic research; to this end, we will briefly review some anecdotal evidence about firms’ basic research strategies, in order to stress the relevance that such an issue might assume in the near future. The review is intrinsically biased towards the US experience, given the fact that that market is well along the way of market liberalisation.

This paper tries to shed some light on the role of industrial turmoil in determining the composition of R&D expenses. Specifically, we study the effect of increased competitive pressure on the relative share of resources allocated to basic and applied research activities. Although the relationship between market structure and R&D has long been a crucial issue in the economic literature, we believe that this specific matter should be given greater attention and its relevance should not be played down in any serious debate on the effects of market liberalisation. Our research question pertains to the balance between basic (long-term) and applied (short-term) activities. We claim that the effects that have been highlighted by the many authors who have studied the causal relationship between R&D and competition exert a different influence on the single components of research activities. Therefore, although we recognise that increased competitive pressure would produce an ambiguous effect on the overall research intensity, we hasten to hypothesise that the effect on composition will be clear and that we should observe an increasing incidence of applied activities in the firms’ research portfolio. We have tested this hypothesis in the telecom market, where we have a very clear example of institutional discontinuity generating increased product market competition, i.e. market liberalisation. We have gathered data for 17 PTOs in European countries where telecom markets were liberalised within the last decade. Measures related to patenting activity and scientific publications have allowed us to empirically decompose the PTOs’ research activities into basic and applied research. We have demonstrated that, in coincidence with (or in anticipation of) the liberalisation date, the PTOs have, on average, drastically reduced their commitment to publication activity and remarkably increased their effort towards patenting activities. We interpret this as a clear indication of a mutation in firms’ attitude towards applied research tasks, leading to a very different composition of their research portfolios. It is worth emphasising that, although we are mainly interested in the effect of competition on R&D composition, our results may also shed some light on the indirect long-term effects of competition on the overall R&D intensity. In a companion paper (Calderini and Garrone, 2000), we demonstrate that a sudden alteration in the balance between the two components of R&D, given an institutional discontinuity, may trigger, under certain conditions, a negative dynamic leading to the complete extinction of the incentives to perform any R&D activity whatsoever. The policy implications of our results are straightforward. The question is whether or not a sharp decrease in the total amount of basic scientific research carried out in-house may weaken the individual firm’s (and industry’s) capacity to sustain an appropriate rate of technological progress in the medium and long term. The issue is that higher incentives to perform applied research may lead to an erosion of the resources that are needed for the firms (industry) to perform basic activities and to produce basic knowledge, a crucial intermediate output. In the short term this might produce a beneficial effect in terms of competitiveness, since exploitation of the existing knowledge stock will improve the innovative performance of the firm (industry). Nevertheless, in the medium long term the depletion of the knowledge stock could negatively affect firms’ (industry’s) efficiency in performing their own applied activities. This is a clear example of the static/dynamic efficiency trade-off, which should be given greater attention by policy-makers. One straightforward option would be to shift the basic research activities that are being dismissed by the industry to the public research system. Two arguments should be raised here. First, the policy-maker, in this very historical circumstance, should seriously consider the adequacy of the public research system to sustain the extraordinary effort that is required to inherit the industrial basic research mission. We are rather doubtful that policy-makers in certain European countries could confidently assert that the national public research system has the resources and the skills to efficiently surrogate the missing quota of industrial basic research. Second, it is very true that, as far as the direct results of research are concerned, the output of research is largely to be considered a public good and, as such, hardly appropriable and easily transferable. Nevertheless, if we consider the indirect results of performing basic research tasks, such as acquiring capabilities in selecting projects and assessing their prospects and diffusion rates, these latter are highly appropriable, idiosyncratic and, most importantly, hardly transferable. Therefore, even assuming that the public research system can efficiently replace in-house industrial research, it is very debatable that indirect results can be transferred to the industry and internalised by individual firms. The relevance of the argument lies in the fact that, for firms that operate in countries that are not on the frontier of scientific research, this latter output (indirect results) is by far and away more important than invention as such. The message is, therefore, that it could be rather hazardous to rely entirely on the public system to provide the national system of innovation with the appropriate stock of scientific knowledge. Policy intervention should be directed at guaranteeing that, even in coincidence of a steep increase in competitive pressure, the information and communication industry could maintain a critical level of in-house basic research activity. Finally, it is important to stress that we have limited our analysis to one specific kind of determinant, the liberalisation process. Nevertheless, we believe that at least two other contingencies are bound to exacerbate the dynamics outlined in the paper. First, liberalisation in European countries has very often been accompanied by privatisation processes. Sobrero et al. (2000) have investigated this problem and their conclusions seem to point in the same direction as the results presented in this paper. Second, PTOs and telecom carriers in general have recently spent enormous amounts of money on third generation mobile phone spectrum rights. In the opinion of many commentators, this is bound to make many firms in the industry severely cash-constrained in the short term, thus exacerbating the tendency to privilege investments that yield high returns in the short term and to use the payback period as a criterion for assessing alternative research projects.16 The issue is crucial to policy-makers. The negative effects (private and social) will become manifest in the medium long term, when it might be far too late to design any kind of policy to mitigate the effects of a collapse in the industry’s knowledge base. We address this comment to policy-makers who operate in countries that are followers in the innovation race and that might therefore be tempted to direct their efforts uniquely to the exploitation of technological knowledge created elsewhere. This strategy is myopic and is likely to lock the national innovation system into utterly vicious trajectories.