Lade Inhalt...

Cooperative and Non-cooperative R&D Behaviour

Seminararbeit 2012 21 Seiten

VWL - Innovationsökonomik










1. Introduction

“ It has become appallingly obvious that our technology has exceeded our humanity. ” 1 Albert Einstein Technological progress is one of the most important factors guaranteeing economical growth in a long term. The present paper deals with a question if cooperative R&D can increase the intensity in research activities. What are the advantages resulting from cooperative agreements? Do they also face any risks? These questions are discussed in the second section of the paper. The intention is to introduce the basic model of cooperative and non-cooperative R&D behaviour, described by D’Aspremont and Jacquemin. In section 3, individual cases of cooperative behaviour are discussed and compared, while in the next step, the impact on the welfare is presented. Section 4 deals with the model by Kamien, Muller and Zang. After a reasonable simple description, each single case is evaluated and the most recommendable solution regarding the research activities is introduced. A conclusion follows.

2. Research and Development

“Research and experimental development (R&D) comprise creative work undertaken on a systematic basis in order to increase the stock of knowledge, including knowledge of man, culture and society, and the use of this stock of knowledge to devise new applications.”2

R&D activity is divided in three main parts: basic research, applied research and experimental development. Basic research is undertaken in order to create new knowledge of basic information and observable facts without leading to specific applications while applied research is focused on more practical issues. Experimental development includes transformation of knowledge received from research activities or practical experience in producing new goods, devices or creating new processes or services. This kind of development can also be used in order to improve the quality of already produced units or existing processes.3

In addition, there are two types of R&D: process innovation, describing a more efficient technology of production goods and product innovation, which is creation of a novel good.4

In the next step, reasons for R&D cooperation between firms are described. Resulting advantages and risks are also taken into consideration.

2.1. R&D Cooperation between Firms: Advantages vs. Risks

One of the main reasons firms decide to start a cooperation project is the argument of scale economies. Becoming more complex, the structure of innovative processes is associated with rising costs. Cooperation gives participating firms a possibility to increase their research activities by directing the whole amount of financial recourses of cooperation partners towards R&D. Moreover, if necessary, for a consortium of firms, it is much easier to obtain a loan than for one single small enterprise. It seems obvious that within collaboration, larger projects can be put into action than in case of firms acting separately. Within cooperation firms have more potential to introduce innovations because of the access to a greater quantity of financial resources.5

Another important argument for a decision in favour or against undertaking R&D depends on access to external knowledge. Using R&D cooperation as an additional source of innovation activities, a probability of achieving corresponding results generally increases. In order to maintain the research intensity, a continuous interaction with external sources of knowledge is needed. Cooperating with other economic operators, firms are exposed to an ambiguous effect of research externalities. On the one hand, firms can benefit from external knowledge generated by a rival company, which is an advantage. On the other hand, there is a risk that the other firm can use their own research results without paying a charge. The example mentioned above illustrates the so called R&D spillover effects, as further described in the following section.6

2.2. R&D Spillovers

R&D spillovers are usually defined as externalities produced as a result of the fact that economic operators are unable to gather the full extent of economic benefits generated by their own R&D activities. That is why other firms, which are active on the market, can benefit from research undertaken by rival firms without paying a charge.

“By technological spillovers, we mean that (1) firms can acquire information created by others without paying for that information in a market transaction, and (2) the creators (or current owners) of the information have no effective recourse, under prevailing laws, if other firms utilize information so acquired.”7

Using information shared by rival firms enables spillovers’ recipients to invest more efficiently in highly promising research activities while avoiding the others. If the information is completely shared by participants, the probability of double research activities can be excluded, which means a reduction of research costs. Undertaking R&D individually, each single firm undergoes the trial and error process searching for the same innovative technology. Sharing the information within the R&D cooperation enables to distribute the innovation process between the single participants avoiding the cost of duplication.8

It is also possible to describe spillovers in a broader way including the option of voluntary transfer of innovation technology.9 However, R&D cooperation does not necessarily include voluntary information exchange if the spillover effects are not internalized by the participating enterprises.10

The degree of the spillover effect between the cooperating firms can be measured by ratio [Abbildung in dieser Leseprobe nicht enthalten]. It describes externality effects firm 1 gets from firm 2 without paying a charge. In case of so called free riding there is a great incentive for firms to reduce their research activity and to take advantage of the rival’s R&D investments for free. The greater [Abbildung in dieser Leseprobe nicht enthalten] the more efficient the R&D undertaken by firm 2 can be used by firm 1 and the higher cost reduction for firm 1 is possible. The most possible spillovers are given if [Abbildung in dieser Leseprobe nicht enthalten] and the spillovers are totally internalized. There are no spillovers if[Abbildung in dieser Leseprobe nicht enthalten]. As a ratio, ߚ has to take a value between zero and one. In this case spillovers are positive but imperfect.11

3. Cooperative and Non-cooperative R&D Behaviour

In following, the cases of cooperative and non-cooperative R&D behaviour are analysed in compliance with the model presented by Claude D’Aspremont and Alexis Jacquemin (DJ).12 In order to simplify the given model, the present analysis is based on the version illustrated by Lynne Pepall, Dan Richards and George Norman.13 In comparison to numerous oligopoly models, the present model illustrates a different view on the relationship between the participating firms within cooperation. Often, cooperation in some fields is given while staying rivals in others at the same time.

There are two different types of R&D cooperation to be distinguished. On the one hand, the collaboration only takes place at the R&D stage itself while firms still compete in the marketplace. On the other hand, firms act in collusion during the whole R&D and the following selling process. In the course of developing common corporate governance and having impact on the production output, cooperating firms get the possibility to take a shared advantage of their investment activity. For both types of such an agreement two possible reasons might be expected. First, it enables companies to reduce costs avoiding double expenditure on R&D. Second, there is more monopoly power because of lower production costs.

Considering the spillover effects, the following excerpt analyses the expectations in three different cases of agreement: non-cooperation in both, output and R&D, cooperating in “precompetitive stage” as well as collaboration during the whole game. The implication on the social welfare is discussed in the next step.

The model analysing the advantages of R&D cooperation is based on following assumptions: an industry consists of two companies 1 and 2 with an inverse demand [Abbildung in dieser Leseprobe nicht enthalten] and describes the total productivity amount.

Abbildung in dieser Leseprobe nicht enthalten

The marginal production costs for the respective enterprise ܿ[Abbildung in dieser Leseprobe nicht enthalten]are presented by a function including its own R&D expenditures[Abbildung in dieser Leseprobe nicht enthalten] as well as the amount of research[Abbildung in dieser Leseprobe nicht enthalten]undertaken by the rival firm. The total value increases with higher marginal production costs ܿ. The amount of research ݔଵ undertaken by the firm 1 has a decreasing effect on the costs as well as [Abbildung in dieser Leseprobe nicht enthalten] which are positive spillovers firm 1 gets from firm 2 without any charge.

Abbildung in dieser Leseprobe nicht enthalten

In order to formulate the model as simple as possible, same research costs for both firms are described as:

Abbildung in dieser Leseprobe nicht enthalten

There are no economies of scale regarding to R&D activities. That is why they become more costly with increasing degree of implementation. The revenue from the R&D expenditure is therefore diminishing and the costs function is quadratic. According to the fact that the first order condition is greater than zero, decreasing returns are given:

Abbildung in dieser Leseprobe nicht enthalten

As a basic assumption considered throughout the analysis, the firms participate in a two-stage game. While choosing their level of research [Abbildung in dieser Leseprobe nicht enthalten] in the first step, firms are exposed to competition in the second step. Looking for an appropriate solution, the game is usually solved backwards. The next step starts by comparing outputs.

3.1. Non-cooperating in both, R&D and Output

In the first case to be analyzed, firms act in a completely non-cooperative way. Both of them undertake their own R&D and compete in output. As a starting point, Cournot equilibrium output is given:

Abbildung in dieser Leseprobe nicht enthalten

According to assumptions of Cournot competition, in the second stage, firms are supposed to maximize their individual profits which are represented as follows. [Abbildung in dieser Leseprobe nicht enthalten], the research costs which influence the profit in a negative way, are taken into account as well. Γ is a ratio describing the efficiency of the technological R&D input. The greater it is the more costly is the research. It has to take on a value greater than zero because the costs of research are still given and cannot be completely reduced.

Abbildung in dieser Leseprobe nicht enthalten

To express the given equilibrium output as well as profit as a function of the firm’s decision made in the first stage regarding to research activities and extent of spillover, it is necessary to replace the respective marginal production costs with equation (2).

Abbildung in dieser Leseprobe nicht enthalten

Equation (7) illustrates each firm’s output depending positively on its own research expenditures. R&D investment contributes to costs reduction as shown in (2) and enables to produce a higher output quantity under more favourable conditions. In contrast to own R&D expenditures, the R&D efforts undertaken by the competitor can influence the output and the profit in two contrary ways. On the one hand, firm 2 reduces firm’s 1 costs by its own R&D expenditures, which is due to positive spillover effect. On the other hand, reducing its own costs by undertaken R&D, firm 2 becomes more competitive and influences firm’s 1 output in a negative way. This process is called competition effect. Which of the both has a greater implication depends on the degree of spillover effects. As shown by the coefficient [Abbildung in dieser Leseprobe nicht enthalten]is positive when spillovers are greater than 0.5 and thus large. In this case the spillover effect dominates. If the value of ߚ is smaller than 0.5, the spillover effect is poor and the competition effect is relatively stronger.

In order to determine the optimal level of research intensity, it is necessary to differentiate the second- stage profit function (8) with respect to [Abbildung in dieser Leseprobe nicht enthalten] and to set the derivative equal to zero. In the next step, the research intensity [Abbildung in dieser Leseprobe nicht enthalten] can be determined.

Abbildung in dieser Leseprobe nicht enthalten

In case of low spillovers with[Abbildung in dieser Leseprobe nicht enthalten]the research intensity reaction functions are downward sloping. It means that firms are strategic substitutes. One can intuitively say that the more R&D research the one firm undertakes, the less the other does and thus, the competition effect dominates. Undertaking R&D activities, firm ݅ǡ in the main affecting, reduces its own costs. It benefits from the competitive advantage in comparison to the rival enterprise. To avoid competitive pressure firm ݆ has to increase its profitability by decreasing the value of expenditure on research.

On the contrary, if the spillovers are relatively large with [Abbildung in dieser Leseprobe nicht enthalten] the research intensity reaction functions are upward sloping. In this case the participating firms are strategic complements. Because of the dominating spillover effect firm ݅ benefits from the research efforts of firm ݆ without paying a charge for it. Increasing its profit, firm ݅ disposes of sufficient financial means to undertake R&D.

The next necessary step is to find the equilibrium level of R&D investment for both enterprises. As known, without any research, both firms have the same costs incurring. The demand function is also identical for both. That is why the equilibrium has to be symmetric and is demonstrated by[Abbildung in dieser Leseprobe nicht enthalten]. After having substituted this equation in (10) and then solved for[Abbildung in dieser Leseprobe nicht enthalten] the Nash equilibrium research intensity is given.

Abbildung in dieser Leseprobe nicht enthalten

Decreasing in [Abbildung in dieser Leseprobe nicht enthalten] the equation shows that because of increasing spillovers, firms are forced to reduce their degree of research. Driven by the tendency of free ride advantages, R&D firms usually reduce their own expenditures on R&D. With the increasing degree of spillovers, the temptation to make use of free riding also gets stronger while the equilibrium research expenditures get lower. However, this does not necessarily mean that total, means own and rival, efforts of research are decreased by a higher value of spillovers.

To calculate the decline of marginal costs caused by the total effort of research, it is necessary to substitute equation (11) into (2) and differentiate with respect to [Abbildung in dieser Leseprobe nicht enthalten]in the next step.

Abbildung in dieser Leseprobe nicht enthalten

Therefore, following relationship between the reduction of firm’s costs and spillover effect is given: if[Abbildung in dieser Leseprobe nicht enthalten]takes a value between 0 and 0.5, cost reduction increases because the numerator is growing. In contrast, if [Abbildung in dieser Leseprobe nicht enthalten] exceeds the value of 0.5, the whole term becomes negative and cost reduction decreases.

Substituting the Nash equilibrium research intensity from equation (11) into (7), there is a possibility to determine how the firm’s output depends on spillovers.

Abbildung in dieser Leseprobe nicht enthalten

Differentiating the equation with respect to [Abbildung in dieser Leseprobe nicht enthalten], one can see that[Abbildung in dieser Leseprobe nicht enthalten] Firm’s outputs are growing with increasing research spillovers indicating falling prices only in the case if [Abbildung in dieser Leseprobe nicht enthalten] takes a value between 0 and 0.5. In contrast, if [Abbildung in dieser Leseprobe nicht enthalten] keeps increasing and takes a higher value than 0.5, produced output decreases and higher prices are generated, which is a disadvantage for consumers. In this case, consumers only benefit from a certain degree of research spillovers while perfect spillovers cause losses. Not only consumers but also firms do not benefit from a perfect degree of R&D spillovers. Substituting (12) in (14) illustrates the equilibrium profit of each firm.

Abbildung in dieser Leseprobe nicht enthalten

According to numerical analysis, profits are growing with increasing degree of research spillovers until the value of 0.88. Profit losses can be noticed if[Abbildung in dieser Leseprobe nicht enthalten] takes a value between 0.88 and 1. Conclusively, one can say that firms are benefiting in case of very large spillovers while generating profit losses in case of perfect research externalities.

3.2. Technology Cooperation

In the second case firms are cooperating in R&D while still competing in their marketing strategy. The main assumptions of the model still remain. In contrast to the previous case firms do not decide for an output separately but reach an agreed solution. In the first step, it is necessary to determine the sum of joint profits, the firms maximize in the second stage. The individual profits are shown in equation (8). After having differentiated the sum with respect to[Abbildung in dieser Leseprobe nicht enthalten] , which is equal to[Abbildung in dieser Leseprobe nicht enthalten] and as well [Abbildung in dieser Leseprobe nicht enthalten] because of the symmetry assumption, following equilibrium degree of research is given:


1 Moffett, Sandra, McAdam Rodney, and Parkinson, Stephen. (2004). Technological Utilization for Knowledge Management, Knowledge and Process Management, 11:3, pp.175-184

2 OECD. (2002). “Frascati Manual. Proposed Standard Practice for Surveys on Research and Experimental Development,” URL: 2002_9789264199040-en [Access: 19.12.2012], p. 30

3 Ibid.

4 Pepall, Lynn, Richards, Dan, and Norman, George. (2010). Contemporary Industrial Organization - A Quantitative Approach. Hoboken, NJ: Wiley., p. 426

5 Hinloopen, Jeroen. (1998). “Strategic R&D Cooperatives,” URL: [Access: 19.12.2012], pp. 6-8.

6 Broekel, Tom, Schimke, Antje, and Brenner, Thomas. (2011). “The Effects of Cooperative R&D Subsidies and Subsidized Cooperation on Employment Growth,” URL: [Access: 19.12.2012], p. 6

7 Grossmann, Gene M., and Helpman, Elhanan. (1992). Innovation and Growth in the Global Economy, Cambridge: MIT Press., p .16

8 Kamien, Morton. I., Muller, Eitan. and Zang, Israel. (1992). Research Joint Ventures and R&D Cartels, The American Economic Review, 82:5, p. 1298

9 Dumont, Michel, Meeusen, Wim. (2000). “Knowledge Spillovers through R&D Cooperation,” URL: [Access: 19.12.2012], p. 1

10 Kamien, Muller, and Zang, op. cit., p. 1294

11 Pepall, Richards, and Norman, op. cit., p. 444

12 D’Aspremont, Claude. and Jacquemin, Alexis. (1988). Cooperative and Noncooperative R&D in Duopoly with Spillovers , The American Economic Review, 78:5, pp. 1133-1137

13 Pepall, Richards, and Norman, op. cit., pp. 444-449


ISBN (eBook)
ISBN (Buch)
509 KB
Institution / Hochschule
Ruhr-Universität Bochum – Applied Microeconomics
R&D F&E Research Development Forschung Entwicklung Kamien Muller and Zang Cooperative R&D




Titel: Cooperative and Non-cooperative R&D Behaviour