Abstract
The practice of the Japanese court in case of a dispute between the employer and employee regarding the amount of remuneration for an employee invention has been to order that the additional profit from the invention be divided proportionally to their respective input contributions. We show that, if the employer’s investment and employee’s effort are weakly complementary, this rule causes the share effect (excessive incentives on the part of each party to expend investment or effort in order to increase his/her share of the surplus) to dominate the probability effect (insufficient incentives arising from the fact that each party obtains only part of the increase in the expected surplus), and thus leads to excessive investment and effort relative to the joint-payoff-maximising levels. If the court cannot capture the employer’s investment as fully as the employee’s effort, the employer’s investment may be too low compared to the joint-payoff-maximising level.
Acknowledgements
We are grateful to two anonymous referees for their comments that greatly improved the paper. We also thank Kazuyuki Motohashi, Jun Suzuki, Schumpeter Tamada, Ryuzo Furukawa, Tagui Ichikawa, Kotaro Suzumura, Akira Yamada, Tatsuo Tanaka, Toshihiro Matsumura, Dan Sasaki, Ikuo Ishibashi, Akira Ishii, Hiroshi Mukunoki, Akira Ogawa, Hiroaki Ino, and Sadao Nagaoka for helpful comments and suggestions on earlier versions. Financial support by Grant-in-Aid of Japan’s Ministry of Education, Culture, Sports, Science and Technology is gratefully acknowledged.
Notes
1Section 35 (4) quoted here has since been revised. We will briefly return to the April 2005 revision of the Patent Law in the concluding section of this paper.
2See, for instance, Tamura and Yanagawa Citation(2003).
3An alternative interpretation of the court rule might be the formulation where the surplus from the invention is distributed in proportion to each party’s respective contribution in increasing the probability of successful invention relative to the hypothetical case where that party does not make any investment or effort, given the level of effort or investment by the other party, i.e. θ (k, l)=[p(k, l)−p(0, l)]/[[p(k, l)−p(0, l)]+[p(k, l)−p(k, 0)]]. Although we do see the rationale behind such a rule, in that it utilises information that accounts for the quality of investment and effort in the form of their effects on the success probability, this formulation requires information on success probabilities at, ([ktilde], [ltilde]), at ([ktilde], 0) and at (0, [ltilde]), where ([ktilde], [ltilde]) denotes the realised values of investment and effort. We feel it reasonable not to assume that the court has the ability to collect and verify this information.
4It may be worth noting that the employee’s effort cost does not have to equal the wage he receives. Thus, the compensation to the claimant employee set by the court in individual cases may well be high above what he had received in wages, as in the Nakamura vs. Nichia Corp. case.
5In this case, p kl =0.
6In this case, p kl =p k p l >0.
7In the C-ownership case in Aghion and Tirole Citation(1994a), the surplus from a successful invention is assumed to be divided equally between the employer and employee. The equilibrium levels of investment and effort are lower than the joint-payoff-maximising levels. This is because the share effect does not arise under this rule. More generally, if the distribution rule is such that each party’s share is independent of its contribution, the share effect does not arise and there are insufficient investment and effort relative to joint payoff maximisation.