Growing Flow of Technology Drain: Considerations in measuring impacts and planning measures

Lately, cases of trade secrets and technology draining from Japanese companies to their foreign rivals have been making headlines in the media. Including those that have gone unnoticed, there may be a significant number of such episodes. Technology drain makes it difficult for originator companies to secure revenue stream from their successful outcomes of research and development (R&D) activities (because the appropriability of technology--i.e., inventing companies' ability to maintain possession of their inventions--decreases), and therefore may seriously hamper innovation activities, which are a key driver of economic growth.

Prompted by such concerns, Japan is currently reviewing the Unfair Competition Prevention Act for amendments to provide greater protection for trade secrets. However, in order to be eligible for protection under the law, trade secrets must fulfill certain criteria, and, in many cases, technological knowledge that has been acquired and become integral part of human knowledge fail to do so (Note 1).

We often hear people say that one big reason behind the decline of Japanese electrical and electronics manufacturers and the rise of their South Korean counterparts in recent years is the draining of Japanese engineers and their technological knowledge to South Korea. What they have in mind is exactly the type of technology that cannot be protected under the Unfair Competition Prevention Act.

However, to what extent can this argument be supported by objective data? Since we can track the migration of Japanese engineers to some extent by using data on the inventors of patent applications, it may be possible to find a relatively precise answer to this question.

In this column, I would like to show, in a simplistic way, how we can approach the problem of technology drain by using patent application data.

The table below lists some of the inventors who moved from Japanese companies to Samsung SDI extracted from data on patent applications in the field of lithium ion batteries, where the market share of Japanese companies has been overtaken by South Korean companies. For instance, H.Y. at the top of the list filed patent applications as an inventor belonging to Sanyo GS Soft Energy Co., Ltd. until some point in 2003, but, thereafter through 2006, his/her affiliation was filed with Samsung SDI in the applications (Note 2).

Similar attempts to capture the state of technology drain from Japanese companies to Samsung SDI have been made in an article published in the Nikkei Business Online (Muto (2013)) and another in the Weekly Diamond magazine (November 16, 2013 issue). However, what is more important is to analyze what types of engineers have moved to South Korean companies and what roles they are playing there. A Japanese engineer doing research alone in a South Korean company may help the company improve competitive advantage, but probably not to the point where it will pose a major long-term threat in terms of technology competition because a significant accumulation of technological knowledge is unlikely to occur.

Though this is just an overview of data, our research has found some general tendencies in the migration of engineers from Japan to South Korea, as described below. (Please note that these findings are hypothetical as they have not been subjected to rigorous econometric analysis or testing).

To begin with, headhunting of top class engineers is surely taking place (for instance, N.Y. listed in the table was among the top in Japan in terms of the number of patent applications in the field of lithium ion batteries). After migration, those engineers typically work on inventions jointly with researchers of their new companies. It is thus believed that they are making considerable contributions to the development of the engineers in their new companies.

However, there have been very few cases of such high end headhunting and the vast majority of engineers recruited by South Korean companies are of an average skill level--i.e., those comparable or slightly inferior to their original colleagues in terms of the number of patent applications filed--who must have been affected by the streamlining of operations by Japanese companies (for instance, the number of patent applications filed by H.Y. and K.M. prior to moving to Samsung SDI was about average at Sanyo GS Soft Energy, which they joined around the time when GS-Melcotec Co., Ltd., their initial employer and a subsidiary of Japan Storage Battery Co., Ltd., was merged into Sanyo Electric Co., Ltd.).

The productivity of those average skill level engineers tends to improve after migration (achieving an average increase of one patent application per year, probably due to severe application quotas imposed by their new employer). As those engineers often work with other Japanese engineers (for instance, H.Y. and K.M. from Sanyo GS Soft Energy and M.H. from Toda Kogyo Corp. worked on inventions jointly at Samsung SDI), technology spillover to the new company would be limited.

By verifying those hypothetically with the quality of patents and changes in the productivity of new company inventors taken into account, we can expect to get a clear picture of what is happening (Note 3).

In Japan, engineers from Japanese companies moving on to their follower companies are usually talked about in a negative context. But this is the kind of practice that has been and is done in any country including Japan in the catch-up process, and this is how new technological opportunities and competition are generated.

In the United States and Europe, research on the migration of human resources tends to focus on the effects of knowledge spillover and brain drain. Here, the term "brain drain" refers to the phenomenon in which highly competent individuals who have acquired advanced knowledge and skills in their own countries emigrate, in pursuit of more favorable treatment and opportunities, to those countries where talented human resources concentrate. Thus, there are some commonalities with technology drain.

However, technology drain differs greatly from brain drain in the following two points: 1) human resources move from countries with a higher level of knowledge and technology to those with a lower level; and 2) the cost of human capital development is mostly borne by source companies. Thus, emotional issues aside, we can say that the focus of criticism on technology drain is on the problem of free riding on the human resources development and R&D borne by others for catch-up purposes (Note 4).

This free riding is difficult to prevent. Human resources development and R&D inevitably involve uncertainty. However, companies poaching established engineers do not bear any part of the cost of their training. They are also free from the risk of all such training producing nothing. That is why those new employers can offer higher wages than source companies and yet secure sufficient profits. Meanwhile, it is only natural for engineers to move on to a company where their technological skills and knowledge are more highly valued, if their current employer has poor performance and they are dissatisfied with how they are treated. If free riding becomes widespread in this manner, incentives for investing in the development of human resources and R&D will diminish.

Of course, migration may turn out to be a new opportunity for engineers to demonstrate their abilities, which have been sealed off and unused at their current workplace due to worsening business performance, at another company launching a new project, and, if so, increase social welfare (measured at the global level) will increase in the short term. However, considering its long-term effects on R&D incentives, it is questionable whether such migration truly contributes to a rise in social welfare.

In the end, the problem of technology drain comes down to the question of appropriability, i.e., to what extent companies are able to secure revenue from the outcome of their R&D activities (Note 5). In order to boost incentives for R&D, many countries have established a system for protecting intellectual property rights (IPRs) as a way to secure appropriability. If we apply the same idea to the development of human resources, a system for allowing companies to retain their trained researchers for a designated period of time should be established.

Increasing control over leaving employees and enforcing the contractual terms of confidentiality and non-competition agreements more strictly would be effective. However, as an additional option and a way to safeguard the freedom of choice in employment, it is worthwhile considering introducing a system that would allow companies to charge transfer fees for employees leaving before completing the period specified in their contract (Note 6). This would have a deterring effect on free riding because companies poaching talented workers from their competitors would have to pay their share of the training cost for such workers. Although this kind of contractual terms is quite common in the world of sports, there will be many obstacles in introducing them into the general labor market. However, if there are sufficiently many Japanese companies that continue to be committed to the development and training of their engineers amid intensifying competition from their rivals from emerging economies, it may not be so long before we see the introduction of this type of contract for engineers in a somewhat modified form.