Efficiency and its determinants in pharmaceutical industries: ownership, R&D and scale economy

Abstract

The purpose of this article is to measure the efficiency of pharmaceutical firms and identify their determinants using Korean and American samples from 1992 to 2004. We document some stylized facts in the patterns and sources of efficiency change in Korean and American pharmaceutical firms. The evidence shows that ownership structure can substantially influence the efficiency of pharmaceutical firms. Especially, institutional ownership rate affects corporate efficiencies negatively, corroborating the myopic institutional investor hypothesis. The hypothesis is supported by both Korean and American samples. However, we find evidence that foreign ownership in Korea promotes efficiency of pharmaceutical firms. It is shown that R&D intensity is positively related to contemporaneous largest ownership rate and prior foreign ownership rate in Korean pharmaceutical firms. In contrast, little evidence is found on the relationship between ownership structure and R&D intensity in the American pharmaceutical industry. These empirical results are robust even after we check the causal links among efficiency, R&D and ownership.

Acknowledgements

We gratefully acknowledge the anonymous referees for their insightful and constructive comments. The paper was initiated when Taewoo You was staying at Kent State University as a visiting scholar. This work was supported by the Korea Research Foundation Grant (KRF-2006-013-100070A).

Notes

¹ The only exception is Gonzalez and Gascon (2004). They analyse the productivity change in the Spanish pharmaceutical industry based on the Malmquist productivity index.

² The relationship between ownership structure and efficiency was examined by Button and Weyman-Jones (1992), and Zelenyuk and Zheka (2006).

³ It is likely that institutional investors in a less developed capital market become more myopic due to their perceived greater uncertainty in attaining long-run investment profit. Moreover, usually their performance is evaluated on an annual basis, thus pressing their short-term oriented investment.

⁴ In January, 1992, foreign investors were permitted to invest in the domestic stock market, subject to the restriction that foreign ownership of listed firms may not exceed 10% of total equity and individual foreign investors may not hold more than 3% of total equity. Foreigners were able to own up to 10% of domestically listed firms. 565 foreign investors registered with the Securities Supervisory Board. In May, 2002, Government approved lifting of foreign ownership restrictions on Korea Electric Power and Pohang Iron and Steel. In July, investments in stocks by resident foreign financial institutions were subject to the same limits as those by national institutions. They were allowed to undertake over-the-counter transactions in listed bonds. In September, the overall limit of $200 000 each emigrant family was allowed to transfer was abolished, and limits would be established on a case-by-case basis. The regulations on direct investments abroad were liberalized. The limit on investments in securities, insurance, and investment trust companies was raised from $10 million to $50 million, and the limit on similar investments by companies with international business licenses and insurance companies with assets of over 5 trillion won was raised from $50 million to $100 million.

⁵ The KIS-FAS is a financial database which is maintained by Korea Investors Service, Inc. (www.kisrating.com). DataGuide Pro is an on-line data retrieval program which is serviced by FnGuide (www.fnguide.com).

⁶ Our sample consists of only pharmaceutical firms which had survived for sufficiently long periods of time. This criterion is desirable due to the volatile and high-risk characteristics of the pharmaceutical industry. Some pharmaceutical firms do not generate sales at early times, or some other firms do not exist after producing sales for several years. Basically, inefficient firms are driven away, thus only efficient firms survive in the long-run. Our sampling strategy is well-suited to the spirit of our study on the efficiency of pharmaceutical firms.

⁷ The amount of materials is calculated by subtracting total labour cost and total capital cost from total cost. Therefore, the materials factor is a sort of residual input factor composed of various input factors other than labour and capital. Moreover, the miscellaneous factor prices are not easily observable by researchers. The heterogeneous and unobservable characteristics of materials factor bring about difficulties with the measurement of the price of materials factor. In order to accommodate these difficulties, we defined the price of materials as unity, and as a result we came to measure the materials factor in monetary amount, not quantity.

⁸ We are grateful to the reviewer who recommended us to consider this variable in the regression model.

⁹ DEA Version 2.1 was used to calculate efficiency scores. I am very grateful to Tim Coelli who generously provided the program, with some modifications for running it under Windows XP environment.

¹⁰ The regression models were estimated using the WinRats Pro version 6.1. Relatively low correlation estimates among variables suggests that there is no severe collinearity problems with the regressions.

¹¹ Our choice of models is slightly different from Sung (2007). Unlike ours, Sung (2007) employ the effects Tobit models as well as OLS. We note that his results did not significantly differ between the OLS and the effects Tobit.

¹² Although we replace log asset for market share, the key regression results do not change significantly.

¹³ This evidence is consistent with Cho (1998) who finds investment affects corporate value, in turn, which affects ownership structure, but not vice versa.