Trade and the margins of innovative technologies: evidence from patent-firm matched data

ABSTRACT

We study the impact of China’s surge in the global economy on the intensive and extensive margins of firm-level innovative technologies. We predict that greater market access following the accession of China to the World Trade Organization (WTO) has positive effects on both the intensive and extensive margins of innovative technologies, and the effects are more pronounced for more productive firms. We confirm our predictions empirically using the United States Patent and Trademark Office patent data matched with Taiwanese firm-level data for the period from 1998–2014.

KEYWORDS:

• Innovation
• patents
• technology scope
• export shock

Acknowledgement

This paper is an expanded version of Chapter 3 from Jeonghwan Choi’s Ph.D. dissertation. Jeonghwan Choi expresses his thanks to his advisors, Steven Matusz and Susan Chun Zhu, for their advice. We are also grateful to an anonymous reviewer and Junbo Wang, the editor, for key suggestions and to participants at various seminars and conferences for helpful comments and suggestions. ByeongHwa Choi acknowledges support from the Ministry of Education and the Ministry of Science and Technology, Taiwan. Any remaining errors are solely our responsibility.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1. In this study, we do not intend to show directly how firms allocate resources to innovation activities. However, because technology scope adjustment implies the reallocation of resources within firms for their best use, how firms adjust their technology scope has important allocation implications.

2. Regarding market expansion and innovation, Acemoglu and Linn (2004), for example, found that the rise in the potential market size from demographic changes in the US heavily encouraged corporations in the pharmaceutical industry to produce new non-generic drugs and molecular entities through innovation. Aghion et al. (2005) produced a classic work on market competition and innovation.

3. Figure 1 describes the distribution of technology diversity and US import diversity at the country-year level. We used US HS product-level import data provided by Peter Schott’s US trade database (https://sompks4.github.io/sub_data.html) to construct country-year level US import diversity. For the country-year level technology diversity measure, we used the USPTO granted patent data.

4. We added the number of patents as an additional control variable to implement a ‘horse-race’ regression between technology diversity and technology quantity. Technology diversity still explains many US import price variations, even after controlling for the number of patents.

5. We do not report figures for recent years after 2014, as they are not free from the censoring issue caused by the patent review process. For international applications, it takes three years of review process on average from the application to the final publication.

6. Studies have found that China’s accession to the WTO had significant effects on Taiwan (e.g. Chen 2003).

7. For a very recent literature review, see Shu and Steinwender (2019).

8. Regarding modeling technology spillover between products in the international trade literature, see Flach and Irlacher (2018), for example.

9. $z_i$ is equivalent to the variety-specific taste shifter in Foster et al. (2008).

10. If $\mathit{\gamma }=0$, all products are homogeneous and the consumer cares only about his aggregate consumption.

11. See Appendix A1 for the detailed steps for deriving demand equation (2).

12. For proofs of all propositions in the paper, see Appendix A2.

13. As $\mathit{c}\left(x_n,0\right)=x_n$, $x_n=\mathit{A}+\mathit{\gamma z}\left(0\right)-\sqrt{\frac{4\mathit{\gamma h}\left(0\right)}{L}}$.

14. $\frac{\mathit{d}\mathrm{\Pi }\left(x\right)}{\mathit{dx}}<0$, $\frac{\mathit{d}\mathrm{\Pi }\left(x\right)}{\mathit{dA}}>0$, $\frac{\mathit{d}{\mathrm{\Pi }}^e\left(\mathit{A}\right)}{\mathit{dA}}>0$, $\frac{\mathit{d}{\mathrm{\Pi }}^e\left(\mathit{A}\right)}{\mathit{dL}}>0$.

15. Note that the weakness of patent data is the selectivity of patenting. While patenting may help to bring new products or processes to the market, it is not a requisite for it, nor is it sufficient to be successful in innovating. Moreover, patents may be applied for strategic reasons to create entry barriers, to be able to cross-license, or as signals of capability in order to attract outside funding (Mohnen 2019). We shall limit ourselves to the use of patents as indicators of innovation.

16. https://bulkdata.uspto.gov/

17. More specifically, the ‘beautifulsoup’ package was used to scrap relevant information from the raw data.

18. https://sites.google.com/site/patentdataproject/Home/downloads

19. http://www.patentsview.org/download/

20. See Table A1 for an example of the firm name cleaning process. See Table A2 for the number of patents and assignees.

21. TEJ data include supplemental information on naming history up to the last three old names.

22. In the TEJ dataset, while the number of firms varies by year, the maximum number of annual observations is 2,750. We successfully matched 1,127 firms to the US patent data. For estimation purposes, firms with incomplete data for all relevant variables were eliminated, leaving a total number of 257 firms for empirical analysis. Since we examine not only the quantity and quality of innovative technologies but also the scope and distribution of these technologies within each firm, the sample was restricted to firms with at least one patent application.

23. Although the usefulness of forward citations as a measure of patent value has been questioned and criticized, recent works of Kogan et al. (2017) and Moser, Ohmstedt, and Rhode (2018) still confirmed the usefulness of citation-based measures.

24. On the relationship between patent value and self-citation measure, see Hall, Jaffe, and Trajtenberg (2005).

25. The time notation $\mathit{t}$ corresponds to the year of patent application by firm $\mathit{i}$. Firms in foreign countries have two options for applying for patents in the US: a direct application and an international application. In the first option, the firm applies to the USPTO directly. When pursuing an international application, the firm files a domestic application first and then starts the USPTO application through the patent office in the host country. For the second option of the international application, an applicant can claim the date of domestic application as the priority date of the patent in the US, and this date is called the foreign priority date. If a foreign priority date is available, it cannot exceed the date of the US application. The foreign priority date should be closer to the timing of the firm’s innovative activity for the patent than the date of the USPTO application. We use the year of the foreign priority date as $\mathit{t}$ if available. Otherwise, the year of the US application is $\mathit{t}$.

26. Using a Fisher-type unit-root test, we find evidence against the null hypothesis of a unit root and therefore conclude that the dependent variables are stationary.

27. The selection of the sample period depends on data availability.

28. An alternative measure of demand shock is $\mathit{Export}\_\mathit{Deman}{d}_{\mathit{it}}=\log \left(\sum _{\mathit{j},\mathit{p}}\frac{{\mathit{X}}_{\mathit{ijpt}0}}{X_{\mathit{ijt}0}}{M}_{\mathit{jpt}}\right)$. This measure gives similar results.

29. This addresses the concern that the exit of low productivity firms may drive the results. Although the estimates are less significant due to smaller sample sizes, the main findings, which are available upon request, are largely unchanged.

Additional information

Funding

The work was supported by the Ministry of Education [108L900201]; Ministry of Science and Technology, Taiwan [MOST 108-3017-F-002-003, MOST 108-2410-H-002-014-MY2].