IFPRI Discussion Paper 01385 November 2014 The Patterns of Patents in China Zhuan Xie Xiaobo Zhang Development Strategy and Governance Division INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE The International Food Policy Research Institute (IFPRI), established in 1975, provides evidence-based policy solutions to sustainably end hunger and malnutrition and reduce poverty. The institute conducts research, communicates results, optimizes partnerships, and builds capacity to ensure sustainable food production, promote healthy food systems, improve markets and trade, transform agriculture, build resilience, and strengthen institutions and governance. Gender is considered in all of the institute’s work. IFPRI collaborates with partners around the world, including development implementers, public institutions, the private sector, and farmers’ organizations, to ensure that local, national, regional, and global food policies are based on evidence. IFPRI is a member of the CGIAR Consortium. AUTHORS Zhuan Xie is a PhD candidate at the National School of Development, Peking University. Xiaobo Zhang (x.zhang@cgiar.org) is a senior research fellow in the Development Strategy and Governance Division of the International Food Policy Research Institute (IFPRI), Washington, DC and a distinguished professor of economics in the National School of Development, Peking University. Notices 1. IFPRI Discussion Papers contain preliminary material and research results and are circulated in order to stimulate discussion and critical comment. They have not been subject to a formal external review via IFPRI’s Publications Review Committee. Any opinions stated herein are those of the author(s) and are not necessarily representative of or endorsed by the International Food Policy Research Institute. 2. The boundaries and names shown and the designations used on the map(s) herein do not imply official endorsement or acceptance by the International Food Policy Research Institute (IFPRI) or its partners and contributors. Copyright 2014 International Food Policy Research Institute. All rights reserved. Sections of this material may be reproduced for personal and not-for-profit use without the express written permission of but with acknowledgment to IFPRI. To reproduce the material contained herein for profit or commercial use requires express written permission. To obtain permission, contact the Communications Division at ifpri-copyright@cgiar.org. Contents Abstract v Acknowledgments vi 1. Introduction 1 2. Stylized Facts about Patents in China 3 3. Match Patent Database with Firm Database 11 4. Some Interesting Patterns Derived from the Merged Database 14 5. Conclusion 20 Appendix: Supplementary Table 21 References 22 iii Tables 2.1 Number of patent applications (1985–2012) 4 2.2 Number of granted patents (1985–2009) 5 2.3 Approval rates for invention, utility model, and design patents (1985–2009) 8 2.4 Duration between the time of application and the time of approval (months) 8 2.5 Number of granted patents, by province (1985–2009) 9 3.1 Distribution of granted patents among firms 11 3.2 Degree of matching between patent data and annual survey of industrial enterprises in China data 12 3.3 Degree of matching between patent data and China Economic Census 2008 12 4.1 Distribution of granted patents, by firm type and year 14 A.1 Gender intensity, by industry 21 Figures 2.1 Ratio of patents granted to domestic applicants for three samples over time (1990–2009) 7 2.2 The composition of patents granted to domestic applicants: Invention, utility model, and design (1990–2009) 7 2.3 Spatial distribution of granted patents, by province 10 2.4 Patents granted to domestic applicants, by industry (1985–2009) 10 4.1 Number of patents per million yuan of value-added, by register ownership type 15 4.2 Number of patents granted per million yuan of value-added, by firm export status 16 4.3 Number of patents granted per million yuan of value-added, by degree of clustering 17 4.4 Number of patents granted per million yuan of value-added, by firm capital intensity 18 4.5 Number of patents granted per million yuan of value-added, by industry gender intensity 19 iv ABSTRACT Innovations are a key driver of long-term economic growth. There has been an explosion of patent filings in China in the past three decades. But empirical studies on the pattern of innovations at the firm level are rather scant primarily due to lack of firm-specific patent data. We have made concerted efforts to match Chinese patent data with a large firm-level database. The matched dataset enables us to examine the patterns of patents at the firm level. Our analysis has revealed several interesting patterns: (1) domestic firms have become increasingly more innovative in terms of patent application; (2) private firms, rather than state-owned enterprises, have been the engine of innovation; (3) rising wages have propelled labor- intensive sectors to become more innovative; and (4) in response to increasing sex ratio imbalances, firms in female-intensive industries have exhibited more innovations than those in male-intensive industries. Keywords: patent, innovation, JEL classification: O12, O31, O53 v ACKNOWLEDGMENTS Financial support from China Natural Science Foundation (grant numbers 71350002 and 7144000034) is gratefully acknowledged. vi 1. INTRODUCTION China has achieved phenomenal economic progress with an average annual GDP growth rate near 10 percent since 1978. During this period, the number of patent filings in China has grown even faster. In 1990, there were only 41,468 patent applications. But by 2012 this number had spiked almost 50-fold to 2,050,649. The number of invention patents, arguably the most important one among three types of patents, surpassed the number in the United States in 2011 (WIPO, 2012). China has become the largest recipient of patent filing globally. Despite the record growth, little is known about the evolutionary patterns and structures of Chinese patents, in particular at the firm level, largely due to lack of firm-level patent data. Innovation is a key driver of long-run economic growth (Schumpeter 1934; Aghion and Howitt 1992). It is particularly important for middle-income countries like China. In the past several decades, China has adopted an open-door policy to attract foreign investment and technologies. Chinese firms have quickly absorbed, imitated, and modified the existing foreign technologies, catching up with the global technologies frontiers. The massive inflow of foreign capital and technology has also brought about hundreds of millions of jobs. However, in the past decade, the seemingly unlimited supply of labor has been exhausted, resulting in escalation of real wages (Zhang, Yang, and Wang 2011; Huang and Jiang 2010). Facing rising wages, Chinese firms have to invest more in technologies to substitute the increasingly more expensive workers. As China becomes a middle-income country and narrows its gap with technological frontiers in developed countries, simply grabbing existing technologies on the shelf of developed countries is no longer a viable option. Instead it has to count more on indigenous innovations to upgrade its industries (Feng and Yao 2014; Lin 2012; Zhuang, Vandenberg, and Huang 2012). Patents are an integral part of modern innovation systems. The patent system is one of the most important institutions to protect intellectual property rights and encourage innovation in a modern economy. It is expected that innovations in the form of patents will play a greater role in propelling Chinese economic growth in the future as China enters an era of rising wages. To better foretell future patent growth, it is critical to understand the historical patterns of patent development. However, due to data constraints, studies on the historical patterns of Chinese patents, particularly at the firm level, are rather scarce. The economic literature on the use of intellectual property in China is much more limited than the legal literature. A few theoretical articles study the optimal patent protection in China (Kou and Zhou 2011; Xu 2010; Dong and Wang 2007; Kou 2004). Most empirical studies use secondary data, such as aggregate data at the provincial or industry level (Hu and Jefferson 2009; Cheung and Lin 2004) or data of publically listed companies (Li, Hong, and Wu 2012; Choi, Lee, and Williams 2011; Lin, Lin, and Song 2010). A few studies based on disaggregated data are limited to only a particular region (Qin, Zhou, and Yin 2012; Dobson and Safarian 2008; Zhu and Xu 2003). A few studies examining the knowledge spillover effect of foreign direct investment have used individual patent data (Wu and Mathews 2012; Awokuse and Yin 2010; Cheung and Lin 2004). None of the above studies, however, has examined patents at the firm level. Some basic questions regarding Chinese patents remain to be answered. To name a few, how important is the role of foreign firms in driving the trend of Chinese patents? Are the state-owned enterprises (SOEs) or privates ones more innovative? In which places or industries are firms more productive in terms of patent applications? Have firms in labor-intensive industries produced more patents than those in capital-intensive industries in response to rising real wages? To answer these questions, we made concerted efforts to match Chinese patent data with the annual survey of industrial enterprises in China (ASIEC). The database covers all the SOEs and above- scale private firms with sales exceeding 5 million yuan from 1998 to 2009. The patent database contains all patents granted between 1985 and 2009. The merged firm patent database enables us to study the historical patterns of China’s patents. 1 This paper also contributes to the general literature on patents in developing countries. Although there is a large strand of economic literature studying patents and innovation in developed countries, the research on developing countries is much more scant (Gonzalez et al., 2013). There are at least two explanations for the disparity. First, innovation is a capital-intensive activity. In the early stage of development, due to lack of capital resources, it makes more economic sense for developing countries to import and imitate some of the outdated technologies that have fallen into disuse in developed countries. Consequently, promoting the development of patents is not high on the policy agenda in developing countries. Second, lack of high-quality patent and firm data is pandemic in developing countries, hindering researchers from conducting empirical investigations on patents in developing countries. This paper is structured as follows. Section 2 introduces China’s patent system and presents some basic stylized facts about China’s patents. Section 3 discusses the methods to match the two databases. Some facts derived from the merged database are listed in section 4. A concluding section summarizes the main findings. 2 2. STYLIZED FACTS ABOUT PATENTS IN CHINA Brief Introduction of the Patent System in China Intellectual property protections involve at least three aspects: patents, trademarks, and copyrights. The intellectual property system in China is administered by several institutions, with the patent system administered by the State Intellectual Property Office of the People’s Republic of China (hereinafter referred as SIPO), trademarks by the Trademark Office of the State Administration for Industry and Commerce, and copyrights by the National Copyright Administration. China established the Patent Office, which later became SIPO, in 1980. The Patent Law was passed on March 12, 1984, and put into effect on April 1, 1985. SIPO began to accept patent filings in 1985. Compared to developed countries’, China’s patent history is rather short. Despite the late start, China joined the Paris Convention, an international treaty with respect to intellectual property, in 1985. The Patent Law has been amended three times—September 4, 1992; August 25, 2000; and December 27, 2008—since its implementation. The Patent Law set clear rules for patent applications, examinations, approvals, protection scopes, and patent prosecution. SIPO is the sole authority to process patent applications in China. In addition to the Patent Law, the state passed the Implementing Regulations of the Patent Law of the People’s Republic of China (hereinafter referred as Implementing Regulations). According to the regulation, patents in China are classified into three types: invention, utility model, and design. Invention patents encompass new technical solutions relating to a product, a process, or improvement; utility model patents represent new technical solutions relating to the shape, the structure, or their combination, of a product, which are mainly for practical use; design patents cover new designs in relation to shapes, patterns, colors, or their combination, of a product. They create an aesthetic feeling and are primarily for industrial applications. The duration of protection for invention patents is 20 years, while the duration for utility model and design patents is 10 years, effective from the application date. China Patent Database We purchased all the records of patents approved as of May 1, 2014, from SIPO. Our patent database includes all the patents approved as of May 1, 2014, and applied as of December 30, 2009. The database contains 4,060,392 observations, including 1,097,000 invention patents, 1,620,069 utility model patents, and 1,343,323 design patents. A typical entry of patent includes the following information: applying number, patent name, applicant, inventor, applying date, publishing date, granting date, main International Patent Classification (IPC) number, filing agent’s name and institution, applicant address, patent origin (provinces in China or other countries), and a short description of the patent. Some variables need more explanation and clarification. Applicant is different from inventor in that the former owns the patent while the latter invents the patent. Applicants can be individuals, firms, or institutions, whereas inventors must be individuals. Applying date is the date of filing patent application. Only an invention patent has a publishing date because an invention patent has to be published within 18 months of its submission. Afterward, it has to undergo substantial examinations. The patents of utility model and design are directly granted and published after preliminary examination. Generally speaking, invention patents embody more technological components than utility model and design patents. According to IPC, all patents could be classified into eight major categories: human necessities; performing operations and transporting; chemistry and metallurgy; textiles and paper; fixed constructions; mechanical engineering, lighting, heating, weapons, and blasting; physics; and electricity. There are subcategories in each major category. One patent can belong to more than one category. The first category is the most relevant and is called main IPC number. 3 Applying patents is a tedious process involving heavy professional work. So some patent applicants hire agents to apply on behalf of them rather than file by themselves. For those entries, the information about filing agent is also displayed. The patent data reveal information about origins of the applicants, that is, province codes for domestic applicants and country codes for foreign applicants. Several General Trends of Chinese Patents In this subsection we present some trends in Chinese patents. It takes time to process patent applications. A patent can be granted one year, two years, or even longer after its submission. We use two yardsticks to illustrate the trend of Chinese patents—the number of patent applications per year and the number of granted patents according to the year of submission. Let’s use an example to illustrate the concepts of the three variables. Suppose 1,000 patent applications in total are submitted in year T. The number of patent applications in year T is 1,000. From the 1,000 patent applications, 50 patents are granted in year T, 200 patents are granted in year T + 1, 320 patents are granted in year T + 2, and 430 applications are rejected. The total number of granted patents corresponding to the application year T is 570.1 This represents the number of eventually granted applications submitted in year T. We define the patent approval rate as the ratio of the number of granted patents and the application number corresponding to the same year. Table 2.1 presents the overall trend of patents applied. As shown in Table 2.1, the application number in China increased rapidly between 1985 and 2012. The annual growth rate of patent application for the whole period averaged at 20 percent, about twice the annual GDP growth rate in the same period. Notably, the annual growth rate accelerated to 28 percent in the last period of 2005–2012. Initially design patents accounted for only 4 percent in 1985. By 2012, the share of design patents had reached 32 percent of total patent filings, reflecting explosive growth. Table 2.1 Number of patent applications (1985–2012) Year Total Invention (%) Utility model (%) Design (%) Foreign (%) 1985 14,372 60 36 4 35 1986 18,509 43 52 4 26 1987 26,077 31 64 5 17 1988 34,011 28 66 6 16 1989 32,905 29 63 8 17 1990 41,469 24 67 9 12 1991 50,040 23 67 11 9 1992 67,135 21 66 12 8 1993 77,276 25 61 13 12 1994 77,735 25 59 17 13 1995 83,045 26 53 21 17 1996 102,735 28 48 24 20 1997 114,208 29 44 27 21 1998 121,989 29 42 28 21 1999 134,239 27 43 30 18 2000 170,682 30 40 29 18 2001 203,573 31 39 30 19 2002 252,631 32 37 31 19 1 570 = 50 + 200 + 320. 4 Table 2.2 Continued Year Total Invention (%) Utility model (%) Design (%) Foreign (%) 2003 308,487 34 35 30 19 2004 353,807 37 32 31 21 2005 476,264 36 29 34 20 2006 573,178 37 28 35 18 2007 693,917 35 26 39 15 2008 828,328 35 27 38 13 2009 976,686 32 32 36 10 2010 1,222,286 32 34 34 9 2011 1,633,347 32 36 32 8 2012 2,050,649 32 36 32 7 Annual growth rate in different periods (%) 1985–1989 23 3 42 41 3 1990–1994 17 17 13 37 19 1995–1999 13 14 7 23 14 2000–2004 20 26 13 22 25 2005–2012 28 25 32 26 6 1985–2012 20 17 20 29 13 Source: Tabulated by authors based on aggregate data downloaded from the State Intellectual Property Office’s (SIPO’s) webpage: http://www.sipo.gov.cn/tjxx/. Notes: The aggregate data do not report the share of applications by firms. Because we did not have access to information about patent applications that were rejected by SIPO, we could not compute the share of applications by firm in Table 2.1 by ourselves, as we do in Table 2.2. Table 2.2 lists the number of granted patents from 1985 to 2009. The overall trend mirrors that in Table 2.1. During the whole period, the number of approved patents grew at an annual rate of 19 percent. Design patents saw the most phenomenal rate of growth (28 percent per year) among the three types of patents. The share of patents granted to foreign applicants dropped from 34 percent in 1985 to 8 percent in 2009. Firms have become a major body of innovation. The proportion of patents granted to firms tripled from 16 percent in 1985 to 49 percent in 2009. As to invention patents, patents applied for by firms have exceeded half since 2005. Table 2.3 Number of granted patents (1985–2009) Year Total Invention (%) Utility model (%) Design (%) Foreign (%) Firm (%) 1985 9,567 48 46 6 34 16 1986 13,329 30 64 6 23 18 1987 19,945 20 74 6 15 19 1988 24,739 18 75 7 13 18 1989 24,804 17 73 10 13 20 1990 30,329 14 75 11 10 23 1991 37,296 12 75 13 7 23 1992 47,662 11 74 15 7 22 1993 51,640 15 70 15 11 22 1994 55,871 21 61 18 18 24 5 http://www.sipo.gov.cn/tjxx/ Table 2.4 Continued Year Total Invention (%) Utility model (%) Design (%) Foreign (%) Firm (%) 1995 62,208 24 53 23 22 28 1996 72,509 25 49 26 24 30 1997 81,886 27 44 29 25 33 1998 91,315 27 42 31 24 34 1999 108,576 25 42 33 22 35 2000 129,030 26 42 32 22 36 2001 153,588 28 40 32 22 35 2002 202,030 29 37 34 21 38 2003 234,839 33 36 31 24 37 2004 267,372 34 33 33 24 37 2005 323,120 33 33 34 22 37 2006 385,688 31 34 36 19 39 2007 451,292 28 33 38 16 41 2008 532,492 26 36 37 13 46 2009 649,263 22 42 37 8 49 Annual growth rate in different periods (%) 1985–1989 27 –2 42 41 0 42 1990–1994 17 29 11 31 37 15 1995–1999 15 16 9 25 15 22 2000–2004 20 28 14 21 24 20 2005–2009 19 7 26 21 –1 34 1985–2009 19 15 19 28 12 27 Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1, 2014, and applied for as of December 30, 2009. Note: The last column is determined by the applicant variable on granted patent applications. Figure 2.1 plots the time trend of the share of patents awarded to domestic applicants in three samples. The top line represents the share of patents granted to domestic applicants in the whole sample. Apparently, the ratio has been rather stable over time. The solid line stands for the share of invention patents awarded to domestic applicants. It first declined from 1990 to the mid-1990s before taking off. By 2009, it reached nearly 80 percent. The bottom line indicates the share of patents granted to domestic firms relative to total firms. It reached the nadir at less than 10 percent in the mid-1990s and spiked to more than 60 percent by 2009. This figure indicates a dramatic improvement in domestic innovation capacity over time. 6 Figure 2.1 Ratio of patents granted to domestic applicants for three samples over time (1990–2009) Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1 and applied as of December 30, 2009. Note: The top curve is the ratio of patents awarded to domestic firms in total patents granted; the middle curve stands for the ratio of patents awarded to domestic applicants in total invention patents; the bottom curve is the ratio of patents granted to domestic firms relative to total approved patents filed by firms. Figure 2.2 presents the composition of patents granted to domestic applicants. We could clearly see the ratio of utility model patents has declined since 1990, while the share of invention and design patents has increased, indicating a shift in the structure of domestic patents. Figure 2.2 The composition of patents granted to domestic applicants: Invention, utility model, and design (1990–2009) Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1 and applied as of December 30, 2009. 7 SIPO examines all the patent filings and makes grant decisions. Table 2.3 presents SIPO’s approval rates according to patent type and origins of application by year. Table 2.5 Approval rates for invention, utility model, and design patents (1985–2009) Grant rate Total Invention Utility model Design Domestic applicants 0.704 0.398 0.786 0.783 Foreign applicants 0.790 0.782 0.843 0.862 Total 0.722 0.585 0.786 0.785 Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1, 2014, and applied for as of December 30, 2009. A few points about Table 2.3 are worth mentioning. First, foreign applicants enjoy a higher approval rate than do domestic applicants. The approval rate among domestic applicants was 70.4 percent, while it was as high as 79.0 percent for foreign applicants. With respect to invention patents, the gap is more striking. The approval rate for domestic invention applications was only 39.8 percent, about half that for foreign invention applications. Assuming that SIPO follows the same criteria for examining domestic and foreign patent filings, we can infer that the foreign patent filings contain more technological intensity than do domestic filings. It takes time to process patent applications. We calculated the number of months from patent application to approval by patent type for domestic and foreign applicants. To gauge the time trend, we divided the sample into five periods and computed average time lags in each period. Table 2.4 presents the average processing time. Two salient features stand out. First, the time lags for invention patents, ranging between three and six years, were much longer than those for utility model and design patents, which averaged about one year. Second, it takes longer for foreign applications than domestic applications to be approved. The difference is mainly with invention patents. On average it took more than five years for a foreign applicant to secure a patent, 16 months longer than domestic applicants. Despite the long process, the foreign applicants were subject to a higher approval rate. Table 2.6 Duration between the time of application and the time of approval (months) Domestic applicants Foreign applicants Year Invention Utility model Design Invention Utility model Design 1985–1989 49 12 17 59 12 17 1990–1994 54 11 14 68 12 14 1995–1999 52 16 12 74 18 13 2000–2004 42 13 8 60 16 10 2005–2009 37 12 12 53 16 13 Total 47 13 12 63 15 14 Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1, 2014, and applied for as of December 30, 2009. There are large regional variations in terms of innovation. Table 2.5 presents the total number of granted patents by province. Of the top 10 provinces, 8 in the table are coastal provinces, including Guangdong, Zhejiang, and Shanghai. However, the numbers in the table neglect the difference in population size. To remedy this concern, we used population to normalize the number of patents, and the spatial distribution of patents per million people is shown in Figure 2.3. Even on a per capita basis, the coastal region still performed better than inland regions. 8 Table 2.7 Number of granted patents, by province (1985–2009) Province Total Firms (%) Invention (%) Utility model (%) Design (%) Guangdong 578,555 45 14 32 54 Zhejiang 401,706 27 9 38 54 Jiangsu 372,255 49 11 38 51 Shanghai 231,676 64 20 39 40 Shandong 226,427 32 10 65 24 Beijing 213,541 40 38 48 14 Taiwan 189,873 55 24 64 13 Liaoning 123,831 22 15 72 13 Sichuan 119,541 29 12 44 45 Fujian 85,868 34 8 41 50 Henan 80,536 33 12 65 23 Hubei 77,949 36 18 62 20 Hunan 76,797 29 15 63 22 Hebei 68,814 27 11 68 21 Tianjin 62,239 47 22 58 20 Heilongjiang 55,146 18 16 71 13 Chongqing 50,853 52 12 47 41 Shanxi 50,412 30 23 61 16 Anhui 47,657 42 14 56 30 Jilin 37,272 21 18 66 16 Yunnan 27,169 28 18 50 32 Shanxi 26,801 27 20 64 16 Guangxi 26,507 30 11 61 27 Jiangxi 26,255 23 12 61 28 Guizhou 18,011 37 18 59 23 Xinjiang 16,774 20 10 70 20 Neimenggu 15,983 26 12 61 26 Gansu 12,843 27 21 64 15 Ningxia 5,985 33 10 51 38 Hainan 5,067 33 17 37 45 Qinghai 2,602 29 16 54 30 Tibet 856 20 13 25 62 Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1, 2014, and applied for as of December 30, 2009. 9 Figure 2.3 Spatial distribution of granted patents, by province Source: Patent data are calculated based on national database (1985–2009) of patents that were granted as of May 1 and applied as of December 30, 2009. Population data are downloaded from the website of the National Bureau of Statistics. Note: Hong Kong, Macao, Xisha, Nansha, and Zhongsha Islands are not shown on the map due to lack of data. Apart from spatial variation, there are large differences across industries in innovation. Figure 2.4 displays the number of patents in each of the eight IPC categories. The operations and transporting industries commanded the largest number of patents, whereas the textile and paper industries ranked at the bottom. Figure 2.4 Patents granted to domestic applicants, by industry (1985–2009) Source: Calculated by authors based on national database (1985–2009) of patents that were granted as of May 1 and applied as of December 30, 2009. Note: The industry is classified according to the International Patent Classification. 10 3. MATCH PATENT DATABASE WITH FIRM DATABASE With rising wages, China’s comparative advantages as a manufacturing powerhouse are eroding. This is especially true in the labor-intensive light manufacturing industry. Therefore, China has to speed up the process of industrial upgrading and reduce its reliance on labor-intensive industries. To gauge the process of industrial upgrading, we need innovation information at the firm level. To achieve this purpose, we managed to match the patent database with the ASIEC database from 1998 to 2009. Patents can be applied for by individuals, firms, or other institutions. Those patents applied by firm record only firm names rather than a unique firm identification code used in the ASIEC database. So we had to use firm names as a bridge to match the two databases. Before matching, we first deleted punctuation, which may be due to input errors. Second, we deleted some Chinese words, including “province,” “city,” “county,” “limited liability company,” “limited company,” “company,” and “factory,” from the firm names. This step eliminates less important information and increases matching precision. Take “Shenzhen Municipal Huawei Technology Limited Company” as an example. This is the official company name. However, in some entries of the patent database and ASIEC database, the company’s name is also listed as “Shenzhen Huawei Technology Limited Company.” After deleting the nonessential words, the firm name changes to “Shenzhen Huawei Technology.” The identical name now can be easily matched in the two databases. Our method increases the matching rate by about 2 to 3 percent. In the matched database, 47,298 firms had at least 1 patent. Table 3.1 presents the distribution of the 47,298 firms according to the number of patents owned. Only about 7 percent of firms in the above- scale firm database possessed at least 1 patent. Among those with patents, 27 percent of them owned only 1 patent, and 16 percent 2 patents. In total, nearly two-thirds of firms owned 1 to 5 patents. There are a few outliers—737 firms (1.6 percent of the matched firms or 0.12 percent of the total number of firms included in the above-scale firm database) held more than 100 patents. Table 3.1 Distribution of granted patents among firms Number of patents per firm Number of firms Percentage 0 635,516 93.0 1 or more 47,298 7.0 Among those with patents 1 12,653 26.8 2 7,474 15.8 3 4,386 9.3 4 3,248 6.9 5 2,581 5.5 1–5 30,342 64.2 6–10 7,760 16.4 11–20 4,490 9.5 21–30 1,633 3.5 31–100 2,336 4.9 100 or more 737 1.6 Total 682,814 100 Source: Calculated by authors based on the merged firm patent database (1998–2009) between the national patent database and annual survey of industrial enterprises in China database. 11 Table 3.2 presents the matching rate by patent type. Since the matched database covers only firms registered in China, foreign firms are naturally excluded. To examine the degree of matching, it is better to use the subsample of domestic firms instead of the total firm sample in the patent database as a benchmark. We define the matching rate as the ratio of matched sample size to total number of patents owned by domestic firms. As shown in Table 3.2, the overall matching rate is nearly 60 percent. There are some slight differences across the three types of patents. The matching rate for invention patents is 55.8 percent, while it is 62.4 percent for design patents. We next provide more evidence to show that the above matching rates are actually high. Table 3.2 Degree of matching between patent data and annual survey of industrial enterprises in China data Type of patent Patents applied for by domestic firms Share of matched patents in total domestic firm application (%) Invention 265,713 55.8 Utility model 557,785 57.0 Design 500,077 62.4 Total 1,323,575 58.8 Source: Calculated by authors based on merged firm patent dataset and national database (1985–2009). Note: The third column stands for the share of patents granted to the matched firms in total patents applied for (and finally granted) by domestic firms. The sample period is 1998–2009. Matching errors can stem from several sources. First, our sample includes only SOEs and above- scale manufacturing firms. Many small and medium enterprises (SMEs) and service firms are excluded. Some of them might have applied for patents but are not included in our sample. To check the degree of bias as a result of omitting SMEs, we made use of China Economic Census 2008, which encompasses not only above-scale firms but also SMEs. We create three subsamples out of the census. The first subsample comprises the above-scale manufacturing enterprises in 2008. The second subsample covers below-scale manufacturing firms in 2008. The third subsample includes firms in the service sector. We matched three subsamples in 2008 with our patent database separately and reported the matching rates in Table 3.3. The matching rate for the above-scale firms, SMEs, and service firms are 45.8 percent, 8.8 percent, and 15.8 percent, respectively. Overall, 70.4 percent of the patents applied for by domestic firms can be successfully matched to the firm sample of China Economic Census 2008. Among the three types of patents, the matching rate for invention patents ranks at the top, as high as 79.3 percent, whereas two- thirds of design patents can be matched. Although the number of below-scale firms is larger than that of above-scale firms, they account for a smaller share of domestic firm applications at 8.8 percent. Service firms contribute 15.8 percent of total applications by domestic firms. Table 3.3 Degree of matching between patent data and China Economic Census 2008 Type of patent Above-scale firms (%) (1) Below-scale firms (%) (2) Service firms (%) (3) Overall (%) (1) + (2) + (3) Invention 48.0 5.6 25.7 79.3 Utility model 44.7 9.5 15.3 69.5 Design 45.8 9.8 11.0 66.6 Total 45.8 8.8 15.8 70.4 Source: The patent data are from the national database (1985–2009) of patents that were granted as of May 1 and applied for as of December 30, 2009, and the firm data are from China Economic Census 2008. Note: The above-scale firms in China Economic Census 2008 follow the same definition as the annual survey of industrial enterprises in China database of the National Bureau of Statistics. The last column stands for the share of granted patents in the matched firm sample in total patents awarded to domestic firms. 12 Second, some firms in the patent database may no longer exist. The matching rate of above-scale firms in 2008 (45.8 percent) is lower than the rate during the longer period from 1998 to 2009 (55.8 percent), as shown in Table 3.2. The result is not surprising. Not all the firms in the above-scale firm database appear in all the years. Some matched firms during the period from 1998 to 2007 may have failed and may have not shown up in the economic census in 2008. Using the ASIEC firm database for the period from 1998 to 2009 can improve the matching rate by 10 percentage points rather than relying on the economic census in 2008. If assuming the same sample attrition rate for the second and third subsamples, we can infer that their matching rates would increase to 11.3 percent and 20.2 percent, respectively, if we could extend the two subsamples to the longer period from 1998 to 2009.2 The overall matching rate would then improve to 90.3 percent (58.8 percent + 11.3 percent + 20.2 percent). In other words, we failed to match less than 10 percent of the patents filed by domestic above-scale manufacturing firms. Given that firms in other sectors, such as agriculture and construction, are not included in the ASIEC database, it is impossible to match all the firms from the two data sources. In sum, our algorithm has done a great job matching the two databases. 2 11.3 = 8.8 * (1 + 28 percent); 20.2 = 15.8 * (1 + 28 percent). 13 4. SOME INTERESTING PATTERNS DERIVED FROM THE MERGED DATABASE Having matched the two datasets, now we are in a position to uncover the patterns of patents along several dimensions that are, origins, ownership, and exporting status, degree of clustering, labor intensity, and gender intensity. Table 4.1 lists the total number of matched patents and the share of different types of firms (private firms, foreign firms, exporting firms, labor-intensive firms, high-clustering industries, and female-intensive firms) over time. However, the statistics in Table 4.1 do not take firm size into account. Firm sizes can differ by employment, capital, and output. Of course, we can normalize the number of patents in different ways. We prefer to normalize the number of patents by million yuan of value added because it better captures the innovation intensity embedded in economic outputs rather than inputs. The results using employment or capital as a denominator reveal similar patterns. So we do not report them here. Figures 4.1 through 4.5 present the normalized number of patents based on value added by exporting status, degree of clustering, capital intensity, and gender intensity, respectively. Table 4.1 Distribution of granted patents, by firm type and year (1) (2) (3) (4) (5) (6) (7) Year Total Private firm (%) Foreign firm (%) Exporting firm (%) High clustering (%) Labor intensive (%) Female- intensive firm (%) 1998 11,714 49 35 53 68 37 63 1999 16,411 50 37 56 66 36 63 2000 19,809 55 34 54 68 50 61 2001 23,579 53 39 59 71 46 60 2002 34,328 54 38 60 74 53 64 2003 40,161 59 35 63 72 56 60 2004 46,867 58 37 70 72 59 60 2005 59,955 60 35 64 74 80 60 2006 80,314 62 33 66 74 82 60 2007 103,167 62 33 63 69 82 61 2008 134,911 63 32 61 68 82 59 2009 178,475 64 31 56 68 83 56 Source: Calculated by authors based on merged firm patent dataset. Note: Column 1 is the share of patents owned by firms registered as privately or collectively owned firms. Column 2 is the share of patents owned by firms registered as foreign or by Hong Kong-, Macao-, or Taiwan-owned firms. Column 3 is the share of patents owned by firms whose state ownership is smaller than 50 percent. Column 4 is the share of patents owned by exporting firms. Column 5 is the share of patents owned by firms in high-clustering regions. See Long and Zhang (2011) for the definition of clusters. Column 6 is the share of patents owned by labor-intensive firms. For each year we define a firm as labor intensive if the capital-labor ratio is less than the median value. Column 7 is the share of patents owned by female-intensive firms (see Table A.1 in the appendix for details. Several interesting patterns emerge. First, we divided firms into three types according to register ownership type—SOEs, foreign firms (including those owned by Hong Kong, Macao, and Taiwan), and privately owned domestic firms. As shown in column 2 of Table 4.1, the share of patents granted to private firms has increased over time, and the share of patents granted to foreign firms has decreased slightly over time as indicated in column 3 of Table 4.1. The result, in number of patents per million yuan of value added, is presented in Figure 4.1. The innovation intensity embedded in state firms decreased during the period, while the innovation intensity of domestic private firms was higher than that of SOEs. Apparently, it is the domestic private firms rather than SOEs that are the engines of innovation in China. 14 Figure 4.1 Number of patents per million yuan of value-added, by register ownership type Source: Tabulated by authors based on merged firm patent dataset. Second, exporting firms are more innovative than non-exporting firms. As shown in column 4 of Table 4.1, exporting firms have consistently contributed to more than half of total firm patents. Trade economists found that exporting firms tend to be more productive than non-exporting firms (Becker and Egger 2013; Bustos 2011; Aw, Roberts, and Xu 2011). One hypothesis is that more productive firms have a higher propensity to export than less productive ones. This is called “selection effect” in the literature. Another hypothesis is that exporting helps improve firms’ productivity. It is beyond the scope of this paper to test which hypothesis is more valid. At the aggregate level, we do find that overall exporting firms are more innovative. This has been particularly true since China joined the World Trade Organization (WTO) at the end of 2001. As illustrated in Figure 4.2, the innovation intensity differed little between the exporting and non-exporting firms prior to 2002, but the gap widened significantly after 2002. The divergence may be explained by trade liberalization after China joined WTO (Yu and Ye 2013; Bustos 2011). 15 Figure 4.2 Number of patents granted per million yuan of value-added, by firm export status Source: Tabulated by authors based on merged firm patent dataset. Note: A firm’s exporting status is defined as 1 if it reports export volume and 0 otherwise. Third, firms in clusters are more innovative. China’s industrialization is largely clustered based (Long and Zhang 2011, 2012). The classical literature on clustering has highlighted several key positive externalities of industrial clusters: better access to the market and suppliers, labor pooling, and easy flow of technology know-how (Marshall [1890] 1920). Long and Zhang (2011) found that with less reliance on external financing, more small firms emerge within clusters, leading to higher total factor productivity. We classified regions into high and low clustering regions according to the median value of the cluster measure in Long and Zhang (2011). As shown in column 7 of Table 3.3, the high clustering region explains more than two-thirds of total firm patents. When comparing the number of patents per million yuan of value added, the same pattern remains, as indicated in Figure 4.3. The firms in clusters were consistently more innovative than those outside clusters. The results provide some supportive evidence that clusters foster firm innovations. 16 Figure 4.3 Number of patents granted per million yuan of value-added, by degree of clustering Source: Tabulated by authors based on merged firm patent dataset. Note: Clustering is defined according to Long and Zhang (2011). Fourth, labor-intensive firms have become increasingly more innovative.3 It is a common perception that capital-intensive firms are more innovative than labor-intensive ones. However, we found the opposite. In 1998, patents invented by labor-intensive firms consisted of only 37 percent of total firm patents. But by 2009, 83 percent of firm patents were owned by labor-intensive firms. This probably reflects the increasing wage pressures. As revealed in Figure 4.4, the gap in innovation intensity between labor-intensive and capital-intensive firms China has enlarged significantly since 2003, coincident with the timing when China reached the Lewis turning point (Zhang, Yang, and Wang 2011). Wage escalations since 2003 have likely induced labor-intensive enterprises to invest more in technology and capital to substitute for labor. 3 We defined a firm as labor intensive if the capital-labor ratio is less than the median value of the whole sample and capital intensive if otherwise. 17 Figure 4.4 Number of patents granted per million yuan of value-added, by firm capital intensity Source: Tabulated by authors based on merged firm patent dataset. Note: For each year we define a firm as capital intensive if the capital-labor ratio is greater than the median value. Fifth, industries with a higher proportion of female workers exhibit more innovation. The combination of the introduction of the one-child policy, the availability of ultrasound sex identification technology, and a son-preference culture has resulted in a severely skewed sex ratio in China as high as 120 men per 100 women (Bulte, Heerink, and Zhang 2011; Wei and Zhang 2011). As a result, there is a shortage of women in both marriage and labor markets. Industries that rely mainly on women would have to look for other options to reduce the reliance on increasingly scarcer female workers. Investing in technologies is one option. To test this idea, we grouped firms into two types, female intensive and male intensive. Since the number of employees by gender is available only for 2004, 2005, 2006, and 2007, we had to impute the gender ratio of male to female workers for the earlier years according to the industrial average calculated for the period from 2004 to 2007. We defined an industry as female or male intensive if the average female intensity (the ratio of female to male workers) from 2004 to 2007 is greater than or less than the median value. Table A.1 in the appendix lists the gender intensity for major manufacturing industries. In general, gender intensity changes slowly over time. So our imputation should have minimal impact on the robustness of our results. Column 7 of Table 4.1 indicates that on average female-intensive industries have secured more patents than have male-intensive industries. With respect to innovation intensity, female-intensive firms consistently perform better their male-intensive counterparts (see Figure 4.5). It seems that sex ratio imbalances matter to firm innovation behavior. 18 Figure 4.5 Number of patents granted per million yuan of value-added, by industry gender intensity Source: Tabulated by authors based on merged firm patent dataset. Note: See Table A.1 in the appendix for the definition of gender intensity. 19 5. CONCLUSION China has made huge progress in intellectual property protection. It has become the largest recipient of patent filings in the world since 2011. To unveil the patterns of the dramatic transformation, we have merged the national patent database with the ASIEC database. The merged database enables us to examine the historical patterns of patents across industries and firm types in China. The explosion in the number of patents in the past three decades means China has greatly improved its innovation capacity. In the 1980s, foreign firms filed most of the invention patents and had a higher approval rate than domestic ones, implying that initially foreign firms had more advanced technologies. However, the domestic firms have caught up over time in both quantity and quality (in terms of invention patents). Firms, in particular domestic ones, have become the key body of innovation in China. Among the domestic firms, it is the private ones that are the engines of innovation. The export- oriented enterprises have witnessed a remarkable spike in innovation since China joined WTO and has had access to a wider market. The trend of patents also coincides with the timing of labor market and demographic transitions. Since China passed the Lewis turning point, labor has become more expensive relative to capital. As a result, labor-intensive firms have shown a stronger propensity to innovate than capital-intensive ones. Facing the shortage of female workers, the female-intensive industries also exhibit more innovations than male-intensive ones. Innovation also has something to do with local industry structure. In areas with higher degrees of clustering, firms applied for and secured more patents than those in less clustered areas. Not surprisingly, the coastal provinces, where most of the clusters reside, dominate patent applications. After all, this paper offers only a descriptive analysis of China’s patents. All the findings are suggestive. More solid in-depth analyses are needed to further test the findings. The merged dataset provides us with a vehicle to empirically undertake the tests. 20 APPENDIX: SUPPLEMENTARY TABLE Table A.1 Gender intensity, by industry Code Industry Gender intensity 6 Mining and washing of coal Male intensive 7 Extraction of petroleum and natural gas Male intensive 8 Mining and processing of ferrous metal ores Male intensive 9 Mining and processing of nonferrous metal Male intensive 10 Mining and processing of nonmetal ores Male intensive 11 Other mining Male intensive 13 Processing of food Male intensive 14 Manufacture of foods Female intensive 15 Manufacture of beverages Female intensive 16 Manufacture of tobacco Female intensive 17 Manufacture of textile Female intensive 18 Manufacture of textile, clothing, and apparel Female intensive 19 Manufacture of leather, fur, and feather Female intensive 20 Processing of timber, manufacture of wood, bamboo, rattan, palm, and straw products Female intensive 21 Manufacture of furniture Male intensive 22 Manufacture of paper and paper products Female intensive 23 Printing, reproduction of recording media Female intensive 24 Manufacture of articles for culture, education, and sport activities Female intensive 25 Processing of petroleum, coking, processing of nuclear fuel Male intensive 26 Manufacture of raw chemical materials and chemical products Male intensive 27 Manufacture of medicines Female intensive 28 Manufacture of chemical fibers Female intensive 29 Manufacture of rubber Female intensive 30 Manufacture of plastics Female intensive 31 Manufacture of nonmetallic mineral products Male intensive 32 Smelting and pressing of ferrous metals Male intensive 33 Smelting and pressing of nonferrous metals Male intensive 34 Manufacture of metal products Male intensive 35 Manufacture of general-purpose machinery Male intensive 36 Manufacture of special-purpose machinery Male intensive 37 Manufacture of transport equipment Male intensive 39 Electrical machinery and equipment Female intensive 40 Manufacture of communication equipment, computers, and other electronic equipment Female intensive 41 Manufacture of measuring instruments and machinery for cultural activity and office work Female intensive 42 Manufacture of artwork Female intensive 43 Recycling and disposal of waste Male intensive 44 Electric power and heat power Female intensive 45 Production and supply of gas Female intensive 46 Production and supply of water Female intensive Source: Tabulated by authors based on above-scale firm dataset. Note: Since we have gender data from only 2004 to 2007, we define an industry as female or male intensive if the gender ratio of workers in the period from 2004 to 2007 is greater than or less than the median value. 21 REFERENCES Aghion, P., and P. Howitt. 1992. “A Model of Growth through Creative Destruction.” Economictra 60 (2): 323–351. Aw, B. Y., M. J. Roberts, and D. Y. Xu. 2011. “R&D Investment, Exporting, and Productivity Dynamics.” American Economic Review 101 (4): 1312–1344. Awokuse, T. O., and H. Yin. 2010. “Intellectual Property Rights Protection and the Surge in FDI in China.” Journal of Comparative Economics 38 (2): 217–224. Becker, S. O., and P. H. Egger. 2013. “Endogenous Product versus Process Innovation and a Firm’s Propensity to Export.” Empirical Economics 44 (1): 329–354. Bulte, E., N. Heerink, and X. Zhang. 2011. “China's One-child Policy and ‘the Mystery of Missing Women’: Ethnic Minorities and Male-biased Sex Ratios.” Oxford Bulletin of Economics and Statistics 73 (1): 21–39. Bustos, P. 2011. “Trade Liberalization, Exports, and Technology Upgrading: Evidence on the Impact of MERCOSUR on Argentinian Firms.” American Economic Review 101 (1): 304–340. Cheung, K.-y., and P. Lin. 2004. “Spillover Effects of FDI on Innovation in China: Evidence from the Provincial Data.” China Economic Review 15 (1): 25–44. Choi, S. B., S. H. Lee, and C. Williams. 2011. “Ownership and Firm Innovation in a Transition Economy: Evidence from China.” Research Policy 40 (3): 441–452. Dobson, W., and A. E. Safarian. 2008. “The Transition from Imitation to Innovation: An Enquiry into China's Evolving Institutions and Firm Capabilities.” Journal of Asian Economics 19 (4): 301–311. Dong, X., and Z. Wang. 2007. “R&D Risk, Innovation Environment and the Optimal Life of Software Patent.” Economic Research Journal 9:011. Feng, Y., and Y. Yao. 2014. “The Middle-income Trap and China’s Growth Prospects.” In Deepening Reform for China’s Long-term Growth and Development, edited by S. I. S. Director, 133–158. China Update Series. Canberra, Australia: ANU Press. Gonzalez, M. C. Fink, B. Hall, and C. Helmers 2013. The Use of Intellectual Property in Chile. Economic Research Working Paper No. 11. Geneva: World Intellectual Property Organization. Hu, A. G., and G. H. Jefferson. 2009. “A Great Wall of Patents: What Is behind China’s Recent Patent Explosion?” Journal of Development Economics 90 (1): 57–68. Huang, Y., and J. Tingsong. 2010. “What Does the Lewis Turning Point Mean for China? A Computable General Equilibrium Analysis.” China Economic Journal 3 (2): 191–207. Kou, Z. 2004. “Patent Scope and Patenting of Intermediate Results in Cumulative Innovation Race.” China Economic Quarterly 3 (3): 744–762. Kou, Z., and M. Zhou. 2011. “Secrecy vs. Patent.” China Economic Quarterly 11 (1): 116–134. Li, S., T. Hong, and C. Wu. 2012. “Patent Output, Intellectual Property Protection and Market Value: Evidence from China’s Listed Companies.” Naikai Management Review 15 (6): 14–24. Lin, C., P. Lin, and F. Song. 2010. “Property Rights Protection and Corporate R&D: Evidence from China.” Journal of Development Economics 93 (1): 49–62. Lin, J. Y. 2012. New Structural Economics: A Framework for Rethinking Development and Policy. Washington, DC: World Bank. Long, C., and X. Zhang. 2011. “Cluster-based Industrialization in China: Financing and Performance.” Journal of International Economics 84 (1): 112–123. Long, C., and X. Zhang. 2012. “Patterns of China’s Industrialization: Concentration, Specialization, and Clustering.” China Economic Review 23 (3): 593–612. Marshall, A. [1890] 1920. Principles of Economics. 8th ed. London: Macmillan. 22 Qin, X.-z., J.-b. Zhou, and Z.-f. Yin. 2012. “Characteristics of Small and Medium-sized Enterprises: Innovation Evidence from Survey on Deyang Municipality’s Enterprises.” China Industrial Economics 4: 63–75. Schumpeter, J. A. 1934. The Theory of Economic Development. Cambridge, MA, US: Harvard University Press. Wei, S.-j., and X. Zhang. 2011. “The Competitive Saving Motive: Evidence from Rising Sex Ratios and Savings in China.” Journal of Political Economy 119 (3): 511–564. WIPO (World Intellectual Property Organization). 2012. World Intellectual Property Indicators 2012. Geneva. Wu, C.-Y., and J. A. Mathews. 2012. “Knowledge Flows in the Solar Photovoltaic Industry: Insights from Patenting by Taiwan, Korea, and China.” Research Policy 41 (3): 524–540. Xu, Z. y. 2010. “On the Optimal Patent Protection of Developing Countries in a Technological Diffusion Model.” China Economic Quarterly 19 (12): 510–532. Yu, M., and G. Ye. 2013. “Trade Liberalization, Product Complexity, and Productivity Improvement: Evidence from Chinese Firms.” World Economy 1:913–934. Zhang, X., J. Yang, and S. Wang. 2011. “China Has Reached the Lewis Turning Point.” China Economic Review 22 (4): 542–554. Zhu, P., and W. Xu. 2003. “On the Impact of Government’s S&T Incentive Policy on the R&D Input and Its Patent Output of Large and Medium-sized Industrial Enterprises in Shanghai.” Economic Research Journal 6: 45– 53. Zhuang, J., P. Vandenberg, and Y. Huang. 2012. Growing Beyond the Low-cost Advantage: How the People’s Republic of China Can Avoid the Middle-income Trap. Manila, Philippines: Asian Development Bank 23 RECENT IFPRI DISCUSSION PAPERS For earlier discussion papers, please go to www.ifpri.org/pubs/pubs.htm#dp. All discussion papers can be downloaded free of charge. 1384. Understanding the rapid reduction of undernutrition in Nepal, 2001–2011. Derek D. Headey and John Hoddinott. 1383. Two-tier asymmetric information as a motive for trade, trade policies, and inefficient trade agreements. Antoine Bouët, David Laborde, and David Martimort, 2014. 1382. The tide that does not raise all boats: An assessment of EU preferential trade policies. Maria Cipollina, David Laborde, and Luca Salvatici, 2014. 1381. The impact of India's rural employment guarantee on demand for agricultural technology. Anil K. Bhargava, 2014. 1380. Towards understanding economic growth in Africa: A reinterpretation of the Lewis Model. Xinshen Diao and Margaret McMillan, 2014. 1379. The nutritional returns to parental education. Harold Alderman and Derek D. Headey, 2014. 1378. Evidence-based research and its effect on policymaking. Nurul Islam, 2014. 1377. The development push of refugees: Evidence from Tanzania. Jean-François Maystadt and Gilles Duranton, 2014. 1376. Food safety and developing markets: Research findings and research gaps. Laurian Unnevehr and Loraine Ronchi, 2014. 1375. Costing alternative transfer modalities. Amy Margolies and John Hoddinott, 2014. 1374. Do girls pay the price of civil war?: Violence and infant mortality in Congo. Olivier Dagnelie, Giacomo De Luca, and Jean-François Maystadt, 2014. 1373. Variable returns to fertilizer use and its relationship to poverty: Experimental and simulation evidence from Malawi. Aurélie Harou, Yanyan Liu, Christopher B. Barrett, and Liangzhi You, 2014. 1372. Drought risk reduction in agriculture: A review of adaptive strategies in East Africa and the Indo-Gangetic Plain of South Asia. Nicola Cenacchi, 2014. 1371. The political economy of MGNREGS spending in Andhra Pradesh. Megan Sheahan, Yanyan Liu, Christopher B. Barrett, and Sudha Narayanan, 2014. 1370. Assessing progress made toward shared agricultural transformation objectives in Mozambique. Todd Benson, Tewodaj Mogues, and Sileshi Woldeyohannes, 2014. 1369. Strategies to control aflatoxin in groundnut value chains. Wojciech J. Florkowski and Shashidhara Kolavalli, 2014. 1368. The agrarian reform experiment in Chile: History, impact, and implications. Alberto Valdés and William Foster, 2014. 1367. What dimensions of women’s empowerment in agriculture matter for nutrition-related practices and outcomes in Ghana? Hazel Jean L. Malapit and Agnes R. Quisumbing, 2014. 1366. The Economywide effects of teff, wheat, and maize production increases in Ethiopia: Results of economywide modeling. Todd Benson, Ermias Engida, and James Thurlow, 2014. 1365. Identifying agricultural expenditures within the public financial accounts and coding system in Ghana: Is the ten percent government agriculture expenditure overestimated? Samuel Benin, 2014. 1364. Environmental migration and labor markets in Nepal. Jean-François Maystadt, Valerie Mueller, Ashwini Sebastian, 2014. 1363. The role of agriculture in the fast-growing Rwandan economy: Assessing growth alternatives. Xinshen Diao, Godfrey Bahiigwa, Angga Pradesha, 2014. 1362. Understanding the context for agriculture for nutrition research: Identifying country typologies of child-stunting determinants. Eduardo Maruyama, Laurian J. Unnevehr, and Jiarui Wang, 2014. 1361. Assessing the economic benefits of sustainable land management practices in Bhutan. Ephraim Nkonya, Raghavan Srinivasan, Weston Anderson, and Edward Kato, 2014. 1360. Examining the sense and science behind Ghana’s current blanket fertilizer recommendation. Antony Chapoto and Francis Tetteh, 2014. http://www.ifpri.org/pubs/pubs.htm%23dp INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE www.ifpri.org IFPRI HEADQUARTERS 2033 K Street, NW Washington, DC 20006-1002 USA Tel.: +1-202-862-5600 Fax: +1-202-467-4439 Email: ifpri@cgiar.org mailto:ifpri@cgiar.org Abstract Acknowledgments 1. Introduction 2. Stylized Facts about Patents in China Brief Introduction of the Patent System in China China Patent Database Several General Trends of Chinese Patents 3. Match Patent Database with Firm Database 4. Some Interesting Patterns Derived from the Merged Database 5. Conclusion Appendix: Supplementary Table References RECENT IFPRI DISCUSSION PAPERS