Knowledge and technology transferring between universities and industries had been an important research focus of innovation management. Bibliometric research on the university-industry knowledge transferring had always used patent collaboration and citation data as indicators. However, patent licensing data were more representative and could target the knowledge transferring directions. This paper had gone through the data of Chinese academic-industry patent licensing and focused on the variances in regional level and geographic distance. Using patent licensing document analyzing method, the academic-industry knowledge flowing patterns had been discovered. Results showed that localization effects had been existed in those Chinese regions and the engagement of knowledge transferring was severely unbalanced.
University innovation output had been considered a crucial source of knowledge and ideas in industry [
In the private sectors, companies have shifted from mainly exploiting internal resources to exploring ideas
and supports from outside firm boundaries. A research stream about absorptive capacities had highlighted the ability of firms to make good use of the scientific knowledge and technology from outside world [
University is one of the important sources of outside scientific innovation resources of industries [
The main purpose of this work is to reveal the regional variance patterns of Chinese university-industry scientific knowledge transfer in the channel of patent licensing. Previous works had investigated the regional variances through other channel: 1) Patent collaboration. Hong (2008) used co-patenting data in Chinese universities in his work and discovered an uneven distribution of knowledge exchange [
So far, no study on the university-industry knowledge transferring had used patent licensing data to investigate regional performance and variance in the Chinese context. Licensing as a main channel of knowledge transfer had not been studied in the bibliometric way. Compared with other channels, the knowledge transferred through patent licensing is solid technologies, which is more closed to application and could be valued in the market place. Results on patent licensing data are more representative to the innovation commercialization process. Also, the licensing data have knowledge transferring directions embedded and the differences in giving and receiving could be identified. Patent collaboration data, on the other hand, treated collaboration partners in reciprocal relationships, thus could not determine the knowledge flowing directions. Paper collaboration and citation were more suitable to studies on the inter-academy knowledge flow but not as well performed as in the field of academy-industry knowledge transferring for the lack of paper co-authorship and citation between universities and private companies. Patent citation data as measurement of knowledge transferring had long been criticized because the coexistence of applicant citation and examiner citation. Citations added up by examiners could not represent the actual knowledge flow from citation patents to the applicants [
The remainder of this paper had been structured as follow: In the Data and Method section, datasets and methods used in this study had been introduced. The Results sections had been organized to analyze the local knowledge transferring patterns, regional knowledge exportation patterns and regional knowledge importation patterns. The Conclusion sections had summarized results and implications of this study.
In this paper, the regional variances of university patent licensing activities had been studied using patent analyzing techniques. Patent licensing data had been a crucial indicator of scientific knowledge transferring. The patent licensing documents data of all the Chinese universities and colleges from 2002 to 2012 had been collected from the online database in SIPO (State Intellectual Property Office of the PRC). In China, the patent licensing contracts between licensors and licensees had to be registered and documented in SIPO system, or they would not be protected by the national patent law. In that case, licensing activities documented in the system were the actual legal patent licensing activities that had happened in the real world. Each licensing documents had information about licensed patents, including the licensers, licensees, terms and territories. One document indicates one licensing activity.
Results on the regional variances of university-industry patent licensing activities had been arranged in the following subsections: First, we look into the local knowledge transferring patterns of each region; second, the patterns of regional universities out licensing patents to industry had been studied; finally, the patterns of regional industries receiving patents from universities had been investigated. These three sections had classified the university-industry knowledge flow into three directions: local universities to local industries, local universities to non-local industries and non-local universities to local industries.
The fourth column in
The second column had shown the local licensing data and the regions in the first column were ranked by this category. Local licensing data represented the total volume of patents that had been transferred from universities to companies in the same ego district. Numbers in this category had two implications. First, they indicated the knowledge contribution of universities to the local economy. Second, they indicated the efficiency of local knowledge utilization of industries. The average local licensing number was 113.98, with a standard deviation of 168.10. Jiangsu, Guangdong, Anhui, Zhejiang and Shanghai ranked top 5 in local licensing numbers. Inference from these data was that both universities and companies in these regions had actively involved in the academic invention commercialization process.
Besides, the average local licensing ratio in this dataset equaled to 56.18% (113.96/2022.86), larger than 50%. In the aspect of academic knowledge out-licensing, 25 of these 28 regions’ universities had transferred the largest amount of patents to local industries, except for Gansu, Shanxi and Hainan. Hainan had no academic patents transferred out therefore was not account. The largest amount of knowledge transferring in Gansu and Shanxi had been generated from Gansu to Zhejiang and from Shanxi to Guangdong. In the aspect of academic knowledge receiving, 26 of 28 regions’ industries had absorbed the largest amount of patents from local universities, except for Hainan and Xinjiang, in which the largest amount of knowledge transferring had been generated from Jiangsu to Hainan and from Jiangsu to Xinjiang. These results indicated that relatively large portion of academic patents bad been transferred into local industries. This result had supported the academic perspective of localization effects on knowledge flow [
Region | Local licensing | Total out-licensing | Total in-licensing | Local out-licensing ratio | Local in-licensing ratio |
---|---|---|---|---|---|
Jiangsu | 818 | 1080 | 1416 | 75.74% | 57.77% |
Guangdong | 386 | 469 | 903 | 82.30% | 42.75% |
Anhui | 293 | 308 | 379 | 95.13% | 77.31% |
Zhejiang | 273 | 433 | 595 | 63.05% | 45.88% |
Shanghai | 158 | 580 | 234 | 27.24% | 67.52% |
Shandong | 146 | 229 | 293 | 63.76% | 49.83% |
Beijing | 138 | 413 | 225 | 33.41% | 61.33% |
Hubei | 122 | 240 | 208 | 50.83% | 58.65% |
Tianjin | 122 | 303 | 144 | 40.26% | 84.72% |
Hunan | 118 | 176 | 191 | 67.05% | 61.78% |
Fujian | 104 | 117 | 207 | 88.89% | 50.24% |
Hebei | 81 | 119 | 184 | 68.07% | 44.02% |
Heilongjiang | 73 | 191 | 89 | 38.22% | 82.02% |
Shaanxi | 68 | 299 | 73 | 22.74% | 93.15% |
Chongqing | 64 | 137 | 71 | 46.72% | 90.14% |
Sichuan | 55 | 137 | 88 | 40.15% | 62.50% |
Liaoning | 47 | 106 | 71 | 44.34% | 66.20% |
Jilin | 34 | 62 | 44 | 54.84% | 77.27% |
Guangxi | 19 | 38 | 72 | 50.00% | 26.39% |
Henan | 19 | 57 | 58 | 33.33% | 32.76% |
Shanxi | 14 | 93 | 25 | 15.05% | 56.00% |
Yunnan | 14 | 27 | 26 | 51.85% | 53.85% |
Jiangxi | 9 | 23 | 24 | 39.13% | 37.50% |
Neimenggu | 8 | 10 | 23 | 80.00% | 34.78% |
Gansu | 3 | 21 | 6 | 14.29% | 50.00% |
Guizhou | 3 | 10 | 8 | 30.00% | 37.50% |
Xinjiang | 2 | 2 | 10 | 100.00% | 20.00% |
Hainan | 0 | 0 | 13 | NA | 0.00% |
Mean | 113.96 | 202.86 | 202.86 | ||
SD | 168.10 | 233.96 | 309.04 |
Data in local licensing number, total out-licensing number, total in-licensing number had been shown in a bar diagram in
bers added up the number of patents that the local universities had transferred to companies in other regions. The total in-licensing numbers were the results of local licensing numbers added up the number of patents that the local industries received from universities in other regions. Therefore, as long as a region had a large number in local licensing, the total out-licensing number and total in-licensing number, which could not be less than local licensing number, would also be large.
To conclude, universities had licensed out more amounts of patents to local industries and the local patents transferred from local universities to local industries through licensing channel were unevenly distributed. Although all 28 regions had local licensing, a small number of regions had been actively involved in local academic licensing than the other regions.
Local licensing numbers could only reveal the academic knowledge transferred from local universities to local industries. Considering each region as a set of academic resources, there had been other patents transferred from ego region to other regions. Some region had transferred more knowledge to other regions and some regions had not. The knowledge out transferring intensity and directions could reflect a region’s ability as the main academic knowledge producer and practitioner. Regions that had transferred more knowledge to other regions had much academic impact on the other region’s industry. The capacity of each region’s knowledge out transferring would be studied in this part.
Label | Non-local out-licensing | Total out-licensing | Non-local out-licensing ratio |
---|---|---|---|
Shanghai | 422 | 580 | 72.76% |
Beijing | 275 | 413 | 66.59% |
Jiangsu | 262 | 1080 | 24.26% |
Shaanxi | 231 | 299 | 77.26% |
Tianjin | 181 | 303 | 59.74% |
Zhejiang | 160 | 433 | 36.95% |
Heilongjiang | 118 | 191 | 61.78% |
Hubei | 118 | 240 | 49.17% |
Guangdong | 83 | 469 | 17.70% |
Shandong | 83 | 229 | 36.24% |
Sichuan | 82 | 137 | 59.85% |
Shanxi | 79 | 93 | 84.95% |
Chongqing | 73 | 137 | 53.28% |
Liaoning | 59 | 106 | 55.66% |
Hunan | 58 | 176 | 32.95% |
Hebei | 38 | 119 | 31.93% |
Henan | 38 | 57 | 66.67% |
Jilin | 28 | 62 | 45.16% |
Guangxi | 19 | 38 | 50.00% |
Gansu | 18 | 21 | 85.71% |
Anhui | 15 | 308 | 4.87% |
Jiangxi | 14 | 23 | 60.87% |
Fujian | 13 | 117 | 11.11% |
Yunnan | 13 | 27 | 48.15% |
Guizhou | 7 | 10 | 70.00% |
Neimenggu | 2 | 10 | 20.00% |
Hainan | 0 | 0 | NA |
Xinjiang | 0 | 2 | 0.00% |
Mean | 88.89 | 202.86 | |
SD | 102.70 | 233.96 |
side means that a large number of regions didn’t transferred patents to outside regions. Hainan and Xinjiang, in particular, had no patents licensed to other regions. The fourth column in
Besides the capacity of local university knowledge exportation, the capacity of local industries absorbing and utilizing academic inventions also mattered. Private companies in some regions could find more academic resources and make good use of them. The absorptive capacity capture by these companies could generate competitive advantage for them. Results in
This paper had gone through the data of Chinese academic-industry patent licensing and mainly focused on the performance of different districts. Using patent licensing document analyzing method, the academic-industry knowledge flowing patterns had been discovered.
The academic patent licensing activities had exhibited a localized trend in Chinese regions. As previous works in the regional patent collaboration, Hong (2008) and Gao et al. (2011) had found localized patterns in patent collaboration channel. Our results supported the traditional view on the impact of distances among knowledge transferring. Majority of those 28 regions had exhibited a local knowledge transferring pattern. Local
Label | Non-local in-licensing | Total in-licensing | Non-local in-licensing ratio |
---|---|---|---|
Jiangsu | 598 | 1416 | 42.23% |
Guangdong | 517 | 903 | 57.25% |
Zhejiang | 322 | 595 | 54.12% |
Shandong | 147 | 293 | 50.17% |
Fujian | 103 | 207 | 49.76% |
Hebei | 103 | 184 | 55.98% |
Beijing | 87 | 225 | 38.67% |
Anhui | 86 | 379 | 22.69% |
Hubei | 86 | 208 | 41.35% |
Shanghai | 76 | 234 | 32.48% |
Hunan | 73 | 191 | 38.22% |
Guangxi | 53 | 72 | 73.61% |
Henan | 39 | 58 | 67.24% |
Sichuan | 33 | 88 | 37.50% |
Liaoning | 24 | 71 | 33.80% |
Tianjin | 22 | 144 | 15.28% |
Heilongjiang | 16 | 89 | 17.98% |
Jiangxi | 15 | 24 | 62.50% |
Neimenggu | 15 | 23 | 65.22% |
Hainan | 13 | 13 | 100.00% |
Yunnan | 12 | 26 | 46.15% |
Shanxi | 11 | 25 | 44.00% |
Jilin | 10 | 44 | 22.73% |
Xinjiang | 8 | 10 | 80.00% |
Chongqing | 7 | 71 | 9.86% |
Guizhou | 5 | 8 | 62.50% |
Shaanxi | 5 | 73 | 6.85% |
Gansu | 3 | 6 | 50.00% |
Mean | 88.89 | 202.86 | |
SD | 148.01 | 309.04 |
Universities had licensed more portion of patents to local industries and local industries had absorbed the majority of licensed technology from local universities. Local universities had made largest technological contribution to local industries. Therefore, public policy should focus on facilitating the knowledge transferring process within regional border.
Different regions had performed differently in various directions of knowledge transferring. This paper had analyzed the patent licensing data in three directions: local universities to local industries, local universities to non-local industries and non-local universities to local industries. The performance of 28 regions had been varied in different directions. Some regions like Jiangsu, Guangdong, Zhejiang, Shanghai and Beijing had engaged in the largest amount of academic patent licensing. Some regions like Neimenggu, Gansu, Guizhou, Xinjiang and Hainan had merely numbers of patents licensed through. Some regions had performed relatively best in local licensing activities, like Anhui. Some regions had licensed out the largest amount of patents to external regions, like Shanghai and Beijing. Some regions had absorbed great amount of knowledge from external universities, like Guangdong and Shandong. Some regions had performed the best in each direction, like Jiangsu. The 28 regions had revealed distinctive patterns in the academic-industry knowledge transferring process. Besides the variances in regional performance, the volume of licensing activities had been spread unevenly in those regions. A very small number of regions had taken up the largest volume of licensing activities in the dataset. The unbalancing patterns of academic knowledge generating and receiving should be taken seriously by policy makers and actions should be made to bridging the gap between knowledge exchange active regions and inactive regions. Since the knowledge transferring patterns had been varied in regional level, the local government of each region should take more responsibility in it. Actions in the local government level should be made to improve regional knowledge transferring performance.
This work was supported by the National Natural Science Foundation of China under Grant Nos. 71202054, the Soft Science Research Project of Anhui Province of China under Grant Nos. 1402052002, and “The Fundamental Research Funds for the Central Universities”.
Inter-regional patent licensing data of 28 regions.