
This replication kit contains the codes and data files to replicate the results in the paper "Mapping U.S.-China Technology Decoupling: Policies, Innovation, and Firm Performance." Specifically, the codes in this replication kit produce the following figures and tables in this paper: Figure 1, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, Figure 8, Figure 9, Table 1, Table 2, Table 3, Table 4, Table 5, Table 6, Table 7, Table 8, Table 9, Appendix Figure A1, Appendix Figure A2, Appendix Table A2, and Appendix Table A3. Each of these figures and tables can be produced by running the do file with the corresponding file name and the underlying data can be found in the dta file with the corresponding file name. For instance, Figure 1 of the paper is based on the dta file "Figure_1_Data.dta" and the results can be produced by running the do file "Figure_1.do." Alternatively, all the results can be produced by running the master do file (titled “Master.do”). 

Patent information in this study is based on the combined patent-level databases from the two countries, i.e., the U.S. patent data is obtained from the United States Patent and Trademark Office (USPTO, https://www.uspto.gov/) and the Chinese patent data is from the Chinese National Intellectual Property Administration (CNIPA, https://english.cnipa.gov.cn/). Firm-level information is based on publicly traded companies in the U.S. and China between 2007 and 2019. On the Chinese side, financial statements information of firms come from the China Stock Market and Accounting Research (CSMAR) database. The CSMAR database provides data on the China stock markets and the financial statements of China’s listed companies. More details about the CSMAR database can be found at https://wrds-www.wharton.upenn.edu/pages/about/data-vendors/china-stock-market-accounting-research-csmar/. We merge the CSMAR data with the Chinese patent database by matching company names. On the U.S. side, financial statements information of firms is gathered from the Compustat, a database of financial, statistical, and market information on global companies throughout the world. More details about the Compustat database can be found at https://wrds-www.wharton.upenn.edu/pages/grid-items/compustat-annual-updates-fundamentals-annual-demo/. we merge the U.S. patent database to Compustat using the procedure developed in Kogan et al. (2017). Firm information for both countries is accessed via Wharton Research Data Services (WRDS). We exclude firms in the financial industry following the common practice. 

The following is a data dictionary of the variables in each data file. 

Figure_1_Data.dta
year: This is a variable for the year information.                     
RnD_CHN: This is the R&D expenditure of China measured in billions of 2005 PPP dollars. 
RnD_USA: This is the R&D expenditure of the U.S. measured in billions of 2005 PPP dollars.   
n_pat_CHN: This is the number of patents in China expressed in thousands. 
n_pat_USA: This is the number of patents in the U.S. expressed in thousands. 

Figure_3_Data.dta
year: This is a variable for the year information.                
p_c_u: This is the propensity for Chinese patents to cite a U.S. patent relative to citing a Chinese one.              
p_u_c: This is the propensity for U.S. patents to cite a Chinese patent relative to citing a U.S. one.

Figure_4_Data.dta
year: This is a variable for the year information.             
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.   

Figure_5_Data.dta
year: This is a variable for the year information.             
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.  

Figure_6_Data.dta
year: This is a variable for the year information.             
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.
category: This is a categorical variable for emerging vs mature technologies. 

Figure_7_Data.dta
year: This is a variable for the year information. 
IPC: This is the three-digit classification code based on the International Patent Classification (IPC) system. 
n_pat_c: This is log number of Chinese patents. 
n_pat_u: This is log number of U.S. patents.  
n_cite_c: This is log number of citations made by the Chinese patents. 
n_cite_u: This is log number of citations made by the U.S. patents. 
treat_yr1-treat_yr12: These are interaction terms between the treatment indicator and the year dummies. 
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.

Figure_8_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier.  
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
treat_yr1-treat_yr12: These are interaction terms between the treatment indicator and the year dummies. 
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.  

Figure_9_Data.dta
year: This is a variable for the year information. 
n_entity: This is the number of sanctioned Chinese entities on the U.S. entity list. 
n_tech_class: This is the number of technology classes exposed to U.S. sanctions. 

Table_1_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier. 
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
decoupling_l1: This is the measure of U.S.-China technology decoupling, lagged by one year. 
decoupling_l23: This is the measure of U.S.-China technology decoupling, the average value of lagged two-three years. 
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Table_2A_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier. 
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
decoupling: This is the measure of U.S.-China technology decoupling. 
crs_decoupling_escalation: This is the interaction term between the decoupling measure and the Escalation Period indicator; the Escalation Period indicator takes the value of one since 2014 and zero otherwise.
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Table_2B_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier. 
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
decoupling: This is the measure of U.S.-China technology decoupling. 
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Table_3_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier. 
asset: This is log book value of firm assets. 
age: This is log one plus firm age since IPO. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
decoupling_l1: This is the measure of U.S.-China technology decoupling, lagged by one year. 
decoupling_l23: This is the measure of U.S.-China technology decoupling, the average value of lagged two-three years. 
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Table_4_Data.dta
year: This is a variable for the year information. 
IPC: This is the three-digit classification code based on the International Patent Classification (IPC) system. 
n_pat_c: This is log number of Chinese patents. 
n_pat_u: This is log number of U.S. patents.  
treat_post: This is the interaction term between the treatment indicator and the post indicator; the treatment indicator takes the value of one for sectors promoted by the SEI and zero otherwise, the post indicator takes the value of one after 2012 and zero otherwise. 
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.

Table_5_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier.  
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
treat_post: This is the interaction term between the treatment indicator and the post indicator; the treatment indicator takes the value of one for sectors promoted by the SEI and zero otherwise, the post indicator takes the value of one after 2012 and zero otherwise. 
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Table_6_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier.  
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
RnD_efficiency: This is the number of patent applications divided by the weighted average of R&D expenditures in recent years. 
breakthrough: This is the share of breakthrough patents filed by the firms. 
explore: This is the share of explorative patents filed by the firms. 
exploit: This is the share of exploitative patents filed by the firms. 
originality: This is the average originality scores of the patents filed by the firms. 
generality: This is the average generality scores of the patents filed by the firms. 

Table_7A_Data.dta
year: This is a variable for the year information. 
IPC: This is the three-digit classification code based on the International Patent Classification (IPC) system. 
n_pat_c: This is log number of Chinese patents. 
n_pat_u: This is log number of U.S. patents.  
post_sanction: This indicator variable takes the value of one for a sector in a year if this sector had been exposed to U.S. sanctions prior to that year and zero otherwise.
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.

Table_7B_Data.dta
year: This is a variable for the year information. 
IPC: This is the three-digit classification code based on the International Patent Classification (IPC) system. 
n_pat_c: This is log number of Chinese patents. 
n_pat_u: This is log number of U.S. patents.  
sanction_exposure_up: This is the weighted average of the sanction indicator of all upstream technology classes of the focal technology class.
sanction_exposure_dn: This is the weighted average of the sanction indicator of all downstream technology classes of the focal technology class.
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.

Table_8_Data.dta
year: This is a variable for the year information. 
IPC: This is the three-digit classification code based on the International Patent Classification (IPC) system. 
n_pat_c: This is log number of Chinese patents. 
n_pat_u: This is log number of U.S. patents.  
n_cite_c: This is log number of citations made by the Chinese patents. 
n_cite_u: This is log number of citations made by the U.S. patents.
sanction_yr1-sanction_yr10: These are event timing indicators representing the years around the sanction event.  
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.

Table_9A_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier.  
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
post_sanction: This indicator variable takes the value of one for a sector in a year if this sector had been exposed to U.S. sanctions prior to that year and zero otherwise.
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Table_9B_Data.dta
year: This is a variable for the year information. 
firm_ID: This is a variable for firm identifier.  
asset: This is log book value of firm assets. 
age: This is log one plus firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
sanction_exposure_up: This is the weighted average of the sanction indicator of all upstream technology classes of the focal technology class.
sanction_exposure_dn: This is the weighted average of the sanction indicator of all downstream technology classes of the focal technology class.
n_pat: This is log one plus the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is log total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is log Tobin’s Q of the firms.

Appendix_Figure_A1_Data.dta
year: This is a variable for the year information.             
decoupling: This is the measure of U.S.-China technology decoupling.        
dependence: This is the measure of China’s technological dependence on the U.S.

Appendix_Figure_A2_Data.dta
year: This is a variable for the year information. 
dependence: This is the measure of China’s technological dependence on the U.S.
n_pat_sh_c: This is the share of China-filed patents out of U.S.-and-China total. 

Appendix_Table_A2_Data.dta
asset: This is book value of firm assets. 
age: This is firm age since founding. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
decoupling: This is the measure of U.S.-China technology decoupling.       
n_pat: This is the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is Tobin’s Q of the firms.

Appendix_Table_A3_Data.dta
asset: This is book value of firm assets. 
age: This is firm age since IPO. 
capex_ratio: This is firm capital expenditures divided by book value of assets. 
ppe_ratio: This is net value of property, plant, and equipment divided by book value of assets. 
leverage_ratio: This is book value of total debt divided by book value of assets. 
RnD_at_ratio: This is firm R&D expenditures divided by assets. 
decoupling: This is the measure of U.S.-China technology decoupling.       
n_pat: This is the number of patent applications a firm files (and eventually granted). 
n_cite: This is the number of citations received divided by the average number of citations received by patents in the same cohort.  
TFP: This is total factor productivity of the firms. 
ROIC: This is the return on invested capital of the firms. 
tobin_q: This is Tobin’s Q of the firms.


References
Kogan, Leonid, Dimitris Papanikolaou, Amit Seru, and Noah Stoffman, “Technological Innovation, Resource Allocation, and Growth,” The Quarterly Journal of Economics, 03 2017, 132 (2), 665–712.

