The Largest Insurance Program in History Tables

Figure 1. NCMS Rollout Notes: This figure plots the fraction of counties implementing the NCMS over time. The number of NCMS counties is obtained from 2005 -2011 China Health Statistical Yearbook and the total number of counties from 2011 China Civil Affairs' Statistical Yearbook. Figure 2. Distribution of the share of agricultural hukou out of the population Note: This figure plots the distribution of the share of population with an agricultural hukou from China's 2000 Population Census. Panel A. Mortality rate Panel B. Life expectancy at birth Figure 3. Mortality rate and Life expectancy around NCMS adoption: subsamples by agricultural share Notes: This figure plots the average value of the residuals from regressing Mortality rate (Panel A) and Life expectancy (Panel B) on NCMS-timing-by-year FE and county-specific linear trend. Counties in the sample are split by the median value of the share of population with an agricultural hukou. Observations more than 3 years (5 years) before (after) the NCMS are binned into groups. Sample period is from 2004 to 2010. Panel A. Mortality rate Panel B. Life expectancy at birth Figure 4. Mortality rate and Life expectancy around NCMS adoption Notes: This figure plots the estimated coefficients on the interactions between year-to-NCMS dummies and 2000 agricultural share from the regression model specified in equation (2). The year before NCMS adoption is omitted, so the estimates are normalized to zero in that year. Observations more than 3 years (5 years) before (after) the NCMS are binned into groups. Results are weighted by county population. Sample period is from 2004 to 2010. The dashed lines represent the 95 percent confidence intervals based on standard errors clustered at the county level. Figure 5. OOP spending vs. Total spending by NCMS enrollment Notes: This figure plots the OOP spending against the total spending from the CHNS. The sample is restricted to those with positive medical spending both before and after enrolling in the NCMS. Table 1. Summary of public health insurance programs in developing countries Year Country Name of Insurance Scheme Number of People (year) 1990 Brazil Sistema Único de Saúde (SUS) 200 million (2018) 1992 Vietnam Social Health Insurance 80 million (2018) 1993 Colombia 34 million (2016) 1995 Philippines Regimen Subsidiado (SR) Philippines Health Insurance Corporation (PhilHealth) 1999 Rwanda 12 million (2018) 2002 Thailand 2003 China 2005 Ghana 2018 India Mutuelles de Santé Universal Coverage Health Scheme (UCS) New (Rural) Cooperative Medical Scheme (NCMS or NRCMS) National Health Insurance Scheme (NHIS) Pradhan Mantri Jan Arogaya Yojana (PM-JAY) up to 104 million (2018) 53 million (2018) 670 million (2015) 11 million (2014) 103 million (2019) Notes: Authors review of material on developing country health insurance expansions. Data sources listed in Appendix 1. For the Philippines, there is larger controversy over the reach of the program, so we list an upper bound here. Table 2. Summary statistics of DSP data Mean Age-adjusted mortality rate Life expectancy at birth Share of Agricultural population in 2000 GDP per capita Wage 546.298 75.325 0.707 22,663 26,705 DSP sample S.D. 126.882 3.642 0.254 14945 10851 Obs 847 847 121 847 847 National statistics 568 74 0.7 22,072 27,102 Notes: a. DSP sample: Age-adjusted death rate is measured as the number of deaths per 100 thousand people. The age adjustment is based on the China's 2000 Population Census. Share of agricultural population in 2000 is obtained from China's 2000 Population Census. GDP per capita and wage are obtained from China City Statistical Yearbook 2005-2011 and measured in 2010 RMB. The statistics are weighted by the county population. b. National statistics: Mortality rate and life expectancy at birth are statistics in 2007, calculated using the China's 2000 and 2010 Population Census. Share of agricultural population in 2000 is obtained from China's 2000 Population Census. GDP per capita and wage are the average values over 2004 through 2010 and measured in 2010 RMB, obtained from 2011 China Statistical Yearbook. Table 3. Summary statistics of the main variables from the CLHLS and the CHNS Mean S.D. Obs Health outcomes Suffering from serious illness (past two years, Dummy) Limited activities of daily living (ADL) (Dummy) Mental health (Dummy) Self-reported health (Dummy) Interviewer-reported health (Dummy) 0.183 0.252 0.228 0.853 0.862 0.387 0.434 0.419 0.354 0.345 53,224 53,224 53,224 53,224 53,224 Healthcare utilization Adequate medical service (at present, Dummy) Positive medical spending (the previous year, Dummy) Total medical spending (the previous year, 1,000 RMB) Financial difficulty in getting medical service (at present, Dummy) 0.920 0.791 2.733 0.059 0.271 0.406 6.064 0.236 58,565 31,378 24,832 52,046 Health behaviours Smoking (Dummy) Alcohol consumption (Dummy) 0.177 0.191 0.382 0.393 58,589 58,589 Health outcomes Cognitive deficit (Dummy) Limited activities of daily living (ADL) (Dummy) Diabetes (Dummy) Hypertension (Dummy) 0.509 0.333 0.019 0.093 0.500 0.471 0.137 0.290 5,942 5,942 41,491 41,491 Healthcare utilization Healthcare use (past 4 weeks, Dummy) Inpatient care use (past 4 weeks, Dummy) Outpatient care use (past 4 weeks, Dummy) Preventive care use (past 4 weeks, Dummy) Total medical spending (past 4 weeks, 1,000 RMB) Out-of-pocket spending (past 4 weeks, 1,000 RMB) Out-of-pocket /Total spending (past 4 weeks) 0.109 0.009 0.095 0.037 1.537 1.112 0.952 0.311 0.096 0.293 0.190 6.546 3.807 0.168 46,519 46,519 46,519 46,519 4,005 680 680 0.241 16.183 0.289 1.551 0.428 9.079 0.454 1.979 40,909 9,624 40,909 10,888 0.543 1.119 0.343 1.462 4.921 5.006 1.329 2.499 1.412 2.803 15.819 14.922 790 Panel A: CLHLS individual sample Panel B: CHNS individual sample Health behaviours Smoking (Dummy) # Cigarettes per day Alcohol consumption (Dummy) Consumption of alcohol per day (liang, 50mg) Panel C. CHNS community sample # Public hospitals (per 10,000 people) # Public clinics (per 10,000 people) # Private clinics (per 10,000 people) # Clinic (per 10,000 people) # Healthcare staff in medical facilities (per 1,000 people) # Beds in medical facilities (per 1,000 people) 790 790 790 790 790 Notes: This table reports the summary statistics of the main variables from the CLHLS and the CHNS. Panel A and Panel B show the individual-level health outcomes, healthcare utilization, and health behaviors from the CLHLS and the CHNS, respectively. The monetary values are measured in 2010 RMB. Panel C summarizes variables used in the analysis of community healthcare supply from the CHNS, weighted by community population. Table 4. Impacts of NCMS adoption on aggregate mortality Baseline Removing covariates Sample from 2004 to 2007 Placebo test (1) Age-adjusted Mortality Rate (2) Life Expectancy (3) Age-adjusted Mortality Rate (4) Life Expectancy (5) Age-adjusted Mortality Rate (6) Life Expectancy (7) Age-adjusted Mortality Rate (8) Life Expectancy Post NCMS × AgriShare2000 -0.195** (0.082) 0.040** (0.015) -0.196** (0.081) 0.040*** (0.015) -0.254** (0.120) 0.051** (0.026) -0.007 (0.204) 0.003 (0.036) County FE Province-by-year FE NCMS-timing-by-year FE County-specific linear trend Covariates Obs Adjusted R-squared Yes Yes Yes Yes Yes 847 0.758 Yes Yes Yes Yes Yes 847 0.705 Yes Yes Yes Yes No 847 0.758 Yes Yes Yes Yes No 847 0.705 Yes Yes Yes Yes Yes 484 0.735 Yes Yes Yes Yes Yes 484 0.634 Yes Yes Yes Yes Yes 704 0.765 Yes Yes Yes Yes Yes 704 0.712 546.298 75.325 546.298 75.325 550.041 75.271 836.383 70.798 Mean Notes: This table shows the estimated coefficients from equation (1) in various specifications. Columns (1) and (2) report the baseline results from equation (1) using DSP sample from 2004 to 2010. Columns (3) and (4) report results from specifications removing economic covariates. Columns (5) and (6) restrict the sample period to 2004 2007. Columns (7) and (8) report the results from a placebo test using the sample from 1994 to 2000. Age-adjusted mortality rate and life expectancy are in logarithm. Results are weighted by county population. Standard errors clustered at the county level are in parentheses. * p<0.10, ** p<0.05, *** p<0.01. Table 5. Impacts of NCMS enrollment on health outcomes and healthcare utilization using CLHLS Mortality hazard NCMS enrollment (1) Baseline (2) Control function 0.554*** (0.025) 0.694*** (0.062) 0.775*** (0.069) 78,446 st 1 -stage Residuals Obs Suffering from serious illness Obs Mean Limited ADL Obs Mean Mental health Obs Mean Self-reported health Obs Mean Interviewer-reported health Obs Mean Adequate medical service Obs Mean Positive medical spending Obs Mean Total medical spending (log) 78,446 Other health outcomes Healthcare Utilization (3) (4) -0.129** (0.063) 53,224 0.183 -0.125** (0.063) 53,224 0.252 0.111** (0.054) 53,224 0.228 0.123** (0.051) 53,224 0.853 0.122** (0.050) 53,224 0.862 0.089** (0.043) 58,565 0.920 0.118* (0.067) 31,378 0.791 0.418* (0.222) Obs Mean Year FE County stratification Covariates County-by-wave FE Hukou-by-wave FE Hukou-by-County FE 24,832 2733.296 Yes Yes Yes No No No Yes Yes Yes No No No No No Yes Yes Yes Yes No No Yes Yes Yes Yes Notes: This table reports the estimation results from equations (3) - (6) using the CLHLS. Column (1) shows how NCMS enrollment is related to one's mortality hazard based on equation (3). Column (2) presents the results from the control function method, with 𝑃𝑜𝑠𝑡⁡𝑁𝐶𝑀𝑆⁡ × ⁡𝐴𝑔𝑟𝑖𝐻𝑢𝑘𝑜𝑢 as an instrument. All estimates are reported after taking exponential to represent proportional increases or decreases in mortality rates relative to one. Standard errors in parentheses are computed based on 1,000 bootstraps and converted via the delta method. Columns (3) and (4) report the results on other health outcomes and healthcare utilization in the sample of survivors. Each cell presents the IV coefficients on 𝑁𝐶𝑀𝑆⁡𝑒𝑛𝑟𝑜𝑙𝑙𝑚𝑒𝑛𝑡 based on equations (5) and (6). Dependent variables in these regressions are shown in the first column. Information on medical spending is available after 2005. Monetary values are in 2010 RMB. Standard errors clustered at the county level are in parentheses. * p<0.10, ** p<0.05, *** p<0.01. Table 6. Impacts of NCMS enrollment on health outcomes and healthcare utilization using CHNS (1) Health outcomes Cognitive Deficit Obs Mean Limited ADL Obs Mean Diabetes Obs Mean Hypertension Obs Mean Healthcare use Obs Mean Inpatient care Obs Mean Outpatient care Obs Mean Preventive care Obs Mean Total medical spending (log) Obs Mean (2) Healthcare utilization (3) Medical spending -0.220* (0.126) 5,942 0.509 -0.257** (0.128) 5,942 0.333 -0.015* (0.009) 41,491 0.019 -0.065** (0.030) 41,491 0.093 0.084** (0.040) 46,519 0.109 0.019* (0.010) 46,519 0.009 0.072** (0.035) 46,519 0.095 0.054** (0.026) 46,519 0.037 0.387 (0.638) 4,005 1537.011 Covariates County-by-wave FE Hukou-by-wave FE Hukou-by-County FE Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Notes: This table shows the results about health outcomes, healthcare utilization and medical spending from CHNS. Each cell presents the IV coefficients on NCMS enrollment based on equations (5) and (6). Dependent variables in these regressions are shown in the first column. 𝐿𝑖𝑚𝑖𝑡𝑒𝑑⁡𝐴𝐷𝐿 and 𝐶𝑜𝑔𝑛𝑖𝑡𝑖𝑣𝑒⁡𝑑𝑒𝑓𝑖𝑐𝑖𝑡⁡are measured for those age 55+ only and the information is available in waves 2000 - 2006. 𝐷𝑖𝑎𝑏𝑒𝑡𝑒𝑠 and 𝐻𝑦𝑝𝑒𝑟𝑡𝑒𝑛𝑠𝑖𝑜𝑛 are measured at all ages above 12. Sample in Column (3) is restricted to those with positive medical spending. Monetary values are in 2010 RMB. Standard errors clustered at the county level are in parentheses. * p<0.10, ** p<0.05, *** p<0.01. Table 7. Impacts of the NCMS on financial protection NCMS enrollment Covariates County-by-wave FE Hukou-by-wave FE Hukou-by-County FE Obs Mean (1) (2) (3) Financial difficulty Log (OOP spending) OOP/Total spending -0.051** (0.026) Yes Yes Yes Yes 52,046 0.059 -0.429 (1.163) Yes Yes Yes Yes 680 1112.364 -0.289** (0.135) Yes Yes Yes Yes 680 0.952 Notes: This table reports the estimation results from equations (5) and (6) and the dependent variables are shown in the first row. Column (1) uses the CLHLS sample and 𝐹𝑖𝑛𝑎𝑛𝑐𝑖𝑎𝑙⁡𝑑𝑖𝑓𝑓𝑖𝑐𝑢𝑙𝑡𝑦 is an indicator for not seeking any medical service due to lack of money. This information is available after 2000. Columns (2) and (3) use the CHNS sample and the sample is restricted to those with positive OOP spending both before and after the NCMS initiation in their communities. Monetary values are in 2010 RMB. Standard errors clustered at the county level are in parentheses. * p<0.10, ** p<0.05, *** p<0.01. Table 8. Healthcare supply Panel A. NCMS effects on healthcare supply (log) Post NCMS × AgriShare2000 Panel B. Controlling for lead NCMS Post NCMS × AgriShare2000 Post NCMS (2-yr lead) × AgriShare2000 Panel C. Interaction effects with baseline supply (log) Post NCMS × AgriShare2000 Post NCMS × AgriShare2000 × Baseline supply (log) P50 vs. P75 of baseline supply (percentage points) Obs Mean (1) Public hospitals (2) Public clinics (3) Private clinics (4) Clinics (5) Healthcare workers (6) Beds 0.461*** (0.159) 0.620*** (0.232) -0.528** (0.249) 0.081 (0.284) 1.031** (0.405) 0.819** (0.389) 0.454*** (0.162) 0.050 (0.183) 0.618** (0.245) 0.017 (0.247) -0.514** (0.253) -0.091 (0.208) 0.092 (0.285) -0.078 (0.262) 1.041** (0.422) -0.066 (0.455) 0.808* (0.408) 0.077 (0.455) 0.673*** (0.147) -0.656*** (0.245) 28 1.451*** (0.285) -0.844*** (0.177) 114 -0.121 (0.199) -1.470*** (0.202) 65 1.070*** (0.342) -0.792*** (0.171) 45 1.855*** (0.435) -0.916*** (0.210) 68 1.444*** (0.400) -0.874*** (0.244) 76 790 0.543 790 1.119 790 0.343 790 1.462 790 4.921 790 5.006 Notes: This table presents the results about the supply-side response to NCMS using the CHNS. Dependent variables are shown in the first row. The number of hospitals and clinics are measured as the number per 10,000 people. The number of healthcare staff and beds are measured as the number per 1,000 people. All supply measures are in logarithm. Panel A shows the estimated coefficients on 𝑃𝑜𝑠𝑡⁡𝑁𝐶𝑀𝑆⁡ × ⁡𝐴𝑔𝑟𝑖𝑆ℎ𝑎𝑟𝑒2000, Panel B includes the two-year lead of NCMS interacted with 𝐴𝑔𝑟𝑖𝑆ℎ𝑎𝑟𝑒2000, and Panel C includes the interactions of 𝑃𝑜𝑠𝑡⁡𝑁𝐶𝑀𝑆⁡ × ⁡𝐴𝑔𝑟𝑖𝑆ℎ𝑎𝑟𝑒2000⁡with the same supply in the first row. 𝑃50⁡𝑣𝑠. 𝑃75 (Baseline supply) shows the differences in the percentage changes of supply for communities at the median of baseline supply relative to those at the top quartile, assuming a median share of agricultural population (0.75). Standard errors clustered at the county level are in parentheses. * p<0.10, ** p<0.05, *** p<0.01. Table 9. Impacts of healthcare supply on healthcare utilization and health outcomes Panel A. Healthcare use NCMS enrollment Healthcare supply (log) Obs Mean Panel B. Limited ADL NCMS enrollment Healthcare supply (log) Obs Mean Panel C. Cognitive Deficit NCMS enrollment Healthcare supply (log) Obs Mean Panel D. Diabetes NCMS enrollment Healthcare supply (log) Obs Mean Panel E. Hypertension NCMS enrollment Healthcare supply (log) Obs Mean Covariates County-by-wave FE Hukou-by-wave FE Hukou-by-County FE (1) (2) (3) Healthcare workers (4) Public hospitals Clinics 0.118* (0.063) 0.029 (0.023) 41,335 0.092 0.110* (0.063) 0.032 (0.031) 41,335 0.092 0.132* (0.069) 0.029 (0.019) 41,335 0.092 0.116* (0.069) 0.029 (0.036) 41,335 0.092 -0.378 (0.295) -0.065 (0.231) 4,671 0.335 -0.351 (0.230) -0.089 (0.137) 4,671 0.335 -0.572 (0.654) -0.170 (0.341) 4,671 0.335 -0.534 (0.994) -0.179 (0.783) 4,671 0.335 -0.478 (0.426) 0.165 (0.319) 4,671 0.534 -0.591** (0.243) 0.030 (0.125) 4,671 0.534 -0.595 (0.514) 0.013 (0.236) 4,671 0.534 -1.264 (2.288) -0.502 (1.759) 4,671 0.534 -0.036 (0.035) -0.008 (0.014) 36,517 0.018 -0.002 (0.049) -0.029 (0.023) 36,517 0.018 -0.037 (0.035) -0.005 (0.009) 36,517 0.018 -0.037 (0.037) -0.007 (0.019) 36,517 0.018 -0.062 (0.055) -0.017 (0.026) 36,517 0.092 Yes Yes Yes Yes -0.010 (0.074) -0.040 (0.033) 36,517 0.092 Yes Yes Yes Yes -0.072 (0.064) -0.024 (0.022) 36,517 0.092 Yes Yes Yes Yes -0.093 (0.117) -0.067 (0.079) 36,517 0.092 Yes Yes Yes Yes Beds Notes: This table shows the results of how NCMS enrollment and local healthcare supply affects individuals' healthcare utilization and health outcomes. We re-estimate equations (5) and (6), additionally including local supply (log) shown in the first row. We instrument 𝑁𝐶𝑀𝑆⁡𝑒𝑛𝑟𝑜𝑙𝑙𝑚𝑒𝑛𝑡 and 𝐻𝑒𝑎𝑙𝑡ℎ𝑐𝑎𝑟𝑒⁡𝑠𝑢𝑝𝑝𝑙𝑦⁡(𝑙𝑜𝑔) with 𝑃𝑜𝑠𝑡⁡𝑁𝐶𝑀𝑆⁡ ∗ ⁡𝐴𝑔𝑟𝑖𝐻𝑢𝑘𝑜𝑢, and 𝑃𝑜𝑠𝑡⁡𝑁𝐶𝑀𝑆⁡ ∗ ⁡𝐴𝑔𝑟𝑖𝑆ℎ𝑎𝑟𝑒2000⁡ ∗ ⁡𝐵𝑎𝑠𝑒𝑙𝑖𝑛𝑒⁡𝑠𝑢𝑝𝑝𝑙𝑦⁡(𝑙𝑜𝑔) as shown in the first row. In addition to controls in equations (5) and (6), we also control for 𝑃𝑜𝑠𝑡⁡𝑁𝐶𝑀𝑆⁡ ∗ ⁡𝐴𝑔𝑟𝑖𝑆ℎ𝑎𝑟𝑒2000, 𝐵𝑎𝑠𝑒𝑙𝑖𝑛𝑒⁡𝑠𝑢𝑝𝑝𝑙𝑦⁡(𝑙𝑜𝑔), and 𝐴𝑔𝑟𝑖𝑆ℎ𝑎𝑟𝑒2000. Each panel shows the IV estimates on NCMS enrollment and healthcare supply for different outcome measures. Standard errors clustered at the county level are in parentheses. * p<0.10, ** p<0.05, *** p<0.01. Table 10. Ex-ante moral hazard: smoking and alcohol consumption CLHLS sample CHNS sample (1) Smoking (2) Drinking (3) Smoking (4) # cigarettes (log) (5) Drinking (6) Amount alcohol (log) NCMS enrollment -0.029 (0.034) 0.011 (0.046) -0.057 (0.035) 0.098 (0.141) -0.054 (0.044) -0.172 (0.228) Covariates County-by-wave FE Hukou-by-wave FE Hukou-by-County FE Obs Mean Yes Yes Yes Yes 58,589 0.177 Yes Yes Yes Yes 58,589 0.191 Yes Yes Yes Yes 40,909 0.241 Yes Yes Yes Yes 9,624 16.183 Yes Yes Yes Yes 40,909 0.289 Yes Yes Yes Yes 10,888 1.551 Notes: This table shows the impacts of the NCMS on smoking and alcohol consumption based on equations (4). Results in Columns (1) and (2) rely on the CLHLS, and results in Columns (3) - (6) rely on the CHNS and restricted to age above 12. Dependent variables are in the first row. Standard errors clustered at the county in parentheses. * p<0.10, ** p<0.05, *** p<0.01.