Transcription

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25 7 Figures and Tables` Death Rate Figure 1: Trends in Cancer Rates in China Year Cancer Rate, Rural Cancer Rate, Urban Source: China DSP. Graphs displayed are created using STATA lowess smoother. Death rates are per 100,000 persons. Figure 2: Water Quality and Digestive Cancer Rates in Select Locations, Water Qualtiy N/A 1 (Best) (Worst) Digestive Cancer Rate N/A < Source: China Center for Disease Control. Disease Surveillance Points (DSP) comprise a nationally representative sample of mortality for China. Rates reported per 100,000 and are age and sex adjusted using the 2000 census population structure. 1

26 Figure 3: Air quality patterns in China Source: NASA satellite imagery. Figure 4: Monthly Precipitation Patterns in China, Source: Global Precipitation Climatology Center. 2

27 Figure 5: Main river basins in China Source: Hydro 1k Project. The above figure reflects the principal 2-digit basins (or watersheds) that comprise the hydrological surface of China. The dark lines reflect the breakdown of the 2- digit basins, and the lighter outline is the breakdown of China into 989 lower-level basins. Figure 6: Example of a River Basin Source: China Hydro 1k. 3

28 Figure 7: Pollution Intensity and the Effective Price Charged for Dumping Pollution Intensity (logs) Effective Price Charged for Dumping (logs) Year Pollution Intensity Effective Price per Ton Soure: China Environmental Yearbook ( ). Note: Pollution intensity is the ratio of dumping in tons to 10,000 yuan of industrial output. The effective price is the value of collected yuan per hundred tons not meeting wastewater discharge standards Figure 8: Age Distribution of Digestive Cancer and Stroke Incidence Share of Deaths Age Digestive Cancer Stroke Source: China DSP. 4

29 Table 1 Age-adjusted Death Rates (per 100,000) by Cause in China, Panel 1: Death Rates by General Cause Males Rural Urban Rural Urban (1) (2) (3) (4) All Causes Cancer Digestive Cancers Lung Cancers Other Cancer Heart Stroke Respiratory Illnesses Accidents / Violence Other Panel 2: Death Rates for Types of Digestive Cancer Females Esophageal Cancer Stomach Cancer Liver Cancer Other Digestive Cancers Source : Chinese Disease Surveillance Points Mortality Registration System (DSP). Note : N=145. Age adjustment is performed by calculating age-specific death rates and creating weighted averages using the population structure in China's 2000 census. Other digestive cancers includes colon cancer, intestinal cancer, and pancreatic cancer. The reported death rates are the average rates for the 145 sites, weighted by the population at each site. These calculations exclude roughly three thousand deaths (of the 500,000 deaths) in the sample where I am missing information on the age or sex of the decedent.

30 Table 2 Sample Means for Disease Surveillance Points by Region Statistic North South Overall (1) (2) (3) Digestive Cancer Rate Overall Water Grade Ammonia Nitrogen Biological Oxygen Demand Oils Permanganate Volatile Phenol Average Years of Education Share in Farming Urban site (1=yes) Air Pollution Reading Monthly Rainfall (mm) Upstream Manufacturing (millions) # of Sites Source : Chinese Disease Surveillance Points Mortality Registration System (DSP), China National Monitoring Center (2004), Global Precipitation Climatology Center (2008) Note : Higher grades reflect lower water quality (1=best, 6=worst) and a greater concentration of the listed pollutants. The water grade measure at each DSP site reflects the average water grade among monitoring sites in the same river basin. The air pollution reading is taken from satellite imagery and takes on values from 0-1, with higher values reflecting more particulates in the air, and is reported as the average reading in the river basin containing the DSP site. The rainfall measure is the average monthly rainfall in millileters in the river basin containing the DSP site from Upstream manufacturing is based on the total value of output in yuan of firms with greater than 500,000 yuan in sales. The sample means are the average values (e.g. average education) among decedents at each site restricted to deaths among persons age 20 and older. Sample means are reported weighted by the population at each site.

31 Table 3 Ordinary Least Squares (OLS) Regressions of Log of Digestive Cancer Rates on Water Grade Statistic No Controls Digestive (all) Esophageal Stomach Liver With Controls Digestive (all) Esophageal Stomach Liver (1) (2) (3) (4) (5) (6) (7) (8) Water Grade (1=best, 0.135** 0.349*** 0.141** 0.090* 0.093** 0.222** ** 6=worst) (0.053) (0.102) (0.065) (0.047) (0.044) (0.089) (0.058) (0.033) Average Education (0.041) (0.101) (0.063) (0.027) Share in Farming *** (0.255) (0.889) (0.391) (0.158) Urban (1=yes) * *** (0.176) (0.656) (0.256) (0.126) Air Pollution * * (0.147) (0.302) (0.176) (0.151) Region Controls No No No No Yes Yes Yes Yes R Squared * significant at 10% ** significant at 5%. *** significant at 1%. Source : China Disease Surveillance Points Mortality Registration (DSP), China National Monitoring Center (2004) Note : N=145. The first four columns represent OLS regressions of the logarithm of the death rate of a cause on the average water grade of the river basin in which the DSP site is located. I add covariates for columns (5)-(8), which are the average values (e.g. education) among decedents at each site restricted to deaths among persons age 20 and older. Standard errors are robust and clustered at the province level. The water grade measure at each DSP ite reflects the average water quality among monitoring sites in the same river basin. Regressions are weighted by the population at each DSP site.

32 Table 4 OLS Regressions of Log of Digestive Cancer Rates on Water Grade by Chemical Men Women Esophageal Stomach Liver Esophageal Stomach Liver Chemical (1) (2) (3) (4) (5) (6) Overall Grade Ammonia Nitrogen Biological Oxygen Demand Oils Permanganate Volatile Phenol 0.219** ** ** (0.10) (0.06) (0.03) (0.11) (0.06) (0.04) 0.214** ** 0.203** ** (0.10) (0.05) (0.03) (0.10) (0.05) (0.03) 0.229*** 0.090* 0.064* 0.170* 0.088* 0.080** (0.07) (0.05) (0.03) (0.10) (0.05) (0.03) 0.214** (0.09) (0.07) (0.04) (0.11) (0.07) (0.04) 0.228*** ** 0.183* *** (0.08) (0.06) (0.04) (0.11) (0.05) (0.04) 0.201** 0.137** * (0.09) (0.06) (0.05) (0.13) (0.07) (0.05) Source : China Disease Surveillance Points Mortality Registration (DSP), China National Monitoring Center (2004) Note : N=145. Each reported coefficient represents a separate regression. The regressions are estimated using the column variable as the dependent variable and the water grade measure in the row as the independent variable. Overall grade and chemical pollution are graded on a 6- point scale, with 1 being the highest quality and 6 being the most polluted. The dependent variable is the logarithm of the age and sex adjusted death rate by cause. The independent variable is a measure of water pollution, which is the amount of the chemical found in readings among water surveillance points in the river basin. All specifications include the control variables shown in Table 2. Standard errors are robust and clustered at the province level. Regressions are weighted by the population at each DSP site.

33 Table 5 Smoking and Dietary Habits by Water Grade in China Smoking Rates Dietary Patterns Water Grade Males Females Caloric Intake Carbohydrates % Fat % Protein Other (1) (2) (3) (4) (5) (6) (7) Level 1 (Best) , Level , Level , Level , Level , Level 6 (Worst) , Source : Smoking rates are taken from the China Household Income Survey (CHIS, 1995). The diet information is taken from the China Health and Nutrition Survey (CHNS, ). Note: The smoking rates are shown for the DSP sites which were in the 19 provinces included in the CHIS (1995), which includes 102 of the 145 DSP sites. Information on diet is shown for DSP sites located in the 9 provinces included in the CHNS, which includes 56 of the 145 sites.

34 Table 6 Two-stage Least Squares (2SLS) Regressions of Log of Digestive Cancer Rates on Water Grade using Annual Rainfall and Upstream Output as Instruments Statistic Monthly Rainfall First-Stage Water Grade Digestive (all) Esophageal Stomach Liver (1) (2) (3) (4) (5) ** (0.005) Upstream Output 0.001** Water Grade (1=best, 6=worst) Average Education Share in Farming Urban (1=yes) Air Pollution F Test of Instruments (0.0005) 0.303** 1.025*** 0.470** (0.131) (0.291) (0.175) (0.122) (0.111) (0.044) (0.097) (0.074) (0.027) *** (1.04) (0.31) (0.85) (0.59) (0.17) *** (0.852) (0.198) (0.548) (0.391) (0.156) 0.593* ** (0.332) (0.165) (0.341) (0.208) (0.172) 11.73*** * significant at 10% ** significant at 5%. *** significant at 1%. Two-stage Least Squares Source : China Disease Suveillance Points Mortality Registration (DSP), China National Monitoring Center (2004), Global Precipitation Climatology Center (2008) Note : N=145. The first column is the first-stage relationship between water grade at the DSP site, the covariates (e.g. urban), and two instrumental variables: the average monthly rainfall in millileters in the basin and output upstream of the basin containing the DSP site. The F statistic for the joint significance of the two instruments is reported (11.73), and the instruments also pass a Sargan-Hansen over-identification test, failing to reject the null hypothesis of their validity. The regressions in columns (2) through (5) represent 2SLS regressions where the dependent variable is the logarithm of the death rate of a cause on the predicted average water grade from colum (1) and the other covariates. Standard errors are robust and clustered at the province level. Regressions are weighted by the population at each DSP site.

35 Table 7 Impact of Water Pollution on Log of Death Rates by Cause in China, All Causes Cancer Heart Stroke Respiratory Violent/ Accidents Other Overall Grade (1=best, 6=worst) Ammonia Nitrogen Biological Oxygen Demand Oils (1) (2) (3) (4) (5) (6) (7) ** (0.02) (0.04) (0.04) (0.05) (0.06) (0.03) (0.02) ** (0.02) (0.03) (0.04) (0.05) (0.05) (0.02) (0.02) ** (0.02) (0.03) (0.04) (0.04) (0.06) (0.02) (0.02) (0.03) (0.04) (0.05) (0.05) (0.05) (0.02) (0.03) Permanganate ** (0.02) (0.03) (0.05) (0.05) (0.07) (0.02) (0.02) Volitile Phenol * (0.04) (0.04) (0.06) (0.05) (0.06) (0.03) (0.02) * significant at 10% ** significant at 5%. *** significant at 1%. Source : Chinese Disease Surveillance Points Mortality Registration System, China National Monitoring Center (2004) Note : N=145. Each reported coefficient represents a separate regression. The regressions are estimated using the column variable as the dependent variable and the water quality measure in the row as the independent variable. Each water reading is graded on a 6-point scale, with 1 being the highest quality and 6 being the most polluted. The dependent variable is the arithemetic average of the death rate by cause for each DSP site (see Table 2). The independent variable is a measure of water pollution, which is calculated as the average chemical reading among the water surveillance points in 2004 for the watershed containing the DSP site. All specifications include the controls listed in Table 3 and are weighted by the population at each DSP site.

36 Table 8 OLS Regressions of Water Grade on Industrial Chemical Dumping and Rainfall Overall Quality Ammonia Nitrogen Biological Oxygen Demand Oils Permanganate Volitile Phenol Chemical (1) (2) (3) (4) (5) (6) Log of Chemical 0.387*** 0.154** 0.262** 0.237*** 0.347*** Dumping (Tons) (0.10) (0.07) (0.12) (0.07) (0.10) (0.07) Monthly Rainfall (mm) *** *** *** *** *** *** (0.005) (0.005) (0.006) (0.004) (0.005) (0.004) Chemical Dumping Measure Total Dumping Ammonia Nitrogen Chemical Oxygen Demand Petroleum Suspended Solid Waste Volitized Phenol Number of Tons Per Site Average Grade 71,900 13, , , R Squared N * significant at 10% ** significant at 5%. *** significant at 1%. Source : China National Monitoring Center (2004), China Environmental Yearbook (2000, 2005) Note: The dependent variable in each regression is the average water grade for the listed chemical within the province. Water quality is graded on a 6-point scale, with 1 being the highest quality and 6 being the most polluted. The independent variables are the log of chemical dumping reported at the province level in China's Environmental Yearbook (2005), except Suspended Solid Waste, which is taken from Standard errors are robust.

37 Table 9 OLS Regressions of the Log of Firm Cleanup (tons) and Firm Spending on Cleanup (yuan) on the Log of the Water Pollution Levy Charge (yuan per ton) Total Industrial Cleanup (tons) Spending on Cleanup (100 million yuan) Chemical (2) (3) Log of Effective Fine Levy 0.815*** 0.137* (0.21) (0.08) Period Available R Squared N * significant at 10% ** significant at 5%. *** significant at 1%. Source : China Environmental Yearbooks ( ) Note : The effective fine levy is yuan collected per ton of wastewater discharge failing to meet regulatory standards. This is reported for all registered manufacturing firm by province and year. The sample sizes correspond to the 29 provinces in China for All regressions include province and year fixed effects and the standard errors are robust and clustered at the province level. Industrial dumping and cleanup measures are taken from the Environmental Yearbooks and gross output is taken from the Statistical Yearbooks. The dependent variable in the first column is the log of total tons of reported water tonnage of cleanup across all chemicals. The dependent variable in the second column is the log of total spending on cleanup (in 100 million yuan). Chongqing is included in the totals for Sichuan, and Tibet is excluded from this analysis since data on levies are unavailable. I censor the bottom and top 1 percent of the observed provincial levy rates (6 observations). The results are very similar whether I include or exclude these values. Regressions are weighted by the population of each province.

38 Table 10 Estimated Benefits and Costs of Raising Fine Levies Benefits and Costs of Doubling Fine Rates DigestiveCancer WaterGrade WaterGrade Dumping Dumping TaxRate Deaths Averted Per Year Extra Compliance Cost Per Year ($) Cost per Averted Death (1) (2) (3) (4) (5) (6) Overall Grade (1)*(2)*(3)*(980,000 + ) (Cost in 2000)*(0.137) (5)/(4) 0.093** 0.387*** 0.815*** 28,746 $ 504,888,032 $ 17,564 (0.044) (0.096) (0.21) * significant at 10% ** significant at 5%. *** significant at 1%. Source : China Disease Surveillance Points Mortality Registration (DSP), China National Monitoring Center (2004), China Environmental Yearbooks ( ) Note : In column 1, I report the relationship between the log of the digestive cancer rate on the overall water grade (See Table 3). In column 2, I report the relationship between the water grade and the log of total dumping (See Table 8). In column 3, I report the relationship between the log of total dumping and the effective fine levy on water dumping (See Table 9). + In column 4, I predict the number of lives saved by raising the fine levy rate, using the fact that the total annual deaths due to digestive cancers is roughly 980,000. In column 5, I predict the cost of compliance of raising the fine levy rate, using the fact that total reported compliance costs in 2001 was roughly $3.7 billion dollars (8 yuan=1 USD), and the estimated elasticity of spending on cleanup of.137 (see Table 9). In column 6, I present the cost to firms in addtional spending required to avert one additional death from digestive cancer.

39 Appendix Table 1 Sample Means (and standard deviatons) for China Disease Surveilance Points by Urbanization, Statistic Poor Rural Medium Rural Rich Rural Urban (1) (2) (3) (4) Share of Deaths Occuring in the Home (0.08) (0.18) (0.09) (0.16) Share of Deaths Occuring in the Hospital (0.05) (0.17) (0.09) (0.15) Share of Decedents Employed in Farming (0.09) (0.20) (0.23) (0.04) Average Education among Decedents (1.11) (1.33) (1.48) (1.27) Total Deaths Recorded 124, , , ,642 Total Person Years Covered 25,016,184 30,227,522 21,918,116 23,584,446 Crude Death Rate (Deaths/Persons) Number of Sites Source : China Disease Surveillance Points Mortality Registration (DSP) Note : The table above summarizes differences across the 145 sites covered by the DSP. The cites are chosen to form a nationally representative sample of deaths for China, with sites being chosen to match China's census in 1990 (see Yang et al. 2005). Employment and education for decedents restricted to deaths among persons age Total deaths recorded is for the entire sample frame from The total person years covered refers to the total number of individuals covered by each DSP site summed over the entire sample frame from

40 Appendix Table 2 Summary Statistics of Water Quality in China's Main River Systems River System Region Overall Ammonia Nitrogen Biological Oxygen Demand Oils Permanganate Volatile Phenol Liao River Hai River Huai River Yellow River Songhua River Fujian/Zhejiang Yangtze River Nu/Yarlung Zangbo Pearl River Northeast North North North Northeast South South Southwest South Inward flowing systems West Source : China National Monitoring Center (2004) Note : Higher numbers reflect lower water quality (1=best, 6=worst) and a greater concentration of the listed pollutants. These readings are taken from 484 water quality monitoring systems across China. The systems are displayed in descending order of overall water quality.

41 Appendix Table 3 Ordinary Least Squares (OLS) Regressions of Log of Digestive Cancer Rates on Water Grade by Category Statistic No Controls Digestive (all) Esophageal Stomach Liver With Controls Digestive (all) Esophageal Stomach Liver (1) (2) (3) (4) (5) (6) (7) (8) Medium Quality * * Water (1=yes) (0.190) (0.391) (0.236) (0.183) (0.130) (0.276) (0.207) (0.114) Lowest Quality 0.429** 0.953*** 0.554** 0.291* 0.323** 0.648** 0.348* 0.283** Water (1=yes) (0.170) (0.344) (0.226) (0.152) (0.146) (0.289) (0.200) (0.109) Average Education (0.042) (0.104) (0.062) (0.027) Share in Farming *** (0.271) (0.933) (0.390) (0.167) Urban Site * *** (1=yes) (0.197) (0.699) (0.261) (0.143) Air Pollution 0.282** 0.649** ** (0.137) (0.290) (0.165) (0.141) Region Controls No No No No Yes Yes Yes Yes R Squared * significant at 10% ** significant at 5%. *** significant at 1%. Source : China Disease Surveillance Points Mortality Registration (DSP), China National Monitoring Center (2004) Note : N=145. See Table 3 for a description of the control variables included in columns 4-6. Standard errors are robust and clustered at the province level. Regressions are weighted by the population at each DSP site. The water grade is measured on a 6 point scale: the base category is drinkable water (grade 1 or grade 2), the medium quality water category (grade 3) is water that is not drinkable but suitable for human contact (grade 3), and the lowest quality water category is water that is not suitable for human contact (grade 4+). The coefficient for medium quality and lowest quality are water reported relative to drinkable water.

42 Appendix Table 4 Comparison of Age-adjusted Death Rates (per 100,000) due to Cancer, China and the United States China United States (Overall) Asian Americans Men Women Men Women Men Women All Cancers Digestive Lung Other % Digestive 66% 58% 25% 22% 34% 34% % Lung 23% 17% 31% 26% 27% 20% % Other 11% 25% 44% 52% 39% 47% Source : China DSP , United States Cancer Statistics: 2004 Incidence and Mortality, Center for Disease Control (CDC) Note : Statistics for the United States reported relative to US standard population in 2000, which is not directly comparable to the rates reported for China, which are adjusted using the age distribution of the 2000 China census, and would be slightly altered if evaluated using the US standard population. The US population is older than the Chinese population: the census for both countries in 2000 reflects that the US median age (41) is 4 years older than the Chinese median age (37).

43 Appendix Table 5 Industrial Waste-water Dumping and Manufacturing Output by Industry, China ( ) Industry Dumping (10,000 tons) Percent of Dumping Output (million yuan) Percent of Output Pollution Intensity (1) (2) (3) (4) (2)/(4) Chemicals and chemical products 1,081, % 1,236, % 2.6 Ferrous metal smelting and pressing 967, % 318, % 9.2 Paper and paper products 707, % 814, % 2.6 Electricity, gas, and water production 610, % 471, % 3.9 Food, beverages and tobacco 512, % 1,512, % 1.0 Mining 377, % 523, % 2.2 Textiles 357, % 1,687, % 0.6 Mechanical and electronic equipment 296, % 4,645, % 0.2 Construction materials 215, % 757, % 0.9 Other 597, % 4,973, % 0.4 Total 5,724, % 16,942, % 1.0 Source : Chinese Environmental Yearbooks ( ), China Manufacturing Firm Database ( ) Note : Manufacturing output is based on the total value of output in yuan of firms with greater than 500,000 yuan in sales. Dumping by industry is reported for firms registered with China's State Environmental Protection Agency (SEPA).

44 Appendix Table 6 Digestive Cancer Rates and (standard errors) at DSP Sites by Region and Urbanization Region Rural Urban Difference (1) (2) (1)-(2) Northern China Southern China ** (6.24) (3.80) (7.82) (5.20) (4.11) (6.79) * significant at 10% ** significant at 5%. *** significant at 1%. Source : China DSP, China National Monitoring Center (2004) Note : Region is as defined in Table 2, with 66 sites classified as Northern and 79 sites classified as Southern. Statistics are reported weighted by the population at each DSP site.

45 Appendix Table 7 OLS Regressions of Digestive Cancer Rates on Water Grade by Period Total Male Female Total Male Female Statistic (1) (2) (3) (4) (5) (6) Water Grade (1=best, 6=worst) Average Education Share in Farming Urban (1=yes) Air Pollution Digestive Cancer Rates Digestive Cancer Rates ** 7.41** 4.83** (3.1) (4.3) (2.2) (2.8) (3.5) (2.1) (2.5) (3.6) (1.8) (2.8) (3.6) (2.1) * (15.5) (19.2) (15.1) (22.4) (28.9) (16.8) * (14.3) (18.1) (13.2) (16.2) (21.4) (11.9) ** 25.17** 7.63 (10.2) (14.5) (6.7) (8.0) (10.3) (5.9) N * significant at 10%. ** significant at 5%. *** significant at 1%. Source : China Disease Suveillance Points Mortality Registration (DSP), China National Monitoring Center (2004), Global Precipitation Climatology Center (2008), China Cancer Database (2004). Note : Columns 1-3 demonstrate the relationship between cancer rates and water quality in data from overall (1) and by gender (2-3). Columns 4-6 demonstrate the relationship between cancer rates and water quality in data from overall (4) and by gender (5-6).The number of observations varies by statistic, due to incomplete data for the period. Regressions are weighted by the population at each DSP site.