Ke Wang, 1 Qiao Zhang, and 2 Shingo Kimura

Transcription

1 Prepared for 9 th OECD farm level analysis meeting, 11 June Ke Wang, 1 Qiao Zhang, and 2 Shingo Kimura 1 Agricultural Information Institute, CAAS 2 Trade and Agriculture Directorate, OECD

2 Outline Introduction Objective Data & Method Results Conclusion Discussion

3 Introduction For the possible global climate change, agricultural risk management has became a hot issue in recent years around the world. An effectiveness risk management need more work focused on risk environment analysis/assessment because risk assessment is a vital step in the process of risk management. AND, It is important to assess agri. risk using farm-level data because the assessment based on aggregation data can underestimate the real risk that farmer faced (Coble and Barnett, 2008; Coble and Dismukes, 2007; OECD, 2010; et al.)

4 Introduction Crop Insurance has considered to be an effective and marketoriented risk management tool for agri. production risk, and has been implemented in more than 40 counties including China. Chinese agri. insurance program has made a great progress in recent year. In 2007, the premium volume was about 830 M USD, the number has raised to 1.7 B USD in 2008 and 2.2 B USD in However, less studies evaluate the effectiveness of Chinese agri. insurance program, especially in farmer s view.

5 Objective To analyze the risk environment of Chinese crop famers using farm-level data. To evaluate the effectiveness of Chinese agricultural insurance program in the view of farmer s welfare

6 Data & Method Sample Region Five Provinces: Shan Dong, Jiang Su, He Nan, ShannXi, and Si Chuan Sample farmers: Coming from 12 counties, which was shown in right map as red marks.

7 Sample Description Description of data used for farm-level analysis Province Jiang Su Shan Dong He Nan Si Chuan Shan Xi Total Farm Type Crop Farm Crop Farm Crop Farm Crop Farm Crop Farm Crop Farm Major Commodity Corn, Oilseed, Rice, Wheat Corn, Oilseed, Wheat Corn, Oilseed, Wheat Corn, Oilseed, Rice, Wheat Corn,Wheat Num of Sample Length of the data 6 year ( ) 6 years ( ) 7 years ( ) 7 years ( ) 7 years ( ) 6 or 7 years Source: Individual farmers data are obtained from China Rural Fixed Observation Office, MOA, China

8 Method In this paper, the basis model for evaluating farmer s welfare under risk environment is expected utility function Power utility function, CRRA(ρ)=2 (C Turvey, H Wang et al)

9 Method The variable values are generated by Monte Carlo approach according to the character of reprehensive farmer in each province. With Crop Insurance π = P i q i L i r L W A n I + G prem + Indem

10 Method π = P i q i L i r L W A n I + G prem + Indem Where, Indem = max yc y,0 CL yc area if y 0 CL 0.7 area if y = 0 CL depicts Coverage level (Yuan/Mu) yc means the guaranteed yield, equal to 0.7 time Yf in China y indicts the actual yield. The policy details (premium and coverage level) are calculated according to the current policy of crop insurance program in China

11 Results Summary of sample data Variable area (Mu) Yield (Kg/Mu) price (Yuan/Kg) It can be seen that the planting size of each crops in China are quite small. JiangSu Shan Dong He Nan Si Chuan Shan' Xi mean sd mean sd mean sd mean sd mean sd corn oilseed na na rice na na na na na na wheat corn Oilseed has the lowest yield, but has the highest price. oilseed na na rice na na na na na na wheat corn oilseed na na rice na na na na na na wheat Cost(Yuan) labor(yuan) Family Income Revenue from off-farm Revenue from crop Other Revenue from farm Profit to plant crops are very low.

12 Risk environment: Price risk and yield risk CV of Yield As expected, CVs of yield in farm level are higher than CV in Agg-level. Assessing yield risk using aggregation data do underestimate the true risk CV of Price CV of Price in farm-level are higher than that in Agg level But, the gaps are more smaller. because price risk are more systemic

13 Risk environment: Price risk VS. yield risk As shown in left fig., it seems price risk is more serious than yield risk but in farm level, the CV of yield risk is as close as, even bigger than that of price risk.

14 Risk environment: Correlation between price and yield corn oilseed rice wheat Correlation of Yield and Price JS SD HN corn oilseed rice wheat corn oilseed wheat SC SX Total rice As expected, correlation in Agglevel is higher Most Case, there are negative relationship between yield and price. corn oilseed rice wheat corn -.39 oilseed rice wheat wheat Farm level corn oilseed rice Agg level Positive relationship exists. Because the yield is not big enough so that they can only accept price. Graphs by Code of provice

15 Risk environment: Correlation between income components Farmer Correlation income between vs. Off Return farm and Cost Farmer income vs. Return from crops

16 Crop Insurance Current crop insurance programs in sample provinces Province Coverage Level (Yuan/mu) Premium Ratio (%) Corn Oilseed Rice Wheat Corn Oilseed Rice Wheat JS SD HN SC SX Crop insurance in China are now only cover the physical cost to plant the crops 2. The proportion of subsidy is 80% for most crop and most region in China

17 Crop Insurance Effective analysis for China s Crop Insurance Program Note: WCI= Without Crop Insurance ; CI_NS= Crop Insurance, but non-subsidy CI_S= Crop Insurance With Subsidy ; DP= Directly Pay the subsidy fund to farmers instead of by crop insurance

18 Effective of China s Crop Insurance program (Cont.) With the increase of subsidy level, Farmer s CE with CI will also be higher. However, very surprisingly, farmer s CE with CI in JS and SD are even lower than that of WCI even the subsidy level is higher to 90%. Possible reason are: 1) Farmer s income are mostly come from non-farm in the two provinces; 2) the coverage level of Crop insurance are very low in China.

19 Effective of China s Crop Insurance program (Cont.) For the farmers in middle and west region of China, CI is better than WCI provided subsidy is available. (65% and 20% premium subsidy are required in SC and SX, respectively ) CI are also need to be improved for its contribution to farmers welfare are less than does DP.

20 Conclusion The results shows that it is necessary to conduct agricultural risk assessment using farm-level data. Gov. Subsidy is necessary to implenmat the crop insurance program because the crop insurance without subsidy will decrease farmer s welfare Although the Effectiveness of crop insurance are different in 5 provinces, Chinese crop insurance program are need to be improved.

21 Discussion and future works Only 7 years time series data were used to assess individual farmer s risk character. We need more data to test the conclusion. Diversification is an important strategy for most farmers to stabile their income. But we ignore the correlation between different crops in simulating reprehensive farmer s household data. That need to be improved in future works. The characters of crop insurance program in 5 provinces are simplified in our study. More robust work need to be conducted to analyze the effectiveness of Chinese crop insurance program.

22 Comments and questions are welcome to