Transcript risk_study_0904

An assessment of farmer’s exposure to
risk and policy impacts on farmer’s risk
management strategy
Shingo Kimura
Trade and Agriculture Directorate
4 September 2009
113th EAAE seminar
Main objectives of the farm level analysis of risk
 What are the sources of risks that individual farmer is facing
and what kind of correlations are they making use of?
- Statistical analysis of risk exposure of individual farm from
micro data (crop farm in Germany)
 How farmers respond to different risk related measures and
what kind of trade-offs that the government and farmer face in
terms of risk and welfare?
- Simulation analysis of risk management strategies and policies
OECD Trade & Agriculture Directorate
2
An assessment of farmer’s exposure to risk
- Statistical analysis -
OECD Trade & Agriculture Directorate
3
Micro data from Germany
 Data source
German national FADN data
 Selection criteria
Single crop farms that cultivated more than 50 ha and have more than 60% of
production from crops between 1995/96 and 2006/07 (12 years)
 Sample size
A panel of 232 crop farms are identified from 3 regions (77 from North , 41 from
Centre/South and 141 from East)
 Data coverage
Yield, output price and planted area for 6 crops
Costs, revenue , net farm income and labor inputs for a whole farm
 Statistical information
Mean and variability (distribution of variance-covariance matrices) by each region
OECD Trade & Agriculture Directorate
4
Risk exposure: Yield variability
Higher yield variability at farm level data
- Yield is location specific and favorable yield in one place is offset
by the unfavorable yield in another place in the aggregated data
(= special aggregation bias).
Coefficient of variation of wheat yield
Centre
All
East / South North regions
0.00
0.05
0.10
0.15
0.20
0.25
0.30
Farm level
Aggregate
Farm level
Aggregate
Farm level
Aggregate
Farm level
Aggregate
*The bracket indicates the mean plus minus the standard deviation of the coefficient of variation across farms.
OECD Trade & Agriculture Directorate
5
Risk exposure: Price variability
Higher price variability at farm level data, but to less extent
- Price is less location specific and the integration of output
market equalize the price across locations.
- If compared at the aggregated level, price risk may seem
larger than the yield risk
Coefficient of variation of wheat price
OECD Trade & Agriculture Directorate
6
Risk exposure: Price-yield correlation
Lower yield-price correlation at farm level data, but still significant
(around -0.2).
-Highly negative correlation may not improve the welfare of risk
averse farmer because it may reduce the expected revenue.
Coefficient of correlation of wheat price and yield
Values are expressed in negative terms
OECD Trade & Agriculture Directorate
7
Risk management strategies and policies
-Stochastic simulation analysis-
OECD Trade & Agriculture Directorate
8
Stochastic simulation model
Farmer maximize his expected utility subject to profit function
~   )(1  )
(

U (~   ) 
(1   )
~ 
 degree of constant relative risk aversion (=2)
 initial wealth (=2664 euro per ha)
6
~
~  c ) * L ]  g( ~
~ , )
[(
p
*
q
p
,
q
 i i i i
i
i
i 1
~
pi uncertain yield of crop i
~
q
i
ci
Li
g

uncertain output price of crop i
variable production cost of crop i
area of land allocated to crop i
transfer from government or benefit from risk market strategy
level of coverage decided by farmer
Farmer’s welfare is computed as certainty equivalent of profit
CE  [(1   ) EU (~   )]1/(1  )  
OECD Trade & Agriculture Directorate
9
Incentive to use risk market strategy:
Crop yield insurance
Proportion of planted area insured
S_Barley
S_beet
W_barley
Wheat
Oilseeds
Rye
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0%
2%
4%
6%
8%
10%
Cost of insurance (% additional cost to fair insurance premium)
OECD Trade & Agriculture Directorate
10
Risk and welfare impact of risk market strategy:
Crop yield insurance
CE profit
CV of profit
CV of per hectare revenue
610
19
609
608
17
607
Euro
606
16
605
15
604
603
14
602
13
601
600
Coefficnet of variation (%)
18
12
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
Cost of insurance (% additional cost to the fair insurance premium)
OECD Trade & Agriculture Directorate
11
Comparison of policy impacts
Impacts of 2 euro expected subsidy through different instruments
Certainty equivalent profit (change in
euro)
Contributing factors
Overall
change
Change in
mean
Change in
variability
CV of profit
(change in
percentage
points)
Single farm payment
2.00
2.00
0.00
-0.06
Subsidy to crop yield insurance
premium
0.39
-0.48
0.87
-2.19
Subsidy to forward price
0.03
-0.11
0.14
-0.35
* Cost of insurance and forward contracting are assumed as 10% and 5%, respectively.
OECD Trade & Agriculture Directorate
12
Interaction between single farm payment and
crop yield insurance strategy
Proportion of planted area insured
Wheat
Oilseeds
S_Barley
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
50
100
150
200
250
300
350
400
Single farm payment per hectare (Euro)
450
500
* Insurance price premium is assumed to be 3%.
OECD Trade & Agriculture Directorate
13
Some preliminary findings
 Farmer faces higher yield variability, and lower negative
yield-price correlation than it looks at the aggregated
data
 Farmer is making use of correlations and crop
diversification to stabilize income
 Participation to risk markets is difficult even for
relatively low administrative costs
 Interaction between government programme and risk
management strategy (e.g., impacts on crop
diversification and crowding out of risk market
instruments)
OECD Trade & Agriculture Directorate
14
Thank You
Trade and Agriculture Directorate
Visit OECD’s risk management project website:
www.oecd.org/agriculture/policies/risk
Contact us:
[email protected]
OECD Trade & Agriculture Directorate
15