Farm Level Risk Management Using Models: VaR and Other Models

1
Farm Level Risk Management Using Models VaR and
Other Models

• Jim Unterschultz
• Rural Economy, U. of A.
• Support IAMF, Agriculture Food Council, AAFRD
  and GE3LS

2
Overview

• Risk
• Primary Agriculture
• Models
• VaR (SR)
• Real Options (LR)
• Other Models

3
Risk

• Uncertain Consequences

Hardacker 1997
4
Risk

• Business
• Production
• Market
• Institutional (policy)
• Human
• Financial
• Debt, Equity

5
Risk Management

• Identify
• Measure
• manage

6
Agriculture and Risk
7
How Many Commercial Grain Farms Are Needed in
AB?

• 32 million acres of crop land in AB
• If size up gt complexity up gt specialized
  functions increase

8

• 2000 Sow9 M

9
Alberta Saskatchewan Feedlot Bunk Capacity
(Head)
21
22
57
Source Canfax 2003 Annual Report
10
Primary Agriculture

• Increasing in Size
• Investment multi-millions
• Business Risk Exposure Specific to Sector
• Diversification may not be feasible

11
Models

• Simplification
• Specific application
• Use when useful

12
Models May Help With

• Identify risk
• Measure risk
• Manage risk

13
Risk
Low
High
Grain, C-C, TSX
T-Bills
Dairy, Broiler
Feedlot
Pork
Biotech startup
0
15
18
25
40
75
Total Risk Approximate Measure Volatility
14
Risk Management

• Identify Build Model
• Measure Run Model
• Manage Test Alternatives

15
Risk, Decision Time Frame and Models

• Short Run
• Cash Flow Management
• VaR (CFaR) Models

• Long Run
• Capital Budgeting
• Real Options Models

16
Value at Risk/Cash Flow at RiskWhat is it and
How Can it be Used?

• Joffre Hotz
• Graduate
• Student

17
What is Value at Risk (VaR)

• summarizes the worst loss over a target time
  horizon for a given probability level

5 VAR
-20,000
Jorion, 2001
18
VaR Example

• A 1000 sow operation has a 20 (1 in 5) chance
  that its cash flow from operations for the next 6
  months will be below -249,500
  (-25.00/hog sold).
• OR
• Over the next 6 months, the probability that
  cash flow from operations will be negative is
  78.

19
Operation Type Model

• 1000 sow farrow-to-finish operation in Southern
  Alberta.

20
Identify Pork Producer Business Risks
21
(No Transcript)
22
Measure Individual Risk Sources

• Severity (range of outcomes)
• Chance (probability occurring)
• Barley price, pork price, production variability

23
CFaR Results First 6 Months 2003

• simulate model 10,000 times
• 20 CFaR and
• probability cash flow is lt 0

24
CFaR Use?

• Benchmark
• compare loss risks across different time periods
• Estimate working capital required
• Risk Management Strategies

25
Manage Risk

• three price risk management strategies
• hog prices
• barley prices
• combination

26
(No Transcript)
27
CFaR Model

• identify the potential downside cash flow
• Compare risk management strategies
• Negotiating strategies

28
Real Option Model Capital Budgeting
Long Run Investment Decisions
29
Standard Cash Flow Investment Model Problems (NPV)

• trouble capturing the value managerial strategic
  decision making adds to an investment

30
Real Options Key Points

• NPV First
• Identify Risks Key Decision Points Through Time
• These have option value
• Build appropriate models
• Value Real Options (From Finance)
• Decision?

31
Case Study Pork Investment

• Background
• 2600 sow farrow-finish
• 8 to 9 M investment
• Duku-Kaakyire graduate student

32
RO Approach

• IdentifyNPV RegressionMonte Carlo
• Measure Option Pricing Models with Cash Flow
  model
• Manage Decision Scenarios

33
Pork Case Study Results

• NPV -1.50 Million
• Real Option
• Abandon Grow 1.6 to 2.0 M
• Staged Investment 1.2 to 1.6 M
• Conclude Invest

34
Applicability of Real Option Models

• Biotechnology Growth
• Irreversible Investments (Pork)
• Identification of options very useful even if
  only doing NPV
• Deng Yu (Cattle), Emmanuel Laate (Bio-technology)
  

35
Other Models

• Value of Riparian Areas in Cattle Ranch
• (Jamie Miller)

36
Wetland Drainage in Cropland (Brett Cortus)
37
AAFRD CropChoice

• computerized budgeting tool to evaluate potential
  risk management strategies.
• Traditional crop budgeting format
• Statistical, mathematical, and computer
  simulation techniques
• 20 years of crop yield and price data
• Premiums, payouts, and underlying logic from crop
  insurance
• Available on the Internet
• Make 2004 cropping decisions.

Source Ted Darling AAFRD
38
Business Level Models Conclusions

• Models a tool to
• Identify risk
• Measure risk
• Assist in decisions to manage risk
• Short Run CFaR
• Long Run Real Options
• Development intensive
• Interpretation simpler