Advanced Business Risk Management Programs Around the World

22
Advanced Business Risk Management Advanced Business Risk Management Programs Around the World Barry J. Barnett Department of Agricultural Economics Mississippi State University November 17, 2010 University of Guelph Policy Education Program University of Guelph, Policy Education Program Guelph, Ontario

Transcript of Advanced Business Risk Management Programs Around the World

Page 1: Advanced Business Risk Management Programs Around the World

Advanced Business Risk Management Advanced Business Risk Management Programs Around the World

Barry J. BarnettDepartment of Agricultural EconomicsMississippi State University

November 17, 2010University of Guelph Policy Education ProgramUniversity of Guelph, Policy Education Program

Guelph, Ontario

Page 2: Advanced Business Risk Management Programs Around the World

C CChallenges with Loss-Based Crop Insurance

Asymmetric information.Moral hazard.Adverse selection.

High transaction costsHigh transaction costs.Spatially correlated risk.Typically requires large premium subsidies (U STypically requires large premium subsidies (U.S., Canada, Spain, etc.)

2

Page 3: Advanced Business Risk Management Programs Around the World

Index InsuranceInsurance product that makes payments based on the 

li d l f ifi d d l i i d threalized value of specified underlying index rather than on losses experienced by the policyholder.Common indexes.Aggregate yield measure.Yield proxy (e.g., weather variable) – can be structured to protect against too little or too much of theto protect against too little or too much of the underlying weather variable.

3

Page 4: Advanced Business Risk Management Programs Around the World

Index Insurance

E l li t t i t i ffi i tExample:  policy protects against insufficient rainfall measured at specific weather station over a three month period.pThreshold = 100 mmLimit = 50 mmS i d $10 000 ti k i i $200Sum insured = $10,000 so tick‐size is $200 per mm

If realized rainfall = 80 mm, indemnity = $4,000If realized rainfall = 50 mm indemnity = $10 000If realized rainfall = 50 mm, indemnity = $10,000If realized rainfall = 40 mm, indemnity = $10,000

Page 5: Advanced Business Risk Management Programs Around the World

Index Insurance

Little potential for adverse selection or moral hazard.e po e a o ad e se se ec o o o a a a dEffectively transfers spatially correlated risk exposure.Low transaction costs relative to loss‐based insurance (no need for policyholder‐specific risk assessment or loss adjustment).

5

Page 6: Advanced Business Risk Management Programs Around the World

f CExample from CanadaProduction Insurance (PI) forage rainfall plan (protects 

i t i ffi i t i f ll)against insufficient rainfall).Underlying index is weighted average rainfall measured at a given weather station from May through August.Purchaser selects weather station and coverage option (that assigns weights to each month).Payments are triggered if realized weighted averagePayments are triggered if realized weighted average rainfall is less than 80 percent of the long‐term average.Realized daily rainfall is capped at 50 mm.  Realized monthly rainfall is capped at 125 percent of the long‐monthly rainfall is capped at 125 percent of the long‐term average.Sum insured varies from $25 to $300 per acre depending on type and value of the forage stand

6

on type and value of the forage stand.

Page 7: Advanced Business Risk Management Programs Around the World

f SExamples from U.S.Rainfall Index Pilot Program (insufficient rainfall).U d l i i d i 12 il b 12 il id i fUnderlying index is a 12 mile by 12 mile grid estimate of rainfall based on weather station, satellite, and radar data.

Vegetation Index Pilot Program (insufficient rainfall).Underlying index is a 4.8 mile by 4.8 mile grid measure of the Normalized Difference Vegetation Index (NDVI)of the Normalized Difference Vegetation Index (NDVI).

Purchaser selects coverage level, period of coverage, and the sum insured per acre (within limits).Pasture, rangeland, and forage and apiculture.For 2010, $415 million sum insured (mostly in rainfall index)

7

index).

Page 8: Advanced Business Risk Management Programs Around the World

f SExamples from U.S.Group Risk Plan (GRP) based on county yield and G Ri k I P t ti (GRIP) b dGroup Risk Income Protection (GRIP) based on product of county yield and futures market price.Purchaser selects coverage level and sum insured per g pacre (within limits).GRIP is available for corn, cotton, grain sorghum, soybeans and wheat GRP is also available for barleysoybeans, and wheat.  GRP is also available for barley,  oysters, and sugarcane.For 2010, $4.2 billion sum insured (mostly in GRIP) out ( y )of almost $78 billion total sum insured in federal crop insurance program.

8

Page 9: Advanced Business Risk Management Programs Around the World

Weather Index Insurance Basis Risk

For any given magnitude of the underlying index  (e.g., rainfall measured at a weather station), there is a conditional probability distribution for the policyholder’s exposure to the same variable (e.g.,policyholder s exposure to the same variable (e.g., rainfall measured at the policyholder’s location).

index               variable measured at farm For any given magnitude of the underlying weather variable measured at the farm, there is a conditional probability distribution of loss.

variable measured at farm               loss  

9

Page 10: Advanced Business Risk Management Programs Around the World

Pricing and Triggering Indemnities

Sufficient quantitative data of an appropriate spatial specificity are required to price the insurance.How much is sufficient?  It depends on the temporal presentation of the riskpresentation of the risk.Is the probability distribution of the index stationary?Is the variance of the probability distribution constant p yover time?Do either of the above exhibit multi‐year cycles?

Understanding the tail of the distribution is critical.

10

Page 11: Advanced Business Risk Management Programs Around the World

S fSahelian Rainfall

11

Page 12: Advanced Business Risk Management Programs Around the World

S fSahelian Rainfall

12

Page 13: Advanced Business Risk Management Programs Around the World

Extreme Weather and Rural Poverty

More than 1 billion people currently live on less than one US dollar per day.Approximately 75 percent of those live in rural areas.Whil l h h ld i i iWhile rural households engage in various income generating activities, these activities are often tied directly or indirectly to weather‐sensitive industriesdirectly or indirectly to weather sensitive industries such as agriculture. In many rural areas the risk of extreme weather events contributes to poverty and low rates of economic growth.

13

Page 14: Advanced Business Risk Management Programs Around the World

Weather Index Insurance in Lower Income CCountriesOver the past few years, weather index insurance has 

i d l t f tt ti f i t ti lreceived a lot of attention from international development researchers and practitioners.Currently, there are between 30 and 40 projects on y, p jweather index insurance being conducted in countries around the world.

Funding has come from the ADB IADB UNDP USAIDFunding has come from the ADB, IADB, UNDP, USAID, the Ford Foundation, the Gates Foundation, GTZ, DFID, etc.

C f i d i h ld f lConferences on index insurance are held frequently around the world.

14

Page 15: Advanced Business Risk Management Programs Around the World

SOENSO Index Insurance

El Niño (years with high Pacific sea surface temperature) are associated with extreme flooding in northern Peru. Index insurance based on sea surface temperature.  

15

Page 16: Advanced Business Risk Management Programs Around the World

ñ SO 1 2 SSEl Niño – ENSO 1.2 SST Anomalies

1982

16

Page 17: Advanced Business Risk Management Programs Around the World

fRainfall in Northern Peru Relative to Normal

1998

17

Page 18: Advanced Business Risk Management Programs Around the World

ñEl Niño Flooding in Northern Peru

18

Page 19: Advanced Business Risk Management Programs Around the World

Weather Index Insurance in Lower Income CCountries

Data limitations are perhaps the greatest challenge.Density of weather stations is very sparse in many lower income countries – especially in Africa.O t id f S th Af i th l bli l il bl d ilOutside of South Africa, the only publicly available dailyweather station data for many African countries is from major airports.More stations exist that report less frequently but density is still sparse – especially in rural areas.

19

Page 20: Advanced Business Risk Management Programs Around the World

SAlternative Data Sources

Spatial interpolation of available weather station data.Satellite‐based Normalized Difference Vegetation Index (NDVI).S lli b d f i f llSatellite‐based measures of rainfall.Satellite‐based synthetic aperture radar (SAR) maps contours of geospatial environments (e g flooding)contours of geospatial environments (e.g., flooding).Reanalysis data.

A class of data products that combines and calibratesA class of data products that combines and calibrates observations from many sources — weather stations, satellites, weather balloons, etc.

20

Page 21: Advanced Business Risk Management Programs Around the World

f SAssessment of Alternative Data Sources

Satellite‐based measures of weather variables still have significant errors when compared to ground‐based measures.Many alternative sources tend to understate outliersMany alternative sources tend to understate outliers.Insufficient spatial and/or temporal specificity (possible exception is NDVI).(possible exception is NDVI).Limited time series of data and challenges with calibrating observations across evolving technologies.Will potential buyers purchase an insurance product based on satellite data?

21

Page 22: Advanced Business Risk Management Programs Around the World

CConclusion

Index insurance has potential to provide insurance coverage for crops/regions that are not well suited to traditional loss‐based crop insurance.Index insurance addresses many of the problems withIndex insurance addresses many of the problems with loss‐based insurance (moral hazard, adverse selection, high transaction costs).gChallenges:

Basis risk.Data limitations.

22