Managing Global Crop Insurance Portfolios€¦ · ©2013 AIR WORLDWIDE CONFIDENTIAL: FOR CAS...

©2013 AIR WORLDWIDE CONFIDENTIAL: FOR CAS Reinsurance Conference only

1

Managing Global Crop Insurance PortfoliosHeidi Wang, FCAS,CCMSenior Manager Business Development

Casualty Actuarial Society2013 Seminar on ReinsuranceSouthampton, BermudaJune 6-7, 2013


2

Agenda

• Crop insurance programs around the world• 2012 year-in-review • AIR agricultural risk model components• Model applications for crop reinsurance and insurance


3

Crop Insurance Markets Align with Crop Production

Corn Soybean Wheat Rice Cotton Rapeseed Sugarcane

U.S. 274 82 61.5 6.5 17 1 8

China 208 12.6 121 143 35 13.5 14.5

Brazil 72.5 83.5 4.5 8 6.5 - 37.5

India 21 11.5 95 101 25.5 7 25.5

Canada 13 5 27.5 - - 13.5 -

Russia 8 2 38 1 - 1 5

France 15.5 - 38 - - 5.5 -

Argentina 26.5 51.5 11 1 1 - 2.1

Mexico 21.5 - 3.5 - 1 - 6

Ukraine 21 2.5 16 - - 1.2 -

Australia 0.5 - 22 1 4.5 3 4.5

Source: USDA Foreign Agricultural Servicehttp://www.fas.usda.gov/psdonline/circulars/production.pdf

Major Crop Production Countries - Millions of Metric Tons*

* 2013 Projected


4

Crop Insurance Premium Is Significant and/or Growing in the Major Agricultural Countries

CountryCrop Insurance Premium

(Millions USD)

United States 11,082 MPCI + 953 for crop-hail (2012)

China 2,700 (2011)

Canada 1,100

Spain 809

India 656 (2012)

Brazil 456 (2009)

France 366

Russia 315

Argentina 240

Mexico 222Source: World Bank Survey 2008* Risk Management Agency, USDA


5

Government Involvement Makes Crop Insurance Feasible

• Impacted areas are large, especially for drought• Losses are highly correlated spatially

– Exception is hail

• Losses are frequent and can be severe for farmers• Premiums are subsidized• Administrative costs for insurers may be subsidized• Losses are limited by government• Gains may be shared with government• Governments typically determine

– Premium rates– Program changes, e.g. perils covered, types of policies– Level of protection


6

Growth of the U.S. Crop Insurance Program and Historical Loss Experience

108%

204%

162%

245%

121%

102% 102%

153%

0%

50%

100%

150%

200%

250%

300%

0

2

4

6

8

10

12

14

1981 1986 1991 1996 2001 2006 2011

Bill

ions

US

D

Total National Agricultural Premiums and Gross Loss Ratio

Agriculturalinsurancepremium

Loss Ratio

Source: National Bureau of Statistics of China


7

Agricultural Insurance Premiums and Claims in China Have Grown Significantly Since 2007

78%

75%

100%

60%

86%

74%

47%

0%

20%

40%

60%

80%

100%

120%

0

2

4

6

8

10

12

14

16

18

20

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Bill

ions

CN

Y

Total National Agricultural Premiums and Claims Sin ce 1998

Agriculturalinsurancepremium

AgriculturalinsuranceClaims

Loss Ratio

Source: National Bureau of Statistics of China


8

Weather Is the Key Source of Uncertainty Affecting Yields

% Crop Loss

Drought & Heat 37%

Excess Moisture 33%

Hail 13%

Cold, Frost & Freeze 5%

Wind & Hurricane 4%

Flood 1%

Subtotal – Directly related to weather 93%

Disease 5%

Insects & Wildlife 1%

Other 1%

Subtotal – Other perils 7%

Total 100%


9

0

10

20

30

40

50

60

70

80

90

100

Percent Saturated

Development of Sub-Surface Moisture Content for 2012 Crop Season

Oct 2011 Nov 2011 Dec 2011 Jan 2012

Apr 2012May 2012 Jun 2012 Jul 2012

Aug 2012Sep 2012


10

The 2012 U.S. Drought Event


11

AIR’s Estimated Losses for the 2012 Drought Compare Well to RMA Data

Range Gross Premium Gross Loss Gross Loss -Premium

Gross Loss Ratio

Upper USD 11 Billion USD 17.2 Billion USD 6.2 Billion 156%

Lower USD 11 Billion USD 15.3 Billion USD 4.3 Billion 139%

Range Retained Premium Net Loss Net Loss -Premium

Post-SRA Loss Ratio

Upper USD 9.3 Billion USD 10.9 Billion USD 1.6 Billion 117%

Lower USD 9.3 Billion USD 10.1 Billion USD 786 Million 108%

AIR 2012 crop insurance industry loss estimates as of November 2012 (Source: AIR)

Gross Loss Ratio (Observed)

2012 153%*

*Source: RMA

AIR’s estimated gross loss ratio range (November 2012) agrees well with the

RMA value (April 2013):


12

DAMAGE

Crop Performance Is a Physical Process and Can Be Modeled in a Fashion Similar to Other AIR -Modeled Perils

HAZARD

FINANCIAL

Insured Loss Calculations

Policy Conditions

Government Reinsurance

Private Reinsurance

Crop Damage /

Yield Estimation

Crop Insurance Program

Crop Requirements

Precipitation

Temperature

Terrain Elevation

Soil

Land Use Land Cover

(Satellite)

Exposure Information

Commodity Price Risk

Stochastic Event Set CatalogAWI

(Agriculture Weather Index)

Risk

Historical Yields

Typhoon Catalog

Historical Prices


13

In a Normal Year, Water Supply and Water Requiremen ts Are Balanced and AWI Indicates Positive Yield Outco me

Eva

potr

ansp

iratio

nR

unof

fS

oil M

oist

ure

Fav

orab

le fo

r Y

ield

sD

etrim

enta

l to

Yie

lds

Apr May Jun Jul Aug Sep Apr May Jun Jul Aug Sep


14

In a Normal Year, Water Supply and Water Requiremen ts Are Balanced and AWI Indicates Positive Yield Outco me

Eva

potr

ansp

iratio

nR

unof

fS

oil M

oist

ure

Det

rimen

tal t

o Y

ield

s


Fav

orab

le fo

r Y

ield

s


15

In a Drought Year, Water Requirements Exceed the Water Supply and AWI Indicates Plant Damage

Run

off

Eva

potr

ansp

iratio

nS

oil M

oist

ure

Fav

orab

le fo

r Y

ield

sD

etrim

enta

l to

Yie

lds



16

AWI Measures County-level Crop Performance During the Season

AWI proportional

to flood damage

AWI indicates significant

drought damage

End of Season AWI estimates

yield variation from normal

AW

I

Calendar Days


171990 1995 2000 2005 2010

8010

012

014

016

018

020

0

x

y

20081990 1995 2000 2005 2010

8010

012

014

016

018

020

0

x

y

20081990 1995 2000 2005 2010

8010

012

014

016

018

020

0

x

y

2008

AWI Removes Weather Variability from the Historic Yields to Reveal True Technology Trend

Yield Time Series

Trend Line Influenced by Weather Variability

True Technology Trend with Weather Variability Removed

17 Bushels

Yie

ld -

Bus

hels

/Acr

e


18

The Choice of Methodology Can Have a Large Impact on Insurance Contract Risk

Probability Density Function of Detrended Yield Time Series

100 150 200 250

0.00

00.

005

0.01

00.

015

Without AWI With AWI

100 150 200 250

0.00

00.

005

0.01

00.

015

100 150 200 250

0.00

00.

005

0.01

00.

015

100 150 200 250

0.00

00.

005

0.01

00.

015

Yield – Bushels/Acre

Pro

babi

lity

Yield Guarantee

100 150 200 250


19

… and all Other Crops Modeled in Detail as well as a ll Potential Interactions

Prices

Corn, Cotton, Sorghum,

Soybeans, Wheat

Historical Time Series on

National Level

Yields

Corn, Cotton, Sorghum,

Soybeans, Wheat

Correlations

Prices

Yields

Prices - Yields

Price / Yield Scenarios

Corn, Cotton, Sorghum, Soybeans, Wheat

Stochastic Catalog

Price Volatility


20

The Price Risk Component Takes Into Account Correla tions Between Total US Production and Harvest Price

40%

60%

80%

100%

120%

140%

160%

180%

60% 70% 80% 90% 100% 110% 120% 130%

Har

vest

Pric

e as

Per

cent

age

of

Pla

ntin

g P

rice

Yield as Percent of Normal

CORN - PRICE RATIOStochastic Catalog Historical

1983

1993

2010

2008

1988

2012


21

County Level Results Are Aggregated to State Level and Subject to SRA Rules and Regulations

6%

Assigned RiskFund

(Each State)after 20% retention

Commercial FundGroups 2 and 3

(Each State)after 35%+ retention

Commercial FundGroup 1 States #

(Each State)after 35%+ retention

500(max)

220

160

100

65

50

Loss Ratio

Company Share of:

Loss

Gain

Portion

Government Share of:

Loss

Gain

# Group 1 States: Illinois, Indiana, Iowa, Minnesot a, Nebraska

Net Book Quota Share of 6.5%Gain/Loss Ceded to Government

7.5%

3%

13.5%

3%

22.5% 97.5%

5%

42.5%

5%

40%

20%

10%

45%

65%

75%

40%

5%


22

AIR Agricultural Model Applications in CATRADER ®

AIR Agricultural Model for

Crop Insurance and Reinsurance

Enterprise Risk

Management

Pricing &Actuarial

Underwriting

Fund Designation

Claims

Risk Transfer


23

Summary

• Crop insurance and reinsurance are increasingly important lines of business worldwide

• Weather is the most important peril for agricultural portfolios; therefore, crop risk models need to account for the weather impact on crop yields

• AIR MPCI models are weather-based and maintain the natural geographical yield correlations that are key for unbiased crop portfolio analyses

• In addition to U.S. and China MPCI models already available, future models will include U.S. crop-hail, India and Brazil/Argentina

Managing Global Crop Insurance Portfolios€¦ · ©2013 AIR WORLDWIDE CONFIDENTIAL: FOR CAS...

Documents

Transcript of Managing Global Crop Insurance Portfolios€¦ · ©2013 AIR WORLDWIDE CONFIDENTIAL: FOR CAS...