See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/241750272

Estimating the Fair Insurance Premium for Dungeness Crab Yields in the
Western U.S. Coast

Article · January 2011

CITATIONS READS

0 33

3 authors, including:

James W. Richardson
Texas A&M University
280 PUBLICATIONS   2,027 CITATIONS   

SEE PROFILE

All content following this page was uploaded by James W. Richardson on 14 July 2015.

The user has requested enhancement of the downloaded file.

Estimating the Fair Insurance Premium for Dungeness
Crab Yields in the Western U.S. Coast

Chia-Lan Liu, James W. Richardson, David J. Leatham

Department of Agricultural Economics, Texas A&M University, College Station, TX

Poster prepared for presentation at the Agricultural & Applied Economics

Association’s 2011 AAEA & NAREA Joint Annual Meeting, Pittsburgh,
Pennsylvania, July 24-26, 2011

Copyright 2011 by Chia-Lan Liu, James W. Richardson, and David J. Leatham. All rights reserved.
Readers may make verbatim copies of this document for non-commercial purposes by any means,
provided that this copyright notice appears on all such copies.
 Estimating the Fair Insurance Premium for Dungeness Crab Yields in the Western U.S. Coast
View publication stats

Chia-Lan Liu, James W. Richardson, David J. Leatham

Department of Agricultural Economics, Texas A&M University, College Station, TX

Introduction and Motivation Results Table 2. Goodness-of –Fit measures and Ranking of Alternative Distributions
Distributions and ranks
The Dungeness is a popular food and the most Tests
Table 1. Yield Data Summaries, 1950-2009 States LogLogistic Lognormal Logistic Normal Weibull Gamma
commercially important crab in the western states in the U.S.
Like all agricultural production, the crab fisherman face yield Time Series Data Detrended Data Chi-Square 1 2 3 4 5 6
risks and must manage these risks. In addition to weather risk, State obs.(n) Mean Std. Dev. Skewness kurtosis Mean Std. Dev. Skewness kurtosis A-D statistic 1 2 3 5 4 6
crab fisherman may experience low yields if the crabs are Alaska K-S Statistic 1 2 3 5 4 6
over fished in previous years. Farmers for many traditional Alaska 59 6,518,320 3,443,387 0.73 0.04 0 2,436,552 1.14 2.19
Sum 3 6 9 14 13 18
agricultural crops can purchase crop insurance to insure Washington 58 12,728,635 7,213,322 0.93 0.38 0 4,947,707 0.60 0.95
against low yields. However, crab fishermen at this time do Rank 1 2 3 5 4 6
Oregon 59 10,685,030 6,048,295 1.47 2.80 0 4,943,273 1.21 2.49
not have this option. The purpose of this paper is to estimate Chi-Square 4 5 3 1 2 6
California 59 10,564,518 6,308,023 1.16 2.35 0 5,786,362 1.21 2.34
a fair insurance premium based on the historical yields of the A-D statistic 4 1 2 6 5 3
Average 58.75 10,124,126 5,753,257 1.07 1.39 0 4,528,474 1.04 1.99
Dungeness crab. This information can then be used in Washington K-S Statistic 2 3 1 6 5 4
risk/return models for crab fishing to determine if it would be Minimum 58 6,518,320 3,443,387 0.73 0.04 0 2,436,552 0.60 0.95
sum 10 9 6 13 12 13
optimal for fisherman to purchase crop insurance. Maximum 59 12,728,635 7,213,322 1.47 2.80 0 5,786,362 1.21 2.49
rank 3 2 1 5 4 5
An important input into the fair insurance premium 5.0% 3.4%
0.01
5.1% 3.4% Chi-Square 1 3 2 5 4 6
91.6% 91.5%
estimation is the yield distribution. Sherrick et al. estimated
10.2% 1.4% 11.9% 1.4%
Normal
2 88.4% 0.008 86.7% A-D statistic 1 2 3 5 4 6
(0,2457467)

Values x 10^-7
alternative yield distributions to evaluate traditional crop Normal 0.006
Oregon K-S Statistic 1 3 2 5 4 6
insurance. However, no one has looked at the yield 1 0.004 Mean 0
sum 3 8 7 15 12 18
distributions for the Dungeness crab nor explored possible 0.002 Input Std. Dev. 2457467
crop insurance. Much of the past literature for the fishing 0
-3.12 5.40
0 -2.90
Normal
5.40 Skewness 0 rank 1 3 2 5 4 6
-6 2 10 -6 -2 2 6 10
Values in Millions Kurtosis 3 Chi-Square 1 2 5 3 4 6
industry has focused on production functions, cost function Values in Millions

3.4%
5.0%
5.1% 3.4%
models, and optimal catching yields for specific fish species. 9.9% 91.6%
88.8%
1.3% 0.01
11.5% 91.5% 1.3% Logistic A-D statistic 2 1 4 5 3 6
2 87.2%
Moreover, most research has focused on the endangered Values x 10^-7
0.008
(-246001,1301148) California K-S Statistic 1 2 4 5 3 6
0.006
Logistic
commercial ocean species such as tuna and swordfish. 1 0.004 Mean -246001 sum 4 5 13 13 10 18
0.002 Input Std. Dev. 2360021
Logistic rank 1 2 4 4 3 6
Data and Method
0
0
-3.12 5.40
-2.90 5.40 Skewness 0
-8 1
Values in Millions
10 -8 -4 0 4
Values in Millions
8
Kurtosis 4.2 sum 20 28 35 55 47 67
The industry
5.1%
13.1% 91.5%
3.4%
2.1% 0.01
5.1%
13.1% 91.5%
3.4%
2.1%
Weibull rank 1 2 3 5 4 6
2 84.8% 0.008 84.8% (2.7055,7142890,
Values x 10^-7

Weibull 0.006 Shift(-6359880))

1 0.004
Mean -7383
Table3. The result of 500 simulated indemnities for the crab industry supposed the
0.002 Input
Std. Dev. 2532661 insurance company compensates 80, 70, 60, and 50% of loss
Weibull
0
-2.90 5.40
0
-8 -4
-2.90
0 4
5.40
8
Skewness 0.27 States Alaska Washington Oregon California
-8 1 10
Values in Millions Values in Millions Kurtosis 2.78
5.1%
The fitted distributions LogLogistic Logistic LogLogistic LogLogistic
8.5% 91.5%
3.4%
0.01
5.1% 3.4%
Lognorm
88.9%
2.7% 8.5% 91.5% 2.7%
80% 1,380,924 702,344 2,823,855 3,470,998
2 0.008 88.9% (8715240,2378647,
Values x 10^-7

0.006
Shift(-8718865)) The fair 70% 1,208,531 615,453 2,471,322 3,037,152
Lognorm
Detrend yield data 1 0.004
Mean -3625 insurance premium 60% 1036812 526,962 2,120,126 2,604,553
t 1 t 1
0.002 Input Std. Dev. 2378647 50% 863,454 439,135 1,765,671 2,169,759
Yt  a   biYt i  u t ; uˆ t  Yt  Yˆt Y t aˆ   bˆiYt i
Lognorm
0
-2.90 5.40
0 -2.90 5.40 Skewness 0.83
i i
-10 0
Values in Millions
10 -10 -6 -2 2
Values in Millions
6 10
Kurtosis 4.27 80% 57.6% 74.8% 57.4% 58.8%
Yt is the annual yield ; uˆ t is used to evaluate the parametric distrubuti ons of the crab . 5.1%
7.6% 91.5%
3.4%
3.2% 0.01
5.1% 3.4% The occurrence of 70% 57.6% 74.6% 57.2% 58.8%
7.6% 91.5% 3.2% LogLogistic
2 89.2%
0.008 89.2% indemnities 60% 57.6% 74.8% 57.4% 58.6%
(-7455023,7063289,5.70)
Values x 10^-7

0.006
LogLogistic 50% 57.6% 74.8% 57.4% 58.8%
1 0.004 Mean -21103
0.002 Input Std. Dev. 2523845
LogLogistic
0 0 -2.90 5.40 Skewness 1.97

t 1 t 1
-8
-2.90
1
Values in Millions
5.40
10

3.4%
-8 -4 0 4
Values in Millions
8
Kurtosis 17.23 Conclusion
5.1% 5.1% 3.4%
Gamma
Ct  h(Yt  Yg )  h[( aˆ   bˆiYt i  et )  (aˆ   bˆiYt i )]  h * et
91.5% 8.6% 0.01
22.1% 22.1% 91.5% 8.6%
69.3%
i i
2 0.008 69.3% (1.9437,2634682, The detrended yields of each state have positive skweness. They are
Values x 10^-7

Ct : the cov erage level ; h : percentage;

0.006 Shift(-5121130)) very different from the negative skewness of the crop yields. Further,
1 Gamma 0.004 Mean 0
Yg : the guaranteed quantities ; and et ~ distributi on ( parameters ) goodness-fit- measures indicated that the Gamma and normal fail to
0.002 Input Std. Dev. 3673221
0
Gamma describe the sample data. The loglogistic distribution is best to estimate
0 -2.90 -2.90 5.40 Skewness 1.43
-6 4
5.40
Values in Millions
14 -6 -1 4
Values in Millions
9 14
Kurtosis 6.08
the indemnities of Alaska, Oregon, and California respectively while the
logistic is best for Washington. If the insurance compensates 80, 70, 60,
Figure 1. Probability and cumulative distribution functions and empirical
50% of loss yields, we find that the occurrence of indemnities is over 50%.
distribution for Dungeness crab detrended yield data in Alaska, 1950-2009

References
Bruce J. S., C.Z. Fabio, D. S. Gary, and H. I. Scott. “Crop Insurance Valuation under Alternative Yield Distribution.” American Journal of Agricultural Economics 86( May,2004) 406-419.