Embed
Email

Presentation NOAA

Document Sample

Shared by: benben zhou
Categories
Tags
Stats
views:
0
posted:
12/9/2011
language:
pages:
16
Markets, Pooling and Insurance

for Managing Bycatch in Fisheries







Daniel S. Holland,

Northwest Fisheries Science Center

Another Perspective on Bycatch

Canary rockfish Canary rockfish

 Individual bycatch quotas NORTH of 40'10'N SOUTH of 40'10'N





strengthen individual Perce



incentives to avoid bycatch 97%

94.1%

0.7%

2.2%



 When bycatch is highly

5%







uncertain individual quotas

markets may be subject to Widow rockfish Widow rockfish

high price variability

NORTH of 40'10'N SOUTH of 40'10'N







(creating substantial financial Perce





risk for fishermen) and may 98.3%

1.2%

97.6%





fail to allocate quota

1.3%







efficiently.

 I explore how pooling Yelloweye rockfish

NORTH of 40'10'N

Yelloweye rockfish

SOUTH of 40'10'N



approaches and possibly Perce

market insurance can be

used to reduce financial risk 99.7% 0.3% 99.8% 0.2%





for fishermen in these cases

Model Assumptions

 Fishing events and bycatch are discrete homogeneous events.

 Each fishing event yields one unit of target catch with certainty

and has a constant probability of catching one unit of bycatch.

 For simplicity, the bycatch is assumed to have no value and the

target catch has a unit net value after harvest costs.

 Bycatch is purely random modeled as a Bernoulli process

where bycatch events are independent over time and across

fishermen

 With this specification the expected value of an individual bycatch

quota allocation is equal to the sum of negative binomial probabilities

of exactly reaching period k before exhausting IBQ holdings, j,

summed over periods (ITQ use) k<=t and IBQ holdings j
probability of reaching the final period without exhausting IBQ times

the profit associated with harvesting in all possible periods.



t 1  (k  1)! 

E   q, t , p    t   

 

k q

* p (1  p) *(k  t ) 

q



k  q  ( q  1)! k  q  ! 

Expected Value of IBQ and ITQ with discrete bycatch

probabilities of 0.01 and 0.05

Maximum value of unit of IBQ=1/p

(a) (b)

Marginal IBQ Value 1.0

Marginal ITQ Value

100

1st Unit IBQ

0.9

90 2nd Unit IBQ

3rd Unit IBQ 0.8

80









Incremental Value of ITQ

Incremental Value of IBQ









4th Unit IBQ 0.7

70

5th Unit IBQ

60 0.6



0.5

P=0.01

50 P=0.01

40 0.4



0.3 Holds 1 Unit IBQ

30 Holds 2 Units IBQ

20 0.2 Holds 3 Units IBQ

0.1 Holds 4 Units IBQ

10

Holds 5 Units IBQ

0 0.0

1 26 51 76 101 126 151 176 201 226 251 276 1 26 51 76 101 126 151 176 201 226 251 276

ITQ Units Already Held

(c) ITQ Units Left (d)

20 1.0

18 1st Unit IBQ 0.9

5th Unit IBQ

16 10th Unit IBQ 0.8

Incremental Value of IBQ









Incremental Value of ITQ



14 15th Unit IBQ 0.7

20th Unit IBQ

12 0.6

10 0.5

8

P=0.05

0.4

Holds 1 Unit IBQ

P=0.05

6 0.3 Holds 5 Units IBQ

4 0.2 Holds 10 Units IBQ

Holds 15 Units IBQ

2 0.1 Holds 20 Units IBQ

0 0.0

1 26 51 76 101 126 151 176 201 226 251 276 1 26 51 76 101 126 151 176 201 226 251 276

ITQ Units Left ITQ Units Already Held

The distribution of ITQ units used (with a maximum of 300) before for (a)

one, (b) two, (c) three units of IBQ is exhausted with p=0.01.



(a) (b) (c)



25% 25% 50%

Probability of Occurence









20%









Probability of Occurence

40%









Probability of Occurence

20%





15% 15% 30%





10% 10% 20%





5% 5% 10%





0% 0% 0%

20-40

40-60

60-80





100-120

120-140

140-160

160-180

180-200

200-220

220-240

240-260

260-280

280-300

0-20









80-100









20-40

40-60

60-80





100-120

120-140

140-160

160-180

180-200

200-220

220-240

240-260

260-280

280-300

0-20









80-100









00





0





0





0





0





0

20





0









14





18





22





26





30

-6



-1

0-



40









0-





0-





0-





0-





0-

80



12





16





20





24





28

Profit Range Profit Range Profit Range







 With only one unit of IBQ the distribution of possible outcomes

50%





40% is skewed to the right but with three units it is skewed to the left

Probability of Occurence









30%  Thus trading away a unit of quota always increases downside

risk, in terms of increased right skew of the new distribution of

outcomes, and may either increase or decrease standard risk

20%





10%

as measured by the standard deviation of expected revenue.

0%

 Sufficient standard risk aversion and or prudence (downside

0-20 20- 40- 60- 80- 100- 120- 140- 160- 180- 200- 220- 240- 260- 280-

risk aversion) could inhibit trading even where it would lead to

40 60 80 100 120 140 160 180 200 220 240 260 280 300



increases in total expected value.

Modeling an IBQ-ITQ Market



 Assume numerous homogeneous participants

 Assume that risk premiums are not large enough to

preclude trading between individuals with different

ratios of IBQ and ITQ, and markets continually

equilibrate the levels and ratios of IBQ and ITQ

across fishermen

 We can model the prices of IBQ as the expected

value of an additional unit of IBQ or ITQ if all the

quota was held by one individual – what would they

pay for another unit of IBQ given their existing ITQ

allocation?

 Need to approximate IBQ value with simulation

techniques since values in equation 1 approach

infinity

Marginal IBQ value for varying starting levels of aggregate IBQ when

aggregate ITQ is 5000 and p is 0.01 (panel a) or 0.05 (panel b)



(a) (b)



100 20

90

80

15

ICQ Per Unit Value









ICQ Per Unit Value

70

60

50 10

40

30

5

20

10

- 0

25 30 35 40 45 50 55 60 65 70 75 200 210 220 230 240 250 260 270 280 290 300

Starting Total ICQ Starting Total ICQ









 When aggregate ITQ is large and ∑IBQ=p* ∑ ITQ, the value of

IBQ is approximately 0.5/p.

 The marginal value of IBQ will asymptotically approach 1/p for

decreasing levels of aggregate IBQ, and the expected value will

asymptotically approach zero for increasing levels of aggregate

IBQ.

Simulated price paths for IBQ from simulations with p=0.01 and 50 fisher

each with allocations of 100 units (aggregate ITQ=5000) and aggregate IBQ

of (a) 45 units, (b) 50 units and (c) 55 units.

(a)

Price (b)

Cumulative Bycatch

110.0 55

100.0 50

90.0 45

80.0 40









Cumulative Bycatch

70.0 35



• Prices vary 60.0 30









Price

50.0 25





widely during the 40.0

30.0

20

15





year as actual 20.0

10.0

10

5

-

bycatch departs

-

1 11 21 31 41 51 61 71 81 91 1 11 21 31 41 51 61 71 81 91

Period Period

(c) (d)

from the 110.0

100.0

55

50



expected level 90.0 45

40









Cumulative Bycatch

80.0

70.0 35

60.0 30

Price









50.0 25



40.0 20



30.0 15



20.0

10

5

10.0

-

-

1 11 21 31 41 51 61 71 81 91

1 11 21 31 41 51 61 71 81 91

Period Period

(e) (f)

110.0 55

100.0 50

90.0 45

80.0 40









Cumulative Bycatch

70.0 35

60.0 30

Price









50.0 25

40.0 20

30.0 15

20.0 10

10.0 5

- -

1 11 21 31 41 51 61 71 81 91 1 11 21 31 41 51 61 71 81 91

Period Period

Simulated distribution of IBQ prices mid season

(a)

45%

 Simulated distribution of 40% Q=45 Q=50 Q=55

IBQ prices half way through









Percentage of Price Observations

35%



the fishing year for varying 30%



bycatch probabilities (a) 25% P=0.01

1%, (b) 5% and with initial 20%



total IBQ allocations equal 15%



to 90%, 100% and 110% of 10%





the expected total bycatch. 5%



0%



 With a very low bycatch 0-10 10-20 20-30 30-40 40-50 50-60

IBQ Price Interval

60-70 70-80 80-90 90-100





probability (p=0.01) the (b)

80%



likelihood of intermediate 70% Q=225 Q=250 Q=275





prices is high.

Percentage Observations



60%



50%

 With higher bycatch 40% P=0.05

probabilities (p=0.05) prices 30%

are more predictable and 20%

tend toward the extremes 10%



0%

2







4







6







8





10







12







14







16







18







20

0-







2-







4-







6-





8-







-







-







-







-







-

10







12







14







16







18

IBQ Price Interval

(c)

120%





100%

CV of individual profit as bycatch probability and quota

vary

1.00





Short 10% Balanced Surplus 10%





0.80

CV of Individual Annual Profit









0.60









0.40









0.20









-

0.01 0.05 0.10

Bycatch Probability







 Although there is little variance in total industry profit there can

be quite a bit of variance in individual profit as some gain and

some lose through transactions in the quota market to cover

bycatch

Pooling

 Pooling bycatch quota can protect pool members from

variability in profit due to individual variability in bycatch and

exposure to price variability in the IBQ market

 Similarly, commitment to a reciprocal trading system at a set

price could also reduce risk and bargaining inefficiencies that

arise from private information (Matouschek and Ramezzana

2007)

 While larger pools decrease price variability they may also

increase problems associated with moral hazard and adverse

selection – so limited pool sizes may be preferable

 Lee and Lignon (2001) show that in the presence of moral

hazard there are typically finite optimal pool sizes. They find

optimal pools sizes between 50 and 100 for a variety of loss

probabilities and utility function specifications.

CV of individual profit with different pool sizes

1.000





Q=45 Q=50 Q=55



0.800





P=0.01

CV of Individual Annual Profit









0.600

∑ITQ=5000



0.400

∑IBQ=45,50 or 55



0.200









-

1 2 5 10 25 50

Pool Size







 Pooling can substantially reduce variability of individual

income even for small pool sizes and low bycatch

probabilities

Insurance

 Risk could be reduced further or even eliminated with market

insurance.

 Insurance might also be important if there was some concern

that a race for fish might develop (e.g., where fishermen, or

pools of fishermen, would race to use up their individual or

pooled allocations of target catch quota early before the total

bycatch quota was exhausted and the fishery shut down).

 For example, the proposed regulations for the West Coast

groundfish trawl ITQ could potentially shut down the entire

fishery in the event of a large bycatch event, a “disaster tow”,

that is greater than or equal to the total remaining quota for

that species.

Market Insurance Against Insufficient IBQ

(All IBQ pooled and individual ITQ allocations insured)



IBQ=p*ITQ IBCQ=p*ITQ*0.8

25.00 25.00

80% insured 90% insured 100% insured 80% insured 90% insured 100% insured





 Insurance

Insurance Premium

20.00









Insurance Premium

20.00









premiums per 15.00 15.00









vessel 10.00 10.00





5.00 5.00





- -

n=25 n=50 n=75 n=100 n=25 n=50 n=75 n=100







 CV expected

Pool Size Pool Size



0.20 0.20

80% insured 90% insured 100% insured



revenue (pre-

80% insured 90% insured 100% insured

CV Revenue Per Vessel









CV Revenue Per Vessel

0.15





insurance)

0.15









per vessel

0.10 0.10







0.05 0.05







-

-





 CV of

n=25 n=50 n=75 n=100

n=25 n=50 n=75 n=100

Pool Size

Pool Size







expected 0.04 0.04

80% insured 90% insured 100% insured 80% insured 90% insured 100% insured







profit (post

CV of Profit Per Vessel









CV Profit Per Vessel

0.03 0.03







insurance) 0.02 0.02





per vessel 0.01 0.01









- 0

n=25 n=50 n=75 n=100 n=25 n=50 n=75 n=100

Pool Size Pool Size

Conclusions

 When bycatch is highly uncertain and rare, markets may not

work effectively or will leave individuals facing substantial

financial risk.

 In such cases it is useful to think about bycatch management

as a risk management problem rather than a joint production

problem.

 Risk reduction mechanisms such as pooling quotas, self

insurance and market insurance may be useful ways for

stakeholders to reduce risk and perhaps increase efficiency as

well.

 There is anecdotal evidence that fishermen do form these

types of risk pools informally with reciprocal trading behavior.

 The risk faced by the industry and the need for insurance might

be partly mitigated in some cases by allowing quotas to be

borrowed or carried forward across years or by setting multi-

year TACs and quota allocations.

(forthcoming in Ecological Economics)

For a copy of the paper email me at







dan.holland@noaa.gov



Related docs
Other docs by benben zhou
dossier pq lucidi Marie
Views: 5  |  Downloads: 0
NYSILC logo
Views: 5  |  Downloads: 0
March 19_ 2010 VML Budget Analys
Views: 8  |  Downloads: 0
Subpart BReclamation of Benefit Payments
Views: 5  |  Downloads: 0
Archival-Based Research Methods in Accounting
Views: 36  |  Downloads: 0
In Naomi House
Views: 5  |  Downloads: 0
King Lear Commentary Act Scene lines scalding
Views: 8  |  Downloads: 0
By registering with docstoc.com you agree to our
privacy policy

You are almost ready to download!

You are almost ready to download!