Full text of DOI:10.1002/(SICI)1520-6688(200024)19:1<23::AID-PAM3>3.0.CO;2-C

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 23
Net Loss: A Cost-Benefit
Richard Schwindt
Aidan Vining
Analysis of the Canadian
Steven Globerman
Pacific Salmon Fishery
Abstract
This article applies cost-benefit analysis to the Canadian Pacific commercial salmon
fishery. It demonstrates that government policies to preserve the fishery have re-
sulted in higher net social costs than would have resulted from a “do nothing” policy,
notwithstanding the rent dissipation associated with unconstrained resource ex-
ploitation. The value of landings and the private costs of the harvest over a cycle
(1988–1994) are calculated. On average, fishers extracted rents of C$34.7 million
(in constant 1995 Canadian dollars) annually. The public costs of enhancing the
resource and organizing and policing the harvest are estimated. When these costs
are included in the calculation, net benefits drop to an average of negative C$55.6
million annually. This translates into a net present value (NPV) of the salmon fish-
ery of negative C$784. The effects on NPV of both modest policy changes imple-
mented in 1996–1997 and of a more dramatic but credible fleet rationalization
program are provided. The results indicate that further policy change is called for.
More generally, the study shows that policy reform that would significantly benefit
both the private sector (through reduced rent dissipation) and the public sector
(through reduced government expenditures) can be surprisingly difficult. © 2000
by the Association for Public Policy Analysis and Management.
INTRODUCTION
It is well known that in situations where property rights to assets are undefined or
insecure, market participants often dissipate most, or all, of the economic rents
(returns above the cost of capital) associated with the exploitation of those assets
[Gordon, 1954]. Rent dissipation is the use of real resources solely to lay claim to the
economic rent generated by some other resource.1 This article documents a less well
established phenomenon: Namely, that the social costs of rent dissipation activities
1 For example, in the context of the fishery, fishers expend extra effort and use sophisticated equipment to
beat others in the race for valuable fish (expenses that would not be incurred if they held a viable property
right to the fish).
Manuscript received January 1998; revise and resubmit recommended March 1998; revision received July 1998; second revise
and resubmit recommended September 1998; second revision received November 1998; paper accepted November 1998.
Journal of Policy Analysis and Management, Vol. 19, No. 1, 23–45 (2000)
© 2000 by the Association for Public Policy Analysis and Management
Published by John Wiley & Sons, Inc.
CCC 0276-8739/00/01023-23

24 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
can considerably exceed the value of the (foregone) economic rents when all costs,
including those associated with government actions ostensibly taken to improve
efficiency, are taken into account.
This article provides a cost-benefit analysis of British Columbia’s (B.C.) wildstock
salmon fishery. It demonstrates that the government’s interventions to preserve the
economic value of the fishery have resulted in substantially higher net social costs
than would have resulted from a “do nothing” policy, even with the rent dissipation
associated with the absence of any government intervention. We believe the B.C.
fishery case is not unique. Conditions in the fishery worldwide can generally be
characterized as a “crisis of dwindling stocks, overcapitalization, and disputes over
jurisdictions” [Grafton, Squires, and Kirkley, 1996, p. 90]. Governments expend
significant resources attempting to protect dwindling stocks, controlling
overcapitalized fleets and resolving jurisdictional disputes. The use of government
resources to regulate private dissipative activities usually can be viewed as part of the
total dissipation, and when they are incurred, they result in “overdissipation.”
Additionally, overdissipation almost certainly occurs in other policy contexts where
there is resource rent potential and where government incurs substantial costs to
attenuate overexploitation.
Our cost-benefit analysis suggests that the net present value of the B.C. salmon
fishery in the mid-1990s approached negative Canadian2 $800 million; however, this
estimate of social value does not include foregone opportunities associated with having
a more efficiently organized fishery.3 It is difficult to estimate this opportunity cost,
but plausible estimates of available cost savings to the public and private sectors
generate a potential net present value of $1.4 billion. This difference between the
current and potential net present values suggests that the existing governance regime
for the fishery costs society more than $2 billion in present value terms.
The analysis proceeds as follows. First, we provide a brief overview of the concept
of efficiency in the context of the salmon fishery, distinguishing between open access
and open effort. Second, we review the history of government attempts to deal with
the open access problem and the subsequent problems once open access was
foreclosed. Third, we provide a general description of the cost-benefit framework
used along with estimates of the annual costs and benefits generated by the salmon
fishery over the 1988–1994 period. These form the basis for estimated costs and
benefits. Fourth, we estimate, under a variety of assumptions, the net present value
of benefits under the recently reformed policy regime and under a more severe
rationalization of the fleet. Fifth, we briefly discuss why this overdissipation has
persisted over time in spite of the large welfare gains that would be realized from
effective reform. Finally, we present a brief set of conclusions.
The analysis contained in the article should contribute to a better appreciation of
the potential consequences of government fisheries policies. It also raises an important
question of political economy. Namely, why do societies forego opportunities to reduce
efficiency losses associated with government fishery policies? The usual explanation
of wealth-reducing government policies is the successful effort of well-organized
interest groups to transfer income to themselves from other groups in society
[Weingast, Shepsle and Johnsen, 1981]. While we believe that interest group activities
2 All values in this article are in terms of Canadian dollars.
3 Schrank [1996] compares the costs of maintaining the landed catch from the Newfoundland fishery to
the value of the catch without incorporating the losses associated with foregoing more efficient regulatory
regimes.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 25
are relevant influences on government policy in the B.C. salmon fishery, this is not
the whole story.
BACKGROUND: THE OPEN ACCESS RESOURCE PROBLEM
The exploitation of wild salmon has the potential to generate rents. In most years,
assuming reasonably efficient harvesting, the value of landed salmon (i.e., the price
paid to fishermen by processors) would far exceed the costs of landing them. Under
many circumstances, especially where the resource is spatially stationary, rents accrue
to those who establish comprehensive property rights to the resource [Vining and
Weimer, 1999; Weimer and Vining, 1999]. But, historically, property rights to salmon
have been attenuated because ownership was not established until the fish were
captured. Because the potential rent stream was not assigned until capture, fishermen
had little choice but to compete for the resource. Absent efficient government
intervention or effective cooperation among common owners, the most extreme result
is an open access resource [Clark, 1973; Gordon, 1954; Scott, 1955].
Open access competition can result in both threats to the survival of the resource
and reductions in the welfare of society. Because no fisherman owns either the current
generation or future generations of salmon, no one has incentives to forego a current
catch to ensure the existence of a future one, even though the net present value of the
rent would be greater by doing so. Competition may eliminate the potential not only
for rent, but also for normal returns to labor and capital.
Competitive exploitation of the resource is particularly dangerous in the Pacific
salmon fishery because of the life cycle of the fish. Pacific salmon spawn in fresh
water, migrate to the sea, and then, after a number of years (depending upon species)
of ocean “grazing,” reach maturity and return to the freshwater stream of origin to
spawn and die. As a result, the returning fish are very easy to harvest because they
need not be hunted. At the extreme, nets can be set at the mouth of rivers, effectively
blocking upstream migration and spawning [Higgs, 1982]. In an unregulated open
access fishery there is no incentive for fishers to allow an adequate number of spawners
to escape, and in the salmon fishery overexploitation is relatively easy. There is ample
evidence that historically this problem has beset the B.C. salmon fishery as well as
U.S. West Coast salmon fisheries [Crutchfield and Pontecorvo, 1969; Higgs, 1982;
Pearse, 1982; Schwindt, 1987; Sinclair, 1960].
Forestalling new entrants into a salmon fishery alone will not solve the rent
dissipation problem or even ensure the viability of salmon stocks if effort is not
constrained (i.e., the fishery is characterized by open effort but limited access).
Incumbents still have incentives to compete against each other in ways that will
dissipate rents and threaten stock viability. Normally, this competition will be
manifested by additional effort on the part of incumbents, either in terms of labor or
capital inputs. Even if some margins of effort relating to these inputs are constrained,
such as restrictions on allowable gear, incumbents will exploit other unconstrained
margins, such as sophisticated sonar [Higgs, 1982]. Unless restrictions on effort are
comprehensive and enforceable, fishermen still have the incentive and means to
increase effort. Moreover, it should be recognized that as government increases the
comprehensiveness of restrictions, the costs of enforcement escalate. Thus, as private
rent dissipation is curbed (assuming the restrictions are effective), public costs
increase. If the restrictions are ineffective, total dissipation increases.
In the end, the only way to protect stocks if all effort dimensions cannot be effectively
restricted is to stop people from catching the fish by keeping the fishery closed. As we
will see, combinations of closed access, effort constraints, and limits on the size of

26 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
the harvest through restricted openings have been imposed in the Canadian salmon
fishery. Nonetheless, it is still subject to significant inefficiencies that dissipate rent.
A SHORT HISTORY OF CANADIAN GOVERNMENT FISHERIES POLICY
Policymakers have long recognized that unconstrained exploitation of the salmon
resource would lead to its destruction. Certain harvesting techniques were potentially
so effective that they could devastate the resource. Land-based terminal fisheries,
using traps, weirs, and fish-wheels, can intercept the fish as they begin their massed
migration up freshwater rivers or streams. Before European contact, aboriginal
peoples relied heavily on terminal harvesting techniques [Jenness, 1977, pp. 64–65],
but within a context of effective private and common property rights that effectively
managed the resource [Higgs, 1982]. Settlers who were effectively unconstrained by
aboriginal property rights adopted these methods in the early commercial fishery. In
time, these devices were banned, ostensibly in the interest of conservation, although
the actual reason probably had more to do with the interest group activities of the
nonnative commercial fishery [Schwindt, 1995]. Whatever the reason, the commercial
fishery came to be a small-boat fishery.
As early as 1887, regulators recognized the open access problem in the small-boat
fishery and recommended entry limits. Over the next 30 years there were repeated,
but futile, attempts to curb entry [Fraser, 1977]. Harvesting capacity increased due
to fleet expansion and technological advances, and this also put increased pressure
on the stocks. By the late 1950s the problem had reached crisis proportions. The
Canadian government appointed a commissioner to study the industry in depth
and provide recommendations. Sinclair [1960] identified three methods to solve
the overexploitation and excess capacity problems: create exclusive private rights
to the rents (that is, impose a monopoly), tax the rents out of the system with fish-
landing taxes, and/or restrict the entry of vessels. Entry restrictions were imposed
in 1968, even though it was recognized at the time that this was only a partial,
short-run solution.
While new fishermen were denied access to the resource (except through the
purchase of licensed vessels), incumbents realized that the private return to increased
effort exceeded its cost. Harvesting capacity expanded through enhancement of the
existing fleet’s fishing power. This process, known as “capital-stuffing,” involved the
replacement of older licensed vessels with technically superior ones and investment
in sophisticated and expensive vessel improvements (e.g., bow thrusters, sonar). As
Wilen [1988, pp. 316–317] points out, “[I]t is clear that the fundamental incentives to
employ extra resources to compete are latent, strong, quickly triggered, and basically
unaffected by limited entry.”
In addition to restrictions on new entrants, the 1968 plan called for a government
funded buyback and retirement of existing vessels. Unfortunately, increasing salmon
prices and harvests raised the value of participation in the fishery. Fishermen rightfully
perceived that closed entry and a reduction in extant capacity through the buyback,
combined with increasing revenues, would increase average rents for those remaining.
The demand price per licensed vessel increased, and buyback funds were quickly
exhausted. Harvesting capacity was not materially reduced.
The end of the commodity boom in the late 1970s precipitated a financial crisis
among both incumbent fishermen who had borrowed to participate in the capital
stuffing and entrants who had paid extremely high prices for licenses and vessels. A
combination of high interest rates and low prices pushed many toward insolvency.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 27
Another government commission (the Pearse Commission) was struck to examine
the industry and make recommendations. The primary problem identified by Pearse
was excess harvesting capacity with concomitant pressure on salmon stocks. He
recommended both taxation on the catch and on fishermen (through a license auction
scheme), with the proceeds going to the government. These proposals, not surprisingly,
were very strongly resisted by incumbents. Then the crisis waned somewhat as landings
and landed values recovered. Ultimately, nothing was done to extract rents. Instead,
gear restrictions were tightened to further curb fishing effort and the harvest was
limited by location and opening restrictions.
By 1995, harvest restrictions were such that the viability of most stocks was not
threatened, but overcapacity was still endemic. In 1982, the year of the Pearse
Commission, the Department of Fisheries and Oceans (DFO) issued 4470 salmon
licenses. In 1995 the number was 4367. While this was a slight reduction in the number
of licensed vessels, fishing power had increased greatly because of technological
improvements and continued capital investment. In March 1995 the then minister of
fisheries reiterated that “a fundamental cause of the increased difficulty in managing
the salmon fishery is that the size of the fleet has remained static while its fishing
power has increased dramatically” [Tobin, 1995].
In March 1996, the federal government implemented a new program formally titled
the Pacific Salmon Revitalization Strategy, but since then known as the Mifflin Plan
after the minister of fisheries and oceans of the day, Fred Mifflin. The program included
two core policies to reduce the size of the fleet. First, $80 million were allocated to,
and soon spent on, a voluntary buyback program. Second, area licensing was imposed.
Under this policy, there are two geographic areas for seine vessels and three areas for
gillnet and troll vessels. Current license holders had to choose a single area in which
they would fish during the 1996 season. However, licensees could “stack” licenses—
that is, license holders could purchase an additional license and use it to fish in another
area. The idea here was that more efficient fishermen would seek to work their vessels
over multiple fishing areas by buying licenses from the less efficient who would then
retire their vessels. The immediate effect of the Plan on fleet size is shown in Table 1.
Table 1. Reduction of the salmon fleet under the Mifflin Plan.
Initial no.
of vessels
Buyback
Stacking
Total
retired
Percent Current
retired
fleet
Seine
Gillnet
Troll
536
2543
1288
48
447
303
67
256
52
115
703
355
21
28
28
421
1840
933
Total
4367
798
375
1173
27
3194
Source: ARA Consulting Group [1996, S-30].
The total fleet was reduced by 1173 vessels, or 27 percent. Relatively more troll and
gillnet vessels were retired than were the more efficient seiners.
While there is no doubt that the fishery was inefficiently organized and managed
prior to the Mifflin Plan, there has been no comprehensive estimate of the actual
value of rent dissipation. Nor are there estimates of the extent to which the recent, or
any future, rationalization scheme might mitigate dissipation. This is not surprising,

28 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
because such prediction is extremely difficult [Boardman, Mallery and Vining, 1994;
Hogarth and Makridakis, 1981].
A COST-BENEFIT ANALYSIS OF THE B.C. SALMON FISHERY
In order to assess the net costs of the current system, it is necessary to provide an
accounting framework for all the costs and benefits of the fishery. This should include
costs and benefits that accrue to fishermen and as well as those that accrue to the rest
of society. The most obvious way to address these costs and benefits is to estimate the
value of the salmon and what it costs to harvest them. However, a serious limitation
of this approach, given that the fishery is being mismanaged, is that it does not address
the potential net value of the harvest if it were properly managed. Put another way, it
does not make explicit the costs specific to the current regulatory system. Examining
only actual costs and benefits does not take into account the benefits that would be
generated by a more efficient fishery.
We assume that over the period 1988 to 1994, the fishery was approximately
managed under a “maximum sustainable yield” regime, and that the quantity and
quality of the harvest were optimized. It is difficult to know whether this is an
accurate assessment given the wide natural variability in salmon stocks [Miller,
1996]. We have no overall evidence that the sustainable yield could be significantly
increased under an alternative management regime. However, there is some evidence
that some species are overexploited. For example, in 1998 DFO announced a total
ban on coho harvesting because of weakness in the stock. As a result, our assumption
that harvests between 1988 and 1994 were at a sustainable maximum is more likely
overoptimistic than not (i.e., the benefits of the fishery are more likely to be
overstated than understated).
On the other hand, there is considerable evidence that total costs (including
governmental expenditures) would be much lower if incentives for capital-stuffing
were eliminated. It is the difference between these lower potential costs and actual
benefits that is a more accurate measure of lost benefits, or, in other words, of the
opportunity costs of the current system. Additionally, the Canadian government spends
considerable resources on managing the fishery and on supplementing incomes in
the fishery. At least some of this cost represents real social costs that should be added
to lost benefits resulting from current management, because such expenditures would
be lower with a more efficient regime.
The goal of economic organization is to maximize the net benefit emanating from
the production of a specific good (or service) over time (i.e., the NPV).4 Estimating
the present value (whether positive or negative) of the current fishery (and the potential
benefits of a more efficient fishery) requires specifying a discount rate. The discount
rate is a difficult conceptual and practical issue [see Burgess, 1981; Jenkins, 1977;
Zerbe and Dively, 1994]. The Canadian Treasury Board Secretariat recommends the
use of a 10-percent real social discount rate, with sensitivity analysis at plus or minus
5 percentage points. A real 10-percent rate was also used by the U.S. Office of
Management and Budget until 1992, at which time the agency switched to 7-percent
[U.S. Office of Management and Budget, 1992]. However, there is now a wide
consensus that even a 7-percent rate is too high (for a discussion of why agencies
may use too high real rates, see Boardman, Vining, and Waters [1993]). Lesser and
4
More technically, the object is to maximize the discounted sum over time of consumer and producer
surpluses.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 29
Zerbe [1998, p. 257] suggest that “in real terms [the] discount rate lies between 3.0
and 5.0 percent.” Boardman and Greenberg [1998, p. 317] conclude that practitioners
should employ a real rate of 4 percent, using plus or minus 2 percent to conduct
sensitivity analysis. Given this range, we use a base real rate of 5 percent, with
sensitivity analysis using 3 percent and 7 percent. Because we extrapolate current
dollars in calculating net present costs and benefits, we use a real, rather than a
nominal, discount rate.
The net benefit is the present value of the positive difference between the value to
society of the good and its cost of production. The task at hand then is to calculate
the value of the salmon harvest and the public and private costs of the harvest. For
biological reasons, the total allowable catch (TAC) varies dramatically from year to
year, with obvious impacts on the value of landings and the cost of the harvest. For
this reason we have estimated both the value and cost of the harvest over a seven-
year historical period (1988–1994), and have based the calculation of the present
value of future net benefits upon the average annual net benefits generated by the
fishery over those years. We somewhat arbitrarily set the discount period at 25 years,
although we also calculate present values for 15 and 35 years.5
The model is as follows.
25
H − (PC + GC)
NPV =
∑
(1+ d)t
t=0
Benefit (B) = value of the harvest (H)
Costs (C) = private costs (PC) + government costs (GC) where
GC = direct government expenditures (GCE) + net unemployment benefits (GCUI)
Discount rate = d
Years = t
Value of the Harvest
We assume that the gross benefits of the salmon harvest in any given year are reflected
in the landed value of the catch (revenues paid to fishermen). This ignores potential
consumer surpluses. However, this is not a significant problem, because most
production is exported (more than 70 percent), and we assume, as is the usual
convention, that only those costs and benefits that accrue to Canadians count. In
other words, only Canadians have “standing” [Whittington and MacRae, 1986]. Under
the circumstances, consumer surpluses accruing to Canadians are likely small and
can therefore be safely omitted from the calculations. We also ignore the value of
salmon taken by the sports fishery, the value of other species that prey on salmon,
and existence values associated with preservation of this resource.
Salmon return to spawn in cycles, and therefore the volume and value of the catch
fluctuate significantly from year to year. Figure 1 shows both the tonnes6 landed and
the dollar values (in 1995 dollars) between 1974 and 1995. Over the 22-year period,
the average annual landing value was $259 million.
5
Although “somewhat” arbitrary, 25 years does appear to approximately correspond to previous policy
cycles in the salmon fishery.
6 “Tonne” refers to a metric tonne (1000 kilograms or 2205 lbs.).

30 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
Source: Derived from British Columbia, Ministry of Agriculture, Fisheries and Food, Fisheries Produc-
tion Statistics of British Columbia, various years; and British Columbia, Ministry of Agriculture, Fisheries
and Food, British Columbia Seafood Industry Year in Review, various years.
Figure 1. B.C. salmon landings and landed value (1974–1995).
The period 1988 to 1994 is representative of a cycle. The average value of landings,
at $236 million, is not significantly different from the 22-year average. We use this
seven-year average as the expected annual revenue for the foreseeable future. This
might be an optimistic assumption given that increasing landings in other jurisdictions
such as Alaska [Miller, 1996] and the global growth of salmon farming are likely to
put downward pressure on all (real) salmon prices over the next decade.7
The Cost of the Harvest
Here we are interested in both the direct private and public sector costs of the harvest.
We ignore indirect costs (e.g., environmental externalities, the cost of lobbying efforts).
Private Costs
The primary source of information on the costs of operating the salmon fleet is the
triennial financial performance survey conducted by the DFO. Table 2 is representative
of these survey results and shows 1988 costs and returns for the salmon fleet.
In this industry, it is reasonable to assume that expenditures on inputs approximate
the opportunity costs of the inputs.8 There is no reason to believe that industry
purchases have a significant impact on input prices or that the markets for most
inputs (such as nets and gear) are distorted.
7 Over the period 1980 to 1995, farmed salmon as a proportion of world production of all salmon (wild
caught plus farmed) increased from 1 to 36 percent [Bjorndal, 1990] (updates provided by Bjorndal in
personal communication, December 1996).
8 Labor costs may be an exception. Presently we discuss the role of unemployment insurance benefits.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 31
Table 2. B.C. salmon fleet, costs and returns (1988, millions of nominal $).
Gross income
Fishing
437.9
427.9
317.8
61.5
Salmon
Herring
Other
48.6
Packing
10.0
Operating expenses
138.3
29.4
4.9
Fuel, oil, grease, and food
Other operating expenses
Wages (excluding captain)
104.1
Fixed costs
77.9
26.9
19.9
31.1
Repairs/maintenance
Net and gear
Other fixed costs
Capital charges
42.3
Net income
179.4
80.1
99.2
Captain returns
Owner returns
Source: DFO [1992b].
Under “noncapital costs” we include operating expenses (fuel, other, and crew
wages), so-called fixed costs (repairs, net and gear, and other), and captain returns.
While captain returns are included under net income in the survey, DFO computes
them as payments the captain would have received if operating a vessel owned by a
third party. Therefore, they are properly viewed as equivalent to the opportunity costs
of these skippers, and we include them as a noncapital cost. Returns to owner-operators
attributable to their capital are reported separately.
With respect to capital, the cost-benefit framework employed here requires an
estimate of the opportunity cost of capital investment annually. It is not straightforward
to convert “capital charges” into appropriate annual opportunity cost. Costs that accrue
as investments should be converted into the value of the stream of consumption that
would have resulted if the investments had not taken place. The appropriate
opportunity cost is the return the capital could have earned in its best alternative
use—the “shadow price of private capital” [Lyon, 1990]. Calculating this shadow price
involves estimating the pretax gross (undepreciated) rate of return applicable to this
capital and the economic depreciation rate [Hulten, 1990].
The capital charges included in the financial performance survey are inappropriate.
First, they do not include the opportunity cost of equity capital. Second, they use
depreciation claimed for accounting purposes, not the economic rate. Third, they
include charges for all capital employed in the industry. The first two problems are
common in this type of analysis; the third deserves some elaboration.
The 1988 fleet financial survey estimated that the market value of capital employed
in the industry was $1.15 billion. However, a significant portion of this (54 percent or

32 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
Table 3. Private costs of the salmon harvest (millions of 1995 $).
1988
1989
1990
1991
1992
1993
1994
Noncapital expense
Capital expense
264.6
69.4
218.9
49.4
206.8
36.5
141.0
35.0
152.9
25.4
148.0
32.2
177.4
35.3
Total private cost
334.0
268.3
243.3
176.0
178.3
180.2
212.7
Source: Published estimates are available for 1988 through 1990 [DFO, 1992b]. Unpublished survey
results were provided by DFO for 1991 and 1994. Estimates were made for 1992 and 1993.
$620 million) represents the value of fishing licenses (primarily salmon and herring
licenses). From the fishermen’s perspective, a license is certainly an asset and a
requirement for participation in the fishery. For those who purchased their licenses
well after the introduction of license limitation, the cost of the license is a significant
component of their capital investment. However, the value of the license should not
be included in the capital stock used to harvest salmon, as it is not a resource that is
used up in the catching of fish. To include this value would involve “double-counting”
the value of the harvest.
To estimate capital costs, we have used only vessel values (i.e., we have ignored
license values). Using industry interviews, we assume that the useful life of a fishing
vessel is 20 years and apply a straight-line depreciation rate. To determine the rate of
return applicable to this capital, we use annual rates of return to capital employed in
the broadly defined Canadian waterborne transportation industry.9
Estimates of capital and noncapital costs were made for each year over the 1988–
1994 period. However, these fleet costs are not uniquely attributable to the salmon
fishery because many of the vessels also harvest roe herring and halibut. In 1988, for
example, 537 salmon vessels also fished herring and 350 fished halibut. Therefore,
some allocation of costs to the various fisheries has to be made. Any allocation of
joint costs is inherently arbitrary; the procedure selected here is to assume that the
proportion of total fleet revenue attributable to salmon landings is applicable to total
fleet costs. Our estimates of the private sector costs of the harvest over the 1988–1994
period are summarized in Table 3.
Direct Government Costs
In addition to the private costs incurred by fishermen, government expends
considerable resources to make the salmon fishery possible. Both the federal and
B.C. governments engage in numerous activities to organize the harvest in any given
year and to enhance the salmon resource to ensure future harvests. These
expenditures can be loosely divided between protection and enhancement of the
stocks. These costs can be approximately represented by government expenditures
on the fishery. However, it is extremely difficult to quantify the costs of managing
and enhancing the salmon resource for a number of reasons. Multiple public agencies
at the federal, provincial, and even the local level are involved (additionally, volunteer
groups engage in conservation programs entailing opportunity costs), and it is
difficult to accurately apportion these various expenditures to the salmon fishery.
9
Annual financial data for the waterborne transportation sector were taken from CANSIM [Statistics
Canada, n.d.] series: D374581, D374582, D374591, D374607, and D374613. The annual rates used ranged
from 9.8 to 4.2 percent over the period and averaged 7.5 percent. This average compares with Shoven and
Topper’s [1992] Canadian estimate of 8.3 percent as the real shadow price of capital.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 33
Table 4. Canada, Department of Fisheries and Oceans, expenditures by program
(fiscal 1988–1989, thousands of nominal $).
Science
28,461
30,652
37,157
17,306
4085
Fish management and enforcement
Salmonid Enhancement Program
Marine support
Inspection
Harbors
Internal support
16,085
16,336
Total
150,082
Source: Personal communication, Finance and Administration DFO, Pacific Region, November 1996.
Additionally, while some costs are direct and derivable from budgets, such as the
Salmonid Enhancement Program, other costs are indirect and extremely difficult
to accurately estimate. For example, provincial forest practices guidelines
incorporate stream protection rules. This imposes costs on forest companies (more
costly harvesting practices are employed) and upon the province, directly through
planning and policing undertaken by forest ministry personnel and indirectly
through foregone timber revenues (e.g., stream protection requires no-cut buffer
zones along the waterway).
There are no comprehensive, species-specific management and enhancement
budgets for all levels of government. However, it is possible to estimate an order of
magnitude with respect to expenditures by the federal government, which has
primary responsibility for the salmon fishery. DFO’s Pacific Region expenditures
by program for fiscal 1988–1989 are set out in Table 4. We perform a number of
calculations to estimate expenditures on organizing, policing, and enhancing the
Pacific salmon fishery.
The Science program involves expenditures on the biological sciences (e.g., stock
assessment required to set the total allowable catch), physical and chemical sciences
(the study of the ocean’s physical and chemical properties) and hydrographic (the
study of the use of Canada’s navigable waters). We assume that only expenditures on
the biological sciences are fishery-related (on average this category accounted for
about half of total science expenditures).
Fish management and enforcement are wholly fisheries-related. The Salmonid
Enhancement Program is associated specifically with the salmon fishery. Marine
support involves the maintenance and crewing of vessels used both for fish
management and enforcement and scientific activities. All of the marine support
dedicated to fish management and enforcement and half of the support of scientific
research are allocated to the fisheries.
Inspection services involve programs to ensure that imported and exported fish
and shellfish products meet standards covering grades, quality, and wholesomeness.
Inspection of imports is not a charge against domestic production. Inspection of
domestic production enhances the wholesale value of the seafood, not the landed
value. Therefore, inspection services are omitted from the costs of the fishery. Both
fishermen and the public at large use harbors. Half of these expenditures are allocated
to the fishery.

34 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
Internal support covers administrative costs of the DFO’s Pacific Region. These
overheads are allocated in the following manner. Expenditures on fishery-related
activities as a percentage of total nonadministrative expenditures were calculated.
This was applied to the costs of internal support to estimate the cost of administering
fisheries-related programs.
These calculations yield an estimate of fisheries-related expenditures in the Pacific
Region. However, these expenditures relate to all fisheries managed by DFO, not just
salmon. In 1988, salmon landings accounted for 59 percent of the value of landings
of all stocks (salmon, herring, halibut, groundfish, and shellfish) managed by DFO.
Applying the (admittedly heroic) assumption that 59 percent of the Pacific Region’s
1988 fisheries-related budget is attributable to the salmon fishery generates an estimate
of federal government expenditures on the salmon fishery in that year. This estimation
procedure is then applied to the Pacific Region’s annual budget for each year over the
1988–1994 period. The results are set out in Table 5.
The estimates of government costs attributable to the salmon fishery are admittedly
crude. On the one hand, they understate public expenditures on this fishery. The
estimates ignore DFO head-office expenses, which are considerable. Other federal
ministries, provincial ministries and local governments incur substantial management
costs that are excluded. Finally, the calculations ignore the marginal excess tax burden,
that is, the costs of taxation to fund these government expenditures.
On the other hand, modest revenues are reaped from the sale of commercial fishing
licenses and charges for the use of harbors. It is also important to note that the
commercial salmon fishery is not the only beneficiary of these management programs.
The sports fishery and the Indian food fishery, for example, benefit directly from
enhancement and management.10 The fish provide food for other species (bears, eagles,
and whales), and the general public reaps some benefit from the knowledge that the
fish stocks are protected (i.e., “existence values”). Notwithstanding these omissions,
we believe our estimates provide a reasonable order of magnitude.
Unemployment Insurance
Transfers to fishermen under the unemployment insurance (UI) program complicate
the harvest cost calculations. UI payments are a significant and continual source of
labor income in the Pacific fishery. Benefits are available to both employees (crew)
and self-employed fishermen (working vessel owners). Fishermen are the only self-
employed workers who can participate in the program and, in the past, had preferential
access to benefit eligibility, since they needed fewer work weeks to qualify for benefits
than did regular claimants. Table 6 shows that payments to B.C. fishermen
10 In the early 1990s, about 93 percent of the total salmon catch was taken in the commercial fishery, 4
percent in the recreational fisheries, and 3 percent in the Indian food fishery [DFO, 1992a].
Table 5. DFO’s Pacific Region expenditures on the salmon fishery (millions of
1995 $).
1988
1989
56.0
1990
56.8
1991
46.2
1992
51.5
1993
50.4
1994
53.2
Salmon share of fisheries 54.9
Salmonid Enhancement
Program
45.7
42.3
98.3
42.9
99.7
34.9
81.1
35.0
86.5
32.3
82.7
30.2
83.4
Total
100.6

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 35
Table 6. UI contributions, benefits and net benefits for the B.C. fisheries (millions of
1995 $).11
1988
1989
1990
1991
1992
1993
1994
Contributions
Benefits
5.8
4.8
5.6
6.1
6.5
6.4
6.2
56.9
61.5
56.1
64.1
64.7
65.1
65.0
Net transfer
51.1
56.7
50.5
58.0
58.3
58.7
58.7
considerably exceed their contributions. Over the 1988–1994 period, annual UI
payments to B.C. fishermen (in constant 1995 $) averaged $61.9 million, and their
contributions averaged $5.9 million, leaving an average net payment of $56 million—
a little more than $5000 per fisherman. Over the 1988–1994 period, UI payments
represented 25 percent of the total labor income (crew wages, captain returns, and
UI payments) of fishermen.12
How should these payments be treated in estimating the total social costs of the
salmon fishery?13 On the one hand, it is reasonable to assume that when seasonal
workers anticipate UI payments, this is factored into wage offers and demands. If
true, fishermen accept lower direct compensation when the job provides reasonably
certain access to UI benefits. However, in calculating the cost of the salmon harvest
in Table 2, it was assumed that fishing income reflected labor’s opportunity costs of
working in the fishery. These payments understate the actual labor opportunity cost
if fishermen base their decision to remain in the sector upon fishing income plus UI
benefits. If they do so, the real opportunity cost is payments plus UI. On the other
hand, it is possible that UI benefits might not be a component of the opportunity cost
of all fishermen. Some fishermen may view UI as a windfall, or bonus, which does
not influence their choice of occupation.
Access to UI benefits certainly induces some fishermen to remain in the fishery.
Many low-income earners (gillnet/troll boat captains and crews) undoubtedly count
on UI benefits as a significant part of their annual incomes. Others, such as captains
and crews of highly productive seine vessels, would remain in the fishery regardless.
Because we do not know what proportion of UI payments should be included in the
opportunity costs of labor, we calculate net benefits of the fishery both with and
without the inclusion of these payments as a cost.14
11
There is a UI program specific to fishermen, and Statistics Canada [n.d., series D730349] reports ben-
efits by province under that program. These data understate the actual benefits, because fishermen can,
and some do, work enough weeks to qualify for normal benefits (which run for a longer period than
fishing benefits). Human Resources Development Canada, personal communication, September 1996,
provided UI contributions.
12 Another source concludes that in the early 1990s, UI payments represented one-third of the earnings of
an average self-employed B.C fisherman [ARA Consulting Group, 1996, p. S-4].
13 The special UI program for fishermen applied to all fisheries, not just salmon. However, because of the
relative importance of the salmon fishery, and the fact that it would generally be difficult for a fisherman
to accumulate enough weeks of work to qualify for UI in any other single fishery, we have attributed all UI
benefits to salmon fishermen.
14 There is also ambiguity surrounding the magnitude of future payments to fishermen (indeed to all sea-
sonal workers). The Unemployment Insurance program has very recently been revised (and renamed Em-
ployment Insurance). The revisions include lower benefits for repeat users, which include most seasonal
workers such as fishermen. Gross benefits paid to fishermen in 1997 were 44 percent of the 1994 level.

36 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
Table 7. Benefits, costs, and net benefits generated by the B.C. salmon fishery (millions
of 1995 $).
1988
1989
1990
1991
1992
1993
1994
Avg.
Landed value (benefits)
Total private costs
Private Rents
383.6 306.9
294.3
182.4
199.9
205.8
262.8 262.3
-333.9 -268.3 -243.3 -175.9 -178.3 -180.2 -212.7 -227.5
49.7
38.6
51.0
-99.7
-48.7
-50.5
6.5
-81.1
-74.6
-58.0
21.6
-86.5
-64.9
-58.3
25.6
-82.7
-57.1
-58.7
50.1
34.7
Public costs
-100.6 -98.3
-50.9 -59.7
-51.1 -56.7
-102.0 -116.4
-83.4 -90.3
-33.3 -55.6
-58.7 -56.0
-92.0 -111.6
Net benefit (excl. UI)
Net UI transfers
Net benefit (incl. UI)
-99.2 -132.6 -123.2 -115.8
Net Benefits
We are now in a position to estimate annual net benefits in the B.C. salmon fishery
over the period 1988-1994. The results are set out in Table 7. The final column shows
the average over the seven-year period, and is the basis for the NPV calculations in
the following section.
The estimates suggest that in every year of the period fishermen did capture some
positive rents from the fishery. As explained earlier, theory predicts that under open
access all rents will normally be dissipated. However, when fishermen have
heterogeneous skills, when the number of vessels is fixed and when there are
numerous, but not complete, restrictions on the physical characteristics of the vessels
(for example, on length and gear type), the equilibrium outcome in terms of rent
dissipation is less clear [Bishop, 1973; Johnson and Libecap, 1982]. It is possible that
these technological constraints “bite” in such a way that some rents are left in the
industry.15 Obviously when prices and harvests are good, as in 1988 and 1994, private
rents are substantial. In below average years such as 1991–1993, rent capture is
significantly less.
When public costs are added in, the annual net benefits generated by the fishery
are negative and substantial. When UI payments are viewed as a component of the
opportunity cost of labor, losses are even greater, exceeding $90 million in every year.
NET PRESENT VALUE OF THE SALMON FISHERY UNDER ALTERNATIVE COST ASSUMPTIONS
Thus far, we have estimated an annual cost or benefit for specific elements of the
fishery: value of the harvest, private harvesting costs, government expenditures on
management, enhancement and enforcement, and UI payments. We extrapolate these
costs and benefits forward over 25 years using the discount rates described earlier.
Table 8 sets out our estimates of the NPV of the B.C. salmon fishery as it operated
15 Hypothetically, net private rents could be a risk premium if they accrued to a subset of fishers or were
reaped by the entire industry sporadically. In this case we would expect that there would be losers (i.e.,
some fishers would suffer negative returns, and/or in some years total private returns would be nega-
tive). In other words, if the supernormal returns were really just compensation for self-insuring, the
industry as a whole would just earn normal returns over the long term. Our evidence indicates that
industrywide, private benefits exceeded private costs (which include a normal return) for every year
over the 1988–1994 period.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 37
Table 8. Present value of the salmon fishery prior to the Mifflin Plan (millions of
1995 $).
5%
3%
7%
Landed value (benefits)
Total private costs
Private rents
3,695.4
-3,206.4
489.1
4,565.7
-3,961.5
604.2
3,055.6
-2,651.2
404.4
Public costs
-1,272.7
-783.6
-1,572.4
-968.2
-1,052.3
-647.9
Net benefit (excl. UI)
Net UI transfers
-789.3
-975.1
-652.6
Net benefit (incl. UI)
-1,572.9
-1,943.3
-1,300.5
between 1988 and 1994 (under alternative interpretations of UI transfers). We
designate this as the “Pre-Mifflin Plan” period—that is, the period immediately before
the fleet rationalization program. Values are in 1995 dollars. The present value of
private rents (at the 5-percent baseline discount rate) is $489 million, but when public
costs are included, net benefits fall to negative $784 million.16 If UI payments are
included in the opportunity cost of labor, net benefits fall to negative $1.6 billion.
In Table 9 we show the sensitivity of the NPV of the fishery (excluding UI payments)
to different private and public cost savings. Under the first stage of the Mifflin Plan
(referred to in the subsequent tables as “Post-Mifflin”), the number of vessels in the
fleet has been reduced by about 27 percent (see Table 1). If this results in a 27-percent
reduction in private costs, the NPV of the fishery would increase to a positive $101
million. If public costs were to be reduced as a result of the fleet rationalization, NPV
would increase even more. We show the results for a 10- and 20-percent reduction in
public costs.
While helpful, fleet rationalization under the Mifflin Plan clearly did not go far
enough. Knowledgeable observers estimate that the harvest could be performed with
a much smaller fraction of the fleet. Pearse [1982] estimated that the fleet was twice
as large as required. DPA Group [1988] estimated that the fleet was two-thirds too
large. Most recently, Walters [1995] estimated that the fleet was from two to four
times too large. Unfortunately, none of these estimates are substantiated by detailed
analysis.
To our knowledge there are no estimates of the optimal size or gear configuration
of a small-boat salmon fleet. Absent such estimates, we have used the experience of
the Southeastern Alaska salmon fleet as a benchmark for achievable harvesting costs
using a small-boat technology (alternative, potentially more efficient techniques, such
as fish traps are not considered).17 Our analysis (discussed below) indicates that the
16
The present values with the same assumptions, except over 15 and 35 years are negative $477 million and
negative $910 million respectively. Full details of the NPV estimates under other assumptions for 15 and 35
years are available from the authors. The engaged reader may also notice that the estimate of the present
value of private rents ($489 million) is considerably less than the value of licenses ($620 million) cited earlier
and might wonder about the discrepancy (license values should reflect the capitalized value of the stream of
rents). The $620 million included salmon, herring, and halibut licenses, hence the high value.
17
The Alaskan salmon fishery is divided into four regions. Of these, the Southeastern region, which in-
cludes the Alaskan panhandle, is most similar to the B.C. fishery in terms of geography and climate.

38 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
Table 9. Net present value of the salmon fishery under alternative cost reduction
scenarios (5% discount rate, millions of 1995 $).
Public costs
Status quo
-10%
-20%
Pre-Mifflin
-784
+82
-656
+209
+1267
-529
+337
+1397
Post-Mifflin
SE Alaskan fleet
+1140
Southeastern Alaska fleet is considerably more productive than its B.C. counterpart.
However, we do not suggest that this fleet has attained minimum costs. There is
evidence that it too suffers from some overcapacity.
The superior performance of the Alaskan fishery is due in part to historical
happenstance. In response to open-access problems, Alaska imposed limited entry
on most of its salmon fisheries in 1975. Licenses were area- and gear-specific at the
outset (i.e., the licenses restricted fishers to specific areas and gears). In contrast, the
B.C. program initially allowed license holders to upgrade gear (e.g., gillnetters
upgraded to more productive seine boats or combination vessels) and, until recently,
to fish the entire coast, thereby encouraging investment in more powerful vessels.
Schelle and Muse [1986] argue that its more restrictive licensing policy probably
spared Alaska the extreme excesses witnessed in B.C. Nonetheless, the Alaskan
program did not stop capital stuffing. Schelle et al. [1982] report significant post-
limitation increases in the fishing power of the Alaskan salmon fleet. In short, both
fisheries suffer overcapacity, but the problem is more severe in B.C. Thus the use of
the Southeastern Alaska fleet as a benchmark reflects an achievable level of fleet
efficiency, not the ideal.
Comparison of harvesting data for the two fisheries reveals two important
differences. First, a significantly larger proportion of the Southeastern Alaskan catch
is taken with seine vessels (as opposed to gillnet and troll). Second, in terms of average
annual catch per vessel, Southeastern Alaskan seiners and gillnetters take nearly three
times the B.C. average for each of these gear types, while their troll vessels are only
slightly more productive. We have estimated the cost of the B.C. harvest over the
1988–1994 period assuming the Alaskan gear mix and vessel productivity.18 The results
indicate that on average the catch could have been taken at a 60-percent lower cost
using such a fleet. NPV estimates based on this cost reduction are shown in Table 9.
The results are dramatic. With no reduction in public cost, NPV increases to positive
18 Estimates of catch by gear type for the Southeastern region were provided by the Alaskan Department of
Fish and Game (personal communication with J. Hartmann) for the years 1993 to 1996 (the only years for
which the data had been compiled). From these data, gear mix and gear productivity were calculated.
Estimates of the per kilogram cost of the B.C. harvest by gear type for the years 1988 to 1994 were derived
from published and unpublished DFO data [DFO, 1992b]. It was then assumed that the B.C. fleet em-
ployed the Southeastern Alaskan gear mix (e.g., a higher proportion of the harvest taken by seine vessels)
and achieved the same productivity per vessel. There were two sources of savings in that seine vessels are
more productive than other gears, and SE Alaskan seiners and gillnetters are more productive than their
B.C. counterparts. These sources of cost savings are consistent with those identified by Dupont [1990].

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 39
Table 10. Net present value of the salmon fishery under alternative cost reduction
scenarios and alternative discount rates (millions of 1995 $).
Public costs
3% discount rate
Status quo
-10%
-20%
Pre-Mifflin
-968
+101
-811
+259
-654
+416
+1723
Post-Mifflin
SE Alaskan fleet
+1409
+1566
Public costs
-10%
7% discount rate
Status quo
-20%
Pre-Mifflin
-648
+68
-543
+173
-437
+278
+1153
Post-Mifflin
SE Alaskan fleet
+943
+1048
$1.1 billion. Modest reductions in public costs would increase the value to the positive
$1.4 billion range.
For completeness, we calculate NPVs based on these private and public cost-saving
scenarios using different discount rates. The results are shown in Table 10; they do
not alter the conclusions. Prior to the Mifflin Plan, the NPV of the fishery was less
than negative $500 million. The Mifflin Plan alone would generate a modest, positive
NPV. Adoption of the Alaskan fleet configuration and achievement of its productivity
would generate a positive and significant NPV.
PROGNOSIS FOR REFORM
Given these negative NPVs, a very relevant question is why has reform not yet been
undertaken? The question is of particular interest because reform does not involve
the usual questions of wealth redistribution. That is, policy reform issues commonly
involve rent eradication to achieve allocative efficiency (i.e., eradication of a
deadweight loss). Clearly, in such cases those who lose rents will obstruct reform. In
this case, the goal of policy reform is to generate potentially available resource rents.
Rent creation, particularly when it is available to private parties, should facilitate
policy reform. Why then has reform not been achieved, given that both private interests
and the public stand to gain?
This question is made more intriguing by considering the state of the world’s major
fisheries. While most remain in a state of extensive rent dissipation [Pauly et al.,
1999], there appear to be a number of fisheries for which governments and interest
groups have either engaged in major institutional reform or managed to arrest
complete rent dissipation. In the former category can be found the fisheries of
Australia, Iceland, and New Zealand [Grafton, 1996; McClurg, 1998; Runolfsson and
Arnason, 1997]; in the latter, the Alaskan salmon fishery, which has already been
discussed. What conditions facilitate reform and why have they, thus far, been absent
from the B.C. salmon fishery? An adequate answer to this question obviously requires

40 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
extensive comparative institutional analysis that goes considerably beyond the scope
of this paper, but here we sketch some preliminary thoughts.19
The work of Robert Bates [1988] and Elinor Ostrom [1990] provides a framework
for addressing the issue of institutional change in open-access contexts. Ostrom
emphasizes that “supplying a new set of rules is the equivalent of providing another
public good; [therefore] the problem faced by a set of principals is that obtaining
these new rules is a second-order collective dilemma” [Ostrom, 1990, p. 42]. Bates
[1988] shows that this result obtains even if the institutional design problem is modeled
as an assurance game, where cooperative outcomes can emerge more easily, rather
than as a prisoner dilemma game. Given this fundamental fact, the question rather
becomes under what circumstances can this second-order, collective action dilemma
be avoided? There appear to be a number of factors that influence the intractability
of this second-order problem.
The first is whether the fishery is part of a “nested” open-access or open-effort
problem. The fishery is a nested problem if there is both an interjurisdictional and an
intrajurisdictional open-access problem. Given the fact that nations have 200-mile
exclusive zones, some fisheries present exclusively intrajurisdictional open-access or
open-effort problems. There is a continuum to other fisheries that have contiguous
national boundaries and migratory fish stocks; these fisheries face major
interjurisdictional open-access problems as well as intrajurisdictional problems.
Nested fisheries present lower benefits to purely intrajurisdictional institutional reform
because some, or indeed in some circumstances most, of the benefits of reform will
be captured extrajurisdictionally. The political costs of allowing rents to be captured
by fishers in other jurisdictions may be higher than the political costs of rent
dissipation. Additionally, international open-access fisheries involve much higher
negotiation costs, since there are fewer opportunities for issue logrolling. Recent
successful reform efforts have been in relatively isolated national fisheries such as
Iceland and New Zealand. In this regard, it is worth noting that there are ongoing
disputes involving Alaska, B.C., Washington State, and Oregon over harvesting rights
to Pacific salmon.
The second factor that appears to influence reform momentum is whether rent
dissipation is a significant percentage of national wealth. In such cases the government
has an important encompassing interest in efficient policies [Olson, 1982]. In countries
or subnational jurisdictions where fisheries’ rent dissipation is a significant component
of jurisdictional wealth, governmental incentives to preserve or rebuild stocks are
enormous, since rent taxes are a significant source of tax revenue, while as a corollary
subsidies are a significant source of expenditures. These benefits are likely to outweigh
the political benefits of inaction. This factor seems to have been an important impetus
to reform in small jurisdictions with large, valuable fisheries such as Iceland and
New Zealand. For example, in Iceland the fishing industry’s contribution to gross
domestic product may be as high as 45 percent and may represent 50 percent of
foreign exchange earnings [Runolfsson and Arnason, 1997]. This is very far from the
case in B.C., where the salmon fishery accounts for less than 1 percent of provincial
19
The important role that institutional and environmental features can play in determining the relative
success of different reform schemes is described in Grafton [1996].
20
Over the decade 1986 to 1995, primary fisheries (the commercial harvesting of all species) plus fish
processing accounted for slightly less than 1 percent of provincial gross domestic product (GDP) and
about 1/10th of 1 percent of national GDP. From the perspective of the national government (which is
responsible for fisheries policy), the industry is of negligible economic importance.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 41
gross domestic product.20 Here the political costs of significant institutional reform
have always outweighed the economic benefits.21
The third factor that affects institutional reform is actually a set of factors relating
to the initial resource problem itself. Over the last decade researchers have identified
those institutional conditions that facilitate resource appropriators such as fishers in
effectively maintaining (historical) common property ownership [Ostrom, 1990; Wade,
1987]. These factors are too extensive to go into detail here, but include: a small
number of appropriators, the same appropriators over time (thus lowering their
discount rate), homogeneity among appropriators (both culturally and
technologically), the ability and power that appropriators themselves have to monitor
rules, and the power and legitimacy that appropriators have to enforce rules.
The absence of these conditions leads to either classic open-access or equally
debilitating open-effort problems. In turn, this creates the requirement for extensive
government involvement in any reformed regime and, thus, a de facto move away
from common property ownership. But the presence of large numbers of appropriators
with heterogeneous backgrounds using heterogeneous technologies also fosters
producer interest groups with opposing interests [Campbell, 1998; Gilligan, Marshall
and Weingast, 1989]. As McClurg [1997, p. 97] puts it: “One of the striking features of
salmon management within British Columbia is the plethora of site and method
controls affecting industry and native Indian harvesters [that] creates sub-groups of
harvesters with hostile interests to each other.” Fishers are also set apart by differing
abilities to capture those rents left in the fishery. Even within subgroups there are
“high-liners” who do better than their peers, and this adds to the heterogeneity of
harvesters. In turn, the presence of opposing producer interest groups also exacerbates
the inevitable uncertainty that surrounds the consideration of any policy change. Of
course, participants would probably be more inclined to accept these uncertainties if
all the private rents were to be squeezed out of the fishery, but as we have demonstrated,
this generally has not been the case. The crucial point is that the same factors that
generate wicked open-access problems also generate a more wicked second-order
collective action problem and reduce government’s capacity to engage in effective
institutional reform.
CONCLUSIONS
The Canadian Pacific salmon fishery illustrates how it is possible not only to dissipate
all resource rents, but to generate substantial net present costs in the exploitation of
a natural resource. This analysis suggests that the net cost (beyond simple rent
21 These political costs (or impediments to reform) involve the following. First, a politically feasible solu-
tion probably requires a costly license buyback which has to be borne “upfront,” whereas many of the
costs of the status quo are borne by the future, when certainly the minister and probably the government
will be gone (on the “political business cycle,” see, for example, MacRae [1977], Mikesell [1978], and
Chappell and Keech [1983]). Fisheries ministers have relatively short tenures because of the “hotseat”
nature of the portfolio. This seems to provide an irresistible temptation for ministers to delay fundamental
reform and engage in Band-Aid reform so as to be seen to be doing something. Second, the current regime
is bureaucracy-intensive, while most reform would require less bureaucracy, at least on some margins.
Third, fishermen are a particularly difficult interest group for the federal government to deal with. They
have developed a reputation for occupying DFO offices and other federal buildings and disrupting politi-
cal meetings when displeased. They are used to physical danger and are generally not deterred by the
possibility of physical confrontations. Given the nature of the salmon fishery, they are mostly in port with
a low opportunity cost of time. Politicians are loath to confront them.

42 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
dissipation) results from the combination of the initial market failure and inefficient
government policies that do not eliminate rent dissipation. Depending on assumptions,
the presence of other government (transfer) programs exacerbates the losses. Under
the system in operation up to 1995, and ignoring unemployment insurance payments,
we estimate the NPV of the fishery to be about -$1.3 billion. If the fleet reduction of
1996 results in a proportionate reduction in private costs, the NPV of the fishery
would increase by about $904 million. The $80 million expended on the vessel buyback
would be money well spent. On the other hand, if fishermen follow past patterns of
capacity expansion (through capital-stuffing), the fishery could continue to extinguish
wealth rather than create it. Finally, the dramatic increases in NPV that would follow
further private and public cost reduction clearly show that policies to reduce these
costs should be vigorously examined. Whether fundamental policy reform is
achievable, however, remains to be seen.
Undoubtedly other commercial fisheries’ policies, particularly in developed countries
with significant fisheries-related government expenditures, suffer the same types of
overdissipation that characterize the Canadian Pacific salmon fishery and confront
the same impediments to policy change. Policy evaluation and reform in those
jurisdictions might well profit, as a starting point, from the type of cost-benefit analysis
done here.
The authors gratefully acknowledge helpful comments from two anonymous referees and
the participants at the Western Economic Association International Conference, Lake Tahoe,
28 June–2 July 1998, especially Bill Niskanen, Gordon Tullock, and David Weimer.
RICHARD SCHWINDT holds a joint appointment in the Department of Economics and
the Faculty of Business Administration at Simon Fraser University.
AIDAN VINING is the CNABS Professor of Business and Government Relations in the
Faculty of Business Administration, Simon Fraser University. He is Co-Director of the
SFU-UBC Centre for the Study of Government and Business.
STEVEN GLOBERMAN is the Ross Distinguished Professor of Canada-U.S. Business
and Economic Relations and Director of the Center of International Business at Western
Washington University.
REFERENCES
ARA Consulting Group (1996). Fishing for answers: Coastal communities and the B.C. salmon
fishery. Report prepared for the B.C. Job Protection Commission, Vancouver, B.C.
Bates, R.H. (1988). Contra contractarianism: Some reflections on the new institutionalism.
Politics and Society, 16, 387–401.
Bishop, R.C. (1973). Limitations of entry in the United States fishing industry An economic
appraisal of a proposed policy. Land Economics, 49, 381–390.
Bjorndal, T. (1990). The economics of salmon aquaculture. London: Blackwell Scientific Pub-
lications.
Boardman, A.E., & Greenberg, D.H. (1998). Discounting and the social discount rate. In F.
Thompson & M.T. Green (Eds.), Handbook of public finance. New York: Marcel Dekker.
Boardman, A.E., Vining, A.R., & Waters II, W.G. (1993). Costs and benefits through bureau-
cratic lenses: Example of a highway project. Journal of Policy Analysis and Management, 12,
532–555.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 43
Boardman, A.E., Mallery, W.L., & Vining, A.R. (1994). Learning from ex ante/ex post cost-
benefit comparisons: The Coquihalla highway example. Socioeconomic Planning Sciences,
28, 69–84.
Burgess, D.F. (1981). The social discount rate for Canada: Theory and evidence. Canadian
Public Policy, 7, 383–394.
Campbell, H.E. (1998). When businesses collide: Producer interest groups and the politics of
regulation. Policy Studies Journal, 26, 310–328.
Chappell, H.W., & Keech, W.R. (1983). Welfare consequences of the six-year presidential term
evaluated in the context of a model of the U.S. economy. American Political Science Review,
77, 75–9 1.
Clark, C.W. (1973). The economics of overexploitation. Science, 181, 630–634.
Crutchfield, J., & Pontecorvo, G. (1969). The Pacific salmon fisheries: A study in irrational
conservation. Baltimore: Johns Hopkins University Press.
DFO (1992a). Pacific fisheries overview. Draft discussion paper, Vancouver, B.C.: Pacific Re-
gion, Program Planning and Economics Branch.
DFO (1992b). Financial performance of the British Columbia salmon fleet, 1986–1990.
Vancouver, B.C.: Pacific Region, Program Planning and Economics Branch.
DPA Group. (1988). British Columbia salmon fleet financial performance, 1981–1985. Vancouver,
B.C.: DFO.
Dupont, D.P. (1990). Rent dissipation in restricted access fisheries. Journal of Environmental
Economics and Management, 19, 26–44.
Fraser, G.A. (1977). License limitation in the British Columbia salmon fishery (Technical Re-
port series No. PAC/T-77-13) Environment Canada, Fisheries and Marine Service, Econom-
ics and Special Industry Services Directorate, Pacific Region.
Gilligan, T.W., Marshall, W.J., & Weingast, B.R. (1989). Regulation and the theory of legislative
choice: The Interstate Commerce Act of 1887. Journal of Law and Economics, 32, 35–61.
Gordon, H.S. (1954). The economic theory of a common property resource: The fishery. Jour-
nal of Political Economy, 62, 124–142.
Grafton, R.Q. (1996). Individual transferable quotas and Canada’s Atlantic fisheries. In D.V.
Gordon & G.R. Munro (Eds.), Fisheries and uncertainty. Calgary, Alberta: University of Calgary
Press.
Higgs, R. (1982). Legally induced technical regress in the Washington salmon fishery. Research
in Economic History, 7, 55–46.
Hogarth, R.M., & Makridakis, S. (1981). Forecasting and planning: An evaluation. Manage-
ment Science, 27, 115–138.
Hulten, C.R. (1990). The measurement of capital. In E. Berndt & J. Triplett (Eds.), Fifty years
of economic measurement: The jubilee of the conference on research in income and wealth.
Chicago: University of Chicago Press.
Jenkins, G.P. (1977). Capital in Canada: Its social and private performance, 1965–1974. Dis-
cussion Paper No. 98, Economic Council of Canada. Ottawa, Ontario: Ministry of Supply
and Services.
Jenness, D. (1977). Indians of Canada (7th ed.). Toronto, Ontario: University of Toronto Press.
Johnson, R.N., & Libecap, G.D. (1982). Contracting problems and regulation: The case of the
fishery. American Economic Review, 72, 1005–1022.
Lesser, J., & Zerbe, R.O., Jr. (1998). A practitioner’s guide to benefit-cost analysis. In F. Thomp-
son & M.T. Green (Eds.), Handbook in public finance. New York: Marcel Dekker.
Lyon, R. (1990). Federal discount policy, the shadow price of capital, and challenges for re-
forms. Journal of Environmental Economics and Management,18 (Part 2), S29–S50.

44 / Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery
MacRae, C.D. (1977). A political model of the business cycle. Journal of Political Economy, 85,
239–263.
McClurg, T. (1997). Bureaucratic management versus private property: ITQs in New Zealand
after 10 years. In L. Jones & M. Walker (Eds.), Fish or cut bait: The case for individual transfer-
able quotas in the salmon fishery of British Columbia. Vancouver, B.C. Fraser Institute.
Mikesell, J. (1978). Election periods and state tax policy cycles. Public Choice, 33, 99 105.
Miller, K.A. (1996). Salmon stock variability and the political economy of the Pacific salmon
treaty. Contemporary Economic Policy, 14, 112–129.
Olson, M. (1982). The rise and decline of nations. New Haven, CT: Yale University Press.
Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action.
New York: Cambridge University Press.
Pauly, D., Christensen, V., Dalsgaard, J., Froese, R., & Torres, F., Jr. (1998). Fishing down ma-
rine food webs. Science, 279, 860–863.
Pearse, P. (1982). Turning the tide, a new policy for Canada’s Pacific fisheries. Ottawa, Ontario:
Supply and Services, Canada.
Runolfsson, B., & Arnason, R. (1997). Individual transferable quotas in Iceland. In L. Jones &
M. Walker (Eds.), Fish or cut bait: The case for individual transferable quotas in the salmon
fishery of British Columbia. Vancouver, B.C.: Fraser Institute.
Schelle, K., & Muse, B. (1986). Efficiency and distributional aspects of Alaska’s limited entry
program. In N. Mollett (Ed.). Fishery access control programs worldwide. Fairbanks: Uni-
versity of Alaska Sea Grant Program.
Schelle, K., Dinneford, E., Fay-Hiltner, V., Hall, S., Muse, B., Singer, J., & Stewart, E. (1982).
Alaska’s fishing fleets: A compilation of data on residence of gear operators, vessel character-
istics and fishery diversification patterns for some major Alaskan fishing fleets, 1969–1980.
Juneau: Alaska Commercial Fisheries Entry Commission.
Schrank, W. (1996). Origins of Atlantic Canada’s fishing crisis. In D.V. Gordon & G.R. Munro
(Eds.), Fisheries and uncertainty. Calgary, Alberta: University of Calgary Press.
Schwindt, R. (1987). Public policy and the Pacific salmon fishery’s harvesting crisis. In T. Gunton
& J. Richards (Eds.), Resource rents and public policy in western Canada. Halifax, Nova
Scotia: Institute for Research on Public Policy.
Schwindt, R. (1995). The case for an expanded Indian salmon fishery: Efficiency, fairness, and
history. In H Drost. B.L. Crowley, & R. Schwindt (Eds.), Market solutions for native poverty.
Toronto, Ontario; C.D. Howe Institute.
Scott, A.D. (1955). The fishery: The objectives of sole ownership. Journal of Political Economy,
63, 116–124.
Shoven, J.B., & Topper. M. (1992). The cost of capital in Canada, the United States and Japan.
In J.B. Shoven & J. Walley (Eds.), Canada-U.S. tax comparisons. Chicago: University of Chi-
cago Press.
Sinclair, S. (1960). Licence, limitation—British Columbia: A method of economic fisheries
management. Ottawa: Ontario: DFO.
Statistics Canada (n.d.). CANSIM. Available online at http://www.statcan.ca/.
Tobin, B. (1995, March 7). Statement in response to the report of the Fraser River sockeye
public review board. Speech delivered in Montreal, Quebec. Available: http://www.ncr.dfo.ca/
communic/newsrel/1995/HQ28E.htm.
U.S. Office of Management and Budget (1992, October 29). Guidelines and discount rates for
benefit-cost analysis of federal programs. Circular A-94 (revised).
Vining, A.R., & Weimer, D.L. (1999). Informing institutional design: Strategies for compara-
tive cumulation. Journal of Comparative Policy Analysis, 1, 39–60.

Net Loss: A Cost-Benefit Analysis of the Canadian Pacific Salmon Fishery / 45
Wade, R. (1987). The management of common property resources: Collective action as an
alternative to privatisation or state regulation. Cambridge Journal of Economics, 11, 95–
106.
Walters, C. (1995). Fish on the line. Vancouver, B.C. The David Suzuki Foundation.
Weingast, B.R., Shepsle, K.A., & Johnsen, C. (1981). The political economy of benefits and
costs: A neoclassical approach to distributive politics. Journal of Political Economy, 89, 642–
664.
Weimer, D.L., & Vining, A.R. (1999). Policy analysis: Concepts and practice (3rd ed.). Upper
Saddle River, NJ: Prentice Hall.
Whittington, D. & MacRae, D. Jr. (1986). The issue of standing in benefit-cost analysis. Journal
of Policy Analysis and Management, 5, 665–682.
Wilen, J (1988). Limited entry licensing: A retrospective assessment. Marine Resources Eco-
nomics, 5, 313–324.
Zerbe, R.O., & Dively, D. (1994). Benefit-cost analysis in theory and practice. New York:
HarperCollins College Pub.