1

2
3
4
Harvesting efficiency in restricted open-access fisheries and the welfare relevance of
5 maximum sustainable yield
6
7
8
9 Martin Quaas
10
11 Department of Economics, Leipzig University and German Centre for Integrative Biodiversity Research (iDiv)
12 Halle-Jena-Leipzig
13 Deutscher Platz 5e
04103 Leipzig
14
Germany
15 (martin.quaas@idiv.de)
16
17
18
19
20
Anders Skonhoft
21
Economic Department
22
Norwegian University of Science and Technology
23 7491 Dragvoll-Trondheim
24 Norway
25 (Anders.skonhoft@svt.ntnu.no)
26
27
28
29 Abstract:
30
Small-scale fisheries often operate under conditions of restricted open access with a limited
31
number of licensed fishers. There are no direct restrictions on catches, such that individual
32
fishers can choose the amount of catch, but harvesting efficiency is limited both by the state of
33
34
technology and by regulations of fishing gear and fishing practices. We study how an increase
35 in harvesting efficiency changes the different components of welfare – consumer surplus and
36 producer surplus – in such a restricted open access fishery, taking the feedback of harvesting
37 on stock dynamics into account. We find that both components of welfare change in the same
38 direction. If, and only if, initial efficiency is low enough so that there is no maximum
39 sustainable yield (MSY) overfishing in steady state, an improvement of harvesting efficiency
40 increases welfare.
41
42
43
44 Keywords:
45
46 Fishing efficiency, myopic exploitation, bioeconomics, maximum sustainable yield
47
48
49
50
51 JEL: Q22
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
1
2
3
4 Harvesting efficiency in restricted open-access fisheries and the welfare relevance of
5 maximum sustainable yield
6
7
8
9
10
11
12
13
14
15
16
17
18 draft, Dec 16, 2019
19
20 Abstract:
21 Small-scale fisheries often operate under conditions of restricted open access with a limited
22 number of licensed fishers. There are no direct restrictions on catches, such that individual fishers
23 can choose the amount of catch, but harvesting efficiency is limited both by the state of technology
24 and by regulations of fishing gear and fishing practices. We study how an increase in harvesting
25 efficiency changes the different components of welfare – consumer surplus and producer surplus
26 – in such a restricted open access fishery, taking the feedback of harvesting on stock dynamics into
27 account. We find that both components of welfare change in the same direction. If, and only if,
28 initial efficiency is low enough so that there is no maximum sustainable yield (MSY) overfishing
29 in steady state, an improvement of harvesting efficiency increases welfare.
30
31
32
33 Keywords:
34
35 Fishing efficiency, myopic exploitation, bioeconomics, maximum sustainable yield
36
37
38 JEL: Q22
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54 1
55
56
57
58
59
60 1 Introduction
61
62
Many fisheries worldwide are operating under restricted or regulated open access, including small-
63
64 scale marine fisheries especially in developing countries (FAO, 2007). About 90% of the world’s
65
66 fishermen and over half of the fish consumed each year are captured by small scale, often inshore
67 fisheries, which are of the restricted open access type, or local common pool resources (Ostrom,
68
69 1990, p. 27; FAO, 2007; World Bank, 2012). Harvesting efficiency in those fisheries depends on
70
the current state of technology, but often it is also limited by regulations of fishing gear and
71
72 practices. Whereas improving efficiency would unambiguously enhance welfare in a first-best
73
74
setting (see, e.g., Clark, 1990, Ch. 2), this is far from obvious in the actual second-best world of
75 most fisheries, with the imperfect regulation and enforcement of resource use. The reason is that
76
77 improving efficiency tends to increase harvesting and reduce the stock, which may undermine
78 long-run productivity of the resource and thus ultimately reduce welfare.
79
80
81 This paper aims to characterize conditions under which a costless improvement of harvesting
82 efficiency will increase or decrease welfare in an imperfectly regulated fishery. To this end, we set
83
84 up a bioeconomic model of a restricted open access fishery. The restricted access of the fishery is
85
considered such that the number of fishermen (or vessels) is fixed. Changing profitability in the
86
87 fishery thus has no effect on the extensive margin, but it will have an effect on the intensive margin,
88
89 as the active fishermen will adjust fishing effort in response to changing profitability. Assuming
90 that the number of fishermen is large – as it often is the case for small-scale fisheries – individual
91
92 fishermen will rationally ignore the effect of their harvest on the dynamics of fish populations
93
biomass (the shadow price is zero). Welfare is derived from the fishery in terms of producer surplus
94
95 and consumer surplus (see, e.g., Copes, 1972; Quaas et al., 2018; Jensen et al., 2019), which is
96
97 certainly relevant for small-scale fisheries in developing countries that serve as an important local
98 source of food supply.
99
100
101
We find that the long run components of welfare – consumer and producer surplus – in the
102 considered restricted open-access fishery change with harvesting efficiency in a non-monotonic
103
104 fashion. This shows how more efficient fishing methods, or more lenient gear restrictions, may be
105 a mixed blessing not only for the catch and fish abundance, but also for the profitability and the
106
107 welfare of the local fishing community. Importantly, we characterize under which conditions
108
109
110 2
111
112
113
114
115
116
increasing efficiency will improve, or reduce welfare. We find that at initially low levels of
117 efficiency an improvement of harvesting efficiency increases both consumer and producer surplus,
118
119 while it reduces both components of welfare when the initial level of efficiency is high. We further
120 find that the turning point is the same for both welfare components. It is the level of efficiency for
121
122 which the resulting steady-state population size would give rise to the maximum sustainable yield
123
(MSY) harvesting. MSY as welfare maximizing harvesting is indeed a very surprising result, given
124
125 that the model includes fishing costs that decrease with fish abundance. This result holds true for
126
127 a large range of specifications for benefits and costs derived from harvesting, allowing for stock-
128 dependency of both benefits and costs. It is demonstrated that the result stems from the stock-
129
130 agumenting property of harvesting efficiency, i.e. that a change in harvesting efficiency is
131
equivalent to a change in resource stock size. Whe show that under this condition, welfare in
132
133 biological and market equilibrium can be expressed as an increasing function of harvest alone and
134
135 thus long run welfare effects of harvesting efficiency depend on the steady-state harvest level only.
136
137 A mixed blessing of more efficient fishing technology is discussed, among others, by Withmarsh
138
139
(1990), Murray (2007), and Squires and Vestergaard (2013a,b). The latter contributions are
140 concerned with different types of technical progress, finding that the rapid technological progress
141
142 in the past contributed to the decline of most, if not all, global fisheries. Gordon and Hannesson
143
(2015) gives an in depth analysis of technological progress and the stock collapse in the Norwegian
144
145 winter herring fishery, and provide evidence that the introduction of the power block technology
146
147
was the principal factor in the demise of the stock. Hannesson et al. (2010) and Eide et al. (2003)
148 find similar results for other fisheries.
149
150
The next section presents our bioeconomic model for a search fishery and for a schooling fishery.
151
152 Section 3 presents the results for both types of fishery. In section 4 the results are illustrated for a
153
154 case study; the Sengalese small scale fishery on Sardinella Aurita. In Section 5 we proceed and
155 show that our main results generalize to a much wider class of fishing technologies and utility
156
157 derived from harvesting the resource, i.e. we consider a general function that describes (net) utility
158
derived from fishing as a function of the catch, stock size, and harvesting efficiency. The final
159
160 section summarizes our findings and concludes.
161
162
163
164
165
166 3
167
168
169
170
171
172 2 Model of a small-scale fishery under restricted open-access
173
174
We consider a single fish stock exploited instantaneously and simultaneously in a myopic manner
175
176 in a local fishing community. The population dynamics in continuous time is described by:
177
178
dX t
179 = F ( X t )  Ht , (1)
180 dt
181
182 where X t is the stock size (measured in biomass) at time t , H t  0 is the harvest, and F ( X t ) is
183
184 the natural growth function, assumed to be density dependent. We adopt the standard assumptions
185
186 F (0) = F ( K ) = 0 , where K > 0 is the carrying capacity of the fish stock, growth is positive for
187
188
biomass positive and below carrying capacity, F ( X t ) > 0 for X t  (0, K ) , F ( X t ) < 0 , and the
189
190
biomass growth function F ( X t ) is single-peaked, i.e. there is a unique stock size X msy that
191
maximizes F ( X t ) , i.e. where F ( X msy ) = 0 .
192
193
194 We first specify and analyze two models of harvesting technology and welfare standard in the
195
196 literature – one closely follows the Gordon (1954) and Schaefer (1957) model of a search fishery,
197 the other one the schooling fishery model where fishing costs are independent of the stock size
198
199 (Bjørndal, 1988; Tahvonen et al., 2013). In both cases, and as we are considering a small-scale
200
fishery, the fish is sold on a local market and where the fish price depends on the amount of fish
201
202 available. We model this by introducing the inverse demand function pt = P( H t ) , with
203
204 P( H t ) < 0 . Fishermen have no market power and hence take the price pt as given.
205
206
207 2.1 Gordon-Schaefer model of search fishery
208
209
210
In this model we consider a harvest function given as H t = qEt X t , such that the harvest, or catch,
211
212
is proportional to fishing effort Et and stock size X t (Gordon, 1954; Schaefer, 1957). The
213
efficiency parameter, or ‘catchability’ coefficient, q is assumed to be exogenous, where a higher
214
215 value indicates more efficient fishing technology. We can think of an increase in q as the result
216
217 of technical progress (Squires and Vestergaard, 2013a,b) or as a result of a more lenient regulation
218
219
of fishing gear, equipment or methods. All the time, changing the value of q is assumed to be
220
221
222 4
223
224
225
226
227
228
costless.
229
230 On the other hand, fishing effort Et is costly, and we use C ( Et ) to denote the effort cost
231
232 function. Marginal effort costs are positive, C ( Et ) > 0 , and non-decreasing C ( Et )  0 .1 We
233
234 think of this cost function as describing the (opportunity) costs of fishing for the whole fishing
235
community, reflecting different productivity (fishing skill) among the various fishermen (Péreau
236
237 et al., 2012; Grainger & Costello, 2016) and different productivities of outside options (Baland &
238
239 Francois, 2005, Okonkwo & Quaas 2019). Therefore, in this fishing community the fishermen
240 with high fishing productivity, or low opportunity costs of fishing, earn high intramarginal, or
241
242 Ricardian, rent, and vice versa. Based on the Gordon-Schaefer harvest function, and the effort cost
243
function, we can formulate the harvesting cost function for the search fishery as:
244
245
246  H 
247 Ct = C GS ( H t , X t , q ) = C  t . (2)
248
 q Xt 
249
250 From the above assumptions, it follows that fishing cost are increasing and (weakly) convex in
251
252 total catch, CHGSt > 0 , CHGSt Ht  0 , decreasing and convex in stock size and catchability coefficient,
253
254 C XGSt < 0 , C XGSt X t  0 , CqGS < 0 , CqqGS  0 . Furthermore, the cross derivatives with respect to H t
255
256 and both stock size and catchability are negative, CHGSt X t  0 , CHGSt q  0 , i.e. marginal fishing cost
257
258 decrease both with stock size and harvesting efficiency. The current profit, or producer surplus,
259 reads then:
260
261
262  tGS = pt H t  C GS ( H t , X t , q). (3)
263
264 The instantaneous welfare derived from the fishery is the sum of consumer and producer surplus,
265
266 and given the inverse demand function pt = P( H t ) it reads:
267
268
Ht  H  Ht  H 
U tGS =  P(h) dh  pt H t  pt H t  C  t  =  P(h) dh  C  t  .
269
(4)
 q Xt   q Xt 
270 0 0

271
272
273
274
275 1 In the following, the univariate function C () will always indicate the effort cost function. In contrast, costs expressed as function of catch,
276
stock size and efficiency differ for the search fishery and the schooling fishery, which is indicated by superscripts.
277
278 5
279
280
281
282
283
284
285
286 2.2 Schooling fishery model
287
288
289
Our model of the schooling fishery is similar to the previous model, except that we consider harvest
290 as independent of the current stock size, H t  qEt . With a similar type effort cost function as
291
292 above, the cost function for the schooling fishery is:
293
294
H 
295 Ct = C SF ( H t , q ) = C  t . (5)
296  q 
297
298
299
Again, from the above assumptions on the effort cost function, it follows that fishing cost also now
300 are increasing and (weakly) convex in total catch, CHSFt > 0 , CHSFt Ht  0 , decreasing and convex in
301
302 catchability, CqSF < 0 , CqqSF  0 , and the cross derivatives with respect to H t and catchability is
303
304 negative, CHSFt q < 0 , i.e. marginal fishing cost decrease with harvesting efficiency. With the current
305
306 profit defined as:
307
308
309  tSF = pt H t  C SF ( H t , q) (6)
310
311 the instantaneous welfare derived from this fishery becomes:
312
313
Ht H 
314
U tSF =  P(h) dh  C  t  . (7)
315 0
 q 
316
317
318 3 Analysis and results for the small-scale fishery under restricted
319
320 open access
321
322
323 As the first step of the analysis, we characterize the harvest under restricted open access for the
324
two models, j  {GS , SF } . Myopic profit maximization for the given stock X t > 0 , and where
325
326 hence the fishing impact on the stock is neglected (the zero shadow price assumption), yields
327
328  t C j ()
= pt   0. This condition defines the (inverse) supply function of fish to the local
329 H t H t
330
331 market. If marginal harvesting costs are increasing, inverse supply is a smoothly increasing
332
333
334 6
335
336
337
338
339
340 function of catch H t . If marginal harvesting costs are constant, inverse supply is a ‘bang-bang’
341
342 curve, i.e. zero whenever marginal costs are above pt and maximum possible if they are below
343
344
pt (Pindyck, 1984). In market equilibrium, it must hold that the inverse supply equals inverse
345
346
demand, and for H t > 0 thus:
347
348 P( H t* ) = CHj *   . (8)
349 t

350
351
Note that the downward-sloping inverse demand function P( H t ) < 0 guarantees a unique market
352
353 equilibrium even for constant marginal harvesting costs.
354
355
356 3.1 Gordon-Schaefer model of search fishery
357
358 We start to analyze the search fishery, and where it follows that market equilibrium harvest in the
359
360 restricted open access fishery is positive or zero according to:
361
362
 H ( X t , q ) = 0 if P(0)  CHGSt (0, X t , q )  0
*
363 H = t
(9)
364  H ( X t , q ) > 0, else.
365
366
367
Therefore, in a search fishery, H t* is an increasing function of fish abundance X t , whenever
368
369
harvest is positive in the first place, H t* > 0 . Differentiating Eq. (8), using the implicit function
370
theorem, yields:
371
372
373 dH t* CHGSt X t
= > 0 for H t* > 0. (10)
374 dX t P( H t )  CHGSt Ht ( H t , X t , q )
375
376
377 The harvest locus may be concave or convex, depending on third order derivatives of inverse
378
379 demand and cost functions.
380
381 Furthermore, more efficient fishing through a higher value of q shifts down the cost function as
382
383 well as the marginal costs, and hence increases the catch for a given size of the fish stock. This
384
follows from differentiating condition (8) with respect to q , which yields, using the implicit
385
386 function theorem:
387
388
389
390 7
391
392
393
394
395
396 dH t* CHGSt q
= > 0 for H t* > 0. (11)
397 dq P( H t )  CHGSt Ht ( H t , X t , q )
398
399
400
401
402 Moreover, we have that, at any given stock size, the restricted open-access harvest varies with
403
404 harvesting efficiency q from zero to a maximum H max   given by P( H max ) = 0 , as stated in
405
406 the following lemma.
407
408 Lemma 1 For any given level H 0   0, H max  of harvest and any given stock size X t ,
409
410
411 1. there exists a q such that H t* < H 0 for all q < q ,
412
413
414 2. there exists a q such that H t* > H 0 for all q < q .
415
416
Proof. The function g ( z ) = zC ( z ) is defined and monotonically increasing on the set of non-
417
418 negative real numbers z  (0, ) . Furthermore, g (0) = 0  C (0) = 0 and lim z g ( z ) =  . Thus
419
420 for any given H 0  (0, H max ) there exists a z 0 that solves H 0 P( H 0 ) = z 0 C ( z 0 ) . Define
421
422 q  q  H 0 /  z 0 X t  . The rest follows from the monotonicity of H t* as a function of q , result
423
424 (11).
425
426
Producer surplus and consumer surplus in the restricted open access market equilibrium are:
427
428
429  t* = CHGSt ( H t* , X t , q) H t*  C GS ( H t* , X t , q) (12)
430
431
and
432
433
H t*
CSt* =  P(h)dh  P( H t* ) H t* ,
434
(13)
435 0

436
437 respectively. We have the following result.
438
439
440 Result 1 In the short-term, i.e. for given fish abundance X t , more efficient fishing technology
441
442
1. increases producer surplus if the elasticity of the inverse demand is smaller than unity,
443
444  H P( H ) / P( H ) < 1 ,
445
446 8
447
448
449
450
451
452
2. increases consumer surplus.
453
454 Proof. See Appendix A.
455
456
Therefore, not surprisingly, we find that in the short run and for a given fish abundance X t , a
457
458 more efficient harvesting technology, or more lenient gear restrictions, improves welfare. For
459
460 producer surplus, more efficient fishing equipment through reduced costs CqGS < 0 works in the
461
462 direction of higher profit and this effect dominates the negative price effect due to a higher harvest
463
and reduced price if the elasticity of inverse demand is small enough. Consumer surplus will for
464
465 sure increase through the reduced market price following more efficient harvesting. More efficient
466
467
technology is therefore not only beneficially for the local fishermen in the short-term, as also
468 shown in a somewhat different setting by Anderson (1986), but it also improves the consumer
469
470 surplus and welfare of the local community in the short term.
471
472 The more interesting question is, however, the welfare effect of increased harvesting efficiency in
473
474 the long run; that is, when the effect of a higher valued q on the stock is taken into account as
475
476 well. With the optimized harvest function H t* = H ( X t , q )  0 from Eq. (9) inserted into the stock
477
growth (1), the dynamics of the harvested fish population is described by:
478
479
480 dX t
= F ( X t )  H ( X t , q ). (14)
481 dt
482
483
484 A steady state is defined as X * when dX t / dt = 0 through:
485
486 F ( X * ) = H ( X * , q ). (15)
487
488
489 Moreover, the steady state described by (15) is locally stable if:
490
491 F ( X * )  H X ( X * , q) < 0. (16)
492
493
494 Therefore, under condition (16), a small deviation from the equilibrium stock size generates a
495
496
dynamic that pushes the stock back to the equilibrium. That is, if the stock is slightly smaller
497 (larger) than the equilibrium value, biological growth exceeds (falls short of) harvest. The stock
498
499 accordingly increases (decreases) again up (down) to the equilibrium level.
500
501
502 9
503
504
505
506
507
508
The given assumptions on the natural growth function, and the result that the harvest locus is
509 strictly increasing in fish biomass, as described by Eq. (10), are not sufficient to characterize the
510
511 number of steady states, i.e. the number of solutions to the equilibrium condition (15). If the inverse
512 demand function is bounded from above, P(0) <  , there will always be an interval of sufficiently
513
514 small stock sizes where marginal harvesting costs exceed the price of fish (see also Nævdal and
515
516 Skonhoft, 2018). Combined with the result that harvest is (weakly) increasing with stock size, Eq.
517 (10), this implies that P(0) <  is a sufficient condition for the existence of at least one steady
518
519 state.
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542 Figure 1: Phase diagram and steady states for the case of the Senegalese Sardinella Aurita small-scale fishery. The
543 green line shows fish population biomass growth, the blue and red lines show a set of indifference curves of
544 instantaneous welfare, and the purple line, connecting the vertical segments of welfare indifference curves, indicates
545
546 market equilibrium harvest. The filled circle marks a stable steady state, the empty circle an unstable one, and the
547 dashed circle the steady state that maximizes welfare.
548
549
550
551
552
Figure 1 illustrates the phase diagram and steady states for the case of the Senegalese small scale
553 fishery (details are presented in section 4). Under the given assumptions and the given value of the
554
555 efficeincy parameter q , there are three interactions between the fish population growth function
556
557
558 10
559
560
561
562
563
564
and the harvest locus and hence three steady states. One at zero stock size and harvest, another
565 stable steady state at a rather high stock size, respectively, and an unstable steady state at an
566
567 intermediate stock size. In the figure also indifference curves of the instantaneous welfare function
568 are shown, to illustrate that market equilibrium harvest, at any given stock size, maximizes
569
570 instantaneous welfare for the search fishery where instantaneous welfare increases with stock size,
571
and has a unique maximum with respect to harvest. The figure also shows the steady state that
572
573 maximizes steady-state welfare, and which is determined by the stock size and harvest where a
574
575 welfare indifference curve and the graph of the biomass growth function are just tangents to each
576 other.
577
578
579 We now turn to the main question studied in this paper; how does increased harvesting efficiency
580 affect consumer surplus and producer surplus in the long run? To this end, differentiate Eq. (15)
581
582 with respect to q and apply the implicit function theorem. This leads to:
583
584
dX *
585  F ( X *
)  H X* ( X * , q ) 
dq
= H q* ( X * , q ). (17)
586
587
588 Through Eq. (11), we have seen that a higher q always increases the harvesting pressure, i.e. the
589
590 right-hand-side of (17) is positive. As we are considering a locally stable steady state, the factor in
591
592 brackets on the left-hand side of (17) is negative. Hence, a small increase in harvesting efficiency
593 will consistently lower the stable equilibrium stock, dX * / dq < 0 . On the other hand, more
594
595 efficient technology may either reduce or increase F ( X * ) , and hence through Eq. (15) also either
596
597 reduce or increase steady-state harvest. The critical steady-state stock size is here X msy , and where
598
599
steady-state harvest increases with efficiency if X * < X msy . As consumer surplus unambiguously
600 increases with harvest, we have:
601
602
 msy dCS* 
603 X* X  0. (18)
604  dq 
605
606
Thus, if the initial harvesting efficiency is low enough such that the stock is not MSY-overfished
607
608 in the considered restricted open access steady state, improved fishing efficiency – be it due to
609
610 technical progress or to a change in fish gear regulation – will increase consumer surplus both in
611 the short run and in the long run.
612
613
614 11
615
616
617
618
619
620
However, the effect of an increase in q on producer surplus is not quite as straightforward, as
621 producer surplus is also directly affected by the decrease in stock size triggered by an increase in
622
623 q . To study the effect of a small change on q on producer surplus at steady state, which is
624
625 defined through Eq. (12), we differentiate this expression with respect to q , and thereby take the
626
627
effect of q on X * and on H t* = F ( X * ) into account. Using Eq. (8) for H t* > 0 , this yields
628
(omitting arguments of functions):
629
630
d *
H *  CqGS   CHX H H X*  C XGS  X q* .
631 GS
= CHH H * H q*  CHq
GS GS
H *  CHH
GS
(19)
632 dq
633
634
635 This expression contains positive and negative terms. Therefore, while Result 1 indicates that the
636 short-run effect of more efficient fishing increases producer surplus in the search fishery (provided
637
638 that the elasticity of the inverse demand function is less than one), the long-term effect is generally
639
ambiguous. The reason is that the long-run effect comprises two opposite forces, and where the
640
641 direct positive effect of a higher q working through the cost function (the first three terms on the
642
643 right-hand side of Eq. 19) is counterbalanced by a negative indirect effect also working through
644
the cost function by a reduction in the fish abundance (the last term on the right-hand side of Eq.
645
646 19).
647
648
Under the given assumptions, we can characterize the sign of the net effect of more efficient
649
650 harvesting on steady-state welfare. To this end, use the cost function (2) for the search fishery in
651
652
condition (8) characerizing the market equilibrium harvest,
653
654 Ht  H   H 
655 H t P( H t ) = C  t  = g  t  , (20)
q Xt  q Xt   q Xt 
656
657
658 where g   is the function we defined in the proof of Lemma 1, which is montonically increasing
659
660
and thus invertible. Thus, q X t = H t /  g 1  H t P( H t )   in market equilibrium. Using this in Eq.
661
662 (3), producer surplus can be expressed as an increasing function of harvest alone.2 This indicates
663
664 that the effect of changing harvesting efficiency on steady state harvest is determining how it
665
666
667
668 2 The function is increasing, as d  t / d  q X t  > 0 and as d  q X t  / dH t = 1 /  g 1  H t P( H t )    H t g (q X t ) > 0 .
669
670 12
671
672
673
674
675
676
influences producer surplus. We have the following result:
677
678 Result 2 The following three statements are equivalent.
679
680 1. The steady state stock size is above/below the stock that generates the maximum sustainable
681
682 yield.
683
684 2. Consumer surplus increases/decreases with harvesting efficiency.
685
686
687
3. Producer surplus increases/decreases with harvesting efficiency.
688
689 Formally,
690
691
 msy dCS*  d * 
692 X *
X  0  0. (21)
693  dq  dq 
694
695 Proof. See Appendix B.
696
697
698
Result 2 shows that in a situation with a high exploitation pressure, channeled through a high
699 harvesting efficiency and X * < X msy , more efficient technology will reduce steady-state producer
700
701 surplus while the opposite happens when initially X * > X msy . Result 2 contrasts the outcome of
702
the standard sole owner biomass model where improved harvesting technology, and hence lower
703
704 catch costs, unambiguously increases the equilibrium rent (again, see Clark, 1990, Ch. 2).
705
706
Both steady-state consumer surplus and steady state producer surplus will therefore be at its
707
708 maximum when the steady stock size is equal to the stock size that generates the maximum
709
710 sustainable yield, X * = X msy . As consumer surplus increases unambiguously with harvest, the
711 maximum of consumer surplus in the search fishery will be higher the larger the MSY is.
712
713 Therefore, the maximum consumer surplus will be higher for larger biological productivity.
714
Maximum producer surplus in the search fishery can be written as
715
716  max = P( H msy ) H msy  C GS  H msy , X msy , q  . For a cost function with a constant elasticity of cost with
717
718 respect to harvest, i.e. where C GS =  H t CHGSt with some constant   1 , we have, using Eq. (8),
719
720  max = P( H msy ) H msy  C GS  H msy , X msy , q  = (1   ) P( H msy ) H msy , which does not depend on
721
722 harvesting efficiency q . We thus have the following result.
723
724
725
726 13
727
728
729
730
731
732
Result 3 For constant marginal harvesting cost, or more generally an iso-elastic harvesting cost
733 function, the maximum steady state welfare is unrelated to fishing efficiency.
734
735
However, while the maximum rent in our model is not contingent upon the technological level, it
736
737 is also clear that there will be a certain value of q that yield the highest rent. This possibly
738
739 efficiency level q = q max solves the stock equilibrium equation F ( X msy ) = H * ( X msy , q max ) and is
740
741 hence contingent upon all the parameters of the model.
742
743
744
745
746
3.2 The schooling fishery
747
748 In a schooling fishery, harvesting costs do not depend on the current stock size. Thus, also the
749
750 level of harvest that fulfills Eq. (8) is independent of X t . Therefore, in the stock – harvest space,
751
752 the restricted open-access harvest schedule H t* = H SF (q ) will be a straight horizontal line. As
753
754 above, the harvest shifts up with higher harvesting efficiency q , H qSF (q ) > 0 . With the
755
756 specification (5), there exists a q for all H such that H SF (q ) > H for all q > q . In particular,
757
758 there exists a q max such that H SF (q ) > H msy for all q > q max .
759
760 A steady state is given by:
761
762
763
F ( X * ) = H S (q) . (22)
764
765 * * * *

766
With F ( X msy ) > H SF (q ) , Eq. (22) yields two equilibria, X and X , with X < X msy < X ,
767 * *
768 where X is unstable, while X is locally stable. Thus, the fish stock will be depleted for sure in
769
770
a schooling fishery under restricted open access with an initial stock size located below that of
771 *
X . In that case, more efficient technology or changing gear restrictions, shifts up the harvest
772
773 locus, H q* > 0 , and drive the stock even faster to extinction. The situation will be of the opposite
774
775 *
with an equilibrium defined through F ( X ) = H (q ) . In this case more efficient harvest
776
777 technology and lower harvesting costs will reduce the stock, but increase the steady state catch,
778
*
779 H SF (q ) = F ( X ) .
780
781
782 14
783
784
785
786
787
788
In the locally stable steady state consumer and producer surpluses can be expressed as:
789
790 H*
CS* =  P(h) dh  P( H * ) H * and  * = P( H * ) H *  C S ( H * , q ) , (23)
791 0

792
793
respectively, and where both are strictly increasing with H * . Thus, we have the following result.
794
795
796 Result 4 In a stable steady state in the schooling fishery with q < q max , more efficient fishing
797
798
increases both consumer surplus and producer surplus.
799
800
801
802
803
4 Case study: Senegalese small scale fishery on Sardinella Aurita
804
805 To illustrate our results, we consider a quantitative case study on the Senegalese small scale fishery
806
807
on Sardinella Aurita, which is operating under conditions of restricted open access. Sardinella
808 aurita are the main target of the Senegalese artisanal fishery. The catch of this small pelagic fish is
809
810 consumed mostly locally in Senegal, and the artisanal purse seine fishery contributes substantially
811 to local food security and livelihoods. Harvesting efficiency is changing due to technical progress
812
813 (e.g., improved motorization of fishing vessels) and climate change, which changes the spatial
814
distribution and accessibility of the resource (Lancker et al. 2019 a,b).
815
816
817 We base the parameterization of our case study on the estimates from Lancker et al. (2019b). They
818
estimate economic parameters for four regions along the Senegalese coastline, here we consider
819
820 the estimates for Thiès Sud. A logistic specification of the natural growth function is used,
821
822 F ( X t ) = r X t 1  X t / K  , and the parameter values are for the entire Senegalese coastline. To
823
824 match spatial scales, we scale down the intrinsic growth rate r by the catches in Thiès Sud
825 relative to total catches by the Senegalese purse seine fleet, and obtain r = 0.78 (1/year) and the
826
827 carrying capacity K = 419 (1000 tons).
828
829 For Thiès Sud, the estimated demand elasticity is 0.3 , and the resulting demand function is
830
831 P( H ) = 160 H 0.3 . Deviating from the generalized Schaefer harvesting function H t  qEt X t and
832
833 c 2
834
the effort cost function C ( Et )  Et , the cost function estimated by Lancker et al. (2019b)
2
835
836
837
838 15
839
840
841
842
843 2
844 c H 
becomes GS
C ( H t , X t , q ) =  t   with  = 0.22 . Market equilibrium catch is thus
845 2 q X t 
846
847 1

 160 2 1 0.3

 q X t0.22   =  q X t0.22  , where the cost parameter is set as
1.54
848 *
H = c = 160 (Central
t
849  c 
850
851 African Francs/effort) by choice of measurement units for the catchability coefficient q , which
852
853
we vary in the subsequent analysis. In the baseline calculation we use q = 4.65 (1/effort) to obtain
854
the observed annual harvest H t = 64 (1000 tons) at the observed biomass X t = 200 (1000 tons)
855
856 (Lancker et al. 2019b, Tables 1 and A5). Figure 2 shows first consumer and producer surplus for
857
858 this Senegalese Sardinella fishery.
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883 Figure 2: Consumer surplus and producer surplus for the Senegalese Sardinella Aurita small-scale fishery, measured
884
in Central African Francs (CFA).
885
886
887
888
889 2
160 1 c  H t 
890 Welfare is the sum of consumer and producer surplus, U GS
= Ht    (million
1  2  q X t 
t
891
892
893
894 16
895
896
897
898
899
900
Central African Francs). and depends on both harvest and stock size in this search fishery.
901 Thus, it makes sense to consider welfare indifference curves in the stock-harvest space.
902
903
Varying the welfare level, we obtain an entire set of indifference curves of instantaneous welfare.
904
905 Note that according to (8), the market equilibrium catch is determined by the condition that
906
907 instantaneous welfare is maximized with respect to H t . For any given X t , the market equilibrium
908
909 harvest thus corresponds to a vertical segment of a welfare indifference curve. This is illustrated
910 in the above Figure 1. The curve showing market equilibrium harvest as a function of fish
911
912 population size connects the vertical segments of the set of welfare indifference curves.
913
914 Steady states of the restricted open-access fishery are obtained as the points of intersection between
915
916 market equilibrium catch curve and the biomass growth curve. The steady state is stable, if the
917
slope of market equilibrium harvest function exceeds the biomass growth function, and unstable
918
919 otherwise. Figure 1 shows a stable steady-state at a high biomass with a filled circle and an unstable
920
921
one at a low biomass with an empty circle.
922
923 Figure 1 also illustrates the point where steady-state welfare is maximized. This is found as the
924
point where the welfare indifference curve is tangent to the biomass growth function, shown as a
925
926 dashed circle in the figure. The steady-state of the restricted open-access fishery is at a higher
927
928 catch, but lower biomass and lower welfare than the allocation that maximizes steady-state
929 welfare, reflecting the well-known issue of resource overuse in restricted open-access.
930
931
932
933
934 5 Generalized model of welfare from restricted open-access fishery
935
936
937 To obtain some indications of how robust the above results are, we will look at a very general
938 model of harvesting and welfare derived from the fishery. In general, aggregate instantaneous
939
940 welfare derived from the fishery is described by a utility function:
941
942 U t = U ( H t , X t , q) , (24)
943
944
945 which is increasing in harvest, U Ht > 0 , nondecreasing in stock size, U X t  0 , and increasing in
946
947 efficiency, U q > 0 . We further assume that marginal utility of catch weakly increases with stock
948
949
950 17
951
952
953
954
955
956
size, U Ht X t  0 , and increases with efficiency, U Ht q > 0 . As above, the assumption U q > 0
957
implies that the immediate effect of increasing efficiency – while keeping stock size constant – is
958
959 positive. The instantaneous welfare function (24) can capture net economic surplus – the sum of
960
961 consumer and producer surplus. The general formulation (24), applied to a commercial fishery,
962 also allows to take into account that consumers’ willingness to pay for fish may include concerns
963
964 for sustainability, for example in a way that the demand for fish positively depends on stock size.3
965
966 The condition determining harvest in restricted open access then simply is:
967
968
969 U H ( H t* , X t , q ) = 0 . (25)
970
971 We assume that the net utility is separable as follows:
972
973
974 U ( H , X , q ) = Uˆ ( H , y ( X , q )) (Assumption 1)
975
976
Harvesting profit with a Gordon-Schaefer cost function is a special case, where y ( X , q ) can be
977
978 interpreted as the catch per unit of effort. Formally, given the separability of utility, the condition
979
980 (25) for harvest under restricted open access fixes the relationship between harvest and y ( X , q ).
981
982 Thus, utility can be expressed as a function of harvest only.
983
984 More intuitively, Assumption 1 imposes a condition how harvesting efficiency affects
985
986
instantaneous welfare: Namely in a way such that an equivalent effect on welfare could be obtained
987 by keeping harvesting efficiency constant, but changing the resource stock size in a specific way.
988
989 This shows that Assumption 1 is related to the idea of labor-augmenting technical change in growth
990 theory (Robinson, 1938): Under labor-augmenting technical change, improving technology has
991
992 the same effect on output as an expansion of labor force. Thus, we can interpret Assumption 1 as
993
994
the property of ‘stock-augmenting’ harvesting efficiency.
995
996 Differentiating (25) with respect to X and q gives, adopting the assumption of stock-
997
998 augmenting harvesting efficiency:
999
1000 U HH H X* = U HX = Uˆ Hy y X (26)
1001
1002
1003
1004
3 The instantaneous welfare function (24) can also model a recreational fishery (see, e.g., Stoeven, 2014) describing the utility of a
representative individual which positively depends on the catch quantity as well as the fish abundance.
1005
1006 18
1007
1008
1009
1010
1011
1012
and
1013
1014 U HH H q* = U Hq = Uˆ Hy yq , (27)
1015
1016
1017 respectively. Under Assumption 1, it follows that U X H q* = U q H X* , where U x = Uˆ y y X and
1018
1019 U q = Uˆ y yq . From the assumptions on U it also follows that H q* > 0 .
1020
1021
1022 Differentiating the steady-state condition F ( X * ) = H * with respect to q gives:
1023
1024
1025 F  H X
*
X q
= H q* . (28)
1026
1027
As stability requires F   H X* < 0 (Eq. 21), this equation implies, X q < 0 .
1028
1029
1030 Now consider a steady state in which changes in q lead to changes in H * giving repercussions
1031
1032 on the steady state stock size X * . Taking these feedbacks into account, we obtain the following
1033
1034 long-term relationship between U and q :
1035
1036 dU 1 F
1037 = U q  U X X q* = U q  U X H q* = Uq . (29)
dq F  HX
*
F   H X*
1038
1039
1040 The first equality holds by virtue of the envelope theorem while the second equality uses Eq. (28)
1041
1042 that states how steady state stock size reacts to changes in q via changes in H . In the last step
1043
we have used the Assumption 1 of stock-augmenting harvesting efficiency, which implies
1044
1045 U X H q* = U q H X* , as shown above. Thus, we can conclude that our main results hold for this
1046
1047 generalized model of welfare as well. This is summarized as follows.
1048
1049
Theorem 1 The following statements are equivalent
1050
1051
1052 1. The steady state stock size is above/below the stock that generates the maximum sustainable
1053 yield.
1054
1055
1056 2. Steady-state welfare increases/decreases with harvesting efficiency.
1057
1058 Formally,
1059
1060
1061
1062 19
1063
1064
1065
1066
1067
 msy dU * 
1068 X* X  0. (30)
1069  dq 
1070
1071
1072
1073
1074 To illustrate this result a bit further, consider the Sengalesian case study again, but for the sake of
1075
the argument we assume a stock elasticity  = 0.65 (section 4) , such that the market equilibrium
1076
1077 harvest is a linear function of stock size.
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095 Figure 3: Effects of increasing harvesting efficiency (from the left to the right), q0 < q1 < q2 . Stock-augmenting
1096
1097 increase in harvesting efficiency compresses the welfare indifference curves to the left. The three blue indifference
1098 curves correspond to the same welfare level in all three graphs. The red indifference curve corresponds to different
1099
welfare levels, as the maximum attainable steady-state welfare level is increasing in harvesting efficiency.
1100
1101
1102
1103
1104 Figure 3 illustrates the effect of increasing harvesting efficiency, q0 < q1 < q2 . Stock-augmenting
1105
1106 increase in harvesting efficiency compresses the welfare indifference curves to the left, i.e. at a
1107
1108 given level of harvest welfare is the same, no matter what the level of harvesting efficiency is.
1109 Thus only the resulting harvest level affects the steady-state level of welfare. For the increase in
1110
1111 harvesting efficiency from q0 to q1 , steady state harvest increases and thus the new steady state
1112
1113 is at a higher welfare level. With the further increase from q1 to q2 the new steady state is at a
1114
1115 lower welfare level.
1116
1117
1118 20
1119
1120
1121
1122
1123
1124
1125
1126 4 Concluding remarks
1127
1128
1129 This paper has theoretically analyzed how increased harvesting efficiency, either through
1130
technological changes or changing gear restrictions, affects steady-state welfare in a restricted
1131
1132 open-access fishery, where a fixed number of fishers harvest a fish stock in a myopic profit-
1133
1134
maximizing manner and where the fishers takes the fish price as given. This model describes a
1135 situation typical for many small-scale fisheries and fishery communities world wide, but especially
1136
1137 in developing countries.
1138
1139 Our most important finding is that the effect of more efficient harvesting goes in the same direction
1140
1141 for the different components of welfare, and that increased harvesting efficiency has an ambiguous
1142 long-term effect on consumer surplus as well as producer surplus, or more generally, on the welfare
1143
1144 in the considered fishing community. Both components of welfare increase with harvesting
1145
efficiency in steady state if the fish stock exceeds the size that would generate the maximum
1146
1147 sustainable yield (MSY). In contrast, if the steady state fish population is smaller than the MSY
1148
1149 stock size, i.e., if there is MSY overfishing, increasing harvesting efficiency reduces welfare. As
1150 the steady state stock size decreases with harvesting efficiency in restricted open access, it depends
1151
1152 on the initial harvesting efficiency whether an increase is welfare-enhancing or detrimental. It is
1153
shown that the critical level of harvesting efficiency depends on demand conditions and the market
1154
1155 price of harvest, and on the biological productivity of the considered fish stock. MSY as welfare
1156
1157 maximizing harvesting is a surprising result as the model includes fishing costs.
1158
1159 We have further shown that the central property of the harvesting technology that leads to this
1160
1161
result is the stock-augmenting harvesting efficiency. This means that for any increase in harvesting
1162 efficiency there is an equivalent increase in the resource stock size that would lead to the same
1163
1164 productivity effect, an assumption that is ubiquitous in resource economics. We have focused on
1165 a steady-state analysis. Our analysis can be easily extended to a dynamic setting with discounting.
1166
1167 As to be expected, the critical stock size below which an increase in harvesting efficiency reduces
1168
1169
welfare is determined by the condition that marginal stock growth equals the discount rate.
1170
1171 Our results have clear policy relevance, especially in situations where fish stocks and resource use
1172
1173
1174 21
1175
1176
1177
1178
1179
1180
are subject to regulations that affect harvesting efficiency, but where no direct catch regulations
1181 can be used (e.g. due to prohibitively high transaction costs). In addition to gear regulations,
1182
1183 regulations of harvesting efficiency may include restrictions on harvesting locations, season
1184 length, and so forth. Our results imply that whenever the stock is MYS-overfished, a regulation
1185
1186 that decreases harvesting efficiency increases welfare. On the other hand, if the current stock size
1187
exceeds the one that would produce the MSY, a marginal increase in harvesting efficiency
1188
1189 improves welfare. Therefore, any increase in harvesting efficiency needs to be considered with
1190
1191 caution. Our analysis also has shown that there may be situations in which critical bifurcations
1192 exist such that with increasing harvesting efficiency a steady state becomes unstable, and the
1193
1194 resource stock eventually collapses to a lower and less productive size.
1195
1196 We have shown that these main results are very general. Although we have basically considered
1197
1198 small-scale fishery, the results hold equally for recreational fisheries and other resources that are
1199
exploited under restricted open access. This may include many cases of wildlife hunting, timber
1200
1201 and non-timber use forests, or rangelands. In all these cases, a stock-augmenting increase in
1202
1203
harvesting efficiency has the described ambiguous effect.
1204
1205 Appendix
1206
1207
1208 A Proof of Result 1
1209
1210
Differentiating (12) with respect to q , using the envelope theorem and doing small
1211
1212 rearrangements, we find (omitting arguments):
1213
1214
d  t* dH *
1215 = CHGSt Ht H t* t  CHGSt q H t*  CqGS (A.1)
1216 dq dq
1217
1218
CHGSt Ht H t* CHGSt q
(11) P( H t ) H t* CHGSt q
1219 =  C H t q H t  Cq =
GS * GS
 CqGS (A.2)
1220 P( H t )  CHt Ht
GS
P( H t )  CHt Ht
GS

1221
1222
1223  H t*2  H t*   H t*  H t*   H t* 
1224 P( H t )  3 2 C      4 3 C  C   
1225
(2)
=  q Xt  q Xt  q Xt  q Xt   q Xt 
(A.3)
1226 1  H* 
P( H t )  2 2 C   t 
1227 q Xt  q Xt 
1228
1229
1230 22
1231
1232
1233
1234
1235
H t*  H t* 
1236
C   
1237 q 3 X t2  q X t   1  H t* 
=   H *
P ( H *
)  C     (A4)
 
1238
 H*
t t
1 q Xt  q Xt 
1239 P( H t )  2 2 C   t 
1240 q Xt  q Xt 
1241
1242
H t*  H t* 
1243 C   
1244 (8) q 3 X t2  q X t    H * P( H t* )  
1245 = P( H t* )1    t   > 0. (A.5)
1  H t*    P ( H *
) 
1246 
P (Ht )  2 2 C   
t

1247 q Xt  q Xt 
1248
1249
The result that consumer surplus increases with q directly follows from the assumption of a
1250
1251 downward sloping inverse demand function according to which consumer surplus increases with
1252
1253 and the result that catch increases with q as stated in Eq. (11).
1254
1255
1256
B Proof of Result 2
1257
1258 We have already proven the result for consumer surplus. Here we show that we obtain the same
1259
1260 for producer surplus. Using (17) in (19), we find
1261
1262
1263
d *
1264
1265 dq
=
1
F ( X )  H X
* C GS
HX
H *  CHH
GS
H H X*  C XGS  H q*   CHH
GS
H * H q*  CHq
GS
H *  CqGS  F ( X * )  H X  
1266
1267
1268
(B.1)
1269
1270 Differentiating condition (8), which determines the market equilibrium harvest, with respect to X
1271
1272
and q , we obtain
1273
1274
1275
 P( H *
)  CHH
GS
 H X* = CHXGS ,
1276
1277  P( H *
)  CHH
GS
 H q* = CHqGS .
1278
1279
1280 Combining these two conditions, we have CHX
GS
H q* = CHq
GS
H X* . Note that the cost function (2)
1281
1282 implies X C XGS = qCqGS and X CHX
GS GS
= qCHq . We thus also have C XGS H q* = CqGS H X* . Using these
1283
1284 results, Eq. (B.1) simplifies to:
1285
1286 23
1287
1288
1289
1290
1291
d * F ( X * )
1292
1293 dq
=
F ( X )  H X
*  GS
CHH H * H q*  CHq
GS
H *  CqGS  . (B.2)
1294
1295
1296
The term in brackets is the short-term effect of more efficient fishing technology on profit, as given
1297 in Eq. (A.1). Under the assumptions stated in Result 1, this term is positive. Thus, the sign of
1298
1299 d  * / dq depends on the sign of the first factor. For a stable steady state, the denominator is
1300
1301 negative. Thus, d  * / dq > 0 if and only if F ( X *) < 0 , which is the case if and only if
1302
X * > X msy .
1303
1304
1305
1306
1307
1308
References
1309
1310 Anderson, L. 1986 , The Economics of Fisheries Management, John Hopkins University Press,
1311
1312 Baltimore.
1313
1314 Baland, J.-M. Francois, P. 2005 , ‘Commons as insurance and the welfare impact of
1315
1316
privatization’, Journal of Public Economics 89(2–3), 211–231.
1317
1318 Bjørndal, T. 1988 , ‘The optimal management of North Sea herring’, Journal of Environmental
1319
1320
Economics and Management 15, 9–29.
1321
1322 Clark, C. W. 1990 , Mathematical Bioeconomics, 2nd edn, Wiley, New York.
1323
1324 Copes, P. 1972 , ‘Factor rents, sole ownership and the optimum level of fisheries exploitation’,
1325
1326 The Manchester School of Economic & Social Studies 40(2), 145–163.
1327
1328 Eide, A., Skjold, F., Olsen, F. Flaaten, O. 2003 , ‘Harvest functions: the Norwegian bottom
1329
1330 trawl cod fisheries’, Marine Resource Economics 18, 81–93.
1331
1332 FAO 2007 , Increasing the contribution of small-scale fisheries to poverty alleviation and food
1333
1334 security, Vol. 881 of FAO Fisheries Technical Paper, FAO, Rome.
1335
1336 Gordon, D. Hannesson, R. 2015 , ‘The Norwegian winter herring fishery: A story of
1337
1338 technological progress and stock collapse’, Land Economics 91, 362–385.
1339
1340 Gordon, H. 1954 , ‘The economic theory of a common-property resource: The fishery’,
1341
1342 24
1343
1344
1345
1346
1347
1348
Journal of Political Economy 62, 124–142.
1349
1350 Grainger, C. A. Costello, C. 2016 , ‘Distributional effects of the transition to property rights
1351
for a common-pool resource’, Marine Resource Economics 31(1), 1–26.
1352
1353
1354 Hannesson, R., Salvanes, K. G. Squires, D. 2010 , ‘Technological change and the tragedy of
1355
the commons: The Lofoten fishery over 130 years’, Land Economics 86, 746–765.
1356
1357
1358 Jensen, F., Nielsen, M. Ellefsen, H. 2019 , ‘Defining economic welfare in fisheries’,
1359
Fisheries Research 218, 138–154.
1360
1361
1362 Lancker, K., Fricke, L. Schmidt, J. O. 2019a, ‘Assessing the contribution of artisanal fisheries
1363 to food security: A bio-economic modeling approach’, Food Policy 87, 101740.
1364
1365
1366 Lancker, K., Deppenmeier, A., Demissie, T. Schmidt, J. 2019b, ‘Climate change adaptation
1367 and the role of fuel subsidies: An empirical bio-economic modeling study for an artisanal open-
1368
1369 access fishery’, PLoS ONE 14(8), e0220433.
1370
1371 Murray, J. 2007, ‘Constrained marine resource management’, Unpublished Ph.D. Dissertation,
1372
1373 Department of Economics, University of California San Diego.
1374
1375 Nævdal, E. Skonhoft, A. 2018 , ‘New insights from the canonical fisheries model: optimal
1376
1377 management when stocks are low’, Journal of Environmental Economics and Management 92,
1378
125–133.
1379
1380
1381 Okonkwo, J.U. and Quaas, M.F. 2019. ‘Welfare Effects of Natural Resource Privatization: A
1382 Dynamic Analysis’. Environment and Development Economics. doi
1383
1384 10.1017/S1355770X19000342.
1385
1386 Ostrom, E. 1990 , Govering the commons: The evolution of Institutions for Collective Action,
1387
1388 Cambridge University Press, New York.
1389
1390 Péreau, J.-C., Doyen, L., Little, L. Thébaud, O. 2012 , ‘The triple bottom line: Meeting
1391
1392 ecological, economic and social goals with individual transferable quotas’, Journal of
1393
Environmental Economics and Management 63(3), 419–434.
1394
1395
1396 Pindyck, R. S. 1984 , ‘Uncertainty in the theory of renewable resource markets’, Review of
1397
1398 25
1399
1400
1401
1402
1403
1404
Economic Studies LI, 289–303.
1405
1406 Quaas, M. F., Stoeven, M. T., Klauer, B., Petersen, T. Schiller, J. 2018 , ‘Windows of
1407
opportunity for sustainable fisheries management: The case of Eastern Baltic cod’,
1408
1409 Environmental and Resource Economics 70(2), 323–341.
1410
1411
Robinson, J. 1938 , ‘The classification of inventions’, Review of Economic Studies 5(2),
1412
1413 139–142.
1414
1415
Schaefer, M. 1957 , ‘Some considerations of population dynamics and economics in relation to
1416
1417 the management of marine fisheries’, Journal of the Fisheries Research Board of Canada 14,
1418
1419
669–681.
1420
1421 Squires, D. Vestergaard, N. 2013 a , ‘Technical change and the commons’, Review of
1422
1423
Economics and Statistics 95(5), 1769–1787.
1424
1425 Squires, D. Vestergaard, N. 2013 b , ‘Technical change in fisheries’, Marine Policy 42,
1426
286–292.
1427
1428
1429 Stoeven, M. T. 2014 , ‘Enjoying catch and fishing effort: The effort effect in recreational
1430
fisheries’, Environmental and Resource Economics 57(3), 393–404.
1431
1432
1433 Tahvonen, O., Quaas, M. F., Schmidt, J. O. Voss, R. 2013 , ‘Optimal harvesting of an age-
1434
structured schooling fishery’, Environmental and Resource Economics 54(1), 21–39.
1435
1436
1437 Whitmarsh, D. 1990 , ‘Technological change and marine fisheries development’, Marine
1438 Policy 14, 15–22.
1439
1440
1441 World Bank 2012 , Hidden harvest: the global contribution of capture fisheries, World Bank,
1442 Washington, DC.
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454 26
1455
1456