Resource Rents, Inframarginal Rents, and the Transition to Property Rights in a Common Pool Resource

Transcription

1 Resource Rents, Inframarginal Rents, and the Transition to Property Rights in a Common Pool Resource Corbett Grainger and Christopher Costello April 4, 2012 Abstract The most common form of rights-based management in sheries, individual transferable quotas (ITQs), allows the capture of rents by allocating shares of the harvest to individual shermen. While the experience with ITQs has varied, the literature suggests that they represent a signicant economic and ecological improvement over the status quo. However, introducing property rights into a common pool resource is often met by political opposition from some of the current users of the resource. We introduce a simple analytical model, within which we derive the rm-level economic gains of ITQs using market prices and information on the distribution of harvest over time. We stress the importance of heterogeneity in skill among shermen which we leverage to estimate separately "inframarginal rents" and "resource rents" by year for both limited entry and ITQ sheries. We use the model to predict welfare eects for policy changes for dierent groups. We illustrate the method using data from the Red Snapper shery in the Gulf of Mexico. As predicted by our model, the transition to rights-based management leads to a conversion of inframarginal rents to traditional resource rents. In aggregate, however, our estimates suggest more than a doubling in the real value of the resource. These ndings have important political economy implications for sheries considering a transition to rights-based management. This is preliminary; please do not cite or distribute without permission. Thanks are due to Ryan Ono for helping with data collection. Department of Ag. & Applied Economics, University of Wisconsin - Madison Bren School of Environmental Science and Management, University of California Santa Barbara; and NBER 1

2 1 Introduction Economists have long advocated the introduction of property rights to address problems pervasive to common pool resources such as sheries (Gordon, 1954). An interesting and understudied feature of this transition is the likely change in the distribution of rents among users of the resource. The literature points to large gains in resource rents from the "rationalization" of sheries, but, somewhat perplexingly, there is often opposition to rights-based management from current users. This reluctance to adopt rights-based management can only be rational if some current users may not benet from changes in management. We argue that this is because heterogeneity in skill under regulated open access (limited entry) will lead to "inframarginal rents" 1, which may be eectively dissipated or transferred to other users when property rights are introduced. Thus, in order to gain the support of current resource users, the allocation of rights must compensate shermen for the changes in rents associated with the introduction of property rights. In this paper, we introduce a simple model to analytically separate resource rents from inframarginal rents under alternative management regimes. We use the framework to calculate empirically the magnitude of these rental values for an important U.S. shery that has recently undergone this transition to property rights. Many sheries around the world are characterized by ecological collapse (Worm, et al. 2006) and rent dissipation (Homans and Wilen, 1997). Catch shares, such as Individual Transferable Quotas (ITQs), have been shown to be associated with the reversal of sheries collapse (Costello et al., 2008), and when rights are secure and exclusive, catch shares add signicant value (Grainger and Costello, 2011). An open question regarding catch shares is the economic gain associated with "rationalization." Estimating the economic gain is generally a dicult task, as a time series of detailed microdata would be necessary to estimate vessel-level costs and revenues before and after rationalization (e.g. Grafton, et al., 2000; Lian, et al., 2010). In the United States, the transition to catch shares is most often preceded by limited entry management where a limited number of permits allow their owners to access the resource but do not grant a right to a share of the harvest. These permits are generally transferable, and the market permit price will reect protability to the marginal sherman. After rationalization, since the transacted goods are units of harvest, the market price of a quota share should reect the 1 Throughout this paper we will refer to "inframarginal rents" as those rents arising from heterogeneity in marginal extraction costs. These could alternatively be referred to as "racing rents", "skill rents", or following Anderson (1989) as "highliner rents". 2

3 protability of share ownership for a marginal unit of harvest. These market prices are generally available to researchers, and combined with other relevant variables, we show that this information can be used to estimate the economic gains of transitioning to a rights-based form of management. Importantly, the market permit price in limited entry does not fully reect the present value of access to all sherman to the shery; it reects only protability to the marginal sherman. The value of a permit to the sher with the highest skill could be signicantly higher, depending on the heterogeneity of skill. In these cases, there are "inframarginal rents" accruing to the permit-holders with high skill. In our model, higher skill shermen also end up harvesting the most in the pre-itq setting. Similarly, under ITQs, the present value of the shery is higher than the aggregate value of quota shares would suggest, because "inframarginal rents" are excluded from that measure. Thus rm-level heterogeneity is critical in determining the value of the inframarginal rents, as well as the overall value of the resource, prior to the implementation of property rights. Earlier authors, such as Copes (1972) and Johnson and Libecap (1982), point to the potential importance of inframarginal rents in the presence of heterogeneous rms. Coglan and Pascoe (1999) indicate the importance of measuring prots, not just resource rents, using survey data. And Johnson (1995) points to the importance of rents beyond "pure" resource rents in his model of taxation in sheries. We build on this foundation by building a model with heterogeneous shermen and compare welfare and distributional outcomes under alternative management regimes. We introduce a parsimonious model, which can easily be parameterized with publicly-available harvest and price data. Our approach relies on market prices for limited entry permits, quota shares, ex-vessel prices, and the distribution of harvest. This method allows one to empirically disentangle "inframarginal rents" from "resource rents" by year and shing rm for both limited entry and ITQ sheries. A signicant advantage of our approach is that we can estimate the change in value due to ITQs with limited information. We illustrate the method using data from the Red Snapper shery in the Gulf of Mexico. Our estimates suggest roughly a doubling in real value due to ITQs. However, by imposing ITQs, management changes may eectively transfer "inframarginal rents" into "resource rents". That is, by assigning property rights in the common pool resource, the "inframarginal rents" may be dissipated and that value is transferred to the (possibly new) owners of the scarce resource. This nding oers insight into the political economy of rights-based management in natural resources. One of the political barriers to the implementation of rights-based management often 3

4 comes from current users who benet from the status quo. The reason for this is clear: if skill is heterogeneous, high-skilled individuals may earn positive rents in limited entry, though marginal individuals earn little above the return to their inputs. The transition to property rights is then marked by the conversion of some of these inframarginal rents into resource rents. If the initial allocation of rights fails to take this into account, disadvantaged shermen may rationally oppose the transition to property rights. Without political support from current users for management change, ITQs are unlikely to be implemented despite their aggregate benets. This paper provides both a theoretical and empirical treatment of this phenomenon. In the following section we introduce a simple model of the transition to rights-based management in a shery. In section 3, we illustrate this model by providing estimates of the model using publiclyavailable data from the Gulf of Mexico Red Snapper shery. Our estimates are consistent with previous estimates in the literature, but require far less information to implement. In section 4 we discuss our results in the context of the political economy of common pool resources and rights-based management. 2 Model For the sake of simplicity, the shery is modeled in steady state so we abstract away from tansition dynamics such as shery recovery. This obviously assumes away some of the complexities modeled elsewhere in the related literature on the transition to ITQs (e.g. Weninger and Just, 1997; Weninger, 1998; Lian, et al., 2010; Schnier, 2009). The shery is initially managed by limited entry with an overall harvest quota. This form of limited entry relies on an overall total allowable catch and tradable permits (licenses) that allow access to the resource, but do not grant individual rights to levels of harvest. We then model the shery under ITQs, and we derive expressions for resource rents and inframarginal rents under both management regimes, which we then use to estimate the empirical changes in rents in a U.S. shery. 2.1 A Model of Ecient Eort There are N license holders. Each sherman i chooses eort, E i, allocated to shing. The marginal cost of eort varies across individuals to reect heterogeneity in skill. Let φ i be i's marginal cost of 4

5 eort. Let E be the total amount of eort in the shery, and E i is the total amount of eort by the other N-1 shermen. We assume that sherman i can increase his share of the harvest through greater eort, but that this comes with decreasing returns. Furthermore we assume that an increase in others' eort leads to a decrease in i's share. Specically, let s i (E i, E i ) be sherman i's fraction of the overall harvest where si E i > 0, 2 s i < 0, and si Ei 2 E i < 0. Fishermen take the total allowable catch (H) and ex-vessel price (P ) as given. Fisherman i chooses his eort to maximize his prots, given by P s i (E i, E i )H φ i E i (1) Dierentiating with respect to E i gives us the (interior) rst-order condition for sherman i: P = φ i s i E i H. (2) Fishermen continue applying eort until the marginal benet of doing so equals the marginal cost. Rearranging, as we have in Equation 2, shermen continue harvesting until the marginal benet of harvest equals the marginal extraction cost. There are N shermen and N corresponding rst-order conditions. Since the optimal harvest levels are dened as shares, the TAC binds with equality. There are some characteristics worth noting, as they will play an important role in what follows. This basic model gives us the intuitive "race for sh" in the intensity of harvest eort. An individual's harvest is determined by the eectiveness and marginal costs of his eort. Moreover, an individual's marginal extraction costs are increasing in his own harvest. Importantly, an individual's eort is increasing in both the ex-vessel price (P) and the total allowable catch (H ). Thus, exogenous increases in either P or H lead to more eort exerted by all shermen. 2 2 To see this, rst note that s i (E i, E i ), the fraction of harvest for sherman i, is increasing in E i. Assuming some regularity conditions hold, dene a function g() as the inverse of s i. Then E E i = g( φ i g ), and i P H P = φ ih g (P H) 2 P. Both terms on the right-hand side are negative, so E i is increasing in P. Showing that E i is increasing in H is analogous. 5

6 2.2 Model with Harvest Choice To gain empirical traction, we must now be more specic about the harvesting technology. For convenience we focus on a version of the model where harvest is the choice variable, for which Equation 2 denes the optimal harvest chosen by sherman i. We assume that total extraction costs are quadratic in harvest. In equation 2 above, marginal extraction costs for the N shermen adjust proportionately to changes in ex-vessel prices or the TAC. We will separate the marginal extraction costs for sherman i into two components: his relative skill (represented by heterogeneous marginal extraction cost parameters), and a parameter common to all shermen. If sherman i harvests h it sh, his total cost is given by 1 2 η tλ it h 2 it. The rst marginal extraction cost parameter, λ it, varies across shermen (i) and time (t). These parameters represent a sherman's relative marginal extraction costs. The regulator's choice of H t may also aect extraction cost through an adjustment parameter, η t. This parameter η t has an intuitive interpretation. Suppose that the ex-vessel price increases (exogenously) by a factor of η. By equation 2, each sherman's marginal extraction costs increase proportionally. Each sherman has an incentive to "race" to capture any marginal rents, resulting in an increase in marginal extraction costs for all shermen. The equilibrium increase in marginal extraction costs is η. Similarly, an increase in the Total Allowable Catch would result in a decrease in marginal extraction costs inversely proportional to that change. 2.3 Limited Entry Management We begin with a model of a limited entry shery with tradable permits and a Total Allowable Catch (TAC), H. Permits in this case are tradable licenses that allow access to the resource, but do not give the holder a right to a share of the overall harvest; this is the predominant form of management in the developed world. Under limited entry management, there are N (tradable) permits and N permit-holding shermen (i = 1,...,N ) in the shery. Assume that i's problem is, given H t, to maximize prots, taking as given other shermen's harvest choices. If the market (ex-vessel) price for sh in year t is given exogenously by world markets by Pt, 3 then 3 Obviously this may not always hold true. This assumption could be relaxed, as long as individuals do not believe 6

7 in equilibrium shermen compete for the harvest until marginal prots are zero (i.e. P t = η t λ it h it ). Let h it be the equilibrium harvest for i in year t. Now the overall TAC binds, such that H t = i h it, and η t adjusts such that each sherman's rst-order condition and the TAC constraint hold with equality, so η t = P t /H t ( i 1 λ i ). Now for convenience dene c it η t λ it. Fisher i's annual rents are given by P t h it 1 2 c it(h it )2. If we dene sherman i's equilibrium share (de facto, not de jure) of the overall harvest in year t as γ it, then his annual rents are given by P γ it H t 1 2 c it(γ it H t ) 2. In steady state, each sherman i expects the same annual rents every year in the future, so the lowest skilled harvester (i.e. the "marginal" sher), indexed by i =1, values his permit at v = 1 + r [(Pt γ 1t H t ) 1 r 2 c 1t(γ 1t H t ) 2 ], (3) which is the market permit price. That is, the market permit price is simply the expected present value of future rents for the marginal sherman. For our purposes, we will dene resource rent as the equilibrium market price for access to the resource, times the number of market participants. Therefore the present value of resource rents in the shery is simply Nv. 4 The present value of "inframarginal rents" for sherman i is given by the present value of returns above the market value of a permit. The present value of sherman i's inframarginal rents is given by P V (S LE it ) = 1 + r [Pt γ it H t 1 r 2 c it(γ it H t ) 2 ] v, (4) This expression is positive for any non-marginal sherman (i.e. i 1). Summing over all shermen and rearranging gives us the present value of inframarginal rents for the shery: P V (S LE t ) = 1 + r [(Pt H t ) (γ it γ 1t ) r i 1 i 1 [ 1 2 c it(γ it H t ) 2 ] + N 1 c it (γ 1t H t ) 2 ]. (5) 2 Now because the marginal sher (i =1) earns no inframarginal rents, the above expression can be they can impact the ex-vessel price, and all shermen face the same price within a season. 4 Suppose for simplicity that a person identical to the marginal sher is in the shery, so that the market price exactly reects protability of the resource for the marginal sher. 7

8 rewritten as P V (S LE t ) = 1 + r r N N [(Pt H t ) (γ it γ 1t ) [ 1 2 c it(γ it H t ) 2 ] + N 2 c 1t(γ 1t H t ) 2 ]. (6) i=1 Importantly, if all rms are identical, then γ it = γ 1t and c it = c 1t for all i. Then the previous expression for the present value of inframarginal rents for the shery equals zero, and the present value of the shery is simply the present value of resource rents. That is, only in the presence of rm-level heterogeneity are there inframarginal rents under limited entry. i= Welfare and Distribution Under Limited Entry Under limited entry, the marginal sher (denoted by i = 1) earns no inframarginal rents, as his prots are completely oset by the market permit price. When skill is heterogeneous (parameterized here through the cost function), the other shermen each earn positive inframarginal rents. In the case of homogeneity, no sher earns inframarginal rents, and the value that each sher places on having access to the resource is equal to the market permit price. In the following section we compare outcomes for these shermen under ITQs. Given the assumptions above, the i th sher earns inframarginal rents equal to 1 2 (P t ) 2 ( 1 c it 1 c 1t ), where i = 1 is the marginal sher. We will use this expression in the following section to compare welfare outcomes under limited entry and ITQs. 2.4 Individual Transferable Quota Now suppose that the regulator imposes an ITQ program in which shermen are granted shares of H t and the xed number of permits is relaxed. Each sherman's marginal extraction costs may change because he now has a right to his share of the harvest. Note that the number of active shermen may change if consolidation occurs due to ITQ management. Dene the equilibrium number of shermen as M. H t is still set by the regulator, and we assume that the market ex-vessel price for sh is greater than the social marginal extraction cost. Each sherman's costs (c it ) are dened as before, but we allow the marginal cost of extraction to change multiplicatively by the rm specic factor α i (0, 1]. This allows for technical eciency gains under ITQs (e.g., due to changes in racing incentives). Now 8

9 given tradable and divisible quota shares, the marginal extraction costs across shermen will be equalized, with the equilibrium lease price of quota shares given by πt = Pt α i c it h it. Then the present value of resource rents for the shery is simply 1+r r π t H t. The annual inframarginal rents for sherman i are now given by S IT Q it = 1 2 (P t π t )h it. (7) The present value of all inframarginal rents under ITQs is then given by P V (S IT Q t ) = 1 + r 1 r 2 (P t πt )H t. (8) Thus, inframarginal rents will persist under ITQs, even in the presence of a homogeneous eet Welfare and Distribution under ITQs Under ITQs, sherman i is allocated some level of harvest, h A i. π h A i + 1 (P π ) 2 2 c i, where π = P 1 M. i=1 1 c i Then i earns annual prots of Lemma 1. Suppose that costs are unchanged when transitioning to ITQs (i.e. α i = 1 i). Then if the ex-vessel price and TAC are unchanged,π = 0. Proof. This follows directly as a result of no cost changes following the implementation of ITQs. Result 1. If costs are unchanged under ITQs (i.e. α i = 1 i), sherman i's prots are higher under ITQs than limited entry, regardless of his/her initial allocation h A i. Proof. Under limited entry, i earns 1 2 ( 1 c it 1 c 1t ), and under ITQs i earns π h A i (P t π ) 2 c it. Comparing these two terms (and dropping t subscripts), 1 2 ( 1 1 ) π h A i + 1 (P π ) 2 (9) c i c 1 2 c i = 1 2 ( 1 1 ) π 1 (P ) 2 (10) c i c 1 2 c i 9

10 which simplies to c 1 c i < c 1 (11) And the result follows that sher i prefers ITQs, regardless of initial allocation. Result 2. If costs under ITQs decrease for all shermen proportionately, such that α i = α (0, 1) i, then equilibrium harvests are unchanged. Proof. Under limited entry, i's harvest share is P c i P N i=1 1 c i, (12) 1 which can be simplied as N. c i i=1 1 c i Under ITQs, if costs proportionately decrease, i's share of the total harvest is given by P α P αc i which simplies to the same harvest share as under limited entry. N i=1 1 c i, (13) Corollary 1. The relative gain from ITQs increases as α 0. Proof. Suppose α i = α for all i. Then the gain from ITQs is represented by the market lease value: π = P αh M i=1 1 c i. (14) Then we have π decreasing in α, and π P as α 0. Corollary 2. ITQs are always better for sher i as long as his/her allocation exceeds his/her historical harvests (h A i > h i ). 10

11 Proof. Because harvest shares do not decrease, as long as α 1, sher i earns at least as much under ITQs. Corollary 3. Let be the maximum dierence between the allocation and historical harvests (h i h A i ) such that ITQs are at least as good for sher i. Then, for (α), we have lim α 0 > 0 and d dα < 0. Proof. Under limited entry, i's annual rents are given by (P ) 2 for all i, and i's allocation is Q A i ), i earns 2c i. And under ITQs (where α i = α < 1 (P π ) 2 2 αc i + π (P π ) αc i π [ (P π ) αc i h A i ], (15) which simplies to (P π ) 2 + π h A i 2αc. i Now for some allocation h A i, sher i is indierent between ITQs and limited entry. This occurs if the allocation is given by Q A i = 1 π α(p ) 2 (P π ) 2 2αc i. (16) In this case, the dierence between i's harvest under limited entry and ITQs, = h i ha i, is given by And the results follow directly. = P 1 α(p ) 2 (P π ) 2 c i 2 αc i π. (17) Result 3. Any decrease in costs for some rm j (i.e. α j < 1 for some j) will strictly increase welfare for rm i j under ITQs. Proof. To see this, consider a decrease in costs for rm 1, holding all other rms' costs equal. Then α 1 < 1 and α i = 1 for all i 1. Then i's surplus under ITQs can be expressed as follows: πh A i + 1 (P π) 2, (18) 2 c 2 i which is strictly increasing in π. 11

12 Together these results suggest that incumbents will always benet from a transition to ITQs if there are no cost changes. When costs change, incumbents are strictly better o under allocations proportional to historical harvests (and for some range of smaller allocations). New entrants, on the other hand, may prefer limited entry if they are of the low-cost type, as there are signicant inframarginal rents under limited entry, and under ITQs new entrants may receive no free allocation. 3 Example: Estimates from the Gulf of Mexico Red Snapper Fishery We have used our model to calculate the economic eects of transitioning to property rights management of a common pool resource. In order to empirically estimate the gains from rationalization and to separate resource rents from inframarginal rents, we need only ex-vessel prices, limited-entry permit prices, ITQ share prices, and the harvest distribution over time. A signicant advantage of our approach is the availability of these data. In this section we illustrate our method using data from the Red Snapper shery in the Gulf of Mexico. The Red Snapper shery transitioned to ITQs in 2007 following years of limited-entry management. 3.1 Data Data come from a variety of sources, including government reports and classied ads. This section briey describes the data and sources. Ex-vessel prices were collected by the Gulf of Mexico Fisheries Management Council and the National Marine Fisheries Service; they were deated using the Consumer Price Index. IFQ share and lease prices for the years come from Gulf Council IFQ reports. Permit prices come from historical issues of Boats and Harbors, a trade publication. Bid and ask prices were both used in the calculation of the median permit price each year, and in cases where the listing was repeated in subsequent months (based on the contact information provided), only the last listing was used in the analysis. In cases where multiple goods were listed for one price (such as multiple permits), the permit price was calculated by subtracting the average price of the second permit, when possible. When this was not practicable, those listings were not included in the calculation. To our knowledge, this is the rst study to use these data to estimate the change 12

13 in rents due to changes in management. Figure 1 shows average ex-vessel prices, average permit prices and average quota lease prices over time. The gure shows Class I and Class II permit values, which allow permit holders to harvest 2,000 and 200 pounds per trip, respectively. The average value for a Class I permit is generally more than ten times as high as a Class II permit, and harvest levels for the shery are generally dominated by Class I permit holders. The ex-vessel prices (product prices) are relatively at in real terms, with a modest increase in value after the implementation of ITQs in This is consistent with the literature on ITQs, which generally shows a product premium post-rationalization (e.g. Grafton, et al, 2000). Finally, lease prices for ITQs increased after the initial implementation in 2007, which suggests an increase in annual protability. 5 Due to privacy concerns, data on the distribution of harvest are binned. For the purposes of our exercise we assume that the harvest for each sher within that bin is the midpoint. 6 This assumption seems reasonable; summing across shermen within any given year yields roughly the total allowable catch for that year. The distribution of harvest for two years, 2005 and 2010, is shown in the following gure. As shown in the gure, the distribution of harvest is far from uniform. Prior to ITQs, many shermen have very small harvest levels, while a few enjoy large catches. A few years after ITQs, once some consolidation has occurred, there are fewer shermen in the extreme ends of the distribution. 3.2 Assumptions and Estimates We assume that costs are heterogeneous and prices are exogenous, as described in Section 2. These assumptions imply that marginal extraction costs for the highliners (shermen who harvest a large fraction of the catch) are lowest, and that the lowest harvest levels accrue to the shermen with the highest marginal extraction costs. Thus the cost function combined with the data on the distribution of harvest, allow us to estimate the cost parameters and calculate the annual inframarginal rents for 5 Not shown here are the average values of reef sh permits, which must be held in addition to ITQ rights in order to land red snapper post These values (roughly $3,500 per permit) are not used in the calculation because a reef sh permit allows access not only to red snapper, but also to other reef sh in the Gulf of Mexico. To the extent that part of the value of the Red Snapper ITQ program is capitalized into reef sh permits, our estimates will lead to an undervaluation of the resource rents. 6 Through a separate Freedom of Information Act request, the maximum harvest each year was also collected, which is used to estimate the value of a permit for the highest harvester. 13

14 Figure 1: Permit-, ITQ Lease-, and Ex-Vessel Prices in the Red Snapper Fishery Note: All prices are in 2010 US Dollars. Permit prices have been scaled by 10,000 and 1,000 for the class 1 and class 2 permits, respectively. Figure 2: Distribution of Harvest Pre- and Post-ITQs 14

15 Figure 3: Total Allowable Catch in the Red Snapper Fishery Gross Weight Total Allowable Catch for the Commercial Sector. Note: each sher and year. As derived above, the market permit price should equal the present value of the stream of expected rents for the marginal sher. Given the assumptions above, we can calculate the annual rents for the marginal sher (i.e. the sher with the lowest harvest this year). Ordering the shermen according to their marginal extraction costs, with i =1 being the marginal sher, the annual surplus for the marginal sher under limited entry management is given by S LE 1t = 1/2 h 1tP t. (19) The market permit price is then given by v t = (1 + r)/r S LE 1t, (20) which is given by the historical average permit price in the classied ads from Boats and Harbors. 15

16 Assuming that the expected future rents in each year equal the rents this year, we can calculate the discount rate that rationalizes the market for permits. The median discount rate for the years that result from this calculation is 11.5%. Assuming that this discount rate is the same across shermen within any given year, we can calculate the latent permit value for each sher in each year. The present value of resource rents should be given by the market permit price times the number of permits. And the present value of inframarginal rents can be calculated using the calculated latent permit values for each sher. The sum of the present values of inframarginal rents and resource rents yields the present value of the shery in any given year. We can then repeat this process to estimate the present value of inframarginal rents, resource rents and the shery in each year of our dataset. As an example, annual inframarginal rents are shown by sherman for two years (under limited entry and ITQs, respectively). Figure 4: Inframarginal Rents by Fisherman Similarly, under ITQ management, the annual rents for a marginal unit of harvest is given by the lease price of a quota share. Given the distribution of harvest in each year, we can then calculate 16

17 the surplus to each sher, again assuming linear marginal extraction costs. For each year under ITQs, we can therefore estimate the present value of the shery, estimating the present values of inframarginal rents and resource rents separately. Our estimates are shown in the following gure, which assumes a ten percent discount rate. 7 Figure 5: Changes in Present Value of the Gulf of Mexico Red Snapper Fishery Note: Authors' calculations using the model described above. The calculations in the gure show a large increase in the overall value of the shery due to ITQ management. Prior to 2000, the real value of the shery was roughly $30 Million, and by 2010 the real value was greater than $150 Million. Comparing the average total value pre- and post-itqs, the shery's value increased by roughly 140%. This increase in real value comes despite decreases in the TAC post-rationalization to rebuild the shery. 8 Interestingly, the total value of the shery increases substantially following the implementation of ITQs, and this is marked by a large increase in "resource rents" and a decrease in "inframarginal 7 A lower discount rate would yield a higher estimated increase in the value of the resource. 8 This is consistent with ndings in Newell, Sanchirico and Kerr (2005), who study asset values after the implementation of ITQ sheries in New Zealand. They nd that an initial TAC decrease after rationalization led to a subsequent increase in the value of quota in those sheries. 17

18 rents". By imposing ITQs, the regulator eectively transfers "inframarginal rents" into "resource rents". That is, by assigning property rights in the common pool resource, the return to being skilled under the race for sh is diminished and that value is transferred to the owners of the scarce resource. This brings to light the importance of the initial allocation in compensating current users of the resource for any inframarginal rents they might have earned under the status quo regime. 4 Discussion This paper contributes in two areas in the economics of natural resources. First, it contributes to our understanding of the distributional eects of changes in management regimes by separating the value of the resource into two types of rents. Second, given limited information it provides a useful way to estimate changes in rents due to changes in management. While economists generally favor market-based approaches to managing natural resources, the current users of resources often oppose so-called "rights-based management." This has often perplexed natural resource economists, who point to signicant increases in resource rents due to the "rationalization" of sheries. While the gain in resource rents can indeed be large, inframarginal rents under current management (such as limited entry management in sheries) can also be significant. Furthermore, the adoption of rights-based management in natural resource settings can lead to a transfer of inframarginal rents to resource rents. 9 The important empirical question here is whether the increase in resource rents can more than oset any decrease in inframarginal rents that could occur because of rationalization. Our estimates suggest a doubling in the overall value of the resource, but further work should be done to calculate these value changes in other sheries. The model shows that resource rents are small under limited entry because that value is a function of the marginal sher's surplus. In the presence of heterogeneous shermen, inframarginal rents under limited entry can be large. If ITQs are implemented and costs decrease, both types of rents change. Because of consolidation, one may expect less variation in skill under ITQs, so inframarginal rents may decrease, while resource rents increase because the value of a harvest right would be high. We then consider the distributional impacts for incumbents and new entrants of dierent abilities. 9 Moreover, our empirical ndings are consistent with arguments as early as Copes (1972), who argued that resource policy should stop the "single-minded" pursuit of resource rent capture. 18

19 The estimates from the Red Snapper shery oer insight into the political economy of ITQ programs. One of the political barriers to the implementation of ITQs is often from highliners in a shery, who support ITQs only under a generous allocation. The reason for this is clear, if under limited entry these shermen enjoy inframarginal rents, which they would lose if the allocation transfers inframarginal rents to resource rents under ITQs. As indicated by Johnson and Libecap (1982), the heterogeneity of skill is critical in understanding the problems facing regulators in natural resources. Other papers point to the importance of heterogeneity in determining the distribution of prots. Others argue that grandfathering permits is even important for economic eciency (Anderson, Arnason and Libecap, 2010). Our approach sheds light on the importance of both resource and inframarginal rents in determining the aggregate value of a shery, and that rm-level heterogeneity is a central determinant of inframarginal rents. Although the model we develop here is very simple, it lends insight into a complex political economy problem. The allocation of rights in catch share programs, or similarly in the allocation of water or forestry rights, determines the allocation of future prots. But it also can take away potential inframarginal rents from those who are very skilled at extracting resources under current management rules. While these distributional implications are not new in resource economics, this parsimonious model shows conditions under which we would expect these problems to be especially prevalent. This paper also provides a useful tool for estimating the change in rents due to changes in management by using market prices. Historically it has been dicult to estimate the change in economic value of a resource like a shery, mainly because data limitations preclude the estimates of revenues and costs at the micro level. We propose a new method, based solely on publicly-available data, to estimate the change in rents due to management changes. Our parsimonious model is based on assumptions about the functional form of marginal extraction costs and well-functioning markets for permits and shing quota. Our estimates suggest roughly a doubling in the real value of the shery due to rights-based management. This is similar in magnitude as some estimates using microdata (e.g. Grafton, et al, 2000; Weninger, 1998; and Weninger and Waters, 2003), but this method requires far less data. Given limited data from a shery, we can then estimate resource rents and inframarginal rents separately, both under limited entry management and ITQs. Our estimates from the Red Snapper 19

20 shery in the Gulf of Mexico suggest a doubling in the present value of that resource due to the rationalization of the shery, much of which is represented by a transfer from inframarginal rents to resource rents. 20

21 References Anderson, G. L. "Conceptual Constructs for Practical ITQ Management Policies," in Rights Based Fishing. P. Neher, R. Arnarson and N. Mollett, eds, pp , Boston, MA: Kluwer Academic Publishers, Anderson, Terry, Ragnar Arnason, and Gary D. Libecap "Eciency Advantages of Grandfathering in Rights-Based Fisheries Management." NBER Working Paper Coglan, Louisa, and Sean Pascoe "Separating Resource Rents from Intra-marginal Rents in Fisheries' Economic Survey Data." Agricultural and Resource Economics Review 28(2): Copes, Parzival "Factor Rents, Sole Ownership and the Optimum Level of Fisheries Exploitation." The Manchester School 40(2): Costello, Christopher, Steven Gaines, and John Lynham "Can Catch Shares Prevent Fisheries Collapse?" Science 321: Gordon, H. Scott "The Economic Theory of a Common Property Resource: The Fishery." The Journal of Political Economy 62(2): Grainger, Corbett, and Christopher Costello "The Value of Secure Property Rights: Evidence from Global Fisheries." NBER Working Paper Grafton, R. Quentin, Dale Squires, and Kevin J. Fox "Private Property and Economic Eciency: A Study of a Common-Pool Resource." Journal of Law and Economics 43(2): Homans, Frances, and James Wilen. "Markets and Rent Dissipation in Regulated Open Access Fisheries." Journal of Environmental Economics and Management 49(2): Newell, Richard, James Sanchirico, and Suzi Kerr "Fishing Quota Markets." Journal of Environmental Economics and Management. Johnson, Ronald N "Implications of Taxing Quota Value in an Individual Transferable Quota Fishery." Marine Resource Economics 10: Johnson, Ronald N., and Gary D. Libecap. "Contracting Problems and Regulation: The Case of the Fishery." American Economic Review 72(5): Lian, Carl, Rajesh Singh, and Quinn Weninger "Fleet Restructuring, Rent Generation, and the Design of Fishing Quota Programs: Empirical Evidence from the Pacic Coast Ground- sh Fishery." Marine Resource Economics 24: Schnier, Kurt E "Spatial Externalities and the Common-Pool Resource Mechanism," Journal of Economic Behavior and Organization 70: Weninger, Quinn "Assessing Eciency Gains From Individual Transferable Quotas: An Application to the Mid-Atlantic Surf Clam and Ocean Quahog Fishery." American Journal of Agricultural Economics 80(4): Weninger, Quinn, and Richard E. Just "An Analysis of Transition From Limited Entry to Transferable Quota: Non-Marshallian Principles for Fisheries Management." Natural Resource Modeling 10(1):

22 Weninger, Quinn, and James A. Waters "Economic benets of management reform in the northern Gulf of Mexico reef sh shery", Journal of Environmental Economics and Management 46(2): Worm, Boris, Edward B. Barbier, Nicola Beaumont, J. Emmett Duy, Carl Folke, Benjamin S. Halpern, Jeremy B. C. Jackson, Heike K. Lotze, Fiorenza Micheli, Stephen R. Palumbi, Enric Sala, Kimberley A. Selkoe, John J. Stachowicz, and Reg Watson "Impacts of Biodiversity Loss on Ocean Ecosystem Services." Science 314(5800):