Quotas, Productivity, and Prices: The Case of Anchovy Fishing

• Date: 01 March 2016
• Author: Gabriel Natividad
• Published date: 01 March 2016
• DOI: http://doi.org/10.1111/jems.12148

Quotas, Productivity, and Prices:

The Case of Anchovy Fishing

GABRIEL NATIVIDAD

Department of Economics

Universidad de Piura

162 Martir Olaya St. Lima 18 Peru

gabriel.natividad@udep.pe

I exploit a 2009 reform that introduced individual fishing quotas (catch shares) for Peruvian

anchovy—the largest fishery in the world—to assess the causal impact of production quotas on

within-firm productivity and market prices. Unique features of the data allow me to create two

alternative counterfactuals: (i) anchovy fishing operations in a region of the country that was

mandated to implement quotas with a delay, and (ii) variation in quota allocations across ships.

I find that quotas do not increase within-asset or within-firm productivity in quantities. Instead,

a 200% increase in anchovy prices benefits extraction firms through higher revenues, consistent

with two mechanisms enacted by individual fishing quotas: more orderly industry operations

reducing excess supply and an increase in bargaining power of extraction firms with respect

to fish-processing. Several market characteristics across geographies differentially affect market

prices after the quota regime. Supplementary evidence on fewer operational infractions, higher

product quality, and a lower banking delinquency observed during the quota regime suggests the

existence of efficiency gains rather than purely rent transfers.

A rising literature in industrial organization has brought into relief the broad market

impact of environmental regulation (Greenstone, 2002; Keller and Levinson, 2002; Davis,

2008; Greenstone et al., 2012; Ryan, 2012). From an economic standpoint, one of the chief

difficulties of regulating the environment consists in addressing the tragedy of the com-

mons in natural resource contexts (Stavins, 2011), a problem often requiring the granting

of property rights to private firms (Vander Ploeg, 2011). This paper presents new causal

evidence about the impact of individual fishing quotas on within-firm productivity and

market prices by exploiting a 2009 reform that radically changed access to Peruvian

anchovy, the largest fishery in the world.

It has been argued that individual fishing quotas—an instrument that restricts free

access to the common fishery (Gordon, 1954), assigning to each ship a pre-established

catch share—can prevent the collapse of global fisheries (Costello et al., 2008; Heal and

Schlenker, 2008). Accordingly, over the last two decades, there has been much interest

in analyzing marine resource property rights in the form of individual fishing quotas

I am grateful to the Ministry of Production of Peru, to IMARPE, and to three fishing firms whose names are

suppressed here for confidentiality for access to data. No confidential information is disclosed in this paper.

Many thanks to John Asker and Natarajan Balasubramanian for valuable insights at different stages of this

project, and to an anonymous co-editor,two referees, Hunt Allcott, Alessandro Gavazza, AviGoldfarb, Deepak

Hegde, Boyan Jovanovic, Virgiliu Midrigan, Joseph Ostroy, Michael Powell, Rob Seamans, Giorgio Zanarone,

Peter Zeitz and audiences at City University of New York, Colegio Universitario de Estudios Financieros,

National University of Singapore, New YorkUniversity, Seoul National University,Universidad de Piura and

the ISNIE annual conference for useful comments. Guillermo Ramirez and Javier Romero Haaker provided

excellent research assistance. All errors are mine.

C2015 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume25, Number 1, Spring 2016, 220–257

Anchovy Fishing 221

(see Scott (1993, 2000, 2008), Libecap (2009), and Branch (2009) for broad discussions

and reviews). This large body of work has often focused on specific studies of fisheries

adopting a quota system1or cross-fishery comparisons of quota systems.2However,this

literature has given relatively little attention to the causal impact of fishing quota regimes

on within-firm productivity and market prices, thus remaining somewhat disconnected

from ongoing research in industrial organization.

As more direct background to my study, a small number of papers have employed

natural experiments to investigate the influence of individual fishing quotas on produc-

tivity and prices. Grafton et al. (2000) and Fox et al. (2003) analyze productivity and

prices around the introduction of individual harvesting rights in the British Columbia

halibut fishery in 1991, utilizing different cross-sectional samples of firms from 1988,

1991, and 1994, and finding changes in outcomes during the quota regime. Brandt (2007)

and Walden et al. (2012) employ the introductionof quotas in the Mid-Atlantic surfclam

and ocean quahog fisheries to study vessel productivity and participation incentives

before and after the reform. Importantly, all this prior work has lacked the econometric

advantage of having a control group to isolate the causal impact of the quota system

treatment. Moreover, some of these studies have lacked disaggregated information on

local market prices and weekly dynamics that would speak to current research on eco-

nomics mechanisms in vertical markets (e.g., Syverson, 2007; Hortac¸ su and Syverson,

2007).

My analysis of the industrial organization of fishing quotas fills an important gap in

the literature because key market outcomes should not be expected to remain unchanged

after quotas are enacted. Specifically,individual fishing quotas can be broadly construed

as a government-mandated reduction of competition in a given market, which may

carry implications for firm productivity and market prices (see Holmes and Schmitz

(2010) for the more frequently studied cases of increases in competition). Moreover,

the market structure analysis of fishery reforms is particularly important given their

potential redistributive effects (Karpoff, 1987). In this paper, I exploit a comprehensive

data set to assess the causal impact of individual fishing quotas, with emphasis on the

joint analysis of productivity and prices in the context of competitive markets.

In essence, I seek to answer two basic questions: How and why do individual

fishing quotas matter for asset-level productivity? And how and why do these quotas

matter for market prices? From a conceptual standpoint, answers to the first question can

1. For example, Geen and Nayar (1988) employ a simulation framework to compare a quota system with

other regulatory systems in the Southern bluefin tuna fishery in Australia and Japan; Casey et al. (1995)

collect descriptive statistics of the impact of a halibut individual fishing quota by conducting a round of 14

interviews with halibut processors in 1993 and by conducting one mail survey of all halibut fishermen in

1994; Herrmann (1996) studies the yearly aggregate time series of halibut prices in Canada before and after

the adoption of a quota system; Campbell et al. (2000) describe broad economic patterns emerging after the

enactment of individual transferrable quotas in Australia’s southern bluefin tuna fishery; Dupont and Grafton

(2000) analyze multi-species individual transferable quotas; Dupont et al. (2005) also analyze multi-species

individual transferable quotas, and although their focus is on productivity and prices, they analyze data in a

more aggregate(industry-species level) fashion; Fox et al. (2006) study the posterior introduction in Australia’s

Southeast trawl fishery of both a brokerage system enhancing the transferability of individual quotas and a

government-sponsored buy-back program for fishing licenses; Sharp and Batstone (2008) discuss the simul-

taneous introduction of total allowable catch (TAC) and individual transferrable quotas in New Zealand’s

fisheries management environment; Sylvia et al. (2008) describe the introduction of a self-government body,

the Pacific Whiting Conservation Cooperative, acknowledged to be quite different from an individual quota

system; Lee and Thunberg (2013) analyze a catch share system creating a simulated counterfactual based on

pre-regime operations

2. For instance, Chu (2009) and Essington (2010) conduct cross-fishery comparisons of ITQ adoption and

subsequent ecological outputs, with a clear biological emphasis rather than a focus on economic outcomes.

222 Journal of Economics & Management Strategy

vary widely. By reducing competition (racing behavior) through guaranteeing a catch

share to all market participants, a quota system may have a positive impact on within-

ship productivity through a smoother, better planned exploitation of the resource. But

when such guarantee helps industry participants reduce a significant portion of their

assets (ships), the remaining assets would need to be operated more intensively than

before, while lacking some of the prior information and incentive mechanisms due to the

presence of excess ships, thus becoming less productive with quotas. The question about

market prices appears to be more intuitively answered: by granting property rights to

fish extraction companies, the price of fish sold downstream should increase. Yet the

mechanism for price increases may vary, perhaps enacting a shift in bargaining power

over the vertical chain or instead more orderly supply operations due to the benefits of

fishing without racing. Generally, there has been much theoretical interest in modeling

fishermen’s racing behavior and the individual quota instrument (e.g., Gordon, 1954;

Arnason, 1990; Boyce, 1992). The causal impact of quotas on productivity and prices is

the open empirical question that I address.

The enactment of individual fishing quotas in the Peruvian anchovy fishery implied

a significant regulatory shift affecting all firms in the industry. Historically, until 2008,

extraction firms operated under a regulatory framework that allowed all ships with

a license to fish anchovy for fishmeal—an intermediate good produced by plants on

land—until a “total allowable catch” was reached. No individual caps on ship output

existed. In this early regime, the primary concern of firms was to find anchovy,extract as

much of it as possible, and deliver it quickly to plants to restart this cycle, aware that fish

not caught by one firm would be caught by another. In 2008, the government suddenly

introduced a new regulatory framework, with the stated goals of reducing overcapacity

and discouraging racing behavior that exerted pressure on the natural environment. In

the new fishing regime, a total allowable catch was still in place, but each ship received

an individual fishing quota, defined as a percentage share of the total allowable catch.

Individual fishing quotas were allocated using a formula based on ship characteristics

and ship output from 2004 to 2007 (i.e., a period ending before the announcement of

the reform). Moreover,the new regulation allowed firms to transfer quotas across ships,

thus fostering the closure of less productive ships once their quotas were transferred

away to others.

Key features of this regulatory reformallow me to study the causal impact of quotas

in a reduced-form fashion.3To be sure, a quota reform is an event among others poten-

tially affecting industry structure, so it is typically difficult to isolate the causal effect of

quotas without a clear counterfactual. By contrast, the reform studied here allows me to

exploit two counterfactuals for the question of interest: (i) anchovy fishing operations in

a region of the country that was ruled to implement quotas with a delay with respect to

the region of interest, and (ii) variation in quota allocations across ships. In a first set of

tests, I exploit data on anchovy fishing in the South region of the Peruvian sea, which

was scheduled to begin its own quota regime a few months later than the North-Center

region (i.e., the region of interest for my study, totaling over 87% of historical anchovy

catches.) The results of within-ship and within-firm panel difference-in-differences re-

gressions with respect to the South region counterfactual show no positive influence

of the advent of the quota regime on productivity measured in quantities. In a sec-

ond set of tests, I exploit ship-specific quota allocations to study the within-ship and

3. See Huang and Smith (2014) for a structural estimation of common pool extraction efficiency in the U.S.

shrimp industry in which policy experiments different from individual fishing quotas are discussed.