Nonpoint pollution control: Inducing first-best outcomes through the use of threats

https://doi.org/10.1016/j.jeem.2005.04.007Get rights and content

Abstract

This paper develops an economic model to analyze the use of a policy that combines a voluntary approach to controlling nonpoint-source pollution with a background threat of an ambient tax if the voluntary approach is unsuccessful in meeting a pre-specified environmental goal. We show that such a policy can be designed to induce cost-minimizing abatement without actually imposing the tax because the threat of imposition of the tax is sufficient to induce voluntary compliance. Thus, the combined policy approach can not only be more effective in protecting environmental quality than a pure voluntary approach but can also be less costly than a pure ambient tax approach. However, whether the cost-minimizing choices are a unique equilibrium or not depends on the design of the tax policy and whether or not it can be applied retroactively. All results still hold if the background threat takes the form of a reducing in government subsidies.

Introduction

While concerns about pollution control originally focused on point sources of pollution, attention has now turned to the control of diffuse or nonpoint-source pollution (NPP).1 Control of NPP is hampered by the fact that emissions of pollutants are not readily observable given their diffuse nature, which implies that traditional policy instruments based on emissions (e.g., emissions taxes or regulation) cannot be used in this context. This has led economists to consider alternative policy instruments for the control of NPP, including input taxes, input regulations, ambient taxes, random fines, direct revelation, and type-specific contracts [7], [11], [12], [15], [26], [30], [31], [39], [40], [41]. Instruments that provide flexible incentives (such as ambient taxes) can be used to induce first-best control of NPP [26], but information about farm-level charactertistics is needed to design these first-best policy instruments. They have thus been criticized as being likely to involve high information and/or transactions costs [3], [7]. This has led some to suggest the use of second best policy instruments instead [14], [35], [36], [38]. Use of these second best instruments involves a tradeoff. While transaction and information costs may be lower, these instruments do not achieve the targeted water quality level at the minimum abatement cost.

In contrast to the theoretical literature of NPP, in the U.S. water quality policy has historically been based on the use of “carrot” instruments designed to entice farmers to use environmentally friendly practices voluntarily or to participate in voluntary programs aimed at improving water quality.2 However, the varying success rates of voluntary programs have led some to question whether reliance on voluntary measures alone will adequately control agricultural pollution [25], suggesting the need for mandatory controls or incentives to ensure adequate environmental protection. Yet, as noted above, mandatory controls3 generally have drawbacks in terms of inflexibility (in the case of second-best instruments) or high transaction or information costs (for first-best instruments). Thus, neither purely voluntary programs nor mandatory approaches by themselves seem to offer a desirable “solution” to agricultural NPP.

There is, however, an alternative role for voluntary and mandatory approaches to NPP, namely, as complementary instruments to be used together rather than as substitutes to be used in isolation. As has become apparent in the context of point source pollution, the threat of the imposition of mandatory controls can be an effective mechanism for inducing firms to participate in voluntary agreements [28], while voluntary approaches have the potential to provide greater flexibility in meeting environmental quality goals and hence lower compliance and transaction costs [8].

A few states have experimented with the use of mandatory approaches as threats to be invoked if reliance on voluntary measures is insufficient to control nonpoint source pollution, and the threat of regulation has been shown to create an incentive for farmers to alter their production practices [25]. Yet to date the economic literature on NPP has not considered the possible use of both voluntary and mandatory approaches as complementary parts of a policy package. Segerson and Miceli [28] develop a model of this type in the context of point source pollution, where the regulator negotiates with an individual firm (or an industry representative) over the level of abatement under the agreement. Their focus is on the negotiated level of abatement that emerges under the agreement. They do not consider an approach under which the regulator sets an environmental quality target and then seeks participation in a program to achieve that target, as is typical in the case of NPP. Dawson and Segerson [10] examine the use of industry-wide threats to induce participation in voluntary approaches, but again the context is point source pollution where the threat is an emissions tax (a policy instrument that is not feasible in the context of NPP). Wu and Babcock [37] compare the relative efficiency of voluntary and mandatory approaches to NPP, but they treat the two as alternatives and do not consider a policy package under which the two are combined.

In this paper we build on principles drawn from the literature on the economics of regulation (e.g., [19]) to understand better the recent debate over the use of voluntary vs. mandatory policies for controlling NPP. We develop a simple game-theoretic model to analyze the use of a policy that combines a voluntary approach to controlling nonpoint-source pollution with a background threat of an ambient tax (or losing government subsidies) if the voluntary approach is unsuccessful in meeting a pre-specified environmental goal. We use the model to examine whether the regulator can use such a policy to induce cost-minimizing abatement decisions without the high transaction or information costs that can accompany the use of first-best instruments. We first consider how a single farmer would respond to such a combined policy and then extend the analysis to the case where the policy is applied to a group of farms. In this context, we ask whether cost-minimizing voluntary abatement is part of a subgame perfect Nash equilibrium (SPNE).

We show that, in both the case where the policy is applied to a single farmer and where it is applied to a group of farmers, such a policy can induce cost-minimizing abatement without actually imposing the tax because the threat of imposition of the tax is sufficient to induce voluntary compliance. In this sense, the combined approach can be both more effective than a purely voluntary approach and involve lower information or transaction costs than a pure tax approach. However, depending on the specific design of the tax policy and whether the tax can be applied retroactively, the cost-minimizing equilibrium may not be unique. When it cannot be applied retroactively (the typical case for most taxes), equilibria under which the target is not met voluntarily can emerge. In addition, equilibria under which the target is met voluntarily but some farmers free ride on the abatement of others emerge. In fact, when applied to a group of heterogeneous farmers we show that the tax cannot be designed to both induce a cost-minimizing equilibrium and eliminate the potential for free riding. However, if the tax can be applied retroactively, then it can be designed to ensure that cost-minimizing abatement (and hence voluntary compliance with the standard) is the unique equilibrium. While we derive the results assuming the regulator threatens imposition of an ambient tax, all of our results still hold if the background threat instead takes the form of reducing government subsidies if a pre-specified environmental goal is not met.

Section snippets

The basic model

Consider n farmers whose agricultural activities pollute a nearby waterbody. Each farmer, indexed i, can engage in abatement practices, such as the use of reduced tillage, establishment of buffer strips, construction of manure storage facilities, and land retirement, to reduce water pollution, or, equivalently, to increase ambient water quality in the waterbody. Let ai=(ai1,ai2,,aim) denote farmer i's abatement vector. Farms are assumed to differ in their characteristics, including both

Voluntary abatement: single polluter case

To establish a benchmark for the study of the combined policy, we first consider the single polluter case (i.e., n=1). In this case, there is no issue regarding the allocation of abatement across farmers; cost-minimization simply requires that the farmer choose the cost-minimizing combination of abatement practices. Throughout this and the following section, we assume that the ambient tax cannot be applied retroactively. We consider the case of a retroactive tax in Section 5.

We begin by noting

Multiple polluters

While the single-farm case provides a benchmark for the study of the combined policy approach, in many contexts there are likely to be several farms whose activities contribute to ambient water pollution in a neighboring lake or stream. We show in this section that a voluntary program coupled with a background threat of an ambient tax can induce cost-minimizing abatement decisions in this context as well. More specifically, we show that under the appropriate design of the ambient tax

Voluntary abatement with a retroactive tax

In the last section we showed that a policy that combines a voluntary approach with a background threat of a tax can induce cost-minimizing voluntary abatement by all farmers. However, such a policy has two potential problems. First, the cost-minimizing abatement choices are not a unique equilibrium. With heterogeneous farms, there will always exist other subgame perfect Nash equilibria under which the water quality target is met voluntarily but not at minimum cost because of free-riding

Reduction of government subsidies

The analysis above assumes that the background threat takes the form of an ambient tax. However, agricultural water quality policy in the United States has historically been based on the use of “carrot” approaches designed to entice farmers to use environmentally friendly practices voluntarily through the use of government subsidies.23 Such a policy can also be used with a background threat to achieve environmental goals at minimum costs. Instead of imposing an

Conclusions

Although the economic literature of NPP has focused on mandatory or stick approaches (e.g., input use taxes or restrictions), water quality policies in the US have historically been based on the use of carrot instruments designed to entice farmers to adopt conservation practices voluntarily. Both the mandatory and voluntary approaches have been criticized in the economic literature. Mandatory approaches generally have drawbacks in terms of inflexibility or high transaction or information costs,

Acknowledgments

The authors thank two anonymous referees for their invaluable comments and suggestions.

References (41)

  • K.J. Arrow

    The economics of agency

  • B. Babcock et al.

    Public fund for environmental amenities

    Amer. J. Agric. Econom.

    (1996)
  • S. Batie, D. Ervin, Flexible incentives for environmental management in agriculture: a typology, paper presented at...
  • D. Bosch et al.

    Voluntary versus mandatory agricultural policies to protect water quality: adoption of nitrogen testing in Nebraska

    Rev. Agric. Econom.

    (1995)
  • J. Braden et al.

    Information problems in the design of nonpoint-source pollution

  • On Environmental Agreements

    (1996)
  • J.C. Cooper et al.

    Incentive payments to encourage farmer adoption of water quality protection practices

    Amer. J. Agric. Econom.

    (1996)
  • N. L. Dawson and K. Segerson, Voluntary agreements with industries: participation incentives with industry-wide...
  • R. Govindasamy et al.

    Nonpoint tournaments

  • Cited by (61)

    • Bilevel optimization of conservation practices for agricultural production

      2021, Journal of Cleaner Production
      Citation Excerpt :

      A number of studies take this approach (Bekele and Nicklow, 2005; Arabi et al., 2006; Rabotyagov et al., 2010a, 2010b, 2014a, 2016; Antle et al., 2015). GA solution methods also provide a way to implement the nested principle-agent structure of ambient pollution policy highlighted in the theoretical nonpoint literature (Segerson, 1988; Segerson and Wu, 2006; Xepapadeas, 2011) within an integrated simultaneous optimization, by formulating the design of nonpoint policy as a bilevel multi-objective optimization problem (Bard, 1998). Bostian et al. (2015a) and Whittaker et al. (2017) use this integrated bilevel multi-objective optimization framework to evaluate the spatial targeting of a standalone policy for fertilizer reduction.

    • Impact of peer comparisons and firm heterogeneity on nonpoint source water pollution: An experimental study

      2021, Resource and Energy Economics
      Citation Excerpt :

      Consequently, designing and implementing policies based on individual behavior is often impossible or prohibitively costly (Xepapadeas, 2011; Miao et al., 2016). In extensions of Segerson (1988) considering homogeneous NPS contributors, numerous researchers have shown both theoretically (e.g., Xepapadeas, 1991; Horan et al., 1998; Segerson and Wu, 2006) and experimentally (e.g., Alpízar et al., 2004; Poe et al., 2004; Spraggon, 2002; Vossler et al., 2006) that ambient pollution policies involving a monetary policy instrument (taxes and/or subsidies) can induce groups to reduce their collective emissions to a socially efficient level. In such schemes, the regulator usually compares the ambient pollution reading to a target level and applies the monetary instrument to every firm in the watershed.

    • A Montero payment mechanism for regulating non-point pollution emissions

      2020, Resource and Energy Economics
      Citation Excerpt :

      The suggested tax on aggregate behavior to induce compliance with regulations adds to the literature on compliance and enforcement building on Becker (1968). With few exceptions (e.g. Segerson and Wu, 2006), this literature builds on the implicit assumption that the individual firm’s behavior that is being regulated can be observed. While this assumption is reasonable in many situations, clearly it is not in others.

    • Graduated stringency within collective incentives for group environmental compliance: Building coordination in field-lab experiments with artisanal gold miners in Colombia

      2019, Journal of Environmental Economics and Management
      Citation Excerpt :

      However, as seen within our results, for strict targets, benefits of collective incentives could be offset by financial costs of compliance alongside risk aversion concerning coordination as well as costs of monitoring and coordination. Future research could consider various other types of policies, e.g., combining carrots and sticks, as proposed by Segerson and Wu (2006): if miners do not comply voluntarily with the limit, they would face credible mandatory punishment (we are not sure whether, in fact, external sanctions can be credible on these frontiers). A combination of internal monitoring and sanctioning across group members, which could facilitate coordination with the tough policies, could also be tested.

    • Increasing the impact of collective incentives in payments for ecosystem services

      2017, Journal of Environmental Economics and Management
    View all citing articles on Scopus
    View full text