Ecological compensation: How much and where?

Highlights

• We show how the individual WTA the ecological compensation depends on the distance to the damage or restoration sites.

• The minimal ecological compensation meeting the NNL condition does not necessarily satisfy the NWO objective.

• In the absence of land cost, the ecological cost is minimized in the gravity center of environmental marginal utilities.

Abstract

We develop a normative approach to optimal environmental compensation in a spatial framework. We determine both the location and the level of compensation that minimize the total cost of restoration while maintaining the social welfare unchanged. In our framework, the policy maker implements a No Net Loss policy that meets the No Worse-Off principle as well as a location constraint on the offset. We describe the additional ecological cost induced by the No Worse-Off principle and how the spatial distribution of individuals, the environment and land costs affect the compensation location. The location constraint is shown to give rise to a trade-off between the compensation cost and the inequality among individuals induced by the policy.

Introduction

Local and national economic development programs are often accompanied by ecological damages. Policy makers have designed mitigation measures in line with the ‘polluter pays' principle to address the adverse impacts of development projects. These compensation measures are perceived as a mean to reconcile development with conservation. However, they face a number of challenges including the amount and the location of ecological compensation, and whether they should reach ecological and social objectives.¹ The aim of our work is to propose a theoretical model that incorporates all these dimensions into a single framework so as to deal with both ecological and social concerns.

While these compensation measures originated in the United States, they now widely spread to other countries like Australia, New-Zealand, Canada, China, European countries, and even South America.² Compensation applies to a wide range of sectors including mining, wind power, hydropower, oil and gas, property development, agriculture, etc. As a priority, developers should avoid and minimize any impacts on biodiversity. When residual impacts cannot be avoided, the creation and the restoration of natural habitats are considered to achieve No Net Loss (NNL), or even a net gain, of biodiversity. This mitigation sequence, referred to as ‘mitigation hierarchy’, cannot be bypassed.³

Biodiversity offset programs developed to compensate for significant and residual biodiversity loss resulting from development projects, involve controversial issues which are still the source of debate in many countries (Clare et al., 2011; Gordon et al., 2015). A major concern is the effectiveness of the system, which fails to consistently achieve the objective of NNL (Bezombes et al., 2019; zu Ermgassen et al., 2019). Another critique is that offsets are usually considered from an environmental perspective only and therefore omit other relevant dimensions such as the quality of life, social values, and the economic well-being. By focusing mainly on the ecological cost, the NNL policy neglects the welfare consequences that the policy has on the impacted population. Governments usually justify the implementation of development projects by arguing that any impact to biodiversity can be compensated. However, the corresponding mitigation measures may underestimate the social harm of the project (Githiru et al., 2015). This means that traditional mitigation planning may have detrimental consequences for people. This calls for offsets that account for the welfare of individuals on top of ecological considerations associated with the NNL objective. According to Griffiths et al. (2019), the No Worse-Off (NWO) principle states that “the social gains associated with the changes in biodiversity caused by development and accompanying offsets must at least equal any social losses”. Consequently, there is a need to develop normative approaches relying on economic modeling. Our work contributes to fill this gap. We develop an environmental compensation policy that meets both NNL and NWO objectives so that social gains and losses compensate each other. By doing so, we integrate the well-being of individuals to the NNL policy. The design of our compensation policy consists of both the location and the level of ecological compensation that minimize the total cost of restoration. Because environmental costs and benefits exhibit spatial decay, the location of the compensation matters as it affects the welfare of individuals.

Individual welfare gains and losses depend on the distance of individuals from the damage and offset sites. For instance, Mandle et al. (2015) use an ecosystem service modeling framework to describe how the impacts of development of the Pucallpa-Cruzeiro do Sul road in the Peruvian Amazon and the benefits of mitigation are spatially distributed. Policy requirements regarding the spatial location of the compensation, when they are any, are essentially motivated by ecological grounds. In practice, the choice of the offset location is often left to the discretion of the project proponent within pre-defined spatial limits. Understandably, that choice is driven by land availability and price (Quétier et al., 2014; Sonter et al., 2020). Because land developers are likely to seek cheap land, offsets may be displaced away from development sites (zu Ermgassen et al., 2020). Moving ecological resources across space has welfare implications even when the ecological impact is limited (Ruhl and Salzman, 2006; BenDor et al., 2008; Kaza and BenDor, 2013; Griffiths et al., 2019).

A strict implementation of the NNL principle would require compensation at the damage location. In that case, the population directly impacted by the damage will reap most of the compensation benefit. However, due to the lack of available or suitable land, on-site compensation is not always feasible. In that case, additional offsets are often required, through ordinances passed by local governments for compensation provided outside local boundaries (Womble and Doyle, 2012). A spatial issue, that we address here, is thus about where to locate offsets within a reasonable distance from the damage site. While ecological compensation practices gain importance in many countries, the location choice of offsets is also becoming a growing issue (Womble and Doyle, 2012).

These practical concerns are not new and have been studied in the recent empirical literature on environmental compensation. Some studies have shown that the distance between the damage and compensation sites significantly influences individual preferences for compensation measures (Borrego, 2010; Burton et al., 2017). Empirical analyses on the North Carolina's ecosystem enhancement Program by BenDor and Stewart (2011), the wetland mitigation banking in Florida by Ruhl and Salzman (2006) and the Chicago's one by BenDor et al. (2007) highlighted the role of offset location constraints and property values on compensation policy. A softer location constraint on offsets (North Carolina case) relocates compensation sites away from urbanized areas due to the availability of cheaper land outside cities. This can induce social impacts on the communities located nearby the damage. These growing concerns have been recently highlighted through an in-depth interview analysis of stakeholders in England (Taherzadeh and Howley, 2018) and also discussed in Griffiths et al. (2019). Other papers dedicated to ecological compensation mainly focus on practical issues related to mitigation banking, mitigation hierarchy in planning, or the way stakeholders could supply biodiversity offsetting (OECD, 2016; Vaissière et al., 2018; Kangas and Ollikainen, 2019). Despite of this growing empirical evidence, the economic principles underlying the compensation design remain a matter of concern (Calvet et al., 2015; Vaissière et al., 2020).

Welfare impacts of compensation measures are an often-overlooked issue. Exceptions are Zafonte and Hampton (2007), Flores and Thacher (2002), or Gastineau and Taugourdeau (2014). These papers consider the consequences of alternative types of compensation (ecological or monetary compensation schemes) on welfare. Zafonte and Hampton (2007) suggest that a pure ecological compensation based on the ecological equivalence principle may provide an acceptable approximation of wealth compensation. A different result is shown by Flores and Thacher (2002) who find that ecological equivalence specified in biophysical equivalents could fail to provide a satisfactory compensation from a welfare perspective. Finally, Gastineau and Taugourdeau (2014) explore the possibility of complementing the ecological equivalence by a monetary compensation in order to restore the social welfare after the damage.

The above existing literature emphasizes spatial implications of various compensation policies and their welfare consequences. To the best of our knowledge, no economic approach addressing the welfare impact of ecological compensation in a spatial context is available in the literature. In our paper, we develop a theoretical framework that integrates the spatial dimension of ecological compensation. To account for the spatial decay of environmental costs and benefits, agents in our model discount spatially ecosystem services. The basic idea is that people value more local public goods that are close to them than those that are far away, e.g. see Hannon (1994) or Fujita (1989). There is ample evidence about the spatial decay of the valuation of ecosystem services, see for instance the overview provided in Table 1 in Yamagushi and Shah (2020). Our framework applies spatial discounting to environmental compensation theory by providing a rigorous approach to the spatial distribution of losers and winners and the location choice of the compensation that achieves both NNL and NWO objectives. Results of our model are in line with empirical findings from the compensation literature on the location of offsets.

The results of our model are the following ones.

First, we show how the individual willingness to accept the ecological compensation depends on the distance to the damage or restoration sites. This means that agents in different locations are impacted differently by the restoration policy. This is because both ecological costs and benefits decay with spatial distance. Second, when the mitigation ratio is low, the minimal ecological compensation meeting the NNL condition cannot satisfy the NWO objective. It is because an additional ecological compensation is actually needed to meet this NWO objective. In the absence of land cost, we show that the ecological cost is minimized in the gravity center of marginal utilities of environmental consumption. Compensating in another location is of course possible but would require a larger ecological offset. This means that the location of offsets matters a lot. In particular, our analysis shows that on-site compensation should not necessarily be favored from a welfare perspective. Third, we show that in the presence of land costs, the compensation relocates towards cheaper land areas as observed in many studies, e.g. see BenDor et al. (2007). When this happens, we show that the level of ecological compensation increases as it is getting away from the gravity center of marginal utilities of environmental consumption, but the total restoration cost falls, as cheaper land costs more than offsets the increase in ecological cost. This result provides an economic rationale for local ordinances requiring additional offsets for compensation implemented outside local boundaries, see Womble and Doyle (2012). Fourth, we show that on-site compensation is welfare neutral. A soft constraint on offset location allows a lower restoration cost because more sites become available but introduces inequality among agents. This establishes a trade-off between the compensation cost and inequality.

This paper is organized as follows. Section 2 describes the spatial economy framework and determines the individual and society willingnesses to accept the ecological compensation. Section 3 introduces the planner's program and characterizes the optimal environmental compensation policy that meets both NNL and NWO objectives. In Section 4 we study the additional ecological cost induced by the NWO condition. We also describe the role of land costs and the location constraint on the optimal offset. The final section summarizes discussions and concludes.