Using local knowledge to model asymmetric preference formation in willingness to pay for environmental services

doi:10.1016/j.jenvman.2010.09.003

Journal of Environmental Management

Volume 92, Issue 3, March 2011, Pages 531-541

https://doi.org/10.1016/j.jenvman.2010.09.003 Get rights and content

Abstract

This paper describes an approach to account for asymmetric preference formation in discrete choice models used for environmental valuation. The paper draws on data from a case study on preferences for environmental change resulting from a hypothetical rural development and conservation programme in Indonesia. Local knowledge on the current state of the environment was used to define an individual-specific status quo that consistently frames changes in a range of environmental services as gains or losses matching the perceptions of the local population living in the vicinity of a National Park. I estimated choice models that included separate parameters for increases and decreases in attribute levels for the environmental services and derived the indicators of local willingness to pay (WTP) corresponding to the bidirectional changes relative to the individual-specific status quo option.

I found clear evidence of an asymmetric response to increase and decrease in attribute levels relative to the status quo. Ignoring asymmetric preference formation can therefore result in biased estimates of WTP indicators and welfare measures of change in cases where the outcomes of environmental programmes can plausibly result in both an increase and a decrease relative to a reference option. Compared to a symmetrical modelling approach, the combination of simultaneously accounting for asymmetric preference formation and preference heterogeneity in the choice model yielded additional insights that may be used to inform the development of local strategies towards biodiversity conservation.

Introduction

Discrete choice models are widely used to estimate willingness to pay (WTP) for changes in the provision of environmental goods and services based on individual responses to a choice experiment (CE). Researchers using these models aim to predict choice behaviour for alternative future states of the environment that result from different management or policy decisions. Commonly, a number of management or policy choice options are offered in conjunction with a status quo option. The status quo option often reflects either the present situation regarding the environment or a predicted future change resulting from a business-as-usual scenario.

The status quo option serves as a reference base against the proposed environmental change and thus frames the changes in attributes as gains or losses relative to the base option. This can be understood in the context of prospect theory (Kahneman and Tversky, 1979), which posits responses to choices to be reference-dependent. Respondents would, as a result of a simplifying heuristic applied in the choice process, relate their decisions to a comparison with a neutral reference, thus framing choices as gains or losses (Tversky and Kahneman, 1981).

Recent evidence from stated choice applications to transportation research suggests that such ‘referencing’ matters in the decision-making process (Hensher, 2008, Hess et al., 2008, Hess and Rose, 2009). Hess et al. (2008) investigate if preferences for trip choice attributes differ for increases and decreases in attribute levels, relative to a reference option. Differences in preferences for increases and decreases would represent an asymmetric response to positive and negative deviations of attribute levels from a reference point. Hess et al. test whether respondents are equally responsive to increases or decreases in attributes such as road toll or trip cost relative to an individual-specific reference option reflecting respondents’ recent or current trip. They found asymmetrical response patterns, which suggest that preference formation in stated preference methods may be related to differences from an individually specified reference option, rather than the result of considering the actual absolute values the attributes takes. WTP estimates for gains or losses in attributes derived from the respective parameter estimates were found to differ and show asymmetries for some but not all of the attributes.

Drawing on data from a CE survey from Indonesia, in this paper I demonstrate an asymmetrical modelling approach. I estimated models that included separate parameters for both increases and decreases in attribute levels for a range of environmental services. The choice options in this study were framed as the outcomes of a public rural development and conservation programme in the buffer region of the Lore Lindu National Park in Central Sulawesi. Especially in early planning phases of such a programme, there can be a considerable degree of flexibility about the objectives, how to achieve them, and how to allocate resources to different programme components under budget constraints. For example, conservation and development objectives would have to be balanced amongst themselves, and against other development needs of the local population. It is therefore plausible that various policy options can result in both increases and decreases in the provision of different ecosystem services.

The approach to asymmetric modelling of responses to a CE follows the one suggested in Hess et al. (2008) but makes use of local knowledge on the environmental services. This knowledge is used to individually specify the status quo option that respondents use as a reference base to a subsequent choice task. This is, to my knowledge, a novel feature in the environmental economics literature. The attributes, for example water availability for paddy rice cultivation, were described in a way that related to the everyday experience of respondents. The use of local knowledge was motivated by a lack of reliable data on the supply of environmental services, together with observations that environmental conditions varied markedly within the Indonesian study area. Because of this approach, the estimates of marginal WTP for changes in attribute levels must be interpreted relative to a status quo option that differs among respondents.

This paper contributes to the emerging literature on asymmetric preference formation in stated preference techniques. The main objective of this paper is to raise awareness of the possibility of relating respondents’ experience to their stated choices, and to model responses to CEs as asymmetrical with environmental economists who use CEs. To my knowledge, this is the first application of an asymmetrical modelling approach in environmental choice modelling.

I compared model results of a symmetrical and an asymmetrical modelling approach. This allowed me to assess whether an asymmetrical approach would result in an improved understanding of respondents’ preferences for environmental services, and if WTP estimates would differ between the two modelling approaches. I also assessed whether differences in preferences for either increases or decreases in attribute levels existed.

Section snippets

Defining a status quo option

The specification of the status quo option is crucial for any approach that incorporates both gains and losses relative to a reference base. If the status quo option is defined as the current state of the world, attributes can be designed to reflect deviations in both directions from the status quo option. This excludes representations of the status quo as a future worst-case business-as-usual scenario that frames all choices as gains (e.g., Colombo et al., 2005, Hanley et al., 2006).

A status

Study background

The data used are drawn from a study carried out in Central Sulawesi, Indonesia. This study elicited preferences for selected environmental services of inhabitants living in the vicinity of the Lore Lindu National Park (LLNP), which is one of the few large remaining mountainous rainforest areas on Sulawesi (Waltert et al., 2004). There were three general objectives of this study.

First, the Central Sulawesi rainforests in Indonesia are part of the global Wallacea biodiversity ‘hotspot’ (Myers

Econometric modelling approach

Stated choice methods have foundations in random utility theory (RUT, e.g. Manski, 1977, McFadden, 1974). According to RUT, the utility associated with alternative j out of a choice set C_i for individual i is partitioned into a deterministic, systematic and observed component V_ij and a random part of utility ɛ_ij: $U_{i j} = V_{i j} + ɛ_{i j} \forall j \in C_{i}$

Observed utility V_ij is a linear and additive function of the attributes and takes the following form for a symmetrical modelling approach, i.e. for an approach that

Results

With respect to the relevance of changes in attribute levels to household well-being, on average, water availability was perceived most important, followed by the preponderance of different shade management regimes in cocoa plantations, rattan availability in the forest and the size of anoa populations. Summary statistics for these variables can be found in Appendix 2.

Of a total of 326 respondents, 22 respondents were omitted because of ‘protest’, payment aversion, or because they had severe

Discussion

I successfully demonstrated the use of perception data for generating an individual-specific status quo option, which allows to model respondents’ choices consistently in terms of individually perceived gains or losses relative to the status quo. Having received some recent recognition in transportation research, this approach is novel to the environmental economics literature. I would like this paper to be conceived as a modest contribution to enriching the SP toolbox for environmental

Concluding remarks

I have demonstrated the potential for using an asymmetrical modelling approach in SP studies aimed at calculation of (local) WTP. In combination with a model that accounted for preference heterogeneity I obtained a more realistic picture of respondents’ preferences for a range of environmental services. It is hoped that these findings will motivate a more widespread application of asymmetric modelling approaches in SP studies for environmental valuation, especially because the informational

Acknowledgements

The author thanks Deutsche Forschungsgemeinschaft (DFG) and The Scottish Government Rural and Environment Research and Analysis Directorate (RERAD) for funding; the villagers of the Lore Lindu Region; the local research team; Jan Barkmann and Rainer Marggraf for research support;Stefan Schwarze and STORMA sub-project A4 for the provision of household data used in Supplemental Material A; Anke Fischer, George Dyer, Kerry Waylen and three anonymous reviewers for useful comments on earlier

References (54)

W. Adamowicz et al.
Perceptions versus objective measures of environmental quality in combined revealed and stated preference models of environmental valuation
J. Environ. Econ. Manage.
(1997)
J. Barkmann et al.
Confronting unfamiliarity with ecosystem functions: the case for an ecosystem service approach to environmental valuation with stated preference methods
Ecol. Econ.
(2008)
E. Barrios et al.
Implications of local soil knowledge for integrated soil management in Latin America
Geoderma
(2003)
I.J. Bateman et al.
Reducing gain–loss asymmetry: a virtual reality choice experiment valuing land use change
J. Environ. Econ. Manage.
(2009)
E. Birol et al.
Using choice experiment to account for preference heterogeneity in wetland attributes: the case of Cheimaditita wetland in Greece
Ecol. Econ.
(2006)
A. Byg et al.
Local perspectives on a global phenomenon – climate change in Eastern Tibetan villages
Glob. Environ. Change
(2009)
V. Cantillo et al.
A Semi-compensatory discrete choice model with explicit attribute thresholds of perception
Transport. Res. B-Meth.
(2005)
S. Charnley et al.
Integrating traditional and local ecological knowledge into forest biodiversity conservation in the Pacific Northwest
For. Ecol. Manage.
(2007)
C.H. Close et al.
A GIS-based protocol for the collection and use of local knowledge in fisheries management planning
J. Environ. Manage.
(2006)
B. Fisher et al.
Poverty and biodiversity: measuring the overlap of human poverty and the biodiversity hotspots
Ecol. Econ.
(2007)

N. Hanley et al.

Estimating the economic value of improvements in river ecology using choice experiments: an application to the water framework directive

J. Environ. Manage.

(2006)

D.A. Hensher et al.

Development of commuter and non-commuter mode choice models for the assessment of new public transport infrastructure projects: a case study

Transport Res. A-Pol.

(2007)

S. Hess et al.

Asymmetric preference formation in willingness to pay estimates in discrete choice models

Transport. Res. E

(2008)

J. Swait

A structural equation model of latent segmentation and product choice for cross-sectional revealed preference choice data

J. Retailing Consum. Serv.

(1994)

J. Swait

A non-compensatory choice model incorporating attribute cut-offs

Transport. Res. B-Meth.

(2001)

J. Wielgus et al.

Including risk in stated-preference economic valuations: experiments on choices for marine recreation

J. Environ. Manage.

(2009)

A. Balmford et al.

Economic reasons for conserving wild nature

Science

(2002)

A. Baranzini et al.

Feel It or Measure It. Perceived vs. Measured Noise in Hedonic Models

(2006)

I. Bateman et al.

Economic Valuation with Stated Preference Techniques – A Manual

(2002)

K.S. Bawa et al.

Tropical ecosystems into the 21st Century

Science

(2004)

J. Belsky et al.

Cultivating Cocoa: implications of sun-grown cocoa on local food security and environmental sustainability

Agric. Human Values

(2003)

L. Bennun et al.

Monitoring important bird areas in Africa: towards a sustainable and scaleable system

Biodivers. Conserv

(2005)

P.C. Boxall et al.

Understanding heterogeneous preferences in random utility models: a latent class approach

Environ. Resource Econ.

(2002)

D.F. Calheiros et al.

Participatory research methods in environmental science: local and scientific knowledge of a limnological phenomenon in the Pantanal wetland of Brazil

J. Appl. Ecol.

(2000)

S. Colombo et al.

Designing policy for reducing the off-farm effects of soil erosion using choice experiments

J. Agr. Econ.

(2005)

S. Colombo et al.

Modelling preference heterogeneity in stated choice data: an analysis for public goods generated by agriculture

Agr. Econ.

(2009)

F. Danielsen et al.

Monitoring matters: examining the potential of locally-based approaches

Biodivers. Conserv.

(2005)

Cited by (21)

Dissonance and commute satisfaction: Which reference point to use?
2021, Transportation Research Part D: Transport and Environment
One of the determinants of travel satisfaction is ‘dissonance’, or the difference between someone’s actual travel and their preferred or ideal travel. Previous studies never questioned the use of ideal travel as a reference point to assess travel satisfaction. We assess how and to what extent changes in commute dissonance, among other determinants, affect commute satisfaction by applying the prospect theory with alternative reference points. Results indicate that negative dissonant commuters (i.e., actual commute time > ideal) are significantly less satisfied with their commute, but the opposite does not hold. By using data from a retrospective survey among staff of the relocated McGill University Health Centre in Montreal, Canada, we found how the reference point of the previous commute time is more salient and impactful. This study provides evidence for policy makers about the positive effects of transit oriented developments in terms of increased levels of commute satisfaction.
Forest owners' preferences for contract-based management to enhance environmental values versus timber production
2021, Forest Policy and Economics
Citation Excerpt :
Consistent with loss aversion (Kahneman and Tversky, 1979; Tversky and Kahneman, 1991), the findings suggest asymmetric preferences for improvements and deteriorations in the attributes describing the potential effects of forest management strategy on environmental values: the losses associated with a decline in the biodiversity and carbon storage attributes were more considerable than gains associated with an equivalent increase in the same attributes. Similar asymmetric findings have been made in the context of travel route choices (Hess et al., 2008), freight transport (Masiero and Hensher, 2010), rural development and conservation programmes (Glenk, 2011), and conditions of water quality (Ahtiainen et al., 2015). According to this result, a forest policy that prioritizes the protection of natural values in forests rich with natural values leads to more significant welfare gains for society than a policy that seeks to improve natural values in forests with few natural values (Ahtiainen et al., 2015).
Forests are sources of multiple ecosystem services (ESs) essential for human wellbeing. Forest owners are critical actors to decide which benefits they produce from their forests. To support the uptake of alternative forest management strategies in a way that is beneficial from the perspective of society as whole, new incentive schemes could be implemented in the future. We applied the choice experiment method to investigate Finnish forest owners' potential participation in an incentive scheme in which they were asked to practice a ‘Timber Oriented’ or a ‘Nature Oriented’ management strategy according to the terms of a hypothetical contract. We found that the majority forest owners are willing to participate in the considered contract-based payment scheme, especially those supporting biodiversity and non-market ESs. Non-profitability attributes including biodiversity, carbon stock, and probability of climate change induced damage were highly valued. Forest owners prefer the management contract with the Nature Oriented strategy. Forest owners' preferences for the contract-based management and associated effects are heterogenous.
Society's preferences when ecological values and health risks are at stake: An application to the population control of a flagship ungulate (Iberian ibex) in Sierra de Guadarrama national park, Spain
2021, Science of the Total Environment
Citation Excerpt :
The analysis of the results also allows analyzing behavior differences between the main users of the national park (visitors) in two different contexts (when they are interviewed either at the park or outside the park), in addition to comparing their preferences with those of other citizens who have not visited the national park within the last year. Moreover, transversally, we will analyze how being more or less informed about the consequences of conducting a management program for the Iberian ibex in this national park could affect the decisions of individuals (Domínguez-Torreiro and Soliño, 2011; Glenk, 2011; Marsh et al., 2010). This latter information could be relevant with respect to designing programs to inform citizens about the Iberian ibex management plan carried out by the national park.
Conflicts derived from the proactive management of ecosystems and wildlife populations abound in national parks, which can prevent the control of some animal populations, consequently causing negative effects to the ecological values and creating health risks for the ecosystems. This work quantifies a conflict related to population control of the Iberian ibex in the Sierra de Guadarrama National Park in Spain where ecological values and health risks are at stake. A discrete choice experiment was conducted of three population samples: 430 on-site visitors, 210 off-site visitors and 210 non-visitors, and two levels of status quo information were considered. The results show that not conducting any ibex management program in the park causes a loss of social well-being and that the design of the management program is shown to be relevant for obtaining greater or lesser acceptance by the surveyed population. In general, better ecological and health levels, as well as avoiding having to kill animals in the park, increase a program's acceptance. Management measures are also shown to take on greater importance to the extent that the results in the health and ecological indicators are worse. Finally, in aggregate terms, additional information about the status quo did not generate major differences in the estimates of a change of well-being.
Valuation when baselines are changing: Tick-borne disease risk and recreational choice
2019, Resource and Energy Economics
Understanding how changes in baseline risk influence preferences for risk reduction is important when valuing the welfare effects of environmental change, including the spread of disease. We conduct a survey-based choice experiment among respondents residing in areas with different prevalence of ticks and incidence of Lyme borreliosis (LB) and tick-borne encephalitis (TBE) in Sweden. Respondents face a trade-off between risk and travel cost when choosing between visiting recreational areas differing in prevalence of ticks and disease incidence. Our study indicates that the presence of ticks and the associated risk of tick-borne diseases significantly influence the choice of recreational area and have substantial welfare effects. The mean willingness to pay (WTP) per trip to avoid areas with different levels of ticks, LB risk and TBE risk ranges from 12 to 78 EUR. The WTP for risk reduction is significantly lower among respondents residing in risk areas compared to respondents in emerging risk areas. Explanations for these differences in WTP for risk reduction between groups with different baseline risks include differences in reference point utility, knowledge and learning, leading to adaptation of behaviour and preferences.
How current lawn attributes affect choices concerning water conserving lawn options: An individualized choice experiment in Kelowna, British Columbia
2019, Landscape and Urban Planning
This paper contributes to the study of water use and policies governing land use in residential landscapes. We employed a discrete choice experiment to examine the effect existing lawn attributes have on stated preferences of residents of detached homes in the City of Kelowna, British Columbia, for four approaches to reducing the irrigation requirements for their residential lawns: restrictions on lawn sizes; the use of drought-tolerant turfgrass; restrictions on outdoor irrigation; and subsidies for removing or replacing lawns. Each respondent’s current lawn (i.e., the status quo) was included as a possible choice in the choice sets, and the attributes of the current lawn were used in constructing the alternatives. Part-worth utilities for a conditional logit and mixed logit model with these status quo predictors were used to analyze the choice data and provide estimates for lawn choices. Results show that status quo conditions contribute to the estimation of utility coefficients for lawn features and lawn choices. Irrigation watering costs and low levels of subsidies to encourage lawn replacement were found to only marginally influence residents’ landscaping decisions. Contrastingly, status quo factors representing the current fraction of turfgrass in the homeowner’s total landscape was the strongest motivating factor driving residents’ lawn choices. Residents with larger proportions of turfgrass were more likely to choose landscaping changes that featured smaller percentages of lawns. Another significant status quo factor in residents’ lawn choices was turfgrass variety, where residents with traditional varieties of turfgrass were more likely to choose landscaping options with water conserving lawns.
Estimating welfare impacts where property rights are contested: methodological and policy implications
2018, Land Use Policy
Where rights over natural resources are contested, the effectiveness of conservation may be undermined and it can be difficult to estimate the welfare impacts of conservation restrictions on local people. In particular, researchers face the dilemma of estimating respondents’ Willingness To Pay (WTP) for rights to resources, or their Willingness To Accept (WTA) compensation for foregoing these rights. We conducted a discrete choice experiment with respondents living next to a new protected area in Madagascar, using a split-sample design to administer both WTP and WTA formats, followed by debriefing interviews. We first examined the differences in response patterns to the formats and their performance in our study context. We also used the two formats to elicit respondents’ attitudes to conservation restrictions and property rights over forestlands. We found that the format affected the relative importance of different attributes: WTA respondents strongly favoured livelihood projects and secure tenure whereas neither attributes were significant for WTP respondents. The WTA format outperformed WTP format on three validity criteria: it was perceived to be more plausible and consequential; led to fewer protest responses; and was more appropriate given very low incomes. Seventy-three percent of respondents did not accept the legitimacy of state protection and strongly aspired to secure forest tenure. The use of a WTP format may thus be inappropriate even if respondents do not hold formal rights over resources. We conclude that estimating the opportunity costs of stopping de jure illegal activities is difficult and coercive conservation lacks procedural legitimacy and may not achieve full compensations. Our findings question the viability of the current conservation model and highlight the importance to conservation policy of locally legitimate property rights over forestlands.

View all citing articles on Scopus

View full text

Using local knowledge to model asymmetric preference formation in willingness to pay for environmental services

Abstract

Introduction

Section snippets

Defining a status quo option

Study background

Econometric modelling approach

Results

Discussion

Concluding remarks

Acknowledgements

J. Environ. Econ. Manage.

Ecol. Econ.

Geoderma

J. Environ. Econ. Manage.

Ecol. Econ.

Glob. Environ. Change

Transport. Res. B-Meth.

For. Ecol. Manage.

J. Environ. Manage.

Ecol. Econ.

J. Environ. Manage.

Transport Res. A-Pol.

Transport. Res. E

J. Retailing Consum. Serv.

Transport. Res. B-Meth.

J. Environ. Manage.

Economic reasons for conserving wild nature

Science

Feel It or Measure It. Perceived vs. Measured Noise in Hedonic Models

Economic Valuation with Stated Preference Techniques – A Manual

Tropical ecosystems into the 21st Century

Science

Cultivating Cocoa: implications of sun-grown cocoa on local food security and environmental sustainability

Agric. Human Values

Monitoring important bird areas in Africa: towards a sustainable and scaleable system

Biodivers. Conserv

Understanding heterogeneous preferences in random utility models: a latent class approach

Environ. Resource Econ.

Participatory research methods in environmental science: local and scientific knowledge of a limnological phenomenon in the Pantanal wetland of Brazil

J. Appl. Ecol.

Designing policy for reducing the off-farm effects of soil erosion using choice experiments

J. Agr. Econ.

Modelling preference heterogeneity in stated choice data: an analysis for public goods generated by agriculture

Agr. Econ.

Monitoring matters: examining the potential of locally-based approaches

Biodivers. Conserv.