Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus Willingness to Work

Environ Resource Econ (2016) 65:697–721 DOI 10.1007/s10640-015-9919-8 Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus Willingness to Work J. M. Gibson3 · D. Rigby1 · D. A. Polya2 · N. Russell1 Accepted: 15 April 2015 / Published online: 9 May 2015 © The Author(s) 2015. This article is published with open access at Springerlink.com Abstract A concern when conducting stated preference valuation studies in rural developing or very low income contexts is the use of monetary willingness to pay (WTP) estimates. In circumstances where cash incomes are extremely low, a significant proportion of the population are not engaged in waged labour and the exchange of goods or services is augmented through barter or work exchange, the role of money is likely to be different from that within an urban developed setting. As such, ability to pay using money may be impaired and downwardly biased when compared with other mediums of exchange. In recognition of this several studies have used hypothetical labour contributions as payment vehicles and a common finding is that households are more often willing to contribute labour than they are money. In this paper we present the results of a split sample DCE using money and labour contributions as payment vehicles for improved drinking water quality in Kandal Province, Cambodia. We find little differences between the payment vehicles in terms of attribute non-attendance, marginal utilities of attributes or derived welfare values. We argue that this provides support for the use of WTP in rural developing areas where there are functioning labour markets. Keywords Choice experiments · Development · Health · Payment vehicles · WTP · WTW · Risk · Water · Latent class · Attribute non-attedance 1 Introduction The valuation of environmental stocks and services is increasingly undertaken in developing countries. Elicitation and estimation of willingness to pay (WTP) in such contexts poses many This paper has not been submitted elsewhere in identical or similar form, nor will it be during the first three months after its submission to the publisher. B J. M. Gibson 1 Economics, School of Social Sciences, University of Manchester, Manchester, UK 2 School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Manchester, UK 3 Manchester Centre for Health Economics, University of Manchester, Manchester, UK 123 698 J. M. Gibson et al. challenges ranging from lack of detailed data for revealed preference studies to language, cultural and logistical barriers for stated preference studies. The challenges are such that the validity of some stated preferences studies in developing countries has been called into question (Whittington 2002). One significant concern highlighted in the literature (see for instance Hardner 1996; Hung et al. 2007; Casiwan-Launio et al. 2011) is the use of monetary WTP measures with the rural poor. Since cash incomes are extremely low, a significant proportion of the population are not engaged in waged labour and the exchange of goods or services is augmented through barter or work exchange, the role of money income in the rural developing setting is likely to be different from that of an urban developed setting. It has been argued that, in these contexts, WTP measures, based on ability to pay using money, will not appropriately represent the economic value of the policy/good/attribute under consideration, given that economic exchanges may occur without the use of money (see for instance Hardner 1996; Hung et al. 2007; Casiwan-Launio et al. 2011). Furthermore, since this downward bias would particularly affect the WTP estimates of the rural poor, those households who are perhaps most in need of assistance may be unduly discriminated against in wider project analysis that utilises WTP estimates. The problem may be further amplified through increased hypothetical biases. Given that rural poor households may not be as integrated with markets compared with households in developed countries, hypothetical contingent markets, which are key for stated preference surveys, may seem less realistic which could lead to unreliable results. Fundamental to this issue is a lack of markets using money as the medium of exchange, the predominant reason why non-market valuation studies are undertaken in the first place. Access to labour, goods and credit markets would permit a household to liquidate current assets to allocate resources efficiently or to access credit for increased current consumption. In these circumstances WTP estimates based on payments using money would be an appropriate welfare measure. In rural developing country settings barter and work exchange can facilitate exchanges in lieu of fully functioning monetary markets.1 If a household does not have the ability to utilise markets to access money, response to stated preference survey questions, which often specifically ask households to consider their money income before responding to stated preference questions, will be made in relation to these liquidity constraints and may thus provide a biased estimate of true economic value. It may therefore be pertinent to consider trade-offs with respect to other resources, rather than, or in addition to, money. Researchers have used a variety of alternative payment vehicles in stated preference valuation studies, for instance Shyamsundar and Kramer (1996) use baskets of rice and Asquith et al. (2008) used in-kind payments such as beehives. However, the most common substitute for money in stated preference studies is labour contributions (Hardner 1996; Kamuanga et al. 2001; O’Garra 2009; Casiwan-Launio et al. 2011; Rai and Scarborough 2013, 2014). Labour is the most commonly used alternative to money due both to the ubiquitous nature of potential working hours, i.e. all households are faced with the decision of how best to allocate their time between productive and leisure activities, and because labour time is arguably the most important asset at the disposal of the rural household. The utilisation of household time for productive activities, however, may be limited by the available labour market or self-employment opportunities, potentially leading to an unsatisfied demand for work and constrained money income. Although utilising these available 1 However transaction costs are likely to be high in terms of matching buyers and sellers which may lead to inefficiencies. 123 Discrete Choice Experiments in Developing Countries. . . 699 working hours, rather than constrained money income, as a medium of exchange in stated preference studies might ease the problem of downwardly biased WTP measures, using working time as a payment vehicle for assessing welfare effects is potentially complicated by the need to apply an opportunity cost of time to achieve monetary measures of welfare and the uncertain impact that the choice of payment vehicle may have on choice behaviour. Further com (...truncated)