Why is it Difficult to Put a Value on Irreplaceable Heritage Assets? Compare Revealed Preference and Contingent Valuation Methods in the Valuation of Such Assets. The process of valuation aims to place a monetary on environmental impact. In this essay I will first outline some core concepts involved within the processes of both Contingent Valuation (CV) and Revealed Preference (RP) methods of valuation. I will then proceed to introduce the CV method, followed by various RP methods.
Following on from this will be a comparison of the two types of valuation and ultimately an evaluation of both. An irreplaceable heritage asset refers to a place of huge historical or social importance, a place that some might call a ‘national treasure’. Examples within England include Stonehenge, The Royal Botanical Gardens, The Tower of London and The City of Bath. It should be clear to see that almost anything, from a pile of rocks to an entire city, can be classified as a heritage asset.
It should also be clear that the benefits that people draw from these assets are not all the same. Broadly, we can categorise these benefits into ‘use value’ and ‘non-use’ value. The use value is the value which an individual places upon the direct use of a resource, in this case an irreplaceable heritage asset. Non-use values “measure aspects of the resource’s value to individuals which are not linked to actual resource use by the individual” (Cicchetti and Wilde 1992, p1121).
These non-use values are particularly prominent when taken in the context of irreplaceable heritage assets: many people may never want to go and visit Stonehenge or the Tower of London, but if these sites were destroyed people would incur some disutility from the notion that perhaps these assets benefited the nation as a whole, and as a result national pride is depleted. Therefore it is vital to have a precise estimate of the value that society places on heritage assets so as to be able to conduct a cost benefit analysis, and therefore aid the decision making process when coming to terms with potential environmental projects.
The CV method is a ‘stated preference’ method of valuation. It uses various types of elicitation methods to ask a specified population sample about their willingness to pay (WTP) for an increase in a certain resource or their willingness to accept (WTA) payment for a decrease in a certain resource. Put simply, a CV study is a survey asking people to put a monetary value on some sort of environmental asset. The main benefit in using a contingent valuation study when aiming to place a value on irreplaceable heritage assets is the ability to capture ‘non-use’ value.
By asking a population sample whether or not they value the Tower of London, irrespective of whether or not they have ever made use of the site, the survey is clearly able to capture non-use value. CV surveys also provide theoretically correct WTP and WTA measures of utility change. The revealed preference method is based on observation of consumer behaviour, and underpinning this method is the assumption that consumers who utilise a resource more place a higher value on it than consumers who use it less. There are various types of RP methods that can be employed.
One method is the Travel Cost Method; this method is used to value the recreational benefits of environmental resources. The Zonal Travel Cost Method, credited to Hotelling, who argued that if you define a series of concentric zones around a target area and then observe and record the number of visitors from each zone, along with data on travel costs, this would provide enough information for an economist to be able to work out the consumer surplus which arises as a result of the availability of the resource.
There are a few variations of the TCM, such as the Individual Travel Cost Method, however for the purpose of this essay they will not be discussed. Another type of RP method is the Hedonic Price Method (HPM). This method relies on the notion that households reveal their preferences for environmental goods through their decisions about where to locate. However, as this principle is more relevant to resources such as air quality and (lack of) noise pollution, rather than irreplaceable heritage assets, the HPM will not be further discussed.
So, if these two distinctly different methods of valuation are available, why is it so difficult to place a monetary value on cultural heritage assets? The answer to this question lies within the significant shortcomings of both the stated preference methods and the revealed preference methods. In their paper ‘Contingent Valuation: Is some number better than no number? ’, Diamond and Hausman launch a stinging attack on the Contingent Valuation Method, claiming that “reliance on contingent valuation surveys in either damage assessments or in government decision making is basically misguided. The basis for Diamond and Hausman’s argument lies in the fact that there are too many inconsistencies with regards to both the elicitation methods used as well as the answers received from respondents. Another example of how different elicitation methods can affect responses is the ‘sequence effect’ (Bishop and Welsh, 1992). The sequence effect occurs when responses to the same questions differ only because the order of the questions differs.
Clearly this is a bias as the answers to each question should not differ simply because the questions are in a different order. Other biases include strategic bias, starting point bias as well as information bias. Diamond and Hausman also identified an embedding effect, essentially insensitivity to scope, which was present in an abundance of responses. This embedding effect, or part-whole bias, refers to the situation in which respondents place the same values on two goods, one being more inclusive than the other.
An example of this would be placing equal values on the benefits of cleaning one lake and the benefits of cleaning five lakes, including the previously mentioned single lake. D&H put forward the notion of credibility; are the respondents answering the question the interviewer is trying to ask? Clearly the embedding effect affects the precision of these answers, and so fundamentally threatens the credibility of CVM. Another threat to the credibility of the CVM is the ‘warm-glow’ effect.
This effect occurs when an individual derives utility simply from making a symbolic commitment to a cause. Again, if the individual is simply putting a value down in answer to a CV question because he or she derives utility from doing so, rather than really taking into consideration the value he or she places on the resource, then the credibility of the CV study is threatened. To sum up this point, a quote “The absence of direct market parallels affects both the ability to judge the quality of contingent valuation responses and the ability to calibrate responses to have usable umbers. It is precisely the lack of experience both in markets for environmental commodities and in the consequences of such decision that makes contingent valuation questions so hard to answer and the responses so suspect. The above are just some of the reasons that D&H believe CVM to be ‘a deeply flawed methodology for measuring non-use values’. While the revealed preference method relies on the analysis of actual data, and so does not suffer from the embedding effects or warm glow effects which plague the stated preference mechanism, there are still significant flaws.
The first of these is multi-purpose utility; often, TC models do not take into account that people make recreational trips with more than one purpose in mind; it would be reckless to attribute the whole travelling cost to the experience obtained from visiting the heritage site. In their paper ‘Travel Cost Analysis of a Cultural Heritage Site: The Case of Historic St. Mary’s City of Maryland’, Poor and Smith attempt to minimise this concern by limiting their data to “planned, day – long visits, such that travel costs are not embedded in multiple purpose or destination costs”.
The problem here is that they therefore have ignored all those people who have made the journey as part of a multi-purpose trip, and so their value estimates will not be reflected in the final estimate. Another limitation of the revealed preference method is the assumption that neither work not travel time confers disutility. While this allows the wage rate to be equated to travel time and hence the derivation of an opportunity cost, this assumption is highly unrealistic.
The revealed preference method is based on observing the preferences that people reveal; hence it captures the value which people place on the use of a particular resource. As a result, there is no way in which a RP method can capture ‘non – use’ value, and this is the biggest criticism of the RP method. When taken in the context of placing a value on an irreplaceable heritage asset, capturing non-use value is an essential part of the valuation, as it could even be argued that most people value assets such as Stonehenge, with only a tiny fraction of those values being use values.
Perhaps the most relevant point to make with regards to irreplaceable heritage assets is that the revealed preference method is unable to capture the bequest element of non-use value. Bequest values are the values placed on the satisfaction of preserving heritage assets for future generations. The fact that the revealed preference method cannot capture non-use values means that it is the least effective method of placing a value on an irreplaceable heritage asset, despite its reported ability to capture use value more accurately than a contingent valuation study.
However, that is not to say that a contingent valuation study will be of more or less use. Both methods have significant drawbacks and can both be very costly to implement. Poor and Smith suggest that their revealed preference study be supplemented with a CVM study to provide a more accurate estimation, however this use of a contingent valuation study leads us back to the very question Diamond and Hausman ask, is ‘some’ number truly better than no number?
Valuing the Benefits of Cleaning Lincoln Cathedral Pollocino & Maddison : Journal of Cultural Economics25: 131 – 148, 2001 Travel Cost Analysis of a Cultural Heritage Site: The Case of Historic St. Mary’s City of Maryland Poor & Smith: Journal of Cultural Economics 28: 217 – 229, 2004 Contingent Valuation: Is Some Number Better than No Number? Diamond & Hausman: Journal of Economic Perspectives, Volume 8, Issue 4.