Chapter 3 Research Design and Methodology
3.2 Econometric Models for Contingent Valuation
3.2.2 Parametric Models for Contingent Valuation
3.2.2.1 Random Utility Model
The Random Utility Model interprets respondentsβ answers to the WTP question as the result of their comparison of the underlying utility of two circumstances, i.e. the status quo and the valuation scenario of paying the offered price for the specified ecosystem service. The utility that respondent j derives from a specified circumstance Uij is assumed to consist of a deterministic, observable component Vij and a random, unobservable error component Ιij.
πππ = πππ + πππ (3.12)
where i = 1 if the respondent answers yes to the offered price tj, and i = 0 if he/she answers no. It is reasonable to suppose that the respondent would answer yes only if U1j > U0j. So the probability of yes answer is
P rοΏ½yesjοΏ½ = PrοΏ½π1π + π1π > π0π+ π0ποΏ½ (3.13)
which can be rewritten as
P rοΏ½yesjοΏ½ = PrοΏ½π1π β π0π > β(π1πβ π0π)οΏ½ (3.14)
Define the difference between the two random components as Ιj = Ι1j - Ι0j, and denote the change in utility as βV = V1j β V0j. Equation 3.14 becomes
P rοΏ½yesjοΏ½ = Pr(ππ > ββπ) = 1 β Pr (ππ < ββπ)
(3.15)
It can be noted that PrοΏ½ππ < ββποΏ½ is the cumulative distribution function of Ιj with regard to -βV. In order to construct a Random Utility Model, assumptions are needed to 1) specify the function of the deterministic utility
component Vij so that the utility change βV can be specified, and 2) to specify the statistic distribution of the random component Ιj.
The most commonly used function for the deterministic utility component is the linear utility function (Hanemann 1984; Loomis et al. 2000; Haab and McConnell 2002; Moreno-Sanchez et al. 2012).
πππ = πΌπππ+ π½ππ¦π (3.16)
where ππ is a vector of m explanatory variables with respect to respondent j, ππ is the vector of m coefficients (i.e. πΌπππ = βππ=1πΌπππππ ), π¦π is the
respondentβs income and Ξ²i is the coefficient of income. If the respondent answers no to the WTP question, π0π = πΌ0ππ+ π½0π¦π; if the answers is yes, it means the respondent is willing to forgo a portion of income (the offered price π‘π) in exchange for the ecosystem service specified in the valuation scenario, i.e. π1π = πΌ1ππ+ π½1(π¦πβ π‘π). So the utility change is presented as βπ = π1πβ π0π = (πΌ1β πΌ0)ππ+ π½1οΏ½π¦π β π‘ποΏ½ β π½0π¦π (3.17)
It is reasonable to assume that Ξ²1 = Ξ²0 (denoted as Ξ² for simplicity), i.e. the marginal utility effect of income is constant in the status quo (no answer) and valuation scenario (yes answer) unless the offered price could cause substantial change to the respondentβs income. Define Ξ± = Ξ±1 β Ξ±0, Equation 3.17 becomes
βπ = Ξ±ππβ π½π‘π (3.18)
Now Equation 3.15 becomes
P rοΏ½yesjοΏ½ = 1 β Pr[ππ < β(Ξ±ππβ π½π‘π)] (3.19)
In order to estimate Ξ± and Ξ², the statistic distribution of Ιj needs to be specified. It is usual to assume that Ιj is independently and identically distributed (IID) with the mean of 0. Accordingly, there are two widely used statistic distributions, i.e. the normal and logistic distributions, both of which are symmetric and exhibit the nature that πΉ(π₯) = 1 β πΉ(βπ₯). So Equation 3.19 can be simplified as
P rοΏ½yesjοΏ½ = PrοΏ½ππ < πΌππβ π½π‘ποΏ½ (3.20)
If Ιj follows a normal distribution with mean 0 and variance Ο2, this distribution of Ιj can be transformed into a standard normal distribution of ππβ with mean 0 and variance 1. Then Equation 3.20 becomes π
P rοΏ½yesjοΏ½ = Pr οΏ½πππ< Ξ±πππβπ½π‘ποΏ½ = π· οΏ½Ξ±ππβπ½π‘π ποΏ½ (3.21)
where π·( )is the cumulative distribution function of standard normal distribution. This is the standard probit model in statistics/econometrics. For
a sample of T respondents, define πΌπ = 1 if respondent j answered yes and πΌπ = 0 if the answer is no, the log-likelihood function for the responses of the
whole sample is
ln πΏ (π¦ππ ) = βππ=1πΌπln οΏ½π· οΏ½Ξ±ππβπ½π‘π ποΏ½οΏ½ + (1 β πΌπ) ln οΏ½1 β π· οΏ½Ξ±ππβπ½π‘π ποΏ½οΏ½
(3.22)
The model parameters, Ξ± πβ and β Ξ² πβ , can be readily estimated by standard probit regression procedure of statistics/econometrics software packages which utilise the Maximum Likelihood Estimation algorithm to find the parameters that maximize the log-likelihood of Equation 3.22.
On the other hand, if Ιj follows a logistic distribution with mean 0 and variance π2ππΏ2β , this distribution of Ι3 j can be converted to a standard logistic distribution of ππβ with mean 0 and variance πππΏ 2β . Then Equation 3.20 3 becomes
P rοΏ½yesjοΏ½ = Pr οΏ½ππππΏ <Ξ±πππβπ½π‘πΏ ποΏ½ = [1 + exp οΏ½βπΌπππβπ½π‘πΏ ποΏ½]β1 (3.23)
This is the standard logit model, and the log-likelihood function of the logit model is ln πΏ (π¦ππ ) = β πΌπln οΏ½οΏ½1 + exp οΏ½βπΌπππβπ½π‘πΏ ποΏ½οΏ½ β1 οΏ½ T j=1 + οΏ½1 β πΌποΏ½ ln οΏ½1 β οΏ½1 + exp οΏ½βπΌππβπ½π‘π ππΏ οΏ½οΏ½ β1 οΏ½ (3.24)
Likewise, the parameters, Ξ± πβ and βΞ² ππΏ β , can be estimated by standard πΏ logit regression procedure of statistic software packages to maximize the log-likelihood in Equation 3.24.
So far, a Random Utility Model (probit or logit) based on the linear utility function in Equation 16 has been established to use respondentsβ characteristics ππ and the offered price π‘π to explain the probability of yes answer in a Contingent Valuation survey. The next step is to calculate respondentsβ mean WTP for the ecosystem service under valuation.
As a welfare measure of the monetary value of the ecosystem service, WTP is the amount of money/income that the respondent is willing to forgo in exchange for the environment service. In other words, WTP is the amount of money that makes U0 = U1, i.e. the loss of utility caused by forgoing that amount of money is equivalent to the utility that the respondent can derive from the environment service. For the Random Utility Model defined by Equations 3.12 and 3.16, U0 = U1 means
As mentioned above, define Ξ± = Ξ±1 β Ξ±0, Ξ² = Ξ²1 = Ξ²2 and Ιj = Ι1j - Ι0j, Equation 3.25 yields
ππππ =Ξ±Ξ²ππ+1Ξ²ππ (3.26)
In probit and logit models where ππ is assumed to follow the normal and logistic distribution respectively, the mean of ππ is 0. So the expected value (mean) of respondent jβs WTP is simply
πΈοΏ½πππποΏ½ = πΌπ½ππ (3.27)
The ratio Ξ±/Ξ² can be easily calculated using the estimated parameters of the probit or logit model, thus the mean WTP of respondents in the whole sample is
πΈ(πππ) =Ξ² πΞ± πββ ππ (probit model) ππ Ξ² πΞ± πββ πΏπΏππ (logit model) (3.28)
where ππ is the sample mean of the explanatory variables.
Equations 3.21, 3.23 and 3.28 construct the widely used linear random utility model (probit or logit) for the estimation of respondentsβ mean WTP which adopts a linear function (Equation 3.16) to specify the deterministic utility component (Vij) in a random utility model (Equation 3.12) and assumes a normal or logistic distribution of the random component (Ιj).
One drawback of the linear random utility model is that the variable of income, which may significantly influence respondentsβ answers in Contingent Valuation surveys, is not included in the probability functions (Equations 3.21 and 3.23). This is because the income term is incorporated in the function of Vij in a linear form (Equation 3.16), and the marginal utility of income is assumed to be constant between the status quo and the valuation scenario (i.e. Ξ²0 = Ξ²1 in Equation 3.17). Thus when the utility difference βV between the two circumstances is specified, the income variable π¦π is removed and only the change in the income, i.e. the offered price π‘π is included in the probability functions. For relaxing the assumption of constant marginal utility of income and retaining the income variable in the probability functions, non-linear forms of income term, such as the log-linear form, can be used in specifying Vij
πππ = π½ ln(π¦π) + πΌπππ (3.29)
where the marginal utility of income is ππππ
ππ¦π =
π½
π¦π, and the corresponding
probability function is
P rοΏ½yesjοΏ½ = Pr οΏ½ππ < Ξ±ππ+ π½ln (π¦ππ¦βπ‘π
An even more complicated form of the income term is the Box-Cox Transformation πππ = π½ lnπ¦π πβ1 π + πΌπππ (3.31)
where the marginal utility of income is ππππ
ππ¦π = π½π¦π
πβ1, Ξ» is the transformation
parameter which can be flexibly chosen by researchers, and the corresponding probability function is
P rοΏ½yesjοΏ½ = Pr οΏ½ππ < Ξ±ππ+ π½(π¦πβπ‘π)
π β π¦ ππ
π οΏ½ (3.32)
Probit and logit models can be constructed depending on the normal or logistic distribution assumption about Ιj in Equation 3.20 and 3.32 (Bateman, Willis and Arrow 2001; Haab and McConnell 2002; Carson and Hanemann 2005).
In practice, the models with complicated, non-linear income terms are rarely used in empirical Contingent Valuation studies because of the consequent complication in WTP calculation. In fact, as the offered prices in Contingent Valuation surveys usually account for a very small portion of respondentsβ income, it is not necessary to suppose the marginal utility of income to vary with this small change (Haab and McConnell 2002). Moreover, since the offered price π‘π in Equation 3.30 and 3.32 is no longer a separate independent variable but a component in the complicated income term, the effect of the offered price on the probability of yes answer, which is important to researchers and policy makers, can no longer be clearly revealed by the model estimation. Additionally, income information is usually prone to a great deal of measure error and often collected in the categorical form (income groups/ranges) rather than exact values in real surveys, which further weakens the rationale of using the models with complicated non-linear income terms.
However, the problem of removing the income variable from the probability function remains if the linear random utility model is preferred to the non- linear models. A more critical but rarely mentioned problem regarding the linear random utility model is that if the assumption of constant marginal utility of the income variable is plausible because the offered price would merely cause a small change to respondentsβ income, why the same assumption is not made for other explanatory variables such as respondentsβ age and education level given these variables are unchanged at all between the status quo and the evaluation scenario? If such
assumption is made (i.e. Ξ±0=Ξ±1 in Equation 3.17), all the explanatory variables should be excluded from the probability functions just like the income variable when the utility difference is specified, then the offered price (the change in income) would be the only explanatory variable in the probability functions. This was indeed the case in Hanemannβs original paper where the variables of respondentsβ characteristics were βsuppressedβ and represented by a constant term in his random utility model (Hanemann 1984). However, Contingent Valuation studies following Hanemann (1984)βs approach, such as Loomis et al. (2000)βs highly cited paper, just added other explanatory variables in their models without addressing why the assumption of constant marginal utility is made only for the income variable. Nevertheless, this does not mean that Hanemann (1984)βs original univariate model is satisfactory because it cannot reveal the effects of respondentsβ characteristics on their answers to the WTP question, and more problematically from the statistical aspect, the univariate model is likely to exhibit poor goodness of fit in practice.
The discussion above indicates that the random utility model exhibit a contradiction between the necessity for retaining income and other explanatory variables in the probability function and the illogicality of assuming inconstant marginal utility of these variables in order to retain them in the probability function. The Random Willingness to Pay Model, an alternative approach of constructing parametric models for Contingent Valuation studies, can avoid this contradiction and thus was adopted by this study.