1
Title: Evaluating Environmental Assets in Developing Countries: Paracas National Reserve in Peru
Author: Roger Loyola,, Universidad Nacional Agraria La Molina, Perú.
Tel: +51 +1 3495647-239 cel: +51 +1 96482648, e-mail:
Abstract: This study assesses the use of two Contingent Valuation formats to determine the value of preserving a protected area in a developing country. To do so, two types of questionnaires were completed simultaneously among the urban populace of Pisco, one using the Open-Ended format and the other using the Single Dichotomous format to ask about Willingness to Pay to preserve the Paracas National Reserve in Peru. These values were related to a range of social and economic characteristics. Discrepancies were noted between these two formats, whereby much care is recommended when interpreting the results of a specific format.
1. Introduction
Over the past few years, there has been an increasing number of studies focused on the evaluation of environmental assets, generally using Stated Preferences methods, particularly Contingent Valuation (CV). However, despite their usefulness and the good prospects offered by this method, most of these studies have been carried out in the developed countries, where the conditions of the populace differ significantly from those in the developing countries. According Whittington (1998) “many economists and survey researchers have been skeptical about the ability to conduct CV surveys in developing countries in the presumed difficulty of understanding and interpreting respondents’s answers to abstracts questions”. This is where the question arises: can their application be valid for the developing countries as well? This question is justified because in most of the studies focused on Contingent Valuation (CV), it is acknowledged that both payment capacity as well as environmental awareness are important elements in the Willingness to Pay (WTP), (Hadker, 1997). As there is a striking gap between the developed and the developing countries with regard to these aspects, there may be a certain discrepancy between their types of conduct. This situation was proven in the CV studies carried out in the developing countries, where most of them took their target populace as being current users of the service, meaning people who had just received or were still receiving direct benefits from the activity at the time of the interview, perhaps because focusing on any other segment of the populace as a target could result in a large number of protest responses[1]. Similarly, as this type of project is already scarce in the developing countries, there are no comparisons among the different CV formats, which is why there are no firm ideas on any possible differences among them and their impact on the results, and more particularly whether or not they coincide with what is expected in theory.
Consequently, this study is designed to determine the value of WTP for preserving an environmental asset among its surrounding populace, although not direct users, through two types of CV format, analyzing whether or not the differences noted between them are significant or not, as suggested by the theory. To do so, a study was carried out on the Paracas National Reserve (RNP – Reserva Nacional de Paracas) in Peru, using two formats simultaneously: Open-Ended and Single Dichotomous. Similarly, the WTP relations were established with the various social and economic variables through the presentation of different functional forms. The results show that the deduced WTP falls within the expected limits. Nevertheless, the use of a specific format may have different effects on the values found.
2. Contingent Valuation
The report drawn up by the NOAA Panel is a milestone in the history of CV. Until then, CV lacked acknowledgement of its effectiveness as a way of measuring the value of environmental assets, although some studies have been carried out since the early 1980s (Bishop & Haberlein, 1979; Bowker & Stoll, 1988). Problems noted when this technique was used, due mainly to the high level of rejection replies, and the marked variability of the results, produced many challenges to its validity (Bateman et al., 1995).
In 1992, headed by Arrow and Solow, the NOAA Panel was assigned the task of discovering whether CV could be useful for finding environmental values, and if so, under what conditions this should be undertaken. Consequently, it was on the basis of this study that the usefulness and validity of the CV, was acknowledged, while also providing a series of conditions that should be taken into consideration when using this type of study. Despite the widespread proliferation of CV studies, most of them have been carried out in the developed countries. In general, CV studies undertaken in the developing nations focus mainly on two areas: water and sewage supplies, and recreation, tourism and national parks, according to Whittington (1998). Actually, there are many studies in developing countries where is showed often studies been related to implemented healthy programs in Africa.
2.1. Conceptual Formulation
In conceptual terms, a form of deriving the measurements of welfare in practice can be found in Hanemann (1984). This discusses how to obtain the Hicksian compensatory equivalent variation based on discrete data, using an example focused on hunting and poaching. Consequently, assuming that the use of any activity may be derived, with z=0 representing the lack of the asset and z=1 its presence, “m” as income and “s” as the other variables, whereby:
u(z, m; s) = v(z, m; s) + ej ; z = 0,1
Given that e0 and e1 are random variables with a 0 average, the amount A is offered, whereby:
v(1, m-A; s) + e1 ³ v(0, m; s) + e0
In the case of the DC format, the individual knows the function, but the researcher does not, nor how the individual response should be considered as a random variable whose probabilities are given by:
P1 º Pr (individual DAP)
= Pr {v(1, m-A; s) + e1 ³ v(0, m; s) + e0}
P0 = Pr (individual non-DAP) = 1 – P1
Defining h º e1-e0, and given Fh( . ), as the accumulated density function of h, so the DAP may be written as:
(1) P1 = Fh( Dv ), where
(2) Dv º v(1, m-A; s) - v(0, m; s)
In the Probit model Fh( . ) is the normal distribution, although in the Logit model, this is the logistical distribution:
P1 = Fh( Dv ) = ( 1 + e-Dv)-1
Considering the form of the functional utility, not included “s” directly:
(3) V(z, m) = az + bm; b > 0, z = 0,1
or
(4) V(j, y) = aj + bLog y; b > 0, z = 0,1
Considering the functional form for (3) and remembering that E(h) = 0, we can arrive at the mean, which is:
E+ = (a0 + a1)/b ou
E+ = m- m ea0 + a1/b E{eh/b}
The median can also be obtained, which will be [2]:
E* = (a0 + a1)/b, or
E* = m -m ea0 + a1/b
From among these possibilities, Hanemann considers the median as the most appropriate measurement, as it is a more robust indication of the core trend of a distribution, and is not so sensitive to the inclusion of the outlying replies, although it is not Pareto-efficient.
2.2. CV Formats
The various formats that can be used for Contingent Valuation include the continuous types, among which are the following: Open-Ended (OE), Iterative Bids and Payment Cards, as well as the discrete types, among which are the Single Dichotomous Choice (SDC) as well as the Double Dichotomous Choice (DDC).
The OE format consists of asking people directly to what extent they would be WTP for specific quantities of an environmental asset or service. In the case of iterative bids, a series of figures was suggested until the person reached his or her maximum WTP. The payment cards consist of showing the interviewee a set of values from which the true WTP must be selected.
In the case of the SDC, the interviewee is asked whether or not he would pay a specific amount for the asset under study. The DDC choice consists of asking a second question about the amount the interviewee would be willing to pay. Consequently, the interviewee is asked if he would be willing to pay a specific quantity B, and if he accepts, a second higher figure is suggested. However, if the first reply is negative, he is then asked if he will be willing to pay a lower amount.
According to Bateman et al. (1995) the OE methods were common in early CV surveys, but today the DC method is very popular, because this apparently avoids a relatively high incidence of non-replies or protest responses found in this type of study.
2.3. Difference between OE and DC
The literature explains the difference between OE and DC in two categories (Halvorsen & Sœlensminde, 1998): differences between the deduction processes with regard to the manner in which the interviewees reply to the valuation question; and the variation in the statistical efficiency and robustness of the WTP estimates for the various formats. In general, DC produces higher values than the OE, although evidence has been found that the deduction processes differ with regard to strategic conduct and the aversion to risk, with the main source of this difference caused by breaches of assumptions made about the functional form and distribution of the random utility. For example in Bishop, Haberlein y Kealy the relationship was 1.9 between OE and SDC, for Kealy and Turner 1,4 a 2,5 and Bateman et al (1995)1.67.
2.4. True WTP Value
Several studies support the idea that the true WTP value could well lie between the results of the OE models and their dichotomous counterparts. For the former, a minimum value is found because the format used “surprises” the interviewee with the valuation of a non-traded asset, requiring an immediate response, while the latter offers an incentive that tends to boost the replies to higher amounts, particularly due to the “yea-saying” bias, caused by the interviewee possibly assuming that the amount mentioned would be the true value of the environmental asset (Kanheman & Knetsh.,1982). Consequently, the presence of this bias increases the probability that the interviewee accepts the bid. This “framing” or “anchoring” effect may occur when interviewees have not analyzed their WTP for a resource, or if they are unsure of its true value. Another explanation is offered by Bateman et al. (1995), who mentioned that according to Kriström, the idea of taking it or leaving it better reflects a market decision, in comparison to the OE. This is why it is suggested that the DC may be higher than the OE.
3. The Paracas National Reserve
The Paracas National Reserve (RNP) lies in the Paracas District of the Pisco Province, Ica Department, covering an area of 335,000 hectares, of which 117,406 hectares are dry land and 217,594 hectares are seawater. There are large groups of sea lions in this Reserve, with a high density of guano-producing birds and an abundance of migratory birds. The climate fosters the formation of sub-tropical desert. In February and August – the summer and winter months – the temperature reaches 22°C and 15.5°C respectively, with an annual average of 18.7°C. The plants and wildlife flourishing in this Reserve are widely diversified and found in abundance. In the case of wildlife, the widely-diversified substratum of the coastal beaches encourages a wide variety of species consisting of fish, birds, mammals and marine invertebrates, contrasting with the sparse wildlife of the desert. On land, plants are divided into biotopes located in the higher areas called lomas or hill vegetation, while those growing along the shoreline are called halophytic plants. Aquatic plants are represented by a wide variety of algae, which flourish in the bio-ecological conditions at Paracas. This area also shelters some 104 archeological sites reflecting every stage of the development of pre-Hispanic Andean societies, as well as major tourist destinations.
According to INRENA (1996), there are polluting industries around Paracas, including fish canneries and flour mills, a steel mill, and the Otuma salt-flats, where almost all inflow and outflow liquids are not treated, discharging wastes directly into the sea. It notes that there are indications of pollution produced by mining activities along the Pisco and Nazca River Basins, north and south of the protected area. The presence of human settlements should also be taken into consideration, including the Port of San Martín and some fishing villages. At the Port of San Martin, sulfuric acid is offloaded on a daily basis, offering a constant threat to the marine environment. Additionally, there are plans to establish the Pisco Industrial Free Zone, which would also constitute a severe threat to the Paracas National Reserve, due to the presence of industrial plants that would be set up here, as well as the consequent human pressures that this would trigger. Much of this industrial activity is related to the fishing industry. Although physically located outside the Reserve, all these enterprises discharge their wastes into their areas of influence. In addition to all the factors listed above, the current discharges by companies should be taken into account, as well as purifying systems and the final disposal of both solid and liquid wastes. Marine pollution in the Paracas Bay caused by the fishing industry is adversely affecting the life and availability of its resources. Furthermore, port activities are also connected to the industries located in this Region.
4. Methodology
A total of 446 OE format questionnaires and 458 DC type questionnaires were applied almost simultaneously. The interviews were held during November and December among the urban populace of Pisco Province in districts located along the shoreline in the metropolitan region, on weekends.[3] The city was divided into several sectors following the urban plans drawn up by its Town Hall. On this basis, the points to be visited were selected at random. The person chosen to answer the questionnaire was the one who answered the door, over the age of 18 years. The study could have been carried out with heads of families, who are presumably responsible for taking decisions. Although many studies have been carried out in this manner, there is no justification in any of them for stating that this should be the correct procedure. In fact, this would only be valid if proof were available that the “purchase decision” for environmental assets were to be made by the head of the family. This is why it was assumed that the purchase decision would be made by adults. In fact, there is no base for ensuring that a selection will be made in a responsible manner after the age of 18, but this cut-off age was chosen as it is the age of responsibility in Peruvian society, when its citizens become legally liable for all their own acts.