The Valuation of Changes to Estuary Services in South Africa as a Result of Changes to Freshwater Inflow
Report No 1304/1/04
Dec 2004

EXECUTIVE SUMMARY

There are just over 250 functioning estuaries in South Africa and of these an increasing number are subject to river inflow deprivation. River water is abstracted upstream of the estuaries for urban and agricultural use, and runoff into rivers is undermined by forestry and the spread of high water consuming alien vegetation. As a result of the reductions in river water inflow, the services rendered by these estuaries are eroded.

The main purpose of this study was to develop a method of valuing changes in freshwater inflow that would better inform estuary managers on economic issues relating to the problem of freshwater deprivation. This aim is pursued by examination of the problem of freshwater deprivation (Chapter One), explaining how economics can better inform estuary managers on the allocation problem (Chapter Two), identifying an appropriate method of valuation (Chapter Three) and applying this method (Chapters Four to Seven). Two secondary objectives of this study were to develop capacity to apply the method of contingent valuation (CV) and inform the public on the nature of the problem of estuary freshwater deprivation. These objectives were pursued by training four students in applying the contingent valuation method (CVM) to value changes in freshwater inflow into selected South African estuaries and holding meetings with identified target populations to share information on the deprivation problem.

The nature of estuaries and the impact on them of freshwater deprivation is discussed in Chapter One. It is shown that some types of estuary are more vulnerable to freshwater deprivation than others. Particularly vulnerable are the type described as temporary open/closed estuaries. It is argued that there is recognition in South Africa that something should be done to counter the problem of freshwater deprivation into estuaries.

Chapter Two explains what Economics deems to be efficient with respect to freshwater inflow into South African estuaries. Efficiency is achieved when the marginal social cost of the freshwater inflow is brought into equivalence with the marginal social value of the inflow. The marginal social costs of freshwater inflow into estuaries are what people are willing to pay to abstract freshwater from (or projects to reduce runoff to) rivers upstream of the estuaries.

The marginal social values of freshwater inflow into estuaries are somewhat more complicated to estimate than the marginal social costs, because people do not directly use this water, but do so indirectly, by consuming the services dependent upon this inflow. These services take many forms, for instance, the area of the estuary available for boating and the quantities of fish, prawn and birds present in and around the estuary. This study focuses most of its attention on the estimation of the marginal social values.

The CVM is selected for the purpose of estimating the marginal social values because it is highly suited to quantifying the benefits to the public of environmental goods and attributes. It serves as an alternative to indirect techniques (the travel cost and hedonic pricing methods) for quantifying these benefits. These values may be used in decision-making with respect to the management of the environment and, if desired, incorporated into cost benefit analyses.

The circumstances to which the CVM is particularly suited are those where market values are difficult to observe, either directly or indirectly. Markets cannot be relied upon to fully incorporate benefit where there are public good characteristics (non-excludability and non-rivalry) present and where passive use is significant. The benefit relating to the latter type of use is not reflected in market behaviour in ways that can be observed because the markets do not provide opportunities for people to show their preference.

The CVM is a highly appropriate technique for estimating value of changes of estuary services induced by changes in freshwater inflow because the environmental services yielded by estuaries are closely connected with quantity of freshwater flowing into the estuary, they are largely public in nature and it is plausible they may give rise to significant passive use benefits (unobservable ones). The public good features of South African estuaries are the tradition of open access to estuaries and the fact that consumption by one person only reduces the services available to others in a minimal way. The probability of there being a passive use benefit is high given that there are a number of ‘uses’ of estuaries that markets provide little scope for revealing, such as the pleasure of seeing it while passing through the area.

The CVM is a controversial valuation technique. It depends on there being a close correspondence between expressed answers given to hypothetical questions (stated willingness to pay) and voluntary exchanges in competitive markets that would be entered into if money did actually change hands. The fact that it has proved very difficult to establish this correspondence has led to CVM being subject to criticism. On account of this difficulty debate continues over whether CVM makes a ‘category’ mistake by attempting to value judgements as if they were preferences.

Many aspects of the criticism of the CVM have been addressed - in the form of using methods to reduce biases, the adoption of conservative elicitation formats and the reporting of tests for consistency. Chapter Three describes some of the ways the criticism has been addressed and some of the recommended guidelines for applying the method that have emerged during the past 25 years.

One of the guidelines for applying the CVM is that pilot studies be conducted prior to the main study. Chapter Four reports the results of the pilot study that preceded the main study. It was undertaken in the year 2 000 at the Keurbooms Estuary; an estuary located on the southern coast of the Western Cape province of South Africa. Estimates were generated of both the marginal social benefit and marginal social cost and two validation tests were administered. Details of the estimates are discussed later in this summary. The expectations-based validity test showed the values generated in the pilot study to be plausible and consistent, but the convergence-type test generated some unexpected results. The unexpected results relate to the impact in an estimated hedonic price equation of view of estuary and distance of property from estuary on selected recorded prices paid for properties in the area.

Other very important guidelines to follow in applying the CVM relate to sample design. These aspects are covered in Chapter Five. The two sample design issues encountered in this study relate to the selection of estuaries to be valued and the selection of users at these estuaries. On the basis of expert opinion a sample of fifty estuaries were selected, and at seven of these the CVM was applied: the Knysna, Groot Brak and Klein Brak Estuaries in the Western Cape and the Kromme, Swartkops, Kariega and Kowie Estuaries in the Eastern Cape. Funding was available to survey about 5% of the estimated user populations at these seven estuaries. In the light of statistical theory, larger sample sizes than these are preferable – ones in the region of 8 to 15% of the target user populations.

For most of the estuaries both high and low forecasts are generated of the impacts of changes in freshwater inflow into estuaries. For proposed increases in freshwater inflow the former relate to optimistic scenarios and the latter to pessimistic scenarios, but for proposed decreases in freshwater inflow, the former relates to pessimistic scenarios and the latter to optimistic scenarios. The forecasts of impacts on estuary services of changes in freshwater inflows were generated on the basis of expert opinion.

Yet another important guideline that should be followed in applying the CVM is that both the person administering the questionnaire and the respondent should be clear about the scenario being valued. In order to achieve this clarity it is necessary that both understand the following: the forecasts of the impacts of the specified changes, the way the payments will be affected and the need for authenticity in response. Chapter Six provides the background information and discussion on selected issues relating to the administration of the CVM surveys.

Chapter Seven summarises what the respondents stated they were willing to pay and relates this to other information elicited from them. A payment card question format was used to elicit the respondent’s willingness to pay for freshwater inflows into an estuary because this question format was expected to yield conservative values. Chapter Eight draws conclusions on the basis of what is reported in Chapter Seven and makes recommendations.