Wayne C. White

Foresight Associates

81 Main Street

Acton,MA 01720, USA

Acknowledgments

Thanks are due to the Rockefeller Brothers Fund, for support which makes possible this program of independent economic analysis of regional hydropower projects; and to Probe International for invaluable support and assistance.

Contents

Executive Summary
I. Introduction
II. Calculating the economic benefit of the Pak Mun Dam
Dependable capacity
Production value
III. Benefits from decommissioning the Pak Mun Dam
Livelihood valuation
Limits to valuation
IV. Benefit comparison
Benefit/cost ratio
Sensitivity
Endnotes

Executive Summary

Conventional wisdom in Thailand, has long regarded the damming of a river for electricity production as beneficial and progressive. The Pak Mun Dam has emerged as a critical challenge to the assumed premium of dam development over preservation of an existing river ecology. It has been suggested that the Dam be decommissioned, that is, removed from active generating service, to allow the river to flow free. This report compares the economics of continuing to operate the dam as it has been since its construction, versus the benefit of decommissioning it to restore the river’s ecology.

There are several factors which combine to make the decommissioning option credible. The project has a markedly lower generating output than projected. Thailand does have alternative sources for electrical generation. The Mun River has been found to have a very rich aquatic culture, and its fisheries, and related fish populations in the Mekong, are significantly degraded by the dam and channel works. Local populations are traditionally heavily dependent on fishing for sustenance as well as income. Alternatives for the affected persons to earn income from farming or from fishing elsewhere are conditioned by pressure on Thailand’s land and river resources.

The present value of the benefit of operating the dam for electricity production is calculated within a range of $43.9 million to $57.2 million, with the $43.9 million figure most closely representing present operation parameters. The average value is $50.6 million.

The present value of the benefit for restoring the river on livelihood alone is calculated within a range of $43.7 million to $117.3 million, with the narrower range of $88 to $117.3 million representing the total number of affected families (6,202) identified by an independent commission. The average value for the calculation covering the total number of families is $102.7 million.

This calculation indicates that the benefit of decommissioning the Pak Mun Dam exceeds the value of leaving it in operation.

I. Introduction

Conventional wisdom in Thailand, as in many countries, has long regarded the damming of a river for electricity production as beneficial and progressive. The resulting power generation is welcomed, since electrification is viewed as a benchmark of development and a prerequisite of industrialization. Consumption and alteration of the natural resource base, including rivers, has been largely accepted as a regular accompaniment of the nation’s rapid economic growth.

The Pak Mun Dam has emerged as a critical challenge to the assumed premium of dam development over preservation of an existing river ecology. Thousands of persons affected by the dam have mounted a campaign of sustained protest, advocating river restoration. Although the dam’s construction is complete and it’s turbines have been on-line since late 1994, they have demanded that its gates be permanently raised and the river allowed to run free. Doing so would effectively decommission, that is, remove from active service, the dam, which requires water impoundment in order to generate power.

While abandoning an expensive new generating system may seem drastic, there are several factors which combine to make the challenge to the Pak Mun Dam’s overall benefit credible. The project has a markedly lower generating output than projected, a function of design limitations and streamflow. Thailand does have alternative sources for electrical generation, and the value of electricity produced must be taken on a comparative basis. The Mun River has been found to have a very rich aquatic culture, and its fisheries, and related fish populations in the Mekong, are significantly degraded by the dam and channel works. Local populations are traditionally heavily dependent on fishing for sustenance as well as income. Alternatives for the affected persons to earn income from farming or from fishing elsewhere are precluded by overall competition for and pressure on Thailand’s land and river resources.

To advance the discussion of whether it would be better to cease allowing the Pak Mun Dam to restrict the river’s flow, application of benefit/cost methodology is helpful to rationally compare options. This report compares the economics of continuing to operate the dam as it has been since its construction, versus the benefit of decommissioning it to restore the river’s ecology.

II. Calculating the economic benefit of the Pak Mun Dam

In examining the economic benefit of the Pak Mun Dam, it is necessary to first identify the range of contributory factors, then to derive a value for them. This will, of course, also be true in the following section, which examines benefits of decommissioning the dam.

Survey of the Pak Mun Dam’s benefits yields the single factor of electricity production. As the project was submitted for appraisal by the World Bank in 1991, its purposes included electricity, irrigation, and increased fish production; some mention was also made of recreation/tourism value.⁽¹⁾ Following implementation, tourism and fish harvests have fallen due to the dam, so it is not possible to accrue any benefit on that account. The project as built has no irrigation component.

The value of the electricity benefit will be calculated by applying appropriate rates to the dam’s observed power output. A survey of the country’s electrical tariffs has been compiled for the National Energy Policy Office of Thailand, and rates from this source, specifically those known as generation marginal costs, will be applied to the electricity production quantities.⁽²⁾

Dependable capacity.

Calculating the value of the dam’s annual generation by this method requires that we know the dependable capacity of the facility, expressed in kilowatts (or megawatts, etc), and the annual production, expressed in kilowatt-hours (or megawatt-hours, etc). The Pak Mun has turbines with a total capacity of 136 Megawatts, but this number is not the dependable capacity. Dependable capacity is an industry term that describes the amount of power that can be relied upon to be available a high percentage of the time; for hydropower it depends not just on the equipment size, but on availability of water to move the turbines. EGAT, like all utilities, has a definition of dependable capacity, as will be referenced below.

The World Commission on Dams report on Pak Mun⁽³⁾ looks into the question of defining the project’s dependable capacity. A long quotation is justified on this important point, including a table:

…. Using EGAT’s definition of Dependable Capacity, and based on daily power output data between 1995-98, the Pak Mun hydropower project Dependable Capacity is 20.81 MW if one assigns all available power to a 4-hour peak demand period.

This is most likely an overestimation of Dependable Capacity….According to the definition of Dependable Capacity used by EGAT the Pak Mun dam, and taking into account the likelihood of being able to generate Peak Power, the Dependable Capacity is 16.16 MW…. the sharp, 4 hour long demand peak of the past has been replaced by a 13 hour plateau that lasts from 0900 to 2200 hours, from Monday to Saturday….It is clear that the Pak Mun does not have enough water flow to be a dependable 13-hour peaking plant.⁽⁴⁾

Summary of Dependable Capacity calculations for 1995 – 98⁽⁵⁾

Dependable capacity assumed by EGAT in its 1988 presentation to the Cabinet of Ministers	75 MW
Dependable capacity calculated from simulated mean monthly data based on the 1967-81 Pak Mun run off record assuming a 4 hour peak.	40.9 MW
Dependable capacity calculated from daily power output in 1995-98, assuming that all available power gets assigned to a 4 hour peak demand period. This calculation and the three summarised below follow the rules for Dependable Capacity Calculations officially declared by EGAT in March 1988	20.81 MW
Dependable capacity calculated from daily power output in 1995-98, assuming that as much as possible available power gets assigned to a 4 hour peak demand period, taking account the likelihood that EGAT is able to assign a maximum amount of power to the peak period.	16.16 MW
Dependable capacity calculated from daily power output in 1995-98, assuming that all available power gets assigned to a 13 hour peak demand period	6.40 MW
Dependable capacity calculated from daily power output in 1995-98, assuming that as much as possible available power gets assigned to a 13 hour peak demand period, taking account the likelihood that EGAT is able to assign a maximum amount of power to the peak period.	4.97 MW

Production value. With this determination of dependable capacity from the WCD, and knowing the total annual energy production at Pak Mun in 1998 and 1999 (listed below), we are ready to calculate the annual value of electricity production at Pak Mun. The National Energy Planning Office document lists the marginal energy cost as .85 baht/kwh, and the capacity cost as 1290 baht per kw. To perform the calculation we multiply the annual production in kwh by the rate of .85 baht/kwh, multiply the dependable capacity by 1290 baht per kw, and then add the two together. The calculation is presented in tabular form below, and also converts the result into US dollars.

Table 1.1, Annual Value, Dependable Capacity 16.16 MW (4 hour peak)

Year	Total Prod’n GwH	Energy benefit, production multiplied by marginal cost of .85 baht/kwh	Capacity benefit at 16.16 MW multiplied by 1290 baht per kw	Total annual benefit in baht	Total benefit in dollars at 37 baht to the dollar
1998	283	240,550,000	20,846,400	261,396,400	7,064,768
1999	189	160,650,000	20,846,400	181,496,400	4,905,308
Average					5,985,038

The values in table 1.1 are for the case of a four hour daily peak period. Thailand now manages its electric grid on the basis of a 13 hour peak (or plateau), so that case will be represented in Table 1.2.

Table 1.2, Annual Value, Dependable Capacity 4.97 MW (13 hour peak)

Year	Total Prod’n GwH	Energy benefit, production multiplied by marginal cost of .85 baht/kwh	Capacity benefit at 4.97 MW multiplied by 1290 baht per kw	Total annual benefit in baht	Total benefit in dollars at 37 baht to the dollar
1998	283	240,550,000	6,411,300	246,961,300	6,674,630
1999	189	160,650,000	6,411,300	167,061,300	4,515,170
Average					5,594,900

These calculations yield an annual value of electricity from the Pak Mun Dam of US$6.0 million, and US$5.6 million, with the lower value being correct in conjunction with the actual length of the peak period.

With the annual value quantified, it is possible to calculate the present value of all the electricity produced over the service life of the dam. The opportunity cost of capital is 12 percent. We will assume that the service life of the project is 25 years without major overhaul or reservoir dredging. For sensitivity we will also look at longer periods of 30 and 35 years.

Table 1.3, Present value calculation of service life electricity generation

Annual value	25 years	30 years	35 years
$5,985,038.00	$46,941,485.63	$48,210,582.14	$48,930,701.59
$5,594,900.00	$43,881,579.02	$45,067,948.78	$45,741,126.84

The case reflective of the 13 hour peaking period presently used in Thailand, for a present value of 25 years operation, yields US$43.9 million. The sensitivity of this value to longer service life is not great, rising only by four percent, or less than $2 million in the case of 35 year life.

III. Benefits from decommissioning the Pak Mun Dam

As for the case of maintaining the dam, benefits from decommissioning must be identified, then valued. The Pak Mun dam has had negative impacts on fisheries, aquatic biodiversity, tourism, social welfare, and livelihood; decommissioning is intended to reverse these effects.

A primary impact of the dam has been on fish populations, and that, in turn, has caused many of the livelihood and social effects:

The Pak Mun Dam has affected aquatic biodiversity and relative abundance of fish populations up and downstream of the dam. The headpond has inundated and destroyed significant spawning habitats such as rapids….Of the 265 species recorded in the Mun watershed before 1994, 77 species were migratory. Furthermore 35 species depended on rapid habitats, now inundated by the Pak Mun reservoir. The latest survey after dam construction recorded only 96 species upstream of the dam. There has been an apparent impact on 169 fish species. A total of 56 fish species have disappeared from known catch since the construction of the dam.⁽⁶⁾

Many of the persons affected by the dam complain of lost fisheries, which not only curtailed their traditional source of cash income, but also a major component of their dietary sustenance. Fishing was a form of safety net, since the rice farming practiced locally was rain-fed, and vulnerable to dry weather years. Some persons also experienced partial or total loss of land. The riverside formerly acted as a commons and was the predominant grazing area; it is now underwater and the reservoir banks are not suitable for this use. Some farmers who had accumulated buffalo as an investment and form of savings had to sell them for lack of pasturage.⁽⁷⁾

A social effect of the dam is the break up of family structures. Many of the young and able bodied from the affected areas have relocated in order to find work. Some of these have gone to the nearby city of Ubon Ratchatani, some to Bangkok, and some abroad to the Middle East.⁽⁸⁾

It would be very difficult to tabulate values for each of the dam’s effects, especially over long periods of time. Fortunately there is one method available to us that can render a value for the bulk of the benefits of restoration. An accepted method for deriving a value is identification of the cost of substitution. Indeed, this is of-a-kind with how markets work: for example, the market value of a house is determined based on evidence of what similar houses have sold for. In the case of Pak Mun, a commission, acting with inputs from stakeholders, has determined a value for replacing the lost livelihood of the affected persons. Although this compensation has not been made, and may never be, it is useful, and appropriate to this calculation.

Livelihood valuation.

In January 1995 the Committee for providing Assistance and Developing Occupation for Fisherman (CAODFF) in the areas affected by Pak Mun was appointed.⁽⁹⁾ The CAODFF ultimately determined that 6,202 families had their livelihoods affected by the dam.

An affected persons group determined that compensation that rose to the level of allowing the affected families a new source of livelihood would best be met by award of 20 rai of land per family. It requested such compensation for 3301 households. The government, on April 29, 1997 decided to award 3,080 families with 15 rai of land each. Later this concept was changed to monetary compensation, calculated at 35,000 baht per rai, or 525,000 baht.

We can tabulate the implications of these numbers on total value.

Table 1.4

	baht/family @ 35,000 baht per rai	total value, baht	total value, dollars @ 37 baht per $
3,301 households per the January 1997 request of affected persons
15 rai land/family	525,000.00	1,733,025,000.00	46,838,513.51
20 rai land/family	700,000.00	2,310,700,000.00	62,451,351.35
3,080 households per the April 1997 government ruling
15 rai land/family	525,000.00	1,617,000,000.00	43,702,702.70
20 rai land/family	700,000.00	2,156,000,000.00	58,270,270.27
6,202 households per the CAODFF total finding
15 rai land/family	525,000.00	3,256,050,000.00	88,001,351.35
20 rai land/family	700,000.00	4,341,400,000.00	117,335,135.14

As stated above, this compensation may never actually be paid⁽¹⁰⁾, but our purpose is to use it to express value, and for that it is very useful. As such, we apply the compensating amount, the lower range figure being 525,000 baht, to the entire 6,202 households, and derive a low value of US$ 88 million, or upper range of US$ 117 million.

Limits to valuation.

An advantage of the livelihood compensation valuation method used is that also treats social effects: the alternative livelihood of rice farming presents the prospect for preserving the family structure.

The livelihood compensation method does not account for biodiversity survival benefits within the Mun River ecosystem, for sustenance of the downstream Mekong ecosystem, or recreation and tourism benefits. Thus the actual benefit of dam decommissioning would be higher than the calculated amount.

IV. Benefit Comparison

Benefit/cost ratio.

The present value of the benefit of operating the dam for electricity production is calculated within a range of $43.9 million to $57.2 million, with the $43.9 million figure most closely representing present operation parameters. The average value is $50.6 million.

The present value of the benefit for restoring the river on livelihood alone is calculated within a range of $43.7 million to $117.3 million, with the narrower range of $88 to $117.3 million representing the total number of affected families (6,202) identified by an independent commission. The average value for the calculation covering the total number of families is $102.7 million.

Restoration of the river may also allow recovery of tourism, survival of the river and shoreline ecosystem, protect the health of the Mekong ecosystem, reverse social disruption, reduce pressure on forests, allow resumption of family structures, and benefit public health by preventing development of schistosomiasis. Globally, and as evidenced by real spending, preservation of a single species has been valued in the millions of dollars or more; the benefits of decommissioning in addition to livelihood may possibly be in the tens of millions of dollars.

Assuming a number of $20 million for the benefits of river restoration over and above livelihood compensation, and taking the average values of $50.6 million for dam operation and $102.7 million for dam decommissioning, calculation of a benefit cost ratio of decommissioning is as follows: ($102.7 mil + $20 mil) / ($50.6 mil) = 2.4.

For construction of a project a benefit cost ratio of greater than 1.0 is generally taken as sufficient justification. Under this set of assumptions, decommissioning has a benefit cost ratio of 2.4. <!–

–>This comparison can be represented graphically, using the value ranges (not their averages), and taking the arbitrary assumption of $20 million present value for non-monetized environmental existence and similar factors as discussed above:

<!–

–>

This calculation indicates that decommissioning the Pak Mun dam is a realistic option for maximizing Thailand’s national best interest.

Sensitivity. This study is short and streamlined; sensitivities of factors that may be included in a more detailed analysis, however, predominately lie in the direction of emphasizing its primary conclusion. The factors which might be known more specifically with more detailed study include optimal steps for river restoration, dam decommissioning costs, future electricity market costs, and environmental recovery rate.

Factors that affect the benefit of dam operation are unlikely to gain in value. The dependable capacity of the facility and its maximum annual generation are fixed by its physical characteristics and the flow quantities of the river; the governing physical laws will not change. The value of electricity in Thailand is unlikely to rise; with the number of new installations coming online, and the competitive force of independent power producers entering the market, the trend is for lower electricity tariffs.

The value of protected and restored habitat can be expected to rise with time; continuing stresses on Thailand’s natural resource base make the remaining assets more valuable.

Costs and benefits of performing the actual decommissioning are mixed, but need not be high and may be of net positive value. Removal of the generators should result in a net income if they can be resold; if their resale value does not exceed the cost of removal, they can be simply left in place. Likewise for the gates: if the scrap value of the steel does not exceed the removal costs they can simply be left in an open position. The submerged rapids can be restored by simply opening the dam and letting the water return to the natural level. Some work would be required to restore the rapids that were removed by blasting. A specific benefit cost study could be performed to determine whether rebuilding each of these is worthwhile. If they are rebuilt simply by dumping broken stone (possibly the removed material), into place, the cost should be low.

The value of the restoration will be affected by how quickly fish stocks rebound. It is thought that the remnants of migrating species still exist in downstream and mainstream Mekong areas sufficient to allow recovery within a few years.

Endnotes

1. Annez, Philippe under contract with the Thai Development Research Institute, The Hydropower Benefits of the Pak Mun Hydropower Dam and Related aspects of the Mekong River Basin Thailand, Circulation Draft, 15 December 1999, p. 12.

2. PricewaterhouseCoopers, London, Review of Electric Power Tariffs, for National Energy Policy Office, Thailand, January 2000. http://www.nepo.go.th/index-T.html

3. Amornsakchai, Sakchai et al, WCD Case Study: The Pak Mun Dam and Mekong River Basin, Thailand, Final Draft, May 2000.

4. Ibid, pp 24, 25.

5. Ibid

6. World Commission on Dams, WCD Case Studies, The Pak Mun Dam & Mekong River Basin, Thailand, Executive Summary, 22 March 2000, pp 5,6.

7. Field interviews by the author, March 2, 2000.

8. Ibid

9. Amornsakchai et al, p 76.

10. One reason that this compensation will not be paid is that the government experienced a change in policy. Even before that, however, many families expressed their lack of preference for undertaking the labor intensive livelihood of rice production. This is a possible explanation of why only 3,301 households applied for compensation so that they might change their livelihood from fishing to rice farming: others were unwilling to contemplate this change in occupation. Other economic realities would come into play in actual practice. If 6,000, or even 3,000, families sought to buy good farmland all in a similar time period, scarcity would cause the price of land to escalate rapidly.