A review of the Se San 3 hydropower project feasibility study

Prepared for Probe International by Wayne C. White, Ph.D., Foresight Associates

October 16, 2000

Contents

Foreword
Executive Summary
1. Introduction
2. Firm power considerations
3. Cost at project completion
4. Geological uncertainty
5. Schedule slippage
6. Social and environmental
7. Sensitivity of viability indicators
Appendix
About Probe International
Probe International’s Power Sector Reform Series
Other publications by Dr. Wayne C. White

Foreword

In the interest of informed debate about the viability of large hydro dams and sound electricity investment decision making in Vietnam, Probe International is distributing the following review of the Se San 3 Hydropower Project Feasibility Study, conducted by Swedish hydropower consultants SWECO, to interested parties in Vietnam and aid-donor
countries.

Out of 24 possible sites along Vietnam’s top three rivers for hydropower potential, SWECO and the Asian Development Bank recommended the Se San 3 hydropower project as Vietnam’s best option on technical, economic, and environmental terms. Based on this recommendation, the Vietnamese government included the Se San 3 dam in its least-cost hydro development plan as a priority investment.

The Asian Development Bank had planned to provide an $80 million start-up loan for the 260-MW dam this year, in the hopes of attracting the remaining $240 million from international investors and dam builders such as Hydro-Quebec (Canada) or Statkraft (Norway). But in June, the ADB deferred its decision following news reports that the Ya Li dam, located 20 kilometres upstream of the Se San site, had caused devastating flash floods in downstream Cambodia’s Ratanakiri province, drowning 32 villagers, destroying livestock, crops and fishing equipment, and threatening the lives of 20,000 people living
along the Cambodian stretch of the river.

In the wake of the Ya Li drownings, ADB had offered to provide an additional $1.8 million to conduct downstream impact studies. But in October, the ADB announced that it had withdrawn its proposed loan for Se San 3 after Vietnam “formally advised ADB that it no longer requires ADB’s assistance to proceed” with the project.

Although Vietnam gave no reason for its decision, ADB staff report that Vietnam is likely concerned that further environmental studies would result in further delays, and there is still no guarantee that ADB would be in a position to finance the project after the studies are completed.

Vietnam is now expected to pursue other sources of financing. ADB staff maintain, however, that as a power project, the Se San 3 dam is “extremely attractive.”

To test the validity of proponents’ claims about Vietnam’s proposed Se San 3 hydro dam, Probe International commissioned the following review of the Se San 3 Hydropower Project Feasibility Study which reveals that the dam’s international proponents have exaggerated its output and economic benefits while ignoring potential costs that are
likely to be imposed on downstream communities beyond Vietnam’s border with Cambodia. Gráinne Ryder

Probe International

Executive Summary

In February 1999 SWECO delivered to the electric utility of Vietnam a report entitled “Feasibility Study on Se San 3 Hydropower Project, Final Report” (hereafter referred to as “the study”). Its objectives, as listed in the study, include “To establish the economic viability of the Project in economic and financial terms.” The present analysis has
been undertaken to examine the fitness of the study, particularly whether the Government of Vietnam is well advised to accept the report’s findings on project viability. It finds that: The study overstates the value of Se San 3 as an electricity producer by disregarding the concept of firm power. River flow quantities measured prior to the construction upstream of Yali Falls Dam are used in the study to calculate Se San 3 power production. Applying firm power criteria, including dry season constraints, to them
indicates an annual power production of approximately 252 Gwh/yr, indicating that the 1,177 Gwh/yr calculated by the study is overstated by 367%. With the Yali Falls Dam in place, lack of outflow from it in the dry season may cause the firm power production rating of Se San 3 to be even lower, possibly at a value of zero.

The study understates the project cost by ignoring that cost at
project completion will include interest charges during the years of
construction. Whereas the study cites project cost under two scenarios
of US$217 million and US$205 million, we calculate the more relevant
‘Cost at end of construction’ as US$510 million and US$479 million,
respectively. Alternate assumptions in cost of capital, geological
risk, and number of years required for construction result in cost at
end of construction in excess of US$720 million and US$600 million.

The net present value of US$ 192 million and internal rate of return
of 22% calculated in the study use the overstated production benefits
and understated construction costs cited above, and exclude other
components of economic analysis including social and environmental
costs. We calculate returns for the Se San 3 project, with the
production benefit and construction cost values corrected, using a
tariff range from 4.5 cents to 7.0 cents US per Kwh; sensitivities to
geological uncertainty, and schedule slippage were also calculated. In
these scenarios, the resulting Net Present Values were negative in
every case, ranging from US$ -146 million to -518 million, Benefit/Cost
values were all less than one, ranging from .28 to .69, and Internal
Rate of Return was less than the opportunity cost of capital, ranging
from 1% to 6%. These indicate that the project is not viable.

Examination of the “Feasibility Study on Se San 3 Hydropower
Project” indicates that it does not identify or discuss social and
environmental factors found to be associated with Yali Falls Dam, which
is inseparably related with Se San 3. The operational parameters of the
design indicate that the Se San 3 will not temper the downstream
environmental effects of the Yali Falls Dam, but will extend them
further down the river channel: its water release schedule will conform
to that of Yali Falls Dam, so it will not temper flows, and its
reservoir will flood any rapids between it and the Yali Falls dam, so
passage through it will not provide aeration to improve the water’s
quality. The study states that it will pass channel scouring and flow
surges further downstream.

Having found that:

• sensitivity analysis of the study indicates that the project is not economically viable, and
• the study does not address downstream effects that may extend across national borders,

we conclude that the SWECO study is an invalid basis for sound investment decision making.

1. Introduction

In February 1999 SWECO delivered to the electric utility of Vietnam
a report entitled “Feasibility Study on Se San 3 Hydropower Project,
Final Report” (hereafter referred to as “the study”). Its objectives,
as listed in the study, are:

• To establish the optimum design…
• To determine the operating regime of the project
• To prepare cost estimates and construction programme for the Project
• To determine environmental constraints, impacts and mitigating measures relating to the proposed project
• To
  determine the benefits of the Project in terms of power generation and
  peak load capacity, as well as benefits relating to other sectors,
  including intangible benefits
• To establish the economic viability of the Project in economic and financial terms
• To analyse alternative financial scenarios for the implementation of the Project.⁽¹⁾

The present report has been undertaken on the behalf of Probe
International to examine the fitness of the SWECO study. Of particular
interest is whether the Government of Vietnam is well advised to accept
the report’s findings on project viability.

This report discusses factors important to project viability
including firm power considerations, cost at project completion,
geological uncertainty, schedule slippage, and social and environmental
impacts. In section seven the effects of these factors are computed on
the major indicators of project viability.

2. Firm power considerations

The study overstates the value of Se San 3 as an electricity
producer by disregarding the concept of firm power. Firm power is
defined by one study of international power sales agreements in this
way:

Firm power is power which can be made available at any instant, with high reliability, often established at 95-98% probability.⁽²⁾

It goes on to say that while the concept of firm power is
internationally accepted, precise criteria varies by country, and
sometimes by major contract.

The study states a recognition that firm power is that which will be available 90 percent of the time.⁽³⁾
It does not, however, apply that concept. In keeping with the firm
power definition it cites, the study should only calculate project
benefit based on an amount of power that the water flow could produce
with 90% reliability; to the contrary, the study applies full tariff
rates to an amount of energy based on water flow that is an annual
average. The average flow value is distinctly different than the
portion of the flow that is reliable enough to be used for calculating
firm power: 1) average annual flow is, by definition, achieved in 50%
of all years, not 90%, and 2) annual averages do not reflect the
constraint of the driest month to limit firm power production
capability. The result is an unrealistically high value for the
project’s benefit.

How should the study have calculated annual firm power production?
While firm power is a universal concept, each utility generally
codifies its own definition. The present policy of the Electric
Generating Authority of Thailand relative to hydropower, for example,
requires it to consider the lowest month’s production over a ten year
span as defining firm power. In that case, annual firm power would
simply be the lowest month’s production over ten years, multiplied by
twelve (the number of months in a year).

It is beyond the scope of this review to rigorously recalculate
foreseeable firm power production of the project, but an indicative
approximation can be drawn from the study. For our calculation of the
sensitivity of project benefit to firm power considerations, we will
use the more lenient 90 percent reliability mentioned in the study (not
the 95 to 98% reliability based on international agreements cited in
the above quote).

Section 11.5 of the study contains a table “Annual Energy Generation
at Se San 3 Hydropower Project, Simulation Period 1976-95.” Since this
table covers a period of twenty years, we can assume for the sake of
our approximation that production in the two lowest years, 788 GWh and
912 GWh, represents the levels at the bottom ten percent of production
probability. As an indicator of the level of production with 90 percent
probability of being achieved, we can reference the third lowest year,
which has an annual generation level of 958 GWh. The value used by the
study to calculate annual benefit, 1,177 GWh, is 23 percent larger than
this value of 958 GWh, so from this reference alone it appears that the
project benefits are overstated by 23 percent. The actual firm power
rating, however, as controlled by monthly production is lower yet.

Taking the total value of energy production over a year to represent
firm power is inaccurate, since firm power is limited by the lowest
month of production. In some cases it is argued that the lowest month
of a single production facility does not control firm power, as the
interplay of the whole grid system may present complementary peaks and
lows of generating facilities; this clearly does not apply to Vietnam.
Hydroelectricity constitutes the majority of the installed generation
capacity in Vietnam. Shortfalls in production occur in the driest
months. As a new power production facility, Se San 3’s value is
manifest in how well it helps meet the power demand, represented by
that shortfall. Therefore, the firm power production of the project is
controlled by its performance in the driest months.

Monthly flows will be affected through regulation by the Yali Falls
dam, which did not exist when the study was performed. The study did
not deal with the problem of anticipating monthly outputs from the Yali
Falls dam, since, as we are pointing out, it sidestepped the issue of
monthly flow variations entirely. We cannot predict the performance of
Yali Falls dam, so we address the issue in the following way. We will
first look at monthly effects based, as the study calculations were, on
flow characteristics without Yali Falls Dam. This discussion will
introduce principles and flow rates that are foundational for the case
including Yali effects. Then we assume a case representing an
optimistic dry season monthly flow, and also discuss the impact if the
pessimistic case is true.

Section 5.4.4.1 of the study states that “based on the monthly data for the period 1960-97… average discharge [is] 268.5 m³/s….In the driest month, April, the average monthly flow drops below some 100 m³/s.”
Again, this information can be used for an indicative approximation, if
we assume a linear relationship between flow volumes and power
production. This section does not state how far below 100 m³/s the April average flow drops, let alone the rate which is not the average but the 10^th
percentile from the lowest month. It does state that the “Minimum
discharge,” that is, the lowest monthly average over the observed
period, is “46.9 m³/s.” If the average April value is assumed at 100 m³/s and the low is 46.9 m³/s,
we can make a linear approximation of the flow having a 90 percent
chance of being achieved in April by dividing the difference between
the low and average by five, and adding it to the low value. This
approximate method yields a value of 57.5 m³/s. Using our
assumption of linear relationship, the corresponding firm power
production as constrained by flows in April is 252 GWh/year. The
average annual production is 1,177 GWh, a number 4.67 times larger than
annual firm power production as constrained by the driest month. From
this construct it appears that project benefits are overstated by 367
percent.

The actual flow available to produce power at Se San 3 will be
controlled by the discharge rate of the Yali Falls Dam. The Yali Falls
Dam is designed to provide inter-seasonal storage. If it could provide
complete and perfect regulation, then the rainfall for the year would
be evenly disbursed among each of the 12 months and the annual
production of 958 GWh would not be further reduced by monthly
constraints. Since regulation is imperfect, a reduction in firm power
estimate will be introduced by a least-flow month; the question becomes
“by how much?”

As a first case with the Yali Falls Dam upstream, we can arbitrarily
assume that the inter-seasonal regulation of the dam is efficient, and
results in a doubling of dry season flows from the case without the
dam. Doubling the previously derived April flow of 57.5 m³/s, yields an annual firm power production of 504 GWh/year.

The ability of the Yali Falls Dam to regulate inter-seasonal flows
differentials is a function not only of the dam’s height and reservoir
size, but also the vertical placement of the inlets. Designers have an
incentive to place inlets higher within the structure in order to
maximize the height differential (head) of the water as it flows
through the inlet to the turbines, thus increasing the amount of energy
produced per flow of water. High inlets can, however, cause stoppages
in dry periods. The height of the dam and size of the reservoir may be
sufficient for the reservoir to retain water throughout the year. If
that water level, however, falls below the inlet elevation, then no
water will flow through the turbines or out of the reservoir. The Yali
Falls Dam has only recently come into existence so does not have an
operational flow history; its performance must be monitored to
determine if the reservoir level will fall below the inlets for
significant periods.

If it does turn out to be the case that Yali Falls’s reservoir falls
below inlet levels for a month in one-in-ten years or more, valuation
of Se San 3 is left with two options. The first is to calculate, the
firm power rating for Se San 3 at zero or near zero, which would of
course cause the project benefit to be valued at zero or near zero. To
preserve the project’s firm power rating, a backup generating station,
probably diesel powered, would have to be added to the scope of the
project, which would of course raise total project cost.

3. Cost at project completion

The study understates the cost of construction by ignoring the fact
that, as a project that will take several years to construct, cost at
project completion will include interest charges during the
construction period.

In section 17.1, the report defines an economic analysis as being
“from the point of view of the country,” as distinct from the financial
analysis, being “from the point of view of the developer of the
project.” Section 17.2 goes on to say that “All financial charges are
also excluded of the economic analysis.”

It is not, in our opinion, possible for the study to adequately
inform the country of Vietnam as to what its full costs for the project
will be, the project’s rate of return, or the project’s comparative
cost to alternative forms of power generation without including
financing costs in the analysis. Interest payments during construction
are quite real and unavoidable; income for repaying the construction
loans is not available until the project is completed and begins its
productive life. Hydropower projects, as the most capital intensive,
and having the longest construction period, of the power generating
alternatives under consideration, are particularly sensitive to
financing costs.

Section 15.6 contains a table indicating when construction costs
will occur over a five year construction period; the table is
reproduced here:

Table 3.1

Year	Distribution of costs, %
	General	Civil	Mech.	Electrical	Transmission
1	50	10	–	–	–
2	20	20	–	–	–
3	10	25	30	25	–
4	10	25	40	40	50
5	10	20	30	35	50

The report, in Chapter 18, in the Conclusions section, lists the
following as the project’s costs, for a Roller Compacted Concrete
structure built by an international contractor, and a Mass Concrete Dam
built by Vietnamese contractors:

Table 3.2

	International Competitive Bidding, MUSD (RCC Dam)	Local Contractors (Mass Concrete Dam)
		MVND (Million Vietnam Dong)	Corresponding MUSD, @ 1 USD = 14,000 VND
General Works	37.8	529,560.0	37.8
Civil Works Dam Structure Waterway Power Station Tailrace Switchyard	49.3 3.5 6.5 3.3 3.0	622,549. 26,513. 63,608. 25,727. 19,692.	44.5 1.9 4.6 1.8 1.4
Total Civil works	65.6	758,089.0	54.2
Mechanical works	40.2	562,800.0	40.2
Electrical works	47.4	663,600.0	47.4
Transmission line	6.4	89,600.0	6.4
Eng. & Sup., 10%	19.7	260,365.0	18.6
Total Capital Costs	217.1	2,864,014.0	204.6

With the information supplied in these two tables, we can calculate
the construction cost per year for each of the two structure/contractor
types. No distribution over the years of construction is given for
‘engineering and supervision,’ so its amount will be added to ‘General’
which should have a similar distribution.

Table 3.3

Expenditure per year for RCC dam built by International Competitive Bidder
	General, MUS$37.8+19.7=57.5		Civil,MUS$ 65.6		MechanicalMUS$ 40.2		ElectricalMUS$ 47.4		TransmissionMUS$ 6.4		Total
Year	%age	amount	%age	amount	%age	amount	%age	amount	%age	amount
1	50	28.75	10	6.56	–		–		–		35.31
2	20	11.5	20	13.12	–		–		–		24.62
3	10	5.75	25	16.4	30	12.06	25	11.85	–		46.06
4	10	5.75	25	16.4	40	16.08	40	18.96	50	3.2	60.39
5	10	5.75	20	13.12	30	12.06	35	16.59	50	3.2	50.72

Table 3.4

Expenditure per year for Mass Concrete dam built by Local Contractors
	General, MUS$37.8+18.6=56.4		Civil,MUS$54.2		MechanicalMUS$ 40.2		ElectricalMUS$ 47.4		TransmissionMUS$ 6.4		Total
Year	%age	amount	%age	amount	%age	amount	%age	amount	%age	amount
1	50	28.2	10	5.42	–		–		–		33.62
2	20	11.28	20	10.84	–		–		–		22.12
3	10	5.64	25	13.55	30	12.06	25	11.85	–		43.1
4	10	5.64	25	13.55	40	16.08	40	18.96	50	3.2	57.43
5	10	5.64	20	10.84	30	12.06	35	16.59	50	3.2	48.33

Now that we have derived expenditures per year, we only need to know interest rates in order to calculate total debt at the end of construction. We will use a uniform cost of capital, representing a weighted average of equity and debt rates. For the project built with international contractors we will apply a cost of capital of 15 percent, for the domestically contracted project we apply a cost of capital of 12 percent.

Table 3.5

	RCC dam built by Internat’l Competitive Bidder			Mass Concrete dam built by Local Contractors
all units mil’ns US$	New expenditure	Previous w/ 15% interest	Sum	New expenditure	Previous w/ 12% interest	Sum
Begin year 1	35.31	0.00	35.31	33.62	0.00	33.62
End of year 1	24.62	65.23	89.85	22.12	59.77	81.89
End of year 2	46.06	149.38	195.44	43.10	134.82	177.92
End of year 3	60.39	285.15	345.54	57.43	256.70	314.13
End of year 4	50.72	448.09	498.81	48.33	400.16	448.49
End of year 5		573.63	573.63		502.31	502.31

From this calculation, we can see that the debt at the end of construction for the “RCC dam built by an international competitive bidder” is 573.63 million US dollars, and that the debt at the end of construction for the “Mass concrete dam built by local contractors” is 502.31 US dollars, or, at an exchange rate of 14,000 VND per dollar, 7.03 trillion VND.

In the transmittal letter accompanying the study, the Vietnamese Ministry for Planning and Investment references a different set of interest rate assumptions than we have made. They are “Amortisation 10%,” “loan cost 8.5%,” “contribution cost 15%,” with the contribution cost described as that “mobilised from home resources,” and refers to
what we have described above as ‘equity.’ Assuming a twenty percent equity share, the weighted average of the “loan,” and “contribution” rates is approximately 10%. The following table recalculates debt at the end of construction using a ten percent cost of capital in keeping with the Ministry for Planning and Investment’s assumptions.

Table 3.6

	RCC dam built by Internat’l Competitive Bidder			Mass Concrete dam built by Local Contractors
all units mil’ns US$	New expenditure	Previous w/ 10% interest	Sum	New expenditure	Previous w/ 10% interest	Sum
Begin year 1	35.31	0.00	35.31	33.62	0.00	33.62
End of year 1	24.62	63.46	88.08	22.12	59.10	81.22
End of year 2	46.06	142.95	189.01	43.10	132.44	175.54
End of year 3	60.39	268.30	328.69	57.43	250.53	307.96
End of year 4	50.72	412.28	463.00	48.33	387.08	435.41
End of year 5		509.30	509.30		478.96	478.96

Even with these interest rates, which are lower than those used in the previous calculation, project costs at end of construction are more than double the project cost figures cited by the study ignoring interest during construction.

4. Geological uncertainty

The study is based on geological research including site visits and a degree of subsurface exploration. Due to the limits, however, of the subsurface exploration, it is wise to examine the sensitivity of the project cost if geological conditions are encountered which require a greater level of effort in construction. We will do so by assuming an
increase of 25% in the cost of the “dam structure” portion of the “civil works” item. This will result in a cost as follows, distributed over the years of construction.

Table 4.1

Expenditure per year for RCC dam built by International Competitive Bidder
	General, MUS$37.8+19.7+ 12.3=69.8		Civil,MUS$ 65.6		MechanicalMUS$ 40.2		ElectricalMUS$ 47.4		TransmissionMUS$ 6.4		Total
Year	%age	amount	%age	amount	%age	amount	%age	amount	%age	amount
1	50	34.9	10	6.56	–		–		–		41.46
2	20	13.96	20	13.12	–		–		–		27.08
3	10	6.98	25	16.4	30	12.06	25	11.85	–		47.29
4	10	6.98	25	16.4	40	16.08	40	18.96	50	3.2	61.62
5	10	6.98	20	13.12	30	12.06	35	16.59	50	3.2	51.95

Table 4.2

Expenditure per year for Mass Concrete dam built by Local Contractors
	General, MUS$37.8+18.6+11.1=67.5		Civil,MUS$54.2		MechanicalMUS$ 40.2		ElectricalMUS$ 47.4		TransmissionMUS$ 6.4		Total
Year	%age	amount	%age	amount	%age	amount	%age	amount	%age	amount
1	50	33.75	10	5.42	–		–		–		39.17
2	20	13.5	20	10.84	–		–		–		24.34
3	10	6.75	25	13.55	30	12.06	25	11.85	–		44.21
4	10	6.75	25	13.55	40	16.08	40	18.96	50	3.2	58.54
5	10	6.75	20	10.84	30	12.06	35	16.59	50	3.2	49.44

With that cost per year now calculated, we can calculate the revised debt at end of  construction.

Table 4.3

	RCC dam built by Internat’l Competitive Bidder			Mass Concrete dam built by Local Contractors
all units mil’ns US$	New expenditure	Previous w/ 15% interest	Sum	New expenditure	Previous w/ 12% interest	Sum
Begin year 1	41.46	0.00	41.46	39.17	0.00	39.17
End of year 1	27.08	74.76	101.84	24.34	68.21	92.55
End of year 2	47.29	164.40	211.69	44.21	147.87	192.08
End of year 3	61.62	305.07	366.69	58.54	273.67	332.21
End of year 4	51.95	473.64	525.59	49.44	421.51	470.95
End of year 5		604.43	604.43		527.46	527.46

For the RCC dam built by an international competitive builder, the increase in debt at end of construction is from $573.63million to $604.43million, an increase of 5 percent.

For the Mass concrete dam built by local contractors, the increase in debt at end of construction is from $502.31million to $527.46, an increase of 5 percent also.

5. Schedule slippage

It is responsible practice in feasibility studies to calculate the sensitivity of project cost to delays in construction. The study did not do this, as it did not consider the project cost at end of construction at all, let alone with sensitivity to variations.

It is common for dams to take longer to build than scheduled. This may occur due to geological reasons as described in the previous section, which require additional construction steps. Other factors may include unfavorable weather, supply problems, or simply that the original schedule was overly optimistic.

During the construction period, interest on the construction loans continues to accrue. Because of this compounding effect, delays in construction can be a major financial factor, significantly raising the amount of debt at the end of construction.

In this section we will examine the sensitivities to accumulated debt if the Se San 3 project requires an additional year to build, for a total of six years, or an additional two years, for a total of seven. In the following calculation we will impose a delay without increasing the total amount of money spent on construction; we do so in order to illustrate the effects of delays in isolation. In a case such as a delay due to the need for greater foundation work because of geological conditions, greater expenditure will be needed, and the total construction time will increase.

Table 5.1 Construction schedule increased by one year

	RCC dam built by Internat’l Competitive Bidder			Mass Concrete dam built by Local Contractors
all units mil’ns US$	New expenditure	Previous w/ 15% interest	Sum	New expenditure	Previous w/ 12% interest	Sum
Begin year 1	35.31	0.00	35.31	33.62	0.00	33.62
End of year 1	24.62	65.23	89.85	22.12	59.77	81.89
End of year 2	46.06	149.38	195.44	43.10	134.82	177.92
End of year 3	60.39	285.15	345.54	57.43	256.70	314.13
End of year 4		397.37	397.37		351.83	351.83
End of year 5	50.72	507.70	558.42	48.33	442.38	490.71
End of year 6		642.18	642.18		549.59	549.59

Table 5.2 Construction schedule increased by two years

	RCC dam built by Internat’l Competitive Bidder			Mass Concrete dam built by Local Contractors
all units mil’ns US$	New expenditure	Previous w/ 15% interest	Sum	New expenditure	Previous w/ 12% interest	Sum
Begin year 1	35.31	0.00	35.31	33.62	0.00	33.62
End of year 1	24.62	65.23	89.85	22.12	59.77	81.89
End of year 2	46.06	149.38	195.44	43.10	134.82	177.92
End of year 3	60.39	285.15	345.54	57.43	256.70	314.13
End of year 4		397.37	397.37		351.83	351.83
End of year 5		456.98	456.98		394.05	394.05
End of year 6	50.72	576.24	626.96	48.33	489.66	537.99
End of year 7		721.01	721.01		602.55	602.55

From this calculation we can see that a two year delay in the construction schedule can increase project cost in excess of US$65 million.

6. Social and environmental

While the study offers some discussion of environmental and social impacts, it fails to factor them into its assessment of project viability. Additionally, recent information about the impacts of Yali Falls Dam indicate that the scope of social and environmental impacts
will exceed those identified in the study. Comprehensive social and environmental assessments are necessary components in assessing the feasibility of a large  infrastructure project such as Se San 3. In terms of the financial analysis, they should identify, and inform the government of Vietnam as to, costs for mitigating and managing impacts (including management efforts that save money in the long run). They may also reveal ways in which social and environmental factors affect project returns, such as environmental sensitivities that could lead to watershed degradation and therefore
decreased usable water flows and shortened service life. The social and environmental assessments have an even farther reaching significance for the economic analysis. As in the financial analysis, they should address the cost and investment considerations just mentioned. In addition, they should capture the full effects on the economy, beyond
project cash flows. This is important because it does not behoove a nation to make investments that generate revenue in one way, without considering if they actually diminish the national wealth in other sectors.

–>

Examination of the “Feasibility Study on Se San 3 Hydropower Project” indicates that 1) it does not identify or discuss social and environmental factors found to be associated with Yali Falls Dam which will be inseparably related with Se San 3, and that 2) social and
environmental effects are not included in calculation of the indicators by which the report declares the project viable. The study includes the following in its description of downstream effects of the dam: The Se San 3 Hydropower Station will not be run independently but will be conditional to the case of Yali Hydropower Station. In the case that the Yali Hydropower Station would be operated on peak demand only, e.g. be closed down, let say between 23:00 hours and 05:00 hours each day, the Se San 3 has to follow suit. The damage to the environment by a ‘peak demand operation mode’ is related to the rapid emptying and filling of the stream course, most severe near the release point at the power station, however, some negative effects probably reach some 20 km downstream of
the dam. The physical impact of a peak operation mode is related to effects on the dynamics of erosion, sedimentation and water quality. Concerning the living environment, it is fair to assume that hardly any “higher” aquatic organisms will be able to cope with the abnormal conditions of the short term regulation of water, i.e. diurnal sequence drought and flooding.

…the Se San 3 Hydropower project will not introduce any new type of environmental impact but only extend the prevailing impact further downstream.⁽⁴⁾

Impacts of the Yali Falls Dam have been reported downstream past the international border and into Ratanakiri Province, Cambodia.⁽⁵⁾

Water quality problems and flash flooding caused by the dam have been blamed for human deaths and deaths of aquatic species and land animals that drink from the river, for damages to property, and loss of income. If, for example, the study is correct and the Se San 3 dam will transfer effects 20 kilometers further downstream, then significant
effects would be carried 20 kilometers further into Cambodia. This is a serious environmental impact, and one whose trans-border character may intensify its economic significance. Vietnam would not be well advised to proceed with the Se San 3 project without at least considering this situation, so the study which does not evaluate it is inadequate for making a decision on project viability.

The Se San 3 will not temper the downstream environmental effects of the Yali Falls Dam, but will extend them further down the river channel. The project design and projected operation indicate that its water release schedule will conform to that of Yali Falls Dam, so it will not temper flows. Its backwater will flood any rapids between it and the Yali Falls dam, so passage through Se San 3’s reservoir will not improve the water’s quality. The study states that it will pass channel scouring and flow fluctuation effects further downstream.

7. Sensitivity of viability indicators

In this section we calculate the returns for the Se San 3 project, incorporating the considerations not present in the study, of firm power, cost at project completion, geological uncertainty, and schedule slippage. Project viability is indicated by the results, expressed in terms of Internal Rate of Return, Benefit Cost, and Net Present Value.
No quantification is presented for environmental and social impacts; valuation of these impacts is beyond the scope if this report, and the results of project viability analysis are clear even without them. We begin with the most optimistic case.

Sensitivity Case 1-Optimistic Case

In this optimistic case we make the following assumptions:

• All civil works constructed at the cost projected in the study, and on time by the study’s projected schedule.
• Inter-seasonal regulation of the upstream Yali Falls Dam is efficient, and results in a doubling of dry season flows from the case without the dam.
• Tariff valued per the study at “7.0 Usc/KWh when the project is  commissioned.”⁽⁶⁾
• The productive life of the dam is 30 years per the study’s assumption.⁽⁷⁾
• Social and environmental costs ignored per the study.

Table 7.1

Sensitivity Case 1aunits millions of US Dollars
Project Cost at construction completion, end of year 5
RCC dam built by International Competitive Bidder	Mass Concrete dam built by Local Contractors
$573.63	$502.31
Benefit from electricity supply
Annual firm production of 504 GWh/year at value $.07/KWh =
$35.28
Internal rate of return, 30 years
4.5%	5.5%
Net present value
Present value of $35.28/yr for 30 years at 12% cost of capital = 8.055(35.28) =
284.18
NPV, PV of benefit minus cost
-289.45	-218.13
Benefit/Cost
0.50	0.57

In case 1b we make the even more optimistic assumption that the cost
of capital will be ten percent, in keeping with the projections of the
Ministry of Planning and Investment.

Table 7.2

Sensitivity Case 1bunits millions of US Dollars
Project Cost at construction completion, end of year 5
RCC dam built by International Competitive Bidder	Mass Concrete dam built by Local Contractors
$509.30	$478.96
Benefit from electricity supply
Annual firm production of 504 GWh/year at value $.07/KWh =
$35.28
Internal rate of return, 30 years
5 1/2%	6%
Net present value
Present value of $35.28/yr for 30 years at 10% cost of capital =9.427(35.28) =
332.58
NPV, PV of benefit minus cost
-176.72	-146.38
Benefit/Cost
0.65	0.69

Both the ADB and the World Bank have advised Vietnam to raised its
electricity price to 7 US cents per kilowatt-hour by the end of 1999
“if Vietnam wants to keep funds coming in.”⁽⁸⁾ The

reason cited for the rate increase is to ensure that EVN meets the
Bank’s self-financing requirement of 30 percent and is able to save
enough to repay loans. This case demonstrates that even if EVN were to
do so, and even if the full retail rate were applied to value power
from this facility, it is not economically viable.

Sensitivity Case 2-Optimistic with reasonable electricity value

The transmittal letter under which the Vietnamese Ministry for
Planning and Investment forwarded the study indicates concern about the
electricity unit cost used in the study:

Preliminarily the project is feasible…based on a
high sale price of power (5.6 to 7 US cent/kWh), this input is not
close with the real situation of Vietnam in the present time. In other
similar projects it is taken in account only at an average sale price
at bus bars of 4.5 – 5.0 USC/kWh.⁽⁹⁾

Analyses of financial efficiency shall be carried
out with a mean sale price of power at bus bars in the region of about
4.5 – 5.0 US cent/kWh.⁽¹⁰⁾

In cases 2a and 2b we calculate project returns with tariffs of 4.5 US cents/kWh and 5.0 US cents/kWh respectively.

Table 7.3

Sensitivity Case 2aunits millions of US Dollars
Project Cost at construction completion, end of year 5
RCC dam built by International Competitive Bidder	Mass Concrete dam built by Local Contractors
$573.63	$502.31
Benefit from electricity supply
Annual firm production of 504 GWh/year at value $.045/KWh =
$22.68
Internal rate of return, 30 years
1%	2%
Net present value
Present value of $22.68/yr for 30 years at 12% cost of capital = 8.055(22.68) =
182.69
NPV, PV of benefit minus cost
-390.94	-319.62
Benefit/Cost
0.32	0.36

Table 7.4

Sensitivity Case 2bunits millions of US Dollars
Project Cost at construction completion, end of year 5
RCC dam built by International Competitive Bidder	Mass Concrete dam built by Local Contractors
$573.63	$502.31
Benefit from electricity supply
Annual firm production of 504 GWh/year at value $.05/KWh =
$25.20
Internal rate of return, 30 years
2%	3%
Net present value
Present value of $25.20/yr for 30 years at 12% cost of capital = 8.055(25.20) =
202.99
NPV, PV of benefit minus cost
-370.64	-299.32
Benefit/Cost
0.35	0.40

Sensitivity Case 3-Two year construction schedule slippage

This case utilizes the same assumptions as case 2b, with the
exception of consideration of a two year increase to the construction
schedule. Conservatively, no additional construction disbursement is
added, although schedule delays are frequently accompanied by increased
costs for work items.

Table 7.5

Sensitivity Case 3units millions of US Dollars
Project Cost at construction completion, end of year 7
RCC dam built by International Competitive Bidder	Mass Concrete dam built by Local Contractors
$721.01	$602.55
Benefit from electricity supply
Annual firm production of 504 GWh/year at value $.05/KWh =
$25.20
Internal rate of return, 30 years
less than 1%	1 1/2%
Net present value
Present value of $25.20/yr for 30 years at 12% cost of capital = 8.055(25.20) =
202.99
NPV, PV of benefit minus cost
-518.02	-399.56
Benefit/Cost
0.28	0.34

1. SWECO, “Feasibility Study on Se San Hydropower
Project, Final Report,” section 1.3.1. All subsequent references are to
the study unless otherwise noted.

2. Minogue, Diane C., World Bank Industry and
Energy Department Working Paper, Energy Series Paper No. 42, A Review
of International Power Sales Agreements, August 1991, p 3.

3. “firm energy, i.e. energy with a reliability of 90%…” section 16.2, p. 16-3.

4. Section 14.5.4 B.

5. The Fisheries Office, Ratanakiri Province, in
cooperation with The Non-Timber Forest Products Project, “A Study of
the Downstream Impacts of the Yali Falls Dam in the Se San River Basin
in Ratanakiri Province, Northeast Cambodia,” May 29, 2000.

6. Paragraph 17.2.3.1

7. “The economic life span of the Project is 30 years with no replacement costs.” Section 17.1, page 17-2.

8. Interview with Hoang Trung Hai, general
director of Electricite du Vietnam (EVN) in “Electric Dreams” Vietnam
Economic Times, August 1998.

9. Letter from the Ministry for Planning and
Investment, Socialist Republic of Vietnam, No. 7061 BKH.VPTD to the
Prime Minister of the Government, dated 10 October 1998, section 2.7.

10. Ibid, section 3.4.

About Probe International

Probe International
educates Canadians about the environmental, social, and economic
effects of Canada’s aid and trade abroad. Together with citizens groups
around the world, we monitor and expose the effects of projects
financed by Canadian tax dollars through international financial
institutions such as the World Bank and the Asian Development Bank. For
more information see https://journal.probeinternational.org/

Probe International is a division of Canada’s Energy Probe Research Foundation,
a leading energy and environmental think-tank which has argued for
breaking up Canada’s electricity monopolies since the early 1980s and
is helping develop new rules for Ontario’s electricity market. For more
information see http://www.energyprobe.org

Probe International’s Power Sector Reform Series

• 2000, Benefit Cost Analysis of Decommissioning the Pak Mun Dam (in Thailand), Wayne White, Foresight Associates.
• What
  Thai Citizens Should Know About Canada’s Nuclear Power Program,
  briefing by Probe International in association with Energy Probe for
  Alternative Energy Project for Sustainability of Thailand, February
  1999.
• 1999, The Three Gorges Dam:
  A Great Leap Backward for China’s Electricity Consumers and Economy,
  Patricia Adams and Gráinne Ryder, Probe International.
• 1999, Thailand’s Flawed Electricity Privatization: The Case for Citizen-Oriented Reform, Gráinne Ryder, Probe International.
• 1999,
  The Theun-Hinboun Public-Private Partnership: A Critique of the Asian
  Development Bank’s Model Hydropower Venture in Lao PDR, Gráinne Ryder,
  Probe International.
• 1997, The Advantages of Combined Cycle Plants: A “New Generation” Technology, Gráinne Ryder, Probe International.
• 1997, Reforming Thailand’s Power Sector: Towards a Sustainable Electricity Future, Gráinne Ryder, Probe International.
• 1996,
  An Economic Critique of Nam Theun-Hinboun Hydropower Project and
  Electricity Development in Laos: Proposal for an Alternative Path to
  Development, Thomas Adams, Borealis Energy Research Association.

Other publications by Dr. Wayne C. White

• Infrastructure Development in the Mekong Basin, for Oxfam America, July 2000.
• Benefit Cost Analysis of Decommissioning the Pak Mun Dam, June 2000.
• A
  review of the Power Purchase Agreement between the Republic of the
  Philippines National Power Corporation and a consortium constituting
  the San Roque Power Corporation concerning the construction and
  operation of the San Roque Multipurpose Project, March 2000.
• Economic viability of Mekong Region Dam projects, delivered at World Commission on Dams Regional Consultation for East and Southeast Asia, Hanoi, Vietnam, 27 February 2000.
• Structuring Finance for Dam Project Sustainability, for World Commission on Dams, thematic review on International Trends in Project Financing, July 8, 1999.
• Making environmental and social assessment relevant for dams, submitted to World Commission on Dams, Web-Conference on EIAs and SIAs for Large Dams, June 3, 1999.
• Review of Economic Impact Study: Nam Theun 2 Hydroelectric Project, for International Rivers Network, September 1997.
• Review of Greater Mekong Task Force Strategies, for International Rivers Network, August 1997.
• Nam Theun 2 project economics, for International Rivers Network, July 1996.