Independent Economic Analysis Board:
Emery N. Castle, Chair
Joel R. Hamilton, Vice-chair
Kenneth C. Boire
Daniel D. Huppert
Lon L. Peters
Jack A. Richards
Anthony D. Scott
Paul C. Sorensen
The Northwest Power Planning Council requested that the Independent Economic Analysis Board (IEAB) provide a summary and review of six existing economic studies of fish recovery policies and measures. The IEAB has prepared conclusions, recommendations, and findings regarding these studies, which are intended to help the Council in (a) its decisions regarding the funding of specific projects in FY98 and beyond, (b) the cost-effectiveness decisions required by the 1996 "Gorton Amendment" to the Northwest Power Act, and (c) its broader policy deliberations. This task has also helped ensure that the IEAB has a basic understanding of the major studies on the economics of fish recovery actions, and can use that information to help formulate its advice to the Council.
The six studies varied in their scopes, objectives, methods, and conclusions.
(1) The earliest of the studies was done for Bonneville Power Administration by Resources for the Future (RFF). [ Paulsen, C.M., et al. (1993).] This 1993 study focused on the cost-effectiveness of 120 propagation alternatives for achieving sub-basin goals above Bonneville Dam. RFF combined several large models to simulate and evaluate the costs and biological effects of the alternatives studied. (2) The most comprehensive study, done jointly by the Corps of Engineers, Bonneville, and the Bureau of Reclamation, is the Columbia River System Operation Review (SOR) [ Bonneville Power Administration, et al. (1995).] . The SOR evaluated the cost-effectiveness and economic benefits of several system operation strategies. (3) A study done by Daniel Huppert and Dave Fluharty for the National Marine Fisheries Service focused on the cost-effectiveness of the March 1994 proposed recovery plan for three Snake River stocks. [ Huppert, D.D. et al. (1996).] Like the SOR, this study took a comprehensive approach to evaluating the economic effects, but relied primarily on the results of other studies for the estimates of costs and effects. (4) Another study of the Snake River stocks was done for the Power Planning Council by the Environmental Defense Fund (EDF). [ Willey, Z. and A. Diamant (1995).] EDF focused on 26 alternative approaches to increasing water particle travel time in the Snake River system. (5) In 1996, Harza Northwest did a study for the Corps of Engineers evaluating the cost-effectiveness of alternative strategies for the recovery of Snake River spring and fall chinook. [ Harza Northwest (1996).] (6) Finally, ECONorthwest, under contract to the Confederated Tribes of the Umatilla Indian Reservation, criticized the SOR methodology and evaluated the Umatilla proposal for reallocation of water to fish. [ ECONorthwest (1995).]
In addition to conclusions, recommendations, and major findings, attachments to this report include narrative summaries of the studies, tables and charts summarizing the studies and their results, and answers to a set of basic questions, which helped organize our reviews of the studies.
Conclusions of Existing Studies
1. There is a notable lack of consensus among the studies concerning the biological effectiveness of different techniques of fish management. In addition, there is substantial variation among the studies as to what measure of biological effectiveness should be used. Notably, none of the studies relied on a biological framework consistent with that recommended by the Independent Scientific Group (ISG) in 1996. Until there is some degree of consensus on these issues, it will not be possible to develop generally acceptable measures of cost-effectiveness.
2. There is variation among the studies regarding the estimated costs of specific recovery measures. A good example of this can be drawn from the EDF and Harza studies of changes in system operations, specifically a seasonal drawdown of the Snake River system. [ This is Option 5B in the SOR.] EDF concluded that the cost of a seasonal drawdown would be about $325 million in present value over the life of the project($1994). [ Willey, Z. and A. Diamant (1995), Table A-4.] Although strict comparisons are difficult, Harza concluded that the same option would cost about $130 million annually ($1996). [ Harza Northwest (1996), Table 8-5, Four Pool Drawdown.] There are many reasons for this difference, including (a) the date chosen for the "constant dollar" base, (b) the "base case" against which the specific alternative is compared, (c) the discount rate used for reducing future costs to the present, and (d) the specific assumptions for a variety of economic cost levels, for example, whether replacement power is obtained from market sources or a specific generation resource. Adjustments would be possible to make the studies more comparable. Generally speaking, economists agree on appropriate techniques for estimating such costs, although there are exceptions (for example, treatment of subsidies and existence values). Even so, assumptions are always required when such estimates are made. Standardized assumptions could remove or reduce major sources of variation among studies of this type.
3. In addition to differences in estimated costs, there is variation across studies regarding the ranking, or relative cost-effectiveness, of similar measures. For example, EDF finds flow augmentation up to one million acre-feet (MAF) from the upper Snake River to be more cost-effective than a four-dam, natural river drawdown on the Snake, when the biological objective is reductions in water particle travel time (WPTT). [ The EDF finding assumes replacement power costs using purchases from the power grid, a 4.5 month flow target, and no "WPTT decay rate" (see Table A-2, page A-3).] On the other hand, Harza finds the four-pool, natural river drawdown to be more cost-effective than one MAF of flow augmentation, when the biological objective is measured per one-percent gain in "salmon survival". [ The Harza finding is from Figure 1-6, page 1-8.] Thus, existing conclusions regarding cost-effectiveness appear to be sensitive to the choice of a measure of biological effectiveness.
4. Based on these studies, some changes in river operations are generally less cost-effective than other activities, such as transportation of fish, changes in passage efficiency, reducing the consumptive use of water for irrigation, and even returns to river operations in effect years before the study was conducted. For example, EDF found that changes in land use by irrigated agriculture would probably be much more cost-effective than changes in river operations, such as drawdowns. [ See Table A-4, for example, which compares 14 measures, including 12 addressing changes in operations.] Harza found that transportation is more cost-effective than a variety of changes in hydro operations and structures, although Harza recognized the significant degree of uncertainty about the effectiveness of transportation. [ See the results summarized in section 1 of the Harza study.]
Recommendations of the IEAB
1. The Council should give high priority to obtaining more reliable measures of achieving biological objectives associated with alternative management methods. Biological goals should have some reasonable prospect of being turned into indexes suitable for cost-effectiveness analysis. Without such indexes, cost-effectiveness analysis cannot be used (a) to inform decisions regarding the allocation of limited budgets, (b) to improve accountability for the use of scarce public funds, or (c) to improve decision-making over time through adaptive management that seeks the best use of scarce resources. Initially biological scientists could be asked to rank measures based on their evaluation of the best information available. Although this will not be adequate for cost-effectiveness analysis, it would be an improvement over current conditions. The Council should encourage the IEAB to work cooperatively with the biological scientists on the ISG and the Independent Scientific Advisory Board to develop biological information amenable to economic analysis.
2. The Council should encourage use of a "base case" tied to current operations (e.g., 1997 conditions projected over time) as a reference point for analytical purposes, both economic and biological. The Council should also encourage and, where possible, require biological and cost-effectiveness analysis of changes in both directions from the base case in all activities (for example, more and less firm power production, more or less spill, more or less transport, larger and smaller harvests, and more and fewer hatchery-produced fish). This would yield more comprehensive analysis than is available from the studies reviewed here.
3. Reliable cost-effectiveness analysis is very difficult to do well. The variations in results across existing studies support the need for a consistent and unbiased examination of both projected and actual costs. The Council should authorize the IEAB to establish guidelines that will encourage greater comparability among future economic studies. This will include establishing principles and procedures for dealing with such subjects as discount rates, existence values, and subsidies. For example, based on the SOR analysis (see paragraph 9 below), "value in use" (harvest and recreation) is not sufficient to justify existing expenditures on fish mitigation. However, there is another possible economic justification for mitigation and recovery actions, called "existence value", which refers to the worth to an individual of knowing that the fish (or some other characteristics of the environment) exist, even if the individual does not derive value from commercial harvest or recreational consumption. Estimating existence values is, however, a very difficult enterprise.
4. The wide range in estimated costs of measures when "normalized" to a particular biological benefit, combined with the likelihood of a budget constraint for the foreseeable future, implies that economic analysis will become more important over time, and that the Council should give serious consideration to the costs of all proposed measures.
Major Findings of Existing Studies
1. In some cases, a single measure of the projected biological benefit can be achieved by measures with widely divergent cost estimates. For example, EDF estimates that water particle travel time can be reduced by 26 hours either by saving about $14 million (net present value) or by spending $196 million, compared with 1992 actual operations. [ See Table A-4. The estimated cost of acquiring 500,000 acres of irrigated land and associated water rights in perpetuity would be $14 million lower than the cost of 1992 actual operations, and the estimated cost of constructing Rosevear Gulch Dam on a tributary of the Snake River would be $196 million more than the cost of 1992 operations. Both yield identical reductions in water particle travel time.] That is, opportunities exist to make economic decisions that will likely save consumers and producers unnecessary costs. This information should be taken into account when redesigning the Fish and Wildlife Program or making specific recommendations to BPA (or Congress) for funding. Graphs are attached to this summary that demonstrate in a simple manner the extremely wide range of estimated costs for a particular measure of biological effectiveness.
2. Various measures of "biological effectiveness" have been used in economic studies, including water particle travel time (EDF), the value of commercial and recreational harvest (SOR), the number of returning adults (Harza), fish-passage efficiency (Harza), simulations of downstream mortality (RFF), harvest goals (RFF), escapement goals (RFF), and increases in the likelihood of juvenile survival (Harza).
3. Economists often rely on assumptions when making estimates and projections. A good example of this is the cost of replacement power for hydroelectric output that could be lost due to changes in operations to protect fish. Two generally used approaches (purchases of power from a market and construction of new sources of generation in the Northwest) can yield significantly different results (EDF).
4. The attached graphs illustrate that the cost-effectiveness results are sensitive to the measure of biological effectiveness that is chosen. For example, Minimum Operating Pool (MOP) appears to be relatively cost-effective in the EDF study when implemented at John Day and the biological goal is reductions in water particle travel time (WPTT), but relatively cost-ineffective in the Harza study when implemented at all Snake River projects and the biological goal is an overall contribution to survival. This points to the need for a comprehensive study that takes an agreed-upon set of measures that are under consideration, and applies economic analysis consistently across all measures and affected economic sectors.
5. Different observers choose different "base cases" given interpretations of legal requirements, political expectations, or other limitations. For example, Harza calculated costs compared with the 1995 Biological Opinion, EDF and the SOR compared costs with 1992 actual operations (as captured in SOR Option SOS2C), Huppert and Fluharty chose the 1993 operating condition in the river, RFF compared proposals made in the late 1980s, and ECONorthwest argued, in part, from rights established by Treaty in 1855. The estimated economic cost is then a function of comparisons with the preferred "base case", and it is not surprising that estimates vary widely because, in part, the "base cases" are not the same. A complete economic analysis of incremental changes should examine the cost-effectiveness of both doing more and doing less of each type of activity. This will become increasingly important if the budget constraint continues to tighten.
6. In some areas, we still face potentially significant uncertainty in the estimates of economic effects. For example, the economic cost of future changes in power production will depend critically on the evolution of markets in that sector, which cannot be reliably predicted at this point. EDF recognized this uncertainty by estimating costs assuming replacement both by the construction of new combustion turbines (CTs) and by greater amounts of buying and selling power in the market.
7. The studies optimize and determine cost-effectiveness within different constraints. For example, the EDF study asks what is cost-effective given a subbasin, its ecology, and the goals set by individual planners for that subbasin; this study examines many options within a subbasin, including propagation, habitat, and passage. In contrast, the SOR analysis asks what is cost-effective among alternatives for reconfiguring the operation of the Columbia/Snake River system, but excludes possible changes harvest management or habitat. The result is that the conclusions of the various studies depend critically on the structure of the question initially posed.
8. The market value of greater numbers of fish, when considered alone, cannot justify either existing or proposed levels of expenditure on recovery actions. For example, the SOR calculates cost-benefit ratios using the commercial harvest plus the recreational value of sport fishing as the measure of "benefit". The SOR?s "preferred alternative" returns $36 million of commercial and recreational value at a system cost of $147 million, for a cost-benefit ratio of just over four. [ The commercial and sport values are taken from Table 4-13, page 4-36 of the SOR, calculated as the average of the "low" and "high" values. The system costs are taken from Table 5-1, page 5-3, column 11. The value in column 3 on page 5-3 under the Preferred Alternative appears to be a typographical error, when compared with the estimates in Table 4-13.] Commercial and recreational harvest is an example of a "use value" or "value in use".
Bonneville Power Administration, U.S. Army Corps of Engineers, U.S. Department of the Interior Bureau of Reclamation (1995), "Columbia River System Operation Review, Final Environmental Impact Statement, Appendix O: Economic and Social Impact" (reviewed by Jack Richards).
Diamant, A. and Z. Willey (1995). "Water for Salmon: An Economic Analysis of Salmon Recovery Alternatives in the Lower Snake and Columbia Rivers" (reviewed by Dan Huppert).
Harza Northwest, Inc. (1996). "Salmon Decision Analysis, Lower Snake River Feasibility Study, Final Report" (reviewed by Ken Boire).
Huppert, D.D. et al. (1996). "Economics of Snake River Salmon Recovery: A Report to the National Marine Fisheries Service" (reviewed by Anthony Scott).
Paulsen, C.M., et al. (1993). "Above Bonneville Passage and Propagation Cost Effectiveness Analysis", Resources for the Future for the Bonneville Power Administration (reviewed by Joel Hamilton).
Niemi, E., et al. (1995). "Economic Consequences of Management Strategies for the Columbia and Snake Rivers" (reviewed by Paul Sorenson).
Additional Recommended Studies
Hamilton, J. et al. (1996). "Cost of Using Water from the Snake River Basin To Augment Flows for Endangered Salmon".
Independent Scientific Group (1996), "Return to the River: Restoration of Salmonid Fishes in the Columbia River Ecosystem".
1. Fundamental Questions
2. Study Summaries (prepared by individual members of the IEAB)
3. Tabulation of Comparison of Study Characteristics
4. Graphs of Cost-Effectiveness Results (EDF and Harza)
This attachment lists some questions that the IEAB considers fundamental for an understanding of the comparative value of the existing studies of cost-effectiveness for policy-makers, and provides brief answers. These questions provide more detail than the summary report.
1. What is the scope of the biological analysis (e.g., an examination of alternative measures confined to one river basin, versus a review of transportation techniques on the mainstem river)?
? Existing studies vary widely in the scope of their geographical and biological analysis. Huppert reviewed the costs of a Recovery Plan; the SOR-EIS considered alternative configurations of the Columbia and Snake River hydroelectric system; the RFF study examines alternative actions within specific subbasins. On the other hand, the ISG report (1996) questions many of the assumptions about biological effectiveness, which means that the conclusions of these studies on cost-effectiveness should also be questioned.
2. How comprehensive is the economic analysis (e.g., a study that explores the direct and indirect economic effects, including opportunity costs properly defined, versus a study that only adds up expected project costs)?
? Some studies are extremely comprehensive in their attention to various economic sectors and the opportunity costs of proposed actions; other studies are not sufficiently documented to permit a clear assessment of their value to policy-makers.
3. Was the biological component of the study a simulation exercise or based on empirical data?
? In almost all cases, computer models and assumptions were used to estimate the biological impacts of proposed measures. Little hard empirical data is apparently available regarding the benefits of any particular actions. Measures of "biological benefit" range from a simple calculation of the value of commercial and sport fisheries to an assignment of essentially infinite "existence values" to restored fish populations.
4. What measures of biological effectiveness have been used?
? Water particle travel time, mortality and population estimates, and size of potential commercial harvest have all been used. In addition, assertions of existence values have been made, with and without attempts at quantification.
5. Do the studies implement acceptable economic accounting standards?
? In most cases, economists studying recovery actions or proposals have attempted to estimate the total economic cost (including both direct and opportunity costs) of proposed measures, consistent with generally accepted standards.
6. Was the study aimed at producing an explicit test of cost-effectiveness?
? In some cases, tabulations of the cost-effectiveness of particular measures permit a comparison of the economic costs of achieving a particular biological goal, or of implementing a proposed Recovery Plan. However, some studies do not attempt such an explicit comparison, and any comparison of results must recognize the assumptions required to produce the estimates of cost-effectiveness.
7. How does the study treat fundamental (but sometimes practically difficult) economic concepts, such as opportunity costs, subsidies, uncertainty, and existence value?
? Opportunity costs have been comprehensively (but variously) addressed for all economic sectors potentially affected by recovery measures. Subsidies, uncertainty, and the potential for existence values of the protected species have generally not been addressed in the reviewed studies, and require additional thought.
8. Does any study provide clear guidance for policy-makers? Conversely, are any studies useless to policy-makers and thus to be avoided?
? Studies that advocate for particular measures and attempt to minimize the logical or legal validity of estimates of economic costs should be avoided. Some studies, on the other hand, provide frameworks within which additional analysis of the cost-effectiveness of alternative proposed measures could be performed.
9. What lessons are possible for the Council?s deliberations regarding FY98?
? The existing studies cannot be applied in their current form to assist the Council in making funding decisions for FY98. Given the limited time to develop recommendations for FY98, the IEAB recommends that the Council concentrate this year on (a) the use of competitive bidding to award funds wherever possible, (b) contract reforms that will shift a reasonable level of risk and responsibility to the entities being funded by the region?s ratepayers through BPA and support improved decision-making in the future, (c) direction to the ISG and ISAB to address the fundamental areas of biological uncertainty and bring recommendations back to the Council, (d) shortening and clarifying the list of potential measures of biological effectiveness, and (e) directing and working with the IEAB to establish a framework for cost-effectiveness analysis that can be applied to annual funding decisions in future years as well as more broadly to the entire Fish and Wildlife Program. Items (a) and (b) can be implemented through decisions made this summer, whereas the remainder will require a longer time horizon.
10. Do the studies support or contradict each other, or can they even be compared?
? Based on the existing studies, the estimated economic costs of changes to the region?s hydropower system are clearly higher than the costs affecting any other sector of the regional economy. However, this conclusion rests on the assumptions made in the various studies, which in some cases may prove unreasonable with further examination.
11. How should economic analysis address or manage the need to "bundle" or combine measures that have complex biological interactions?
? Simulation exercises are possible that suggest the biological impacts of combinations of measures. However, these exercises are driven by assumptions that may or may not be valid.
12. What lessons can be learned regarding the sequence of measures?
? It is reasonable to assume that some amount of "learning by doing" will be possible, and that more economical adaptations to constraints on existing activities will be found, the longer the affected parties have to respond. The real challenge for environmental policy-makers will be to encourage (or even require) similar adaptations on the part of those directly managing the environmental resources. One task for the IEAB could be to review the track record of adaptive management, to find ways to encourage more efficient management of limited resources over time.
13. Can or should a cost-benefit analysis be conducted in addition to a cost-effectiveness analysis? If so, where does cost-benefit analysis appear to have a role?
? Cost-benefit analysis is possible where clear dollar values can be attached to the biological results. For example, the SOR-EIS ranks proposed system operations in terms of their value measured by additional commercial and sport fisheries. However, this approach is of limited applicability in the case of endangered species.
14. What areas of economic effect remain subject to considerable uncertainty?
? Given the imminent restructuring of regional power markets, the economic cost of changes to the operation of the hydropower system may change dramatically over time, and may vary more widely from year to year.
15. For a given objective, can we conclude that certain measures are clear not cost-effective?
? Based on existing studies, which are imperfect, two conclusions appear clear. First, drawdowns of Snake River reservoirs are more expensive by far than other passage alternatives. Second, certain proposed levels of flow augmentation are also clearly significantly more expensive than others.
16. Do existing studies provide any guidance if the Council moves to an "ecosystem approach", as proposed by the ISAB?
? Generally speaking, no. The RFF analysis may be an exception, because of its emphasis on habitat. In addition, some parts of the existing studies may provide some useful information in the context of an ecosystem approach, but none addresses the economic effects of such an approach comprehensively.
17. What fundamental economic questions remain unanswered?
? Areas for useful research would include (a) the economic impacts of shifting from existing definitions of property rights in the natural resources to other definitions, (b) the costs and benefits of establishing markets for water rights where none currently exist, and (c) measures of the existence value of the protected species.
Study Title: Salmon Decision Analysis Lower Snake Feasibility Study: Final Report
Authors: Harza Northwest, Inc.
Date: October 1996
Reviewer: Ken Boire
Report Purpose - The purpose of this October 1996 report prepared by HARZA Northwest Inc. for the Walla Walla District Corps of Engineers, was to develop a decision analysis process regarding the feasibility of salmon recovery measures, including reservoir drawdown, on the lower Snake River. The report applies only to Snake River spring/summer chinook. The scope is limited to options for improving the hydropower system, so harvest and habitat are not included in the fifteen major "tools" given consideration. The "tools" include: surface bypasses and collectors, sluices, baffled spillways, extended length screens, spillway improvements, turbine improvements, fish guidance curtains, juvenile bypass improvements, spill, barges and trucks, storage, dam removal, drawdown, minimum operating pools, sound repulsion.
Report Summary - The report looks at three major juvenile migration paths, transportation and in river, and a third that combines elements into a mixed path. The process presented in the report relies on first seeking the best answer in biological terms and then follows this with an evaluation of what the report calls benefits, costs, and risk. The second part of the analysis contains elements of a cost effectiveness analysis. In the first part of the report, evaluation based on biological objectives is given a priority because according to the report options that take the longest time to implement produce "cost savings" over those that can be implemented quickly, and this is said to make no sense if the goal is to reduce extinction. This is a basic position taken in the report, but this reviewer does not agree with the logic.
The report presents a decision tree incorporating arbitrary thresholds to decide which migration path is best. The path selected (transportation or in river, or a mix) is the one which consistently returns 30% more adults than the other. The selected path is given a second threshold test to see if it can return enough salmon to reverse population declines. The threshold is established as a minimum smolt-to-adult return rate (SAR) of 1.5%. So if a path returns 30% more adults than another, and can produce a SAR of 1.5% it is preferred. Other paths leading to drawdown, or a mix of in river and transportation, are identified for SAR thresholds of .0.7%, 0.9%, and 1.2%. Thresholds are arbitrary, and cost is ignored in the decision tree.
After attempting to test the decision tree model, the report moves into a cost effectiveness approach. For the sake of a cost effectiveness example, the report assumes the decision tree has selected a mixed path, and presents "tools" consistent with a mixed path to increase survival and meet the 1.5% SAR. It is proposed the tools would be compared by use of cost effectiveness analysis which inputs goals of 80% Fish Passage Efficiency (FPE), and 95% juvenile survival, at each project. Output is shown as annual cost per each 1% gain in salmon survival provided by each of the tools (shown for 12 of the 15), but the cost effectiveness idea is not developed to the point of supporting a preferred plan. Cost per one percent gain in salmon survival range from a high of $5 million for flow augmentation (1MAF), to a low of $0.1 million for a sound repulsion system.
The report concludes biological information regarding SAR for in river and transport paths, is insufficient to allow use of the decision tree. Decision Analysis is said to provide a useful framework, but complexity of the options and variability of data is not likely to allow Decision Analysis to provide a clear and simple answer as to which path to choose. Cost and incremental survival for various tools is shown, but the report does not use, or recommend, cost effectiveness outside of, or separate from the Decision Analysis framework.
What Kind of Economic Study is Done? - Economic costs associated with the various options are examined in Appendix C of the report. The basis for economic cost is the National Economic Development (NED) approach, and is taken from studies done for the SOR, and Phase 1 System Configuration Study (SCS). The discount rate is 7.75%. Current system operation is the basis from which changes are measured, and is different from the SOR Preferred Alternative in that John Day MOP, and all of the cost associated with John Day MOP, are excluded.
For purposes of demonstrating economic costs of a full transportation strategy, it is assumed the hydro system will be returned to a system operation similar to operation prior to implementation of the 1980 Northwest Power Act.
A cost effectiveness example is given for the transportation sub-path. Survival is related to annual cost, and a least cost frontier is illustrated in Figure 10-3.
In a more complete presentation for all paths, Appendix B, Risk Assessment, presents incremental benefits (% change in juvenile Snake River spring chinook survival) for 29 different measures, actions, or tools. The report shows surface collectors at Lower Granite/Little Goose (LG-LG), w/ transportation, and 90% FGE, is at the top of the list for incremental benefits, see page B-7. Cost data in the report supports that this would be the cost effective choice, but the report does not take the analysis to a conclusion.
What Measures of Effectiveness are Assessed? - Within the Decision Analysis framework, branches in the decision tree are based on SAR, and TIR values. Within the portions of the report which discuss economics and cost effectiveness, percent survival of spring chinook above Lower Granite pool to the estuary, is the measure of effectiveness. Other stocks are ignored. The CRiSP 1.5 Model is used with low and average water conditions so results cannot be viewed as expected annual values. The report does not attempt to measure non-biological effects such as associated economic benefits, or cultural or social effects.
What Areas of Economic Opportunity Cost Are Assessed? - Areas assessed include project implementation, power, reservoir recreation, navigation, and water supply.
? Reservoir Recreation - Reservoir recreation is not in accord with the NED concept as it does not adequately account for alternative recreation opportunities. The result is to show economic cost of "lost" recreation is greater than cost associated with losses in navigation or hydropower. In one case recreation "losses" are shown to be greater than the implementation cost of a natural river.
? Hydropower - This report relies on the analysis in the SOR EIS, and is out of date in that energy and capacity prices have fallen. Power replacement cost is based on gas fired turbines, or Intertie supply. Addition of gas turbines to replace a loss of average hydro energy actually adds some firm capacity to the system and it is not clear how this has been evaluated.
? Implementation Cost - In this report, implementation costs are converted to equivalent annual values, so those furthest off into the future have smaller present values. Ordinarily this effect would be countered by a similar present worth adjustment on the benefit side of a benefit cost analysis but this report is oriented towards a cost effectiveness analysis so benefits are absent. One way to adjust for this would be to express equivalent annual cost effectiveness in terms of cost per fish. Since impacts on fish would also be off in the future, the decision maker would be aware of the cost and the effect.
? Navigation - Alternative cost of transportation is used as a basis for estimating economic cost. Effect of increased transportation cost on quantities shipped does not appear to have been accounted for.
What Major Conclusions Were Reached, If Any? - The report concludes, at present, information regarding SAR and TIR is insufficient to allow use of the Decision Analysis model. Complexity of options, combined with variability of biological data, suggests Decision Analysis is not likely to provide a clear and simple answer as to which path to choose
What are the Most Relevant Policy Aspects of the Study, and What are the Principal Policy Dimensions? - As a conclusion, Decision Path analysis is not supported as an analytic solution fitting the requirements for application of cost effectiveness criteria. Parts of the report, however, highlight presentation of incremental effects of various measures, and illustrate a possible basis for developing a cost effectiveness analysis based on cost per % survival, or cost per fish. The cost effectiveness procedure in the report is not fully developed, as measures are presented as bundles not having undergone incremental evaluation themselves, and costs have not been assigned though they appear to be available within the report.
Though not stated in the report, a principal policy implication is that Decision Path Analysis should not be pursued, and attention should be given to procedures which match identifiable, separable measures with specific incremental effects.
What were the Important Cost and Effectiveness Estimates Produced by the Study?
Application of the CRiSP 1.5 Model resulted in estimation of incremental effects on spring chinook, for 29 different measures (Table B-2, page B-7). Application of CRiSP demonstrates effects for low and average water conditions instead of annualized flows, so results could change with a more complete application.
Did The Study Have Any Appreciable Impact on Policy Decisions? - Failure of the Decision Analysis Model to formulate a salmon recovery plan emphasized weakness of the model where cost effectiveness is a requirement.
Reviewer?s Overall Impression - After reviewing the report, my thoughts focus on the following areas:
? According to the report, in river juvenile survival is now at 43%, and the maximum potential is 56%. Given this, in river strategies may not be able to match effectiveness offered by other tools, at any cost.
? Figures 11-1, 11-2, and 11-3 indicate that transported fish do better.
? Table B-2 shows insight into relative merits of some paths. Surface collectors at LG - LG w / transport, provides highest incremental survival.
? Figures 1-6 and 7-6 show annual cost per 1% gain in survival. In river strategies are more costly.
? Cost effectiveness measures need to recognize stocks other than the target species will benefit. Methodology should allow for crediting each measure for benefiting other stocks. A credit based on alternative cost may be appropriate as incremental cost data will be available.
? Bundling of measures should be avoided so as to allow each separable measure to be evaluated on it?s own merits.
? Evaluation of separable measures should identify those which can be considered first added, and those which depend on other measures being added before them.
Study Title: Above Bonneville Passage & Propagation Cost Effectiveness Analysis
Authors: Resources for the Future
Date: January 1993
Reviewer: Joel Hamilton
Geographical Applicability: The study addressed 26 sub-basins of the Columbia and Snake systems above Bonneville Dam.
Problem Analyzed: The study was an attempt to demonstrate the feasibility of using cost/ effectiveness analysis to study the propagation and passage alternatives for meeting salmon and steelhead harvest and escapement goals set by fisheries planners for the 26 sub-basins.
Principal Techniques: The study methodology links 4 models. The Systems Analysis Model (SAM) is used to show storage, flow and hydropower responses to basin water conditions. The Columbia River Salmon Passage Model (CRiSP)is used to model downstream migration mortality, given the flows from the SAM model, for all possible combinations of propagation and passage alternatives. A Deterministic Life Cycle Model (DLCM) is used to account for numbers of fish of each stock, age, and subbasin, and to test whether these numbers meet sub-basin harvest and escapement goals. Finally a System-Wide Optimization Model was used to find the least cost set of recovery actions from among the set of actions that meet the subbasin goals.
Summary of Results: The study concluded that the levels of harvest and escapement called for in many of the subbasin plans could not be achieved with the passage and propagation alternatives considered, so the report concluded that the planners had set unrealistically high goals.
They conclude that the "current" mainstem flows representative of conditions in the mid- 1970s to the early 1980s, are generally more cost-effective than the augmented flows proposed in 1991 by the Northwest Power Planning Council under the Phase II amendments to the Fish and Wildlife Program. They note that this rests not only on the estimates of hydropower costs, but also on the assumptions about biological effectiveness dam fish bypasses, and the fish transportation program. Hence they "argue forcefully for a comprehensive biological research program on the relationship between flows and downstream survival".
It is useful to remember that this study was the earliest of the six reviewed by the IEAB, so it was breaking new ground. To produce a tractable model, RFF drastically reduced the number of alternative actions considered, and the number of alternative levels of these actions. Fully implementing model that addresses the range of alternatives currently under discussion, the interactions among these alternatives, and the uncertainty about the biological effectiveness of some of these alternatives, would be an intimidating task.
Study Title: Economics of Snake River Salmon Recovery: A Report to the National Marine Fisheries Service
Authors: Daniel D. Huppert and David L. Fluharty with Eric E. Doyle and Amjoun Benyounes
Date: October 1996
Reviewer: Anthony Scott
Brief summary: This study estimates the costs and other economic consequences of an NMFS team?s proposals for recovery of three endangered Snake-River salmon species. The team's proposals had addressed each stage in Snake River salmon's life history: spawning, rearing, downstream migration, ocean maturation, fishing, and migration upstream. Of these, this study finds the costs of life history: spawning, rearing, downstream migration, ocean maturation, fishing, and migration upstream. Of these, this study finds the costs of improvements in : upstream habitats, migration down through reservoirs, dams and mainstream Columbia; fishing regulation; and migration up. The costs are those of reduced hydropower (around $200 m. yearly) and reduced earnings in farming, forestry, fishing plus four other categories( adding up to around $ 100 m. yearly.) Details are provided for each. Hydropower costs of alternative reservoir-level and drawdown plans are compared. Otherwise no alternatives are examined. Emphasis is laid on future cost reductions as managers and operators gain experience. Also includes estimates of benefits from salmon recovery.
Type of study: Several recent agency Snake studies proposed sets of measures to ensure salmon survival. Taking the agencies' measures, the present authors do their best to identify a single program, for economic study. They look for "incremental" costs: those beyond the expense of existing, present, measures. Their answer: in the neighborhood of $250 million per year.
What measures of effectiveness are assessed? The study outlines, at several places, three different ways of using measures of cost and benefit of salmon-recovery measures:
a. Safe Minimum Standard (Bishop) : Study the costs or net costs of preservation action and let society decide whether preservation is worth this cost.
b. Cost effectiveness of outlays in order to guide design of mix of recovery measures.
c. Benefit-cost analysis.
Discussion: Centering on p. 2-16, the authors consider which of these ways of using estimates of cost and benefit they are working toward. All three face two overwhelming difficulties: the physical effectiveness of a measure is very uncertain, and the assessment of effectiveness as dollar benefits is even more uncertain. The authors conclude that only the first use, to try measuring the net costs of a given vector or mix of recovery measures is feasible. That is what their report does.
What is the scope of the fish recovery issues addressed? They use as their given mix of fish recovery measures all seven elements in the March 1995 NMFS Proposed Recovery Plan for three Snake River salmon stocks. These address improvement of each stage from spawning to up-migration.
a. Snake tributary habitat improvement (federal plus private lands.)
b. Issues: choice of baseline for increment of improvement; selection of projects: all from US Geological Survey & EPA models of river reaches --- no field studies performed.
c. Mainstream modifications, primarily flow augmentation selected by 1995 Biological Opinion.
d. Issues: choice of baseline conditions; choice among many flow-augmentation scenarios; choice among structure changes of fish passage facilities.
e. Harvest measures, primarily buy-back.
f. Issues: buy-back not really a fish-recovery measure.
h. Issues: effectiveness and genetic side-effects: "serious questions about the net costs of artificial propagation can be raised."
i. Enforcement (primarily education and patrolling.)
j. Institutional (primarily for coordinated analysis and modeling.)
k. Research and data ( primarily for flow-augmentation efforts.)
What areas of opportunity cost are assessed?
a. Tributary habitat: Mitigation and lost output of grazing, forest practices and recreation. (p. 3-4)
b. Flow augmentation: mainly lost hydropower ( purchased energy) and
c. higher flood damage; also lost recreation (might be ameliorated), irrigation
d. and water-supply benefits.
e. Harvest measures: nominally lost catches, but actually mostly buy-back, a
f. redistributive expense.
g. Hatcheries: artificial propagation may eventually reduce value of runs;
h. otherwise no significant opportunity cost over baseline.
i. enforcement: no opportunity cost.
j. institutional: no opportunity cost.
k. research and data: no opportunity costs.
Summary: Most opportunity costs arise from (a) tributary habitat and (b) mainstream flow augmentation measures. (d) Hatcheries may also cause long-run loss of fishery values. There are other, cash, costs for additional personnel or for construction.
? Much of total cost is hydropower opportunity cost.
? Of this total cost, only about $100 million is for habitat and other measures.
? Of this, about $30 million is for buy-back (not a restoration measure.)
? These are all annual. The hydropower opportunity costs would be permanent;
? some of the others would fall with research and mitigation.
? Wide range of total cost of all seven measures, from $ 250 to $ 390 million per year. The wideness results from unknown future decision to select either a blend of "flow augmentation" measures or a set of reservoir level draw-downs.
Some strengths and weaknesses:
? Strength: the authors investigate many alternatives yet keep exposition short and clear.
? Attention to trend: authors predict that if the whole package goes ahead, many costs will decline with experience ( trial and error in hydropower management), and because they will not recur after initial installation.
? Attention to uncertainty: frequently, authors usefully mention that the total costs being estimated are for a package of uncertain content and unknown effectiveness. This is a different problem from mere lack of cost data.
? Omission: although authors do not make clear what decision procedure will actually be followed in choosing between alternative flow regimes, it may be suggested that their study might have had influence if it had emphasized economic values to be kept in mind when sub-optimizing large "survival" decision.
? Learning by doing and sensitivity information: the authors might have laid greater on imperative need for information on effectiveness. Not for cost-cutting but for more effective survival program (Holling.) Authors' data suggests the large research and modeling budget is mostly to be aimed at cutting the hydropower opportunity costs of existing program. I got the impression that the Corps and Bonneville intend to model power effects of hypothetical alternatives. It seems the program needs more observation of fish-survival effects of experimental alterations of water-level and drawdown regimes.
Principle policy dimensions?
It seems that the policy being followed did not really require this study. If it had been a condition that the cost measurement be confined to actual "cash" outlays that NMFS, other agencies or the government would have to pay for, then the study could be used for auditing: to make sure that the intended program of measures kept within the funds available. But as it was to include opportunity costs, many uncompensated and impacting well beyond government, the study could not be used for this auditing purpose. (I suppose it might be used to publicize the extent of the campaign to save the fish. ) As the authors suggest, their study's cost items are well-defined to play a role in a possible cost-effectiveness study of the mix of measures. But cost-effectiveness has not guided agency spending so far. Consequently refinement of and experience with the survival program has not produced data on the beneficial effects of certain groups, vectors or mixes of costly measures. Clearly, for this economic study to have had a "policy dimension" would first have required that biologists and engineering experts have better-specified policy goals for their day-to-day river management and for their studies. With these, large-scale biological trial-and-error or sensitivity research could be productive of data that would contribute to cost-effectiveness calculations.
Study Title: Water for Salmon: An Economic Analysis of Salmon Recovery Alternatives in the Lower Snake and Columbia Rivers.
Authors: Environmental Defense Fund (Adam Diamant and Zach Willey)
Date: April 1995
Reviewer: Daniel Huppert
Nature of the Report This report is the product of a contract between the Environmental Defense Fund and the Northwest Power Planning Council. It analyzes the cost-effectiveness of measures to aid Snake river salmon recovery based on data and information contained in existing reports. This report briefly summarizes the types of salmon recovery measures being considered; it outlines the analytic methods used to evaluate the cost-effectiveness of each measure; it explains the EDF analysis of reservoir drawdowns on the Snake River, the analysis of proposed new storage dams in the upper Snake River basin, and the analysis of several water marketing and flow augmentation programs; it provides a concise summary of results; and it concludes with some recommendations to the NPPC. The report is written in a concise, clear manner and is well laid out with an Executive Summary.
Type of Study - This is a cost-effectiveness analysis of twenty-six (26) alternative Columbia river hydroelectric power system project configurations which may promote the recovery of endangered Snake River salmon populations. The analysis focuses on 15 reservoir drawdown configurations, 6 new water storage projects in the Snake River Basin, and 5 "non-structural" water management measures. All of these options are drawn from existing project proposals or analyses, and the report is based largely upon secondary information from (1) the System Operation Review (SOR) of the Columbia River Hydroelectric Power System conducted by the U.S. Army Corps of Engineers, Bonneville Power Administration, and U.S. Bureau of Reclamation, (2) the System Configuration Study (SCS) conducted by the Corps of Engineers, (3) a report by Bookman-Edmonston Engineering prepared for BPA in 1994, and (4) a US Bureau of Reclamation/Army Corps of Engineers report on Snake River Basin Storage projects. Bringing some cohesion to the analytical approach is the authors? own detailed assumptions concerning the distribution of costs over project lives and use of a spreadsheet model of regional hydropower production and delivery (obtained from BPA) to evaluate hydropower costs and salmon enhancement benefits for every project alternative. The authors claim their analysis to be superior to those of the SCS and Storage project reports because they use 50-year hydrologic model consistently to evaluate both costs and benefits.
Measure of Project Effectiveness - The reduced water particle travel time from Lower Granite dam reservoir to below Bonneville dam (reckoned in hours) is selected as a measure of effectiveness. For any given draw down, water storage, or flow augmentation project, the expected flow of water through the system is modeled through a 50-year hydrological record, and the change in water particle travel time is calculated using equations explained in the text. The authors calculate a present value of reduced water particle travel time over each project?s lifetime, calling this a measure of "benefits", and using alternatively a 0% or 10% as a "decay rate" of benefits with time. On p. 38 it is declared that this "decay rate is biological in nature" and is not equivalent to an economic discount rate. It is used to reflect the urgency with which salmon recovery measures are needed. [Maybe this could be interpreted as a reflection of biologists? impatience with delay in recovery planning.] Using the decay rate as one would an interest rate, the authors calculate an annualized, or "levelized", travel time change.
The Economic Analysis is limited to a thorough examination of direct project costs and opportunity costs associated with each project configuration. The cost estimates are developed for the following categories: (a) construction costs associated with reservoir draw downs, new storage projects, and irrigation system changes; (b) annual operation, maintenance, and pump power costs; (c) interest during construction; (d) hydroelectric power impacts (negative and positive) associated with draw downs, flow augmentation, and shifting of water releases to assist salmon downstream migration; (e) municipal and industrial water supply impacts; (f) decreased farm income associated with land fallowing and/or dry year leasing of water rights which reduces consumptive us of water in the upper Snake River basin, permitting release of water to increase flow; (g) flood damages; (h) impacts to shallow draft navigation (barges); (i) recreation impacts; and (j) costs of trucking logs around Dworshak reservoir. The categories and methods of cost assessment are similar to and often drawn from the SOR, SCS and related reports. The authors apparently improve upon the previous cost and effectiveness analyses for Snake River storage projects by modeling the effects over the 50-year hydrological record rather than evaluating effects for an "average year".
Much of the cost analysis follows methods already understood and well documented in the SOR and SCS analyses. Two exceptions are worth noting -- the hydropower costs of draw downs and the net farm income loss associated with flow augmentation at Lower Granite dam. The hydropower costs are based upon replacement cost estimates used in the draft SOR analysis (e.g. 37mills/kWh for gas combustion turbines), which are significantly higher than cost estimates in the final SOR analysis (which uses a cost of 26.3 mills/kWh for generation by combustion turbines). The EDF report was finalized before the SOR was completed. Hence, the amended estimates were not available to EDF.
The evaluation of land fallowing and dry year leasing apparently includes new analysis, although it derives from Huffaker, Whittlesey, and Wandshneider?s "Institutional Feasibility of Contingent Water Marketing to Increase Migratory Flows for Salmon on the Upper Snake River (Natural Resources Journal, Vol. 33, Summer 1993). For the land fallowing option, EDF assumes that 500,000 acres of land are purchased and taken out of production in order to release 1 MAF of water every year. In the dry year leasing option EDF calculates the deficit of flow at Lower Granite Dam relative to flow targets set by NMFS. They assume that sufficient water rights are leased to meet the flow target ? with the proviso that no more than 1 MAF is leased in any given year. EDF models the water augmentation needs for each year of the hydrologic record, then determines how much water is needed during either a 2-month spring period or the 4.5 month April 16 - August 31 period. Then the water is "dispatched" during the 2-month or 4.5-month period to best meet flow objectives, the consequences for water particle travel time is calculated, and the hydropower production and values associated with this water are estimated. Using a 0% decay rate for TT benefits, for the 2-month flow augmentation period, EDF estimates that the net cost (farm income loss minus hydropower value gain) would be $212,000 per hour of travel time increase. For the 4.5 month period, EDF estimates the cost as -$44,000/hour (i.e., a net economic gain).
Cost-effectiveness is defined as a project?s costs divided by its benefits. Costs are measured as annualized net present value of the stream of costs associated with a project. Benefits are defined as the levelized present value of the number of hours of reduced water particle travel time over the project?s life. The authors apply a discount rate to water particle travel time reductions which they term a "decay rate". The decay is supposed to reflect the relative value of delayed improvements in smolt survival during downstream migration. The authors recognize that a ratio of project cost to reduced water particle travel time is an incomplete decision criterion ? especially when a project with a high cost-effectiveness ratio has a small overall impact on the smolt travel time. To more fully inform the decision makers, the authors provide actual hours of reduced travel time, economic costs, and cost-effectiveness ratios for each project.
Base Case All costs and benefits are measured relative to the System Operation Review option SOS2C which reflects the operation of the Federal Columbia River Power System with interim flow improvements adopted in response to the listing of Snake River salmon under the Endangered Species Act.
Sensitivity analysis covers four separate variables: (1) two different river flow augmentation period or drawdown periods are analyzed -- April 16 - June 15 and April 16 - August 31, (2) two real discount rates are used in calculating present values and annualized equivalent costs ? 3% and 8%, (3) the "replacement cost" of electric power is based upon either construction of combined cycle gas combustion turbines (37 mil/kwh) or purchases on the power grid, and (4) two travel time "decay" rates are used -- 0% and 10%. The first three of these are common to other regional economic analyses associated with the Columbia River basin. The last is apparently unique to the EDF analysis.
The major conclusions are displayed as histograms and tables showing the costs, cost-effectiveness, and travel time reductions for each recovery measure. These are summarized below, choosing the results calculated for an interest rate of 8%, a travel time decay rate of 0%, and using power sales to estimate hydropower costs/benefits (rather than using combustion turbines to make up for hydropower losses).
Summary of EDF Cost-Effectiveness Analysis. From Tables A-1 and A-2.
||Two Month Flow Period
||4-1/2 Month Flow Period
||TT decrease (hours)
||TT decrease (hours)
||Dry year leasing (up to 1 MAF)
||Basinwide land fallowing (500k acres)
||33? Drawdown of 4 Snake R dams (using MacClean report cost estimate)
||Galloway Dam storage project
||John Day dam operated at MOP
||33? Snake R 4 dam drawdown (ACE cost estimates)
||43? Draw downs
||33? Draw down, Lower Granite dam (MacClean cost estimate)
||52? Draw downs, 4 dams.
||Natural River draw down, 4 dams
Strengths and Weaknesses
The report has several strengths. It evaluates a wide variety of hydroelectric system projects consistently using a common method tied to the 50-year hydrologic record. While most other reports on Columbia River salmon recovery also adhere to this approach, not all extend the 50-year simulation analysis to the upper basin storage projects. Further, the report leads decision makers to think hard about selection of salmon recovery projects based upon both the project cost and the project?s contribution to some measurable aspect of salmon survival in the river.
Complex relationships between hydrologic variables, such as water particle travel time, and biological variables, such as migration rate and survival rate for salmon smolts, has lead other scientific teams to caution that this measure may not reflect accurately increased salmon smolt survival. For example, smolts are not simply carried down stream like neutrally buoyant particles at average rates of flow. Smolts behave in ways that, to some extent, determine their rate of travel. Second, specific mortality factors such as (a) nitrogen supersaturation at dam sites, (b) concentration of predators in dam forebays and tailwaters, and (c) the effects of passage through dam turbines, smolt passage facilities, and/or Corps of Engineers barges are largely unaffected by travel time per se. Nevertheless, the search for a single "biological benefit" measure is a difficult one which the authors have resolved in a manner consistent with the thinking of many regional fish and wildlife agency biologists. Still, it is not clear that the magnitude of travel time improvements expected with the projects examined will be a significant factor in the recovery of Snake River salmon. A decision analysis must have much better guidance than this.
I found some aspects of the land fallowing cost analysis to be puzzling. It is unclear, for example, why the cost-effectiveness of fallowing 500 thousand acres shifts from $212,000/hour to -$44,000/hour when the flow augmentation season is extended from 2 to 4.5 months. Since the amount of land fallowed is unchanged, the expected costs to the farm sector should be independent of the flow augmentation period length. Hence, the decreased cost per hour must come from increased travel time reduction when the flow augmentation period is extended or from an increase in the value of hydropower produced by the additional water. Neither of these is explained in detail within the report. One would expect less flow rate increase for a given water volume when the flow season is extended. Hence, the travel time should drop less on average when the flow season is extended to 4.5 months. If power were more valuable in the summer, when sold to California during a higher power demand season, that could decrease the cost per hour of travel time reduction. But, the model explanation in Chap. IV suggests that all power sales are valued initially at 21 mils/kWh with price dropping for larger sales.
Overall, the report covers most of the important hydropower system changes being proposed for Snake River salmon recovery. Like other reports prepared for agencies in the region (including my own) it does not provide decision makers with a realistic notion of the magnitude of salmon recovery realized per million dollars committed to recovery. Also, like other reports, it does not develop a vision of how salmon-related actions affect the economy of the Columbia basin, nor of the relative socio-economic importance of the fragments of salmon populations that could be affected by these measures. It is perhaps a result of the manner in which the debate over salmon recovery has been engaged ? one project at a time to address one specific salmon population decline.
We are lacking an overview of the regional economic importance of the resources and structures linked to salmon survival, and we are consequently unable to address the core issue of what makes a reasonable trade-off between salmon preservation and economic development. This issue may be imbedded in the Northwest Power Planning Council?s need to give fish and wildlife equal consideration to that of power. What constitutes an "equal" consideration? This issue is clearly connected to the Independent Scientific Group?s declaration that ecosystem policy decisions "include trade-offs between salmon and important regional social and economic factors". (Return to the River, p. 7) Rather than engaging in more cost-effectiveness analysis that uses specific, but necessarily narrow and partial measures of salmon enhancement in comparison to estimated costs of specific projects, we might consider developing a document that reviews the conceptual foundations of regional tradeoffs between salmon enhancement and other economic uses. This could lead both the NPPC and other analysts to ask the more important questions.
Study Title: Economic Consequences of Management Strategies for the Columbia and Snake Rivers
Authors: ECO Northwest (Ernie Niemi, Ed MacMullan and Ed Whitelaw)
Date: July 1995
Reviewer: Paul Sorensen
Report Purpose - The Confederated Tribes of the Umatilla Indian Reservation (CTUIR) is proposing a strategy for managing the rivers that "restores tribal member?s access to resources guaranteed by the treaty (of 1855)". These proposals include two phases:
1. Phase 1: Immediate drawdown of John Day reservoir and three reservoirs on the Snake River to minimum operation level and Lower Granite to 710 feet,
2. Phase 2: By 2000, drawdown of Snake River reservoirs to natural river elevations and by 2005 drawdown of John Day reservoir to natural river.
The Economic Consequences of Management Strategies for the Columbia and Snake Rivers was prepared by ECONorthwest for CTUIR in July 1995 to discuss the proposed strategy. The report advances two separate but interconnected analyses:
1. Critique of the methodology employed in the SOR DEIS
2. Estimate of the economic consequences of the CTUIR?s water allocation proposal
The two analyses are reviewed below.
1. Critique of SOR DEIS Methodology - The framework for the critique consists of stages of impact, including:
? Stage 1: Society acts to devote additional water to anadromous fish and restricts activities that would degrade fish habitat.
? Stage 2: the change in policy sends signals to four groups (group 1 are industries that benefit from water restrictions; group 2 are industries that incur costs when water is restricted; group 3 includes those who see fish/habitat as elements in the quality of life; group 4 are those who place intrinsic value on fish/habitat).
? Stage 3: The economy responds to the signals.
? Stage 4: The economy reaches its long-run equilibrium
The authors suggest that there are several important analytical issues in assessing the economic impacts of policy alternatives and that these issues are not dealt with properly in the SOR DEIS:
1. Weigh each alternative from a With-Without Perspective - critiques the SOR DEIS procedures as Before-After and promoting "a seriously biased defense of the status quo" (Page 7).
2. Incorporate the With-versus-Without perspective into the definition of any base case - the authors would establish a base case that either meets the minimum legal requirements (which are not defined in the report) or (alternatively and preferably) set the base case for conditions that preclude violation of the 1980 Northwest Power Planning Act. The authors argue that any other base case automatically assumes that existing activities (called illegal activities that harm fish and degrade habitat) with respect to fish/habitat are legal.
3. Specify property rights and responsibilities clearly - the authors postulate that the value that is set on the resource "may" depend on who is assigned ownership rights. "Conventional analyses of the past have tended to ignore the possibility that the value of a resource depends on who owns the resource. There remains much research to be done to clarify the relationship between ownership and value, but, until it is completed, analysts, at a minimum, should acknowledge the possibility that conventional analyses may underestimate the values of endangered species" Page 11 The authors opine that Native Americans may have superior rights and that irrigators may have inferior (or secondary) property rights. [The assertion that factors such as ownership et al. should be considered even though there is no meaningful way to measure relative importance is a frequent but insupportable refrain in this report].
4. Examine all the economic impacts, including impacts related to the region?s quality of life - the authors argue that PNW area residents receive a "second" paycheck from the quality of life in the area and that fish survival in the Columbia/Snake River is linked to this paycheck. They critique BPA?s focus on non-resident anglers in recreation benefit analysis done for the SOR DEIS. They postulate that some people will relocate to the PNW if the impacted fish populations are strong.
5. Provide a full accounting of all major subsidies and externalities that distort prices - the authors assert that BPA customers do not pay the explicit cost of the dams (subsidized debt) nor the implicit cost (externalities and loss of fish). In addition, the authors argue that agriculture and forestry also receive substantial subsidies. They state that these subsidies should be treated in the SOR DEIS analysis and that BPA wrongly excludes them. Without treatment of subsidies, the report is "disingenuous, seriously misleading and grossly biased" (page 17)
6. Define geographical areas consistent with the theory and reality of the linkages among metropolitan and nonmetropolitan areas - the authors dislike the SOR DEIS definition of subregions and declare that rural subregions should be linked with the nearest metropolitan region due to economic integration.
7. Demonstrate the dynamic character of the economy - the authors state that SOR DEIS suffers from "dumb-person bias" because the analysis does not allow people to change after imposition of changes in policies and economics. In particular, the authors state that the I-O model used in the SOR DEIS "yields a worst case, static estimate of the adverse, initial impacts of each alternative on the regional economy" (page 22). Instead, the SOR DEIS should include a dynamic view of the economy. In latter sections of the report, the authors argue that the "water-intensive" industries are unimportant to the PNW economy.
8. Address the principles of sustainability - the authors stress that the concerns of future generations should be included. This implies that the assessment of substitutes should be undertaken for current users of the Rivers while taking into account that the Native American religious value of the salmon has no substitute. They deride the SOR DEIS for not taking sustainability into account.
9. Do not assume that quantity means quality - the SOR DEIS focuses too heavily on the resource-extraction industries and too little on tourism. They argue that economists rely too heavily on data and that impacts which are not easily or cannot be modeled get short shrift.
ECONorthwest argues that the water-management agencies should explicitly discuss these issues and "compensate for these tendencies that may bias the evaluation of the alternatives" (Page 25)
Discussion of Water-intensive industries - ECONorthwest defines water-intensive industries to include firms that are electricity-intensive (DSIs, pulp and paper, irrigated agriculture, primary metals, chemical industries, lumber and wood processing). The authors state that these industries are subsidized, account for a small portion of the PNW economy and will continue to decline as a share of employment and income despite management policies of the Columbia River. Due to a host of factors (e.g., labor-saving efforts, international competition and soil erosion) these industries are assessed to have questionable future sustainability.
This summary analysis fits into the framework for a dynamic model and sets the stage for ECONorthwest?s statement that these firms will cease to exist or are utterly unimportant in either the existing or future PNW economy. In addition, these industries are said to be heavily subsidized and generate substantial externalities. This is ECONorthwest?s argument which they quantify with a disclaimer that the numbers "are illustrations, not precise, comprehensive estimates" (page 42). The subsidies are, however, quantified to illustrate the magnitude of the damage:
1. Depletion of the Columbia - $150-300 million per year
2. Subsidized electricity - $46 million/year
3. Sediment-related damage - $143 million/year
4. Unauthorized irrigation - $35 million/year
5. Subsidized irrigation facilities - $592 million/year.
6. Transportation subsidies - undetermined
7. Government payments - $336 million in 1994
There is no attempt to verify any of the numbers used in this analysis.
2. Economic Consequences of CTUIR?s proposal
The impact on agriculture is evaluated within the context of subsidies. The authors basically conclude that the loss of income to farmers is far less than the combined subsidies/externalities gained/generated by farmers. In addition, the authors conclude that the farmer?s loss of subsidies is a wealth transfer back to the taxpayers and after removal of subsidies only those farmers will continue that have positive net revenues after all costs.
Regarding the impacts on waterborne transportation, the authors argue that while the value of the subsidy to barges is unknown, the subsidy could be re-applied to trucks and railroad system and this would apparently solve any access/transportation problems.
Regarding the value to commercial and recreational fisheries, the authors postulate the following:
1. Total fish population will increase to between 5 and 10 million fish under the CTUIR strategy. There is apparently no uncertainty regarding these numbers.
2. Commercial fishing will generate $36 to $37 million in net output, assuming that harvesting costs are 10% of revenues as was allegedly the case under Indian in-river set net fisheries. ECONorthwest "assume that the region is capable of finding a way to manage the commercial fishery efficiently" (page 55)
3. Recreational fishing will generate $18.7 to $189.1 million in fishing expenditures (assuming 5 to 10 million fish respectively) and about the same amount in consumer surplus.
Regarding quality of life, ECONorthwest argues that surveys have shown that quality of the environment is more important than the economy as a major reason for moving into Oregon. The assumption is that stronger returns of salmon is linked in the respondents minds with quality of life. They stretch the argument to conclude that improved fish runs (of the impacted runs) could induce greater population migration.
Finally, the authors argue that the salmon have an intrinsic value: "Countless statements of the religious, cultural, and biological importance of salmon represent economic values that appear to be large. These values are difficult to measure, but they are no less important for this. One should not ignore them simply because analysts have failed to develop adequate tools for measuring them." (emphasis added - Page 67)
What Kind of Economic Study is Done? - This is a provocative report that is high on theory but very low on quantification. It is an advocacy piece that is extremely subjective and intended to stage the best case for salmon restoration.
What Measures of Effectiveness are Assessed? - There is little of value in the measures of effectiveness because little attention was given to quantification.
What Major Conclusions Were Reached, If Any? - The major conclusion are:
1. all activities that harm salmon/habitat are illegal
2. all firms that use water are unimportant to the economy
3. the value from restored runs is extremely high
4. the certainty of salmon restoration is not questioned
Reviewers Overall Impression - I was not impressed with the report because it makes no attempt to address the complicated issues of multiple uses on the river system. Instead, it is an advocacy report with unhelpful accusations directed against water users (especially agriculture). The report does very little to further the debate on how best to address the impacts of salmon restoration.
Study Title: Columbia River System Operation Review: Final Environomental Impact Statement Appendix O: Economic and Social Impact
Authors: Bureau of Reclamation, Corps of Engineers and Bonneville Power Administration
Date: November 1995
Reviewer: Jack Richards
Type of Study
The System Operation Review (SOR), Appendix O, is the Final Environmental Impact Statement prepared by three Federal Agencies; Bureau of Reclamation, Corps of Engineers and Bonneville Power Administration with assistance from numerous other government agencies, Indian Tribes and private organizations. The SOR process, however, began in 1990 prior to ESA listing of Snake River stocks. Its purpose is to analyze future operations of the Columbia River system and river use issues. Analysis of power production, irrigation, navigation, flood control, recreation, fish and wildlife habitat, and municipal and industrial water supplies are included in the SOR process. The study analyzes the impacts of alternative system operation strategies (SOS) relative to national economic development, regional economic impacts, social impacts and environmental impacts.
Sixteen work groups were organized to manage the SOR study and to analyze the environmental, economic and social impacts. The Economic Analysis Group, that was responsible for the socio-economic analyses and for Appendix O, consisted of 18 core members from the three sponsoring agencies and 38 others. Anyone interested was encouraged to participate in the work group meetings.
The goal of the SOR is to achieve a coordinated system operation strategy for the river that better meets the needs of all river users.
Appendix O addresses only the economic and social effects of alternative system operating strategies for managing the Columbia River system.
During the screening phase a total of 90 system operation alternatives were considered. A total of 13 alternatives were considered in the final EIS. Six of the 13 alternatives considered are unchanged from the draft EIS, one is a revision of an original strategy and six are straties added in the final EIS. Many of the alternatives evaluated in the final EIS emphasize fish recovery issues. Some of the SOS's could not be implemented now and in some cases design changes would require several years to complete. The most important SOS's relative to fish recovery are summarized in the final section of this review.
Other Fish Recovery Emphasis
Appendix C related to Anadromous Fish and Juvenile Fish Transportation.
Measures of Effectiveness Assessed
The opportunity costs of all other water-based products (power, flood control, irrigation, M & I pumping, navigation) are compared to the benefits to anadromous fish and recreation. Each SOS is compared to the base case (SOS 2c).
The costs of each SOS (estimated dollar value) is compared to the effectiveness or benefits (in terms of fish harvested) that accrue to anadromous fish and recreation. This provides a measure of the effectiveness of expenditures and opportunity costs to the number of fish harvested. Since values of fish harvest are also developed elsewhere in the report, it seems that it is only a simple matter to convert the available data to a cost-benefit analysis. Consequently, the results reported in this study go beyond cost-effectivenesses measures. Nonetheless, the benefit-cost comparison may suggest such a minimum return to expenditures that cost-effectiveness measures may be preferred. Only in the case of ESA species, however, are managers justified in using cost-effectiveness measures when information is available for a benefit-cost comparison. This study of system operation strategies seems to confuse the requirements of endangered species with the broader goal of optimizing system operation.
Areas of Economic Opportunity Costs Assessed
All major ares of opportunity costs (to fish resources) are addressed in this study since all watew-based uses are carefully analyzed to extent that information is available, and in some cases could be developed as part of the SOS process. Many of the competing water uses are discussed separately in other appendices (Appendix C - Anadromous Fish; E - Flood Control; F - Irrigation/Municipal and Industrial Water Supply; H - Navigation: I - Power: J - Recreation; K- Resident Fish; N - Wildlife). Each Appendix presents a detailed description of the work group's analysis of alternatives.
Anadromous Fish: Economic and social impacts related to commercial, recreational, and tribal harvests are included. Low and high value estimated were included for commercial and sport catch.
Resident Fish and Wildlife: From an economic perspective, the monetary value of the resident fish and wildlife is captured in the recreation analysis. These values were determined for different SOS's.
Wildlife: Brief discussion related to recreational use.
Flood Control: Economic and social impacts related to changes in annual damages from flooding.
Irrigation and water supply: Economic and social impacts are related to changes in net farm income. Up to six reservoirs would experience lowered pools under alternative SOS's considered. Impacts on municipal and industrial water users were also evaluated.
Navigation: Economic and social impacts related to changes in sthe cost of transportiang commodities. Both deep and shallow draft navigation impacts along with Dworshak log transport impacts were evaluated.
Power: Economic impacts related to changes in rates; impacts of rates on power demand; net system costs; and changes in consumer surplus are evaluated. Substantial changes were made in power evaluation between the DEIS and FEIS due to increased competitiveness of the power system during that time. During the 1990's, the wholesale market for electricity has broadened to become much more of a West Coast rather than a Pacific Northwest regional market. Deregulation of natural gas is seen as indicative of trends that are to be expected for power markets.
Recreation: Economic and social impacts related to the economic value of recreation activity are evaluated. Both direct and indirect economic values are estimated where indirect effects are the impacts on local economic activity.
Water Quality: Economic and social impacts related to impacts on water quality.
What Important Cost Effectiveness Estimates Produced by the Study?
Although cost-effectiveness information was developed in terms of the costs of each SOS relative to changes in fish harvest, the information is available to simply use a cost-benefit analysis if current or past market prices are utilized to estimate benefits.
Did the Study Have Any Appreciable Impacts on Policy Decisions?
The impacts of this study on future policy direction remains to be seen. Managers will almost certainly utilize this information in policy decisions, but the extent that it will be used in unknown.
The goal of this study is to achieve a coordinated system operation strategy for the river that better meets the needs of all river users. Given the support provided to this study by the three lead agencies and the assistance provided by other cooperating entities, the study can be expected to impact policy decisions.
What Does the Study Contribute to Policy Dimensions?
This study attempts to coordinate the complex management of the Columbia River system and to analyze the impacts of alternative operating strategies on different users of the river. Much important information relating to social, economic and environmental impacts have been developed through this process.
Regional economic impacts were estimated for each of the System Operation Strategies. These were developed from the estimates of direct economic impacts. The most common indicators of changes in regional economic activity are adjustments in employment and earnings where were measured from the input-output models in this study.
Were the Assumptions Realistic?
Numerous assumptions were made as the study proceeded. Some of these are more realistic than others, but most resulted from practical limitations relative to time and cost. Some examples of these assumptions are:
The biological fish models and the economic analysis assume a fixed ratio over time of the number of fish harvested and the number of fish that escape harvest to spawn.
Water rights were considered an inviolable institutional constraint. Consequently, the analytical assumption was that current water availability will remain the same under all operating strategies.
Numerous similar assumptions were necessary to simplify the analytical process. Clearly the assumptions made must be carefully related to the analytical results in all cases. For example, the above assumption of a fixed ratio between harvest and spawning is typical, but represent long run expected values that may not be consistent with management decisions typically reflecting an annual analysis and in-season adjustments. It may also be useful to reduce harvest to increase spawning to improve productivity. Furthermore, the effectiveness of any given level of spawning escapement can vary with changes in habitat conditions of a particular year. This example is only intended to demonstrate that use of the results summarized in Appendix O, obviously, would have to be carefully related to the assumptions utilized in the estimates.
Conclusions Reached, If Any
Conclusions were reached regarding the costs and benefits associated with the different SOS's. However, this study involved the use, to varying extents, of several extensive models (IMPLAN, Columbia River Salmon Passage Model, Empirical Life Cycle Model, HYDROSIM, Stochastic Life-Cycle Model) plus estimates developed as part of the study (e.g. recreation survey). It is difficult to assess the value of any of the conclusions reached unless the reader accepts the validity of all of the models utilized in the study. This complex process may limit the usefulness of this information for management decisions since managers may be reluctant to base decisions on information that they do not fully comprehend.
The estimated values are probably reasonable, but they often must be accepted on face value due to the limitations imposed on the reader who is not likely to be familiar with all of the complex models utilized in the study.
In spite of any limitations the reader faces in assessing the validity of many of the estimates, there is a vast amount of economic, social, and technical data presented that can serve as base information needed for fishery management.
Summary of SOS's with Major Benefit to Anadromous Fish
SOS 2d 1994-98 Biological Opinion -- This strategy was added after the draft EIS analyzes the 1994-98 Biological Opion operation that includes flow augmentation on the Columbia, flow targets at McNary and Lower Granite, specific volume releases from Dworshak, Brownlee, and the Upper Snake, meeting sturgeon flows 3 out of 10 years, and operating lower Snake projects at MOP and John Day at MIP.
SOS 5b Natural River Opeation draws down the four lower Snake River projects to near river bed levels for four and one-half months during the spring and summer salmon migration period, by assuming new low level outlets are constructed at each project.
SOS 5C Permanent Natural River Operation -- This strategy was added since the draft EIS. It would operate the four lower Snake River projects at near river bed levels year round.
SOS 6b Fixed Drawdown Operation draws down the four lower Snake River projects to near spillway crest levels for four and one-half months during the spring and summer salmon migration period.
SOS 6d Lower Granite Drawdown Operation draws down Lower Granite project only to near spillway crest level for four and one-half months.
SOS 9a Detailed Fishery Operation Plan is also an added alternative. This alternative includes flow targets at The Dalles based on the previous year's end-of-year storage content, specific volumes of releases for the Snake River, the drawdown of Lower Snake River projects to near spillway crest level for four and one-half months, specified spill percentages, and no fish transportation.
SOS 9b Adaptive Management. This is also an added alternative that establishes flow targets at McNary and Lower Granite based on runoff forecasts, with specific volumes of releases to meet Lower Granite flow targets and specific spill percentages at run-of-river projects.
SOS 9c Balanced Impacts Operation This is an added alternative that draws down the four lower Snake River projects near spillway crest levels for two and one-half months during the spring salmon migration period. Flow augmentation and other fishery enhancement requirements are included in this alternative.
SOS PA Preferred Alternative represents the operation proposed by NMFS and USFWS in the Biological Opinions for 1995 and future years; this SOS operates the storage projects to meet flood control rule curves in the fall and winter in order to meet spring and summer flow targets for Lower Granite and McNary, and includes summer draft limits for the storage projects.
Summary of Studies of the Economics of Salmon Recovery Actions
||Scope of Measures
||Type(s) of Economic Analysis
||Type of Biological Analysis or Objective
|Harza Northwest, for the Corps of Engineers (1996)
||Lower Snake River
|Hydropower only; no consideration of other components of the eco-system
||Comparison of three "system paths": transportation, in-river movement of fish, and "mixed"; 15 projects compared
||Measures selected that meet arbitrary thresholds for returning adults, fish-passage efficiency, and juvenile survival; CRiSP v1.5 to determine biological effectiveness; Decision Tree used to evaluate alternative paths
|Diamant and Willey, EDF for the Council (1995)
||Snake River (Upper and Lower)
ed Snake River salmon
|Alternative configurations (26) of Snake River projects, including various drawdown options, new storage, and purchase of water rights
||Cost-effectiveness of various project configurations, measured in dollars/hour of reduced travel time; total travel time reductions for each configuration
||Direct project and opportunity costs: construction, O&M, power, water supply, irrigated agriculture, flood control, navigation, and recreation
||Measure of benefit across all proposed configurations is reduced water particle travel time (WPTT)
|Niemi, et al., ECONorth-
west for Umatilla Confederation (1995)
|Snake River and John Day Reservoir
||Various drawdown scenarios
||Defines "value" in terms of ownership of resources; argues that Native Americans have superior ownership rights over fish, stemming from 1855 Treaty
||Postulates consumer surplus related to the region's quality of the environment; emphasizes roles of subsidies and ability of economic actors to respond to relative price changes
||Essentially no independent analysis; proposed drawdown strategies are assumed to yield 5-10 million fish
|BPA, COE and BOR, Appendix O of the SOR-EIS (1995)
||Columbia River System
||Columbia and Snake River
||Alternative system operating strategies for the Columbia/Snake system (19, including drawdowns)
||Estimates of opportunity costs of power, flood control, irrigation, water supply, navigation, compared with dollar value of fish harvest
||Cost-effectiveness analysis (although cost-benefit analysis is possible, given a dollar value of harvest)
||Model-driven estimates of biological benefits under different operating scenarios; benefits measured solely in terms of the value of commercial and sport fisheries
|Paulsen, et al. Resources for the Future, for BPA (1993)
||26 subbasins above Bonneville Dam
||79 chinook and steelhead stocks above Bonneville Dam
||120 propagation alternatives, including hatcheries, habitat changes, . . .; 8 passage strategies; 5 terminal harvest rates
||Linkage of 4 models (SAM for flows, CRiSP for mortality, DLCM for population estimates, and SWOM for cost minimization)
||System-Wide Optimization Model (SWOM) determined least cost set of recovery actions within set of measures meeting subbasin goals
||Simulation of downstream mortality for combinations of measures; testing of population sizes against harvest and escapement goals by subbasin (CBFWA 1990)
|Huppert et al. (1996)
||Three Snake River stocks
||NMFS Proposed Recovery Plan (1995)
||Economic analysis of those measures proposed in the Recovery Plan
||Cost-effectiveness, based on opportunity costs of proposed measures
||No independent analysis; study relies on the Recovery Plan for estimates of effectiveness
Summary of Studies of the Economics of Salmon Recovery Actions (Cont.)
||Applications to Future Policy Choices
|Harza Northwest, for the Corps of Engineers (1996)
||Ratio of costs of most- and least-effective measures is 50 to 1; biological information is insufficient to use a decision-tree approach; transportation combined with pre-fish flow operations is cheapest; drawdowns are generally most expensive
||Marginal benefit of additional measures aimed at in-river survival may be low, given current configuration; surface collectors with transport may be most cost-effective
||Overstates cost of lost recreation; subsidies to agriculture ignored; power analysis outdated by changes in market conditions; elasticity impacts on transportation not accounted for; analysis not comprehensive biologically or economically
||Suggestion of a cost-effectiveness measure: incremental benefits expressed as "cost per % survival"
||Decision Path analysis as presented does not support cost-effectiveness analysis; marginal analysis may point in new directions; drawdowns probably do not fit with budget constraints; possible avoidance of rehabilitation to Snake River dams after 2000
|Diamant and Willey, EDF for the Council (1995)
||Estimated changes in net farm income suggest opportunities for "gains from trade" (i.e., payments to farmers to release water); dry-year leasing and basin-wide fallowing are the most cost-effective actions; future reductions in WPTT should be discounted
||Ratio of most- to least-cost-effective measure may be as high as 28 to 1; drawdown alternatives are generally not cost-effective, yield the largest reductions in WPTT, but also take the longest to implement
||Power cost estimates outdated by market changes; assumes WPTT is correlated with biological success; some results counter-intuitive and unexplained
||Examines both Upper and Lower Snake
||Limited applicability, due to narrow scope of analysis and definition of index of biological benefit
|Niemi, et al., ECONorthwest for Umatilla Confederation (1995)
||Many, if not most industries using water in the Northwest are subsidized, relatively unimportant economically, and will eventually prove unsustainable; Native American fisheries are sustainable
||Elimination of subsidies means that irrigated agriculture can be shrunk at relatively modest cost; subsidies to barging could be applied to trucking instead; drawdowns will support growth in commercial and recreational fisheries
||Estimates of subsidies to existing industries, and economic costs generally, are not documented or explained
||From an NED perspective, subsidies should be taken into account; however it is not clear that NED is the most appropriate basis for Northwest decision-making, given the expectation that Northwest residents will pay for these programs
||Limited, due to clear "advocacy" nature of the work; does not offer a broad perspective in terms of biology and is not documented in terms of economics
|BPA, COE and BOR, Appendix O of the SOR-EIS (1995)
||Compared with Corps' 1993 Flow Improvement EIS, and using harvest as measure of biological benefit, some configurations are arguably less expensive than "current operations"
||Assumes existing institutional constraints, such as distribution of water rights
||Comprehensive economic analysis based on opportunity costs in affected sectors
||Possibility that SOR-EIS could be broadened to encompass more proposed measures
|Paulsen, et al. Resources for the Future, for BPA (1993)
||Subbasin goals (set in 1990) were unrealistic, and could not be achieved by the set of passage and propagation alternatives considered; mainstem flows of 1989-91 were more cost-effective than the 1991 Phase II proposal of the Council
||Genetic risk reduction conflicts with subbasin escapement and terminal harvest objectives; planners' preferred actions emphasize more propagation and cost more than the cost-effective set of actions
||Deterministic modeling ignores risk, variability of results over time, and uncertainty associated with data and inputs; insufficient documentation of nature of economic costs included; apparent exclusion of irrigation, land use, recreation, navigation
||Simulation permits testing of biological interactions among proposed measures, although subject to lots of assumptions
||Limited in its existing form, due to exclusions of various economic sectors (e.g., harvest) and number of simplifying assumptions, but potentially significant, given application of cost-effectiveness methodology and comprehensive modeling approach
|Huppert et al. (1996)
||About 2/3 of the cost of the Plan is associated with changes to the hydropower system; total costs vary widely depending on assumptions regarding combinations of measures
||Economic benefits of enhanced harvests (or "use value" generally) are not sufficient to justify the costs of recovery actions; estimates of "existence value" may also not be sufficient to justify existing expenditures
||Scope of analysis was limited by definition to those measures proposed in the Recovery Plan, so conclusions are not comprehensive
||Many alternatives examined; reminder of "learning by doing"; uncertainty recognized; solutions not bound by existing institutional constraints; explicit recognition of assumptions
||Potential for application of the approach more broadly than the 1995 Recovery Plan, but would require much more extensive analysis