PHASE I FINAL REPORT:

 

OPTIONS FOR PERFORMING SAMI’S

SOCIOECONOMIC ASSESSMENT

 

 

 

 

 

 

Prepared by:

 

GHL Incorporated                                                                Mathtech, Inc.

1020 Nineteenth Street NW, Suite 520                                  202 Carnegie Center, Suite 111

Washington, DC  20036                                                       Princeton, NJ  08540

 

 

 

 

 

Prepared for:

 

Southern Appalachian Mountains Initiative (SAMI)

Socioeconomic Assessment Workgroup

 

 

 

 

 

 

 

 

 

August 1998


 

CONTRIBUTING CONSULTANTS

 

The project for which this serves as the final report involved the collaboration of two firms and several consultants.  The following list identifies the specific individuals who, as a group, developed the consensus recommendations that are described in this report, and who helped lead one or more of the workshops mentioned in this report.

 

 

Thomas Dietz, George Mason University

A. Myrick Freeman III, Bowdoin College

Robert L. Horst, Jr., Mathtech Incorporated

Daniel J. Mullarkey, Mathtech Incorporated

Anne E. Smith, Charles River Associates

 


INTRODUCTION

 

            This report presents suggested methods and the associated costs for conducting a socioeconomic assessment of alternative environmental control strategies applied in the SAMI region.  The report relies on information presented by the authors at Workshops held in May and July of 1998 but modifies that material in several ways.  Specifically,

 

·                                            The list of topics potentially addressed quantitatively is narrowed, based on feedback provided by the Socioeconomic Assessment Workgroup subsequent to the May Workshop;

 

·                                            Descriptions are provided for a low cost (minimal) quantitative method and an incrementally enhanced approach (in cost and quality) for each identified topic. The most thorough and complete quantitative approaches possible are not described here, recognizing that these are not options within SAMI's budget and requirements for coverage of more than a single topic;

 

·                                            An alternative approach is also described that would rely on a qualitative assessment in return for coverage of the broader set of topics discussed at the May Workshop;

 

·                                            Cost estimates are provided by topic for each assessment method.

 

 

The objective of this effort is to provide the Socioeconomic Assessment Workgroup with information that will assist in the development of a clear Statement of Work for the Phase II analysis.  This report does not offer a recommended approach or package of topics for Phase II.  That decision is to be made by the Workgroup in the future.

 

TOPICS FOR QUANTITATIVE ANALYSIS

 

            Table 1 summarizes our qualitative judgments about the potential relative magnitude of impacts and the ability to develop credible estimates of these impacts in a socioeconomic analysis.  These judgments were presented at the May Workshop, and supported by the two days of presentation materials used in the Workshop.  As a result of that part of the project and associated discussions with the Workgroup, the Socioeconomic Assessment Workgroup has developed a narrowed list of topics which have been determined to be of greatest interest to


Table 1.  Summary of Consultant Judgments on Topic Selection Criteria

(The judgments in this table are intended as relative to others within same column and were made with only very limited information about control strategies to be assessed.  The judgments on potential estimate quality and analysis cost were made assuming a single best methodology reasonably available, or sensible to consider in the absence of severe budget constraints.  Lower cost options than those reflected in this table are available, but usually entail more qualitative analysis than column 2 indicates.  To account for SAMI’s budget constraints, many of the analysis options described in the text of this report are of a lower-cost nature than the analysis methods assumed in this table.)

 

 

EFFECT OF CONCERN

 

$,T,L (*)

 

IMPACT

MAGNITUDE

 

DISTRIBUTIONAL

DIFFERENCES

QUALITY

OF

ESTIMATE

 

ANALYSIS

COST

Impacts of AQRV-related Changes

Fishing

$

Low to moderate

Mild

Moderate

High

Hiking/enjoying scenery

T/$

Moderate

Demog group, states

Moderate

High

Stewardship

L

Fairly large

Mild

Moderate

Fairly low

Community sense of place

L

Fairly large

Demog ; communities

Moderate

Fairly low

Other Environmental Impacts

Agric/forestry/fish prod.

$

Relatively low

Sectors and communities

High

Low to mod.

Materials – building

$

Relatively low

Mild

Fairly low

Very high

Materials – cultural

L

Low to moderate

Mild

Low

Fairly high

Health:  Mortality

$

Potentially large

Demog group

Controversial

Fairly low

Health:  Morbidity

$

Potentially large, but

less than mortality

Demog group

Moderate

Fairly low

Impacts of Emissions Management Strategy

Lifestyle changes

$

Potentially important

Demog; communities

Low to moderate

Mod. to high

Reliability

$

Low

Some sectors

Low to moderate

Mod. to high

Administrative req’ts

$

Low (but potentially important to policy choices)

Mild

Fairly high

Low

Infrastructure req’ts

$

Low

Mild

High

Low

Nat. resource impacts

$

Fairly low

Communities

Moderate

Moderate

Economic Impacts of Emissions Management Strategy’s Costs

Inter-state, inter-regional, and inter-sectoral change

T

Low to moderate

Communities and sectors

Fairly high

High

Number of jobs; pay level

T

Low to moderate

Substantial for all

Fairly high

High

Job dis/relocations

T

Potentially important

Substantial for all

Fairly high

High

Tax implications

L

Low to moderate

Mild

Moderate

Low to mod.

Change in tourism

L

Low

Moderate

Fairly low

Fairly low

 (*) $ = Assumes a method that gives monetized estimate, T= Assumes a method that generates quantitative, non-monetized estimate, L = Qualitative method


SAMI for possible quantitative analysis.  One objective that guided Workgroup consensus on the narrowed list of topics was the desire to maintain balance across the selected topics in terms of the types of values considered and the stakeholders affected.  This report provides suggested scopes and budgets for the shorter list of topics, which includes:

            1.            Fishing

2.            Hiking/Enjoying scenery

3.            Stewardship of environmental resources

4.            Community sense of place

5.            Human health (mortality)

6.            Competitiveness impacts (interstate, interregional and intersectoral)

7.            Effects on jobs and real income levels (wages, prices)

8.            Lifestyle changes associated with regulation.

 

METHODS FOR QUANTITATIVE ANALYSIS

 

            This section describes methods for quantitatively assessing each of the selected topics identified above.  Each subsection provides an overview of the benefit or cost impact associated with a topic, a description of a low-cost method for quantifying the benefit or cost, an alternative “incrementally-enhanced” method, and a critical assessment of the suggested methods.

 

            For several of the topics discussed, the low cost quantitative assessment option is identified as “benefits transfer” from existing studies.  The benefits transfer method involves using existing estimates of the economic value of one impact or topic to approximate the economic value of a similar impact or topic.  For example, a variety of studies have been conducted on the economic value of changes in fish stocks or fishing opportunities.  One or more of these existing studies may be used to obtain information on the value of changes in fish stocks in the SAMI region.  Ideally, value estimates would be transferred from carefully conducted studies that involve similar impacts and similar geographic and temporal locations.  In practice, adjustments to a study’s unit WTP values often need to be made to account for differences between the conditions in the original study and conditions in the area of interest.  Adjustments may need to be made for differences in the size of the population affected, differences in the socioeconomic characteristics of the populations, and differences in the timing of the impact.  In the case of ecological services, differences in the specific ecological sites or species being considered, as well as differences in the availability of substitute services may also need to be considered.  The primary appeal of the benefits transfer approach to valuation is that it is generally faster and less expensive than conducting an original study.  The major disadvantages are that quality of the valuation information gained is heavily dependent on the quality of existing estimates and on the comparability of existing studies to the current application (because economic values can be very context-dependent and thus not fully transferable).

 

1.            Recreational Fishing

 

            Reductions in air pollutant emissions may reduce acidification of fishable waters in the SAMI region.  This may lead to a variety of measurable chemical or biological impacts, one or more of which are to be projected by the Effects Group as part of the Integrated Assessment.  For the socioeconomic assessment, the key relationships of concern are how these chemical and biological impacts affect the demand for fishing.  For example, an increase in the density of fish of a desired species increases the probability of catch success, which provides value to those who fish.  The objective is to quantify the change in value brought about by a selected control strategy.

 

            Based on preliminary discussions at the May Workshop, we assume this topic area is limited to freshwater recreational fishing anywhere in the eight-state SAMI region.  Both resident and non-resident fishers within the SAMI region would be counted.  Any benefits realized outside the SAMI region, from coastal fishing or from commercial fishing activities would not be measured.

 

Low-Cost Method

 

            The low-cost method for deriving quantitative estimates of fishing benefits would involve a benefits transfer approach, where estimates of the willingness-to-pay (WTP) relationship are tied to a study performed previously, and not with SAMI's specific conditions in mind.  The results of the pre-existing study would be "transferred" by attempting to account for the specific socioeconomic conditions of the SAMI region, relative to those in the original study.  The measure to be transferred would be the willingness of an angler to pay for a unit change in a given impact (e.g., density of species population).  The recommended approach would focus on a single study for benefits transfer to avoid inappropriate aggregation.  Available studies are often site-specific and there likely exist confounding factors that make aggregation across studies problematical.

 

In order to select one study from which to transfer fishing benefits, a limited literature review is required to identify the most appropriate study from which to obtain the transfer value.  Travel Cost and Contingent Value applications are possible sources of information.  Multiple criteria can be used to select the transfer study, including relevance to the SAMI region, quality of the original study, and associated data costs.  In selecting the preferred study, consideration must also be given to the impact valued and its relationship to effects considered in the Integrated Assessment.  It must be possible to scale the transferred values to reflect the impact changes associated with specific SAMI control strategies.  The evaluation of the criteria must be clearly presented to demonstrate the tradeoffs associated with the selection of one study.

 

            Scaling of the transfer value is only one of the adjustments required to implement the benefit transfer approach.  It is also necessary to identify the level of participation by anglers in the SAMI region for each year of the analysis.  This requires data on participation over time. Recent surveys conducted by the U.S. Fish and Wildlife Service may provide a credible basis for gauging these magnitudes.  More simply, one can assume that changes in participation rates over time will be associated with changes in population, not changes in behavior.  This approach would estimate benefits for the entire SAMI region and not for individual subregions such as Class I areas.

 

Incremental Method

 

            The benefits transfer approach described is performed using a single “best” study that provides a value measure that can be associated with chemical and biological outcomes.  Reliance on a single study puts full weight on that study for the credibility of the SAMI analysis.  An alternative approach which could incrementally improve the quality of the transferred value would be to develop a synthesis of the WTP relationships from several leading studies identified in the literature review.  Additional effort would be directed to ensuring that the impacts being valued were consistent across the various studies.

 

            Use of multiple studies in the benefits transfer exercise may allow for value estimates to cover a wider range of fish species than relying solely on one study.  Consideration of several studies may also provide an opportunity to define disaggregate values according to the type of area in which the angler fishes.  It may be more appropriate to treat area-specific values in this disaggregate fashion, since the studies may value different types of fishing experiences.  In this case, participation data would also need to be available at a more disaggregate level by geographic region.

 

Critical Assessment

 

            As noted in Table 1, we expect that the magnitude of benefits for fishing will be relatively low to moderate in comparison to other benefit categories.  The quality of the estimate is conditional on the participation data, the quality of the study from which the transferred value is taken, and the conformity between what is being valued and the impacts being measured.

 

2.            Hiking/Enjoying Scenery

 

            An important aspect of Class I areas is the provision of aesthetic pleasure, as part of hiking activities, other outdoor activities such as photography and bird/wildlife watching, and opportunities to enjoy ambient vistas in general.  One component of the value derived from these activities is associated with visibility, or the "view".  Noticeable changes in flora and fauna may also affect the value.  Emission control strategies to be considered by SAMI may improve visual range conditions in the SAMI region, and these changes are to be estimated as part of the Integrated Assessment.  There may also be some information provided about changes in tree health, although the nature of such information is less certain. 

 

Low-Cost Method

 

            Addressing values associated with activities that have a relationship to aesthetic conditions is complex if done thoroughly.  A low-cost option would have to rely on pre-existing estimates of WTP for perceptible visual range improvements, which we believe to be the most salient of the potential forms of aesthetic change likely to be associated with SAMI control strategies.  The presumption would be that the elicited statements of WTP for improvements in visibility, as represented in a contingent valuation survey, encompass the many activities (hiking, photography, etc.) through which survey respondents might expect to derive value.

 

One existing study for use in an assessment of recreational values associated with visibility changes in the SAMI region is a 1990 study by Chestnut and Rowe[1] that addressed conditions in the Shenandoah National Park.  This study used the Contingent Valuation Method (CVM) to assess use and non-use values associated with visibility improvements for a sub-area of the SAMI region.  Another existing study that should also be used is a 1983 study by Rae[2]  that considered values for visibility improvements in the Great Smoky Mountains National Park.  This is an earlier study and may have greater limitations because the methods of contingent valuation were advancing rapidly during the interval between these two studies.  Nevertheless, a comparison of estimates from both studies would make sense, since both specifically address visibility values for two key parks in the SAMI region.  Differences between the results of the studies, combined with discussion of possible study-design causes for such differences, could be helpful in demonstrating a range of uncertainty for this value. 

 

Visibility valuation is a particularly complex issue that has greater remaining potential biases and methodological difficulties than some other valuation issues such as recreational fishing.  Therefore, we believe that this task should emphasize a qualitative discussion of all remaining potential sources of bias that the two studies cited above may have not addressed completely, and the implications of these issues for the quantitative results that would be drawn from the two specific studies.  This would provide better context for the interpretation of the quantitative results.  An important element of this discussion would be an examination of how visibility is projected to change on particular days (as compared to the annual or seasonal average change) and a discussion of implications for perceptibility of change, which is a pre-requisite for imputation of value.

 

            As is the case with fishing, it is necessary to ensure that there is conformity between the impacts being estimated in the Integrated Assessment and the endpoints being valued at this step of the analysis.  Furthermore, participation data for outdoor non-consumptive activities will need to be identified across time for the SAMI region to extrapolate from individual WTP values to aggregate regional benefits estimates.

 

Incremental Method

 

            We have identified two enhancements to the low-cost method described above that would help improve the quality of a quantitative assessment of hiking/enjoying scenery.  First, a synopsis of the range of WTP results in the literature could help identify the degree of uncertainty and enhance the credibility of quantitative results.  The low cost option covers only two studies, selected because of their relevance to the SAMI region specifically.  However, there are many more studies for other regions of the country.  These studies present a range of values that can help reflect the overall uncertainty in obtaining estimates of visibility values, and how dependent these values can be on the particular survey design used. 

 

            We recommended that even the low-cost analysis option should include some discussion of the projected magnitude of visibility changes at the daily level, in addition to changes summarized only on an annual average basis.  This is an important issue because the valuation studies are based on instantaneous changes that are quite large, not on annual average changes that are quite small.  The imputation of the resulting willingness-to-pay to an annual average change is a methodological convenience that has not been supported by empirical findings.  Especially for use values, it is important that the changes in question be large enough to be perceived by people.  However, it is not possible to determine from an annual average alone whether the underlying daily changes are large enough to be perceived.

 

Few visibility benefits estimates to date have attempted to address this methodological complexity, but it remains an important issue for credibility.  Thus, the next increment of funding should also involve a more thorough quantitative assessment of this “variability and perceptibility” issue.  This would involve providing a more detailed analysis of the day-to-day patterns of visibility change based on episodic information from the Integrated Assessment.  The analysis would explicitly estimate the number of days for which SAMI’s integrated modeling projects perceptible change in visibility.  It would discuss what the annual benefits might be if the WTP for visibility improvements in the original CVM studies were recalculated on a per-day basis and then attributed only to the fraction of days where change can be expected to be large enough to be perceived.  This should be compared to the results based on the annual average change alone, and be used to help further define the range of uncertainty in the benefits estimates.

 

Critical Assessment

 

            Table 1 identifies recreational effects as having a moderate potential magnitude of impact relative to other effects and a moderate quality of estimate.  Key issues are the careful definition of what is being valued and the extrapolation of unit values to obtain annual values for given geographic areas.  Also, CVM is a maturing method that has critics because it is based on stated rather than observed choices.  A good analysis would carefully review the studies used and discuss possible remaining sources of biases.

 

3. and 4.            Stewardship/Sense of Place

 

            Environmental regulations provide both benefits and costs to society.  Some of the beneficial impacts can be associated with a direct use by individuals of an environmental resource (e.g., consumption of agricultural products) or linked to activities sensitive to environmental quality (e.g., recreation activities).  A common element of these benefits is that they involve values based on use of the environment.  Individuals may also have intrinsic values for environmental resources and environmental quality that do not involve a use value.  Stewardship and community sense of place are examples of such non-use values that may play a substantial role in the case of SAMI concerns.

 

            Stewardship refers to the notion that there is a fundamental ethical responsibility for humans to tend to nature and to pass on to succeeding generations a world that reflects a sustainable pattern of consumption of nature’s resources.  Stewardship values are reflected in individual preferences but do not overlap with any present or future use value held by an individual.  They are distinct values, often associated with unique, irreplaceable environmental assets, for which an individual has a willingness to pay to maintain positive supply now and in the future.  Stewardship values can have local, regional and global dimensions.

 

            Community sense of place refers to values that reflect the amalgam of quality-of-life attributes associated with living in a specific area.  In the context of the environment, sight attributes, including viewscape are especially important to community values, but would be captured under aesthetic changes.  Another aspect of sense of place may be closer to the idea of preservation of a particular complex set of quality of life attributes.  Given the close relationship of this with stewardship, and the overlap that would occur in trying to assess the two, we recommend combining these two topics and treating them together in an assessment. 

 

Low-Cost Method

 

            The low-cost option would consist of a qualitative summary of the evidence on significance and trends in community/stewardship values.  The assessment leading to this summary would include a systematic literature review that would emphasize the identification of the types of issues that are often expressed in explorations of social concerns with environmental changes.  One source of this information may be existing environmental risk assessments that include input from a range of stakeholders.  Of interest is the identification of prior assessments that involve proposed environmental changes that may be similar to changes expected under alternative SAMI control strategies.  Prior studies conducted in the SAMI region may be especially helpful to better understand impacts that may be unique to the SAMI region.  Although this review would not lead to even a relative ranking of issues in terms of monetized values, the assessment would describe the types of impact that have the greatest social significance.  The qualitative nature of the assessment also means that it will be difficult to attach relative values to alternative SAMI control strategies.

 

Incremental Approach

 

            The alternative incremental approach would use focus group interviews to elicit information from stakeholders within the SAMI region, and for the types of changes specific to SAMI control strategies.  The focus groups would involve in-person information exchanges and structured group discussions to assess the overall significance of stewardship/community values.  As feasible, specific SAMI control strategies would serve as scenarios for the focus groups.  The socioeconomic assessment would include qualitative discussion emphasizing primary data collected in the interviews.

 

Critical Assessment

 

            Table 1 lists stewardship and community values as potentially large relative to other topics.  However, there remains uncertainty about value magnitudes especially for incremental changes in environmental assets, which is what the SAMI control strategies will entail.  The methods described above will not reduce this uncertainty but should provide SAMI with more concrete information on the social issues of importance to stakeholders, and on differences in values among stakeholder groups that may be sources of conflict.

 

5.         Human Health: Mortality

 

            Statistical analyses have shown an association between elevated levels of particulate matter (PM) and increased mortality.  These studies have examined daily time series data as well as long-term (chronic) exposures.  These results have led to critical discussion involving issues of causality, specification error and measurement error.  Despite the continuing controversies, these statistical functions have been used to develop quantitative estimates of potential impacts of changing ambient concentrations of PM, most recently in the process of setting National Ambient Air Quality Standards.  These efforts rely on a two-step process.  First, premature deaths avoided are predicted using the statistical functions.  Second, the estimated count of avoided premature deaths is multiplied by an estimate of the value of reducing the risk of premature mortality.  Setting questions of causality aside, the first step raises many questions of uncertainty in appropriate functional form that the statistical analyses cannot illuminate.  Further, the second step has been subject to questions regarding the appropriate measure of value.

 

The net effect of the state of the art is that quantitative estimates can be relatively easily derived, but any quantitative estimate, however derived, will be controversial.  Alternative methods of deriving quantitative estimates lead to extremely different numbers.  No amount of further analysis that SAMI might conceivably perform would be able to reduce this controversy. 

 

Low-Cost Method

 

An assessment of mortality benefits should emphasize developing a range of potential impacts, explaining how alternative assumptions affect whether an estimate would be nearer the high or low end of the range.  The suggested low-cost method would include (1) a more detailed summary of the mortality valuation issues alluded to above, and (2) estimates of the upper and lower reaches of the quantitative range of value.  The range would be derived via calibration to national studies, combined with information from sensitivity analyses on those results that have already been performed.  The range would be scaled to account for relative differences in baseline concentrations and differences in changes from the baseline for those scenarios analyzed in the national studies vis-à-vis the control strategies to be considered by SAMI.  Furthermore, scaling for exposed population would also be conducted. 

 

Sensitivity analysis of the statistical relationship and the risk of premature mortality would also be summarized to illustrate the elements of uncertainty in the analysis.  As part of the sensitivity analysis, there should be a discussion of the clinical and laboratory evidence, and the extent to which it does or does not support the epidemiological (statistical) studies on which the risk estimates to-date have been solely based.  Although it would add somewhat to the cost of the task, one might consider having this qualitative discussion written by a clinician or other appropriate member of the medical profession.

 

Incremental Method

 

            The proposed incremental method for estimating mortality benefits would replace the calibration to pre-existing studies with direct estimation of the range of value, using changes in concentrations predicted from the Integrated Assessment.  The evaluation would be conducted for sub-regions (e.g., the model cell, or the county level) to better reflect any relationship between concentration changes and the geographic distribution of population.  This analysis would estimate the range of uncertainty by using multiple combinations of alternative assumptions about functional form, relative risks, and WTP to avoid the associated type of premature death, as consistent with the literature.

 

As with the low cost option, the credibility and breadth of perspective in this part of the benefits assessment could be enhanced by bringing a clinician or other appropriate member of the medical profession into the project effort.  This would add breadth because to-date, all benefits estimates for mortality impacts have been based solely on epidemiological or statistical studies.  The expert judgment of such informed individuals could provide greater context and credibility for both the estimated risk ranges and the underlying assumptions associated with the sensitivity analysis.

 

Critical Assessment

 

            This endpoint may be one of the topics selected for SAMI's assessment because it is expected to generate large impacts under some sets of assumptions that are possible to derive from the literature.  Nevertheless, controversy and uncertainty are a dominant attribute associated with this topic, and that fact must be part of the summary of results.  The proposed approaches will not reduce existing uncertainties; however, better characterization of the uncertainty in qualitative and quantitative terms, may assist decisionmakers in their reviews of the strategies.

 

6. and 7.            Competitiveness, Employment, And Real Income

 

            At the May Workshop, the discussion on estimation of economic impacts emphasized model-based approaches.  We have, in the course of this project, learned more about what is likely to be undertaken to address economic impacts.  We feel that it would be appropriate at this time to provide a more detailed discussion about how the various economic impacts interrelate.  This will be useful in better understanding the trade-offs and limitations in choosing alternative approaches.

 

            Environmental regulations impose direct costs on owners of emissions sources affected by control strategies, as they have to invest in technology or process changes to achieve emissions reductions.  One of the potential economic impacts of these control costs is that they can translate into reduced sales or profitability impacts on businesses, particularly when the sources face increased costs relative to their competition.  The effects on competitiveness may be inter-regional, inter-sectoral or inter-state.  For example, competitors may include producers of the same product in regions that do not face comparable emissions requirements.  Consumers may shift to these lower cost suppliers to avoid having to incur increased costs themselves.  Competitors may also be producers of alternative, substitute products from within the region that do not have the same magnitude of cost, by virtue of having a different manufacturing process altogether.  The substitute products may be associated with totally different economic sectors, as in the case of wood products and plastic products, creating an inter-sectoral shift in economic activity as consumers shift to now-cheaper alternatives.  Even within the region, some states may be more affected than other states because of differences in the mix of sectoral activities in each. 

 

Businesses will pass costs through to consumers to the extent that they can do so.  Significant competition from other regions or from substitute products prevents this from happening, as described above.  However, in some cases, consumers cannot find acceptable alternative supplies or good substitutes for the affected products.  In this situation, they may (1) curtail their consumption rather than absorb price increases.  This also leads to business impacts similar to competitiveness shifts, but with the offsetting economic gains not going to the affected producer's competitors.  Rather, consumers may use the funds for a totally different type of purchase.  Alternatively, consumers may (2) absorb the price increase because they view the affected product as relatively essential.  In this case, the business impacts to the owner of the emissions source may be minimal, and the brunt of the cost is shifted to consumers.  Consumers, in turn, will reduce their consumption of other goods that they view as less essential, and the ultimate business impact resulting from the control cost may manifest itself in sectors or businesses far removed from the sector that bears the emissions control cost.

 

Another complicating aspect of economic impacts needs to be mentioned.  Whenever a business incurs a control cost, it is effectively spending that money either on other sectors or on additional employees or hiring contractors.  That spending represents an economic boost to some sectors and some labor categories, while serving as a detrimental impact to either the directly affected business, or to consumers and providers of less essential goods and services.  In summary, there are some winners as well as losers, and the most significant aspects of economic impacts from spending on emissions control probably will be observable at the disaggregate level. 

 

            To the extent that some businesses contract due to competitiveness effects there may be reductions in employment in sectors that face significant control costs.  There may also be indirect employment impacts in sectors that provide less essential types of goods and services, as a result of price effects (e.g., recreation, vacation travel, fashion apparel, family restaurants, etc.).  These employment losses will be partially offset by increased employment in the regions, sectors or states that are competitors in price, or providers of pollution control activities.  Job impacts are therefore more difficult to estimate than relative shifts in competitiveness.  However, they cannot be estimated at all without first developing some quantitative understanding of the competitiveness effects.  Thus, we combine jobs and competitiveness effects into a single topic here.

 

            It should be apparent that quantitative estimates of these impacts would require an integrated, multi-sector, multi-region model that addresses demand as well as supply conditions.  Also, it should be apparent that if one of the impacts (competitiveness, jobs, price effects) is assessed quantitatively, then information is also simultaneously generated for the others.  Models exist which evaluate these economic impacts in an integrated fashion, but none of the options is feasible within the low-end of SAMI's socioeconomic analysis budget.  Thus, for this report, we suggest a low-cost method that involves some quantitative calculations, but which only generates qualitative insights about potential impacts.  In going to this low-end analysis, it becomes impossible to assess the degree to which consumer welfare is affected.

 

Low-Cost Method

 

            One method that has been used to understand the potential significance of a regulation to an industry is to compare the average cost of meeting regulatory requirements to the per-firm gross revenues (sales) for that industry.  All else constant, this ratio provides a first-order estimate of the sectors that are most likely to risk sustaining significant impacts.  For a given ratio, the risk is greater for sectors that have less ability to raise prices and for sectors with lower profit margins.  Sectors would be sorted according to the value of the ratio and judgments provided concerning the potential risk-enhancing factors of sectors. 

 

            It would be possible to also estimate and report other ratios than those strictly based on revenues.  For example, one could report a ratio of cost per unit of output, such as dollars per megawatt-hour for electricity, and dollars per ton of aluminum.  These ratios are closely related to costs-to-revenues ratios, but may be more easily interpreted by some individuals in terms of their likely market impact.  However, it should be noted that alternative ratios such as these would not be in comparable units.  This would make it difficult to compare across different sectors, or even to develop a single ratio for a single sector (because there may be many different specific products generated by a single sector).  Thus, we believe that alternative ratios would be useful to provide as a supplemental but not primary form of information.

 

            Qualitative discussions should be provided on the degree to which cost increases may be possible to pass through to users of the product produced by the controlled source.  This is relevant because it implies a price impact to consumers and discretionary consumption patterns rather than an impact to the sector bearing the cost.  Jobs "at risk" also can be estimated by providing information on the jobs in each sector and the associated ratio of cost impact.  Neither of these steps will suggest the actual degree of consumer welfare or job impact, but they will at least highlight the range of ways in which control costs may manifest themselves as economic impacts and distributional impacts.

 

Although it would increase the cost of this task, the process of interpreting the sector’s risks given these ratios would be strengthened by interviews with individuals from some of the key sectors who are familiar with marketing issues and market trends in their industry.  Such individuals may be able to provide market-specific information that could help in understanding the degree to which a particular product can sustain some or any price increase.  We suggest a case study approach that would involve approaching marketing specialists in three to five key sectors as part of this task.  The case study sectors should be selected based on information regarding cost incidence and sectors that appear to have a combination of high cost-to-revenue impacts and low profit margins. 

 

The electricity sector should probably receive particular emphasis for more in-depth investigation and qualitative discussion because of the pending effects of electricity market deregulation.  Pending deregulation means that historical information on electric sector profit rates is probably not a valid basis for assessing potential impacts of future costs to this sector because historical profit rates were supported by rate regulation.  In future years, particularly by 2010, there is likely to be a substantially different market structure, and profitability is likely to drop, along with many other likely changes in the structure of this sector.  Shifts in competitiveness are likely to occur across regional boundaries, and it would be important to explore whether the expected market shifts will be exacerbated or attenuated by the potential added costs of a SAMI-initiated emissions management strategy.  Explicit exploration and discussion of the unique issues for estimating electric sector impacts should be included in this task even if the case study approach suggested above is not undertaken.

 

            The low-cost approach suggested above will not provide an integrated assessment of the effects of controls across industry categories.  It will not be able to identify secondary impacts to businesses that result from price increases, yet these could be as significant as direct impacts to business that pay for the control measures. 

 

Incremental Method

 

            There appears to be no good intermediate option between the screening analysis described above and more detailed economic modeling.  The recommended incremental approach therefore consists of using one of the existing multi-sector, multi-regional models which simulate both business and consumer behavior in an integrated setting.  These models use microeconomic principles to characterize both price and output changes under different scenarios (e.g., with and without a control strategy) and link these changes in equilibrium price and quantity to other economic impact categories such as jobs and wages.  These models are well developed and appropriate for the present application.  However, it can be costly to conduct such an integrated analysis, even at the most limited scope feasible.

 

            If one were to invest in the integrated modeling, it would still be important that the modelers recognize and address the uncertainties for estimating electric sector impacts that are created by pending deregulation of that sector.  Ideally the analysis should employ a model that simulates the electric sector as a deregulated system, to the best extent possible.  Econometric models based on historical relationships would be of particularly dubious merit for this situation.  In addition, the analysis should include a careful summary of the likely implications of deregulation for uncertainties in the impact assessment in this sector.

 

Critical Assessment

 

            Table 1 identifies the expected magnitude of incremental impacts to economic variables as low to moderate relative to other topics, but with potentially significant effects in distributional impacts.  The suggested low-cost analysis would provide some information to try to identify risks of significant impacts to sectors.  It would also help elucidate qualitatively the ways that the impacts might be manifested, while not providing any quantitative estimates of these effects.  If the low-cost approach is selected, there is a risk that it might appear to overstate the impacts to businesses with large control requirements, while understating the impacts to consumers and to businesses that are not directly involved in emissions control or in providing emissions control equipment and services.

 

8.            Lifestyle Changes

 

            Some of the impacts that might be thought of as lifestyle changes are those related to price increases and job market restructuring resulting from control cost measures.  These were discussed above and are best thought of as part of the overall economic impact analysis.  They cannot be separated from the assessment of business impacts.  There are, however, additional types of impacts (both positive and negative in nature) that might fall under the term "lifestyle changes" and that would require a somewhat separate set of analysis steps.  These are discussed in this section.

 

            One impact that would fall under this topic is hidden costs that are borne by consumers in the form of time that they need to invest as individuals because of regulations.  An example might be vehicle inspection requirements.  If an enhanced inspection program requires more time on the part of car owners than would be spent without the enhanced inspection program, then the value of this additional time should be estimated in addition to the control measure's technical costs. 

 

Another type of hidden cost comes in the form of constraints on consumer choices that are directly altered by control strategies.  An example of the latter would be carpooling requirements.  If mandated, consumers will lose significant welfare associated with the convenience and independence of driving one's own car.  The potential magnitude of the welfare impact is reflected in the fact that direct control of consumer behavior is rare in the U.S. for environmental measures.  However, more subtle forms may arise.  For example, a control strategy that includes a parking tax will, to the degree that it is effective in altering behavior, have hidden costs to consumers in the form of lost convenience, et cetera, in addition to any tax collections.  The key point is that the hidden costs may be as significant as the direct financial expenses and these should not be ignored in assessing the socioeconomic impacts of control strategies that directly affect consumers' choices.

 

Lifestyle changes resulting from control strategies may not all be negative.  If a transportation control measure is effective in reducing congestion, it may result in less time consumption on the part of some individuals.  Other options, such as flextime and telecommuting, can alter worker lifestyles in positive ways that should also be accounted for.  Any assessment of lifestyle changes under this category should seek to identify both positive and negative potential effects, and give them equal consideration in discussion or valuation steps.

 

            Types of emissions control measures that are most likely to produce lifestyle changes are transportation control measures, energy conservation requirements and direct modification of product choice.  An analysis of such strategies must first determine the nature and degree of the lifestyle adjustment.  This can be a significant endeavor in itself.  Then, the value of the adjustment to the individual must be assessed.  There are quantitative tools for estimating these hidden costs.  In the case of time burdens, the quantitative cost may be reasonably well approximated by the number of hours times the affected person's wage rate.  If the hour burden is known, estimates are possible to develop.  In the case of behavioral alterations, more complex analysis is necessary, and beyond the scope of the SAMI budget. 

 

Low-Cost Method

 

            The suggested low-cost approach for addressing lifestyle changes would be limited to a qualitative review of control actions to identify the potential for lifestyle changes other than price pass-through effects.  Possible areas of concern and possible beneficial lifestyle changes would be highlighted, and evidence regarding the significance of the value of such changes would be described.  Quantitative costs would be provided for measures that involve only a direct added time-burden.  Coordination with SAMI would be important to ensure a clear understanding on how such control actions are evaluated in other parts of the Integrated Assessment.

 

Incremental Method

 

            The incremental method for these additional lifestyle changes would combine the results of the low-cost alternative (above) with an integrated quantitative analysis of economic impacts as described for the incremental method under topics 6 and 7.  To the extent feasible within the selected economic modeling framework, any monetized elements of lifestyle impacts and identified amenity changes in lifestyles would be translated into additional scenario input assumptions of the integrated economic model.  The non-quantifiable aspects of lifestyle changes would remain an important part of the summary of scenario results, however.

 

Critical Assessment

 

            Table 1 shows that lifestyle changes have been judged to be potentially important relative to other impact categories.  It is difficult to judge the quality of quantitative estimates of lifestyle values at this time because the set of control strategies has not yet been defined completely.  The level of complexity and the quality of any estimate would be closely tied to the manner in which the control strategy is defined and emission reductions are determined in the rest of SAMI’s Integrated Assessment effort.

 

COST ESTIMATES FOR ALTERNATIVE OPTIONS

 

            Table 2 provides ranges of cost estimates for the alternative options described above.  Individual estimates are provided by topic for both the low-cost and incremental methods described in the sections above.  The costs in Table 2 for the “Next Incremental Level” are total costs, and should not be added to the costs for the “Low-Cost Level”.  These cost estimates reflect our best judgment based on our own cost structures and experience.  Other estimates may be appropriate for contractors operating under different circumstances.


Table 2.  Summary of Analysis Cost Estimates

 

Topic

Low-Cost Level

Next Incremental Level               (total cost by topic)

Fishing

$40-60K; benefits transfer based on single best study in literature

$60-80K; region-specific data analysis & multi-study WTP estimate

Hiking/Enjoying Scenery

$40-60K; use two visibility studies referenced in text and qualitatively discuss issues of remaining bias & perceptibility of change

$60-80K; quantitative analysis of WTP by day types & review of uncertainty ranges based on additional studies

Sense of Place & Stewardship

$20-30K; qualitative review

$50-70K; focus groups

Mortality

$20-40K; summary of range based on existing national studies plus existing sensitivity analysis studies

$30-50K; quantitative ranges specific to data for SAMI region (ranges will be just as wide as in low-cost option)

Competitiveness effects (& jobs in at risk sectors)

$60-100K; cost-to-sales ratios and qualitative discussion of potential implications for businesses, jobs, and price effects, possibly supported by electric and other sector-specific case studies.

$150-200K; economic modeling:  would provide detailed quantitative info on these and other economic impact categories (e.g. income distribution effects, # jobs, etc).

Lifestyle changes

$20-40K; qualitative review of control strategies for potential hidden costs.

$20-40K; review of control strategies for potential hidden costs; to extent feasible, incorporate value of lifestyle or amenity changes into economic modeling step described above dotted line.

           

           

It can be seen that only a subset of these topics can be addressed within a budget of $130,000.  The final package to devise from among the options shown in Table 2 must come from the SAMI Subcommittee, and may vary in content over the range possible budgets for Phase II.  In addition, the Subcommittee has expressed interest in an alternative approach that is purely qualitative, but provides for coverage of the broader range of topics discussed in the May Workshop.  That option would be an alternative to any combination of the approaches summarized in Table 2, and is discussed in the next section of this report.

 

QUALITATIVE ASSESSMENT OF ALL TOPICS

 

            One quite different alternative to analyzing a subset of the above topics would be to rely on expert judgment regarding qualitative estimates of the impacts of scenarios.  Because this approach would avoid the steps of developing models and complex data linkages, it would be possible to consider most or all of the topics discussed in the May Workshop.  In fact, one might think of this option as consisting of a second iteration on the ranking matrix that we developed and discussed at the May Workshop.  In the first iteration, we found it possible to develop a general consensus on the priorities for analysis.  This was based on our significant prior experience on the full range of impact assessment areas.  A second iteration, based on better-defined information from the Integrated Assessment, should further improve the quality of those judgments. 

 

The second iteration on the ranking matrix would not start until Integrated Assessment results become available for SAMI control strategies.  At that point, there will be more clarity regarding the baseline, and size and specific types of air quality changes that are in question.  Even the qualitative discussion surrounding the matrix could then be more precise in terms of experts’ assessment of which of the associated socioeconomic impacts are likely to be significant.  It may be more possible to identify impact categories that are insignificant as well as those that merit careful discussion as input to the development of SAMI's final policy recommendations.

 

To keep such an effort within the bounds of a budget of approximately $130,000, we recommend structuring the effort in the following ways:

 

·                    No more than two control strategies would be assessed at the socioeconomic stage.  We would recommend that these be two of the most stringent options, but differing in significant ways in the mix of types of control measures included in each strategy.

 

·                    All judgments would involve comparisons of each selected control strategy relative to a single baseline.

 

·                    For each of the topics, the experts would identify the most relevant Integrated Assessment results that would drive these impacts, and obtain tabular summaries of those particular inputs.  The socioeconomic assessment team would define the results tables needed and these would be provided to them by SAMI for the one or two strategies to be addressed.

 

·                    Based on the observed absolute and percentage changes in relevant determining factors, the experts would discuss how projected changes are likely to affect socioeconomic activities and values.  Where possible, relevant quantitative estimates of socioeconomic impacts may be developed based on readily available data, but such results would only serve to help illustrate, inform, and guide the qualitative summary of impacts.

 

·                    Once the experts have come to some consensus on results and issues to emphasize, they would develop a brief written summary about the most salient social costs and benefits that could be reasonably anticipated for each of the strategies addressed.  These narratives would be designed to accompany any quantitative results from the Integrated Assessment that also would be presented to decisionmakers.

 

·                    The final product would also include a discussion about the degree to which any insights or conclusions of the above “bounding analysis” may be scaled down to approximate impacts/values for less stringent control strategies of similar structure.

 

One value of this qualitative approach is that it will provide a thorough written discussion about assessment issues for each of the topical areas.  This builds on the work the Phase I consultants have already communicated via presentation and slide packages.  There is another reason that the purely expert-judgment approach may provide more specific insights than have been possible in Phase I.  Different types of air quality changes affect each impact category (“topic”).  Once SAMI’s strategies have been defined and analyzed, the socioeconomic experts will be able to determine whether some of the air quality attributes are likely to change in more significant ways than others.  For example, ozone may be affected much less than visual range, and fish populations may be found to change in ways that are not particularly relevant to fishing values.  These examples are purely for illustrative purposes and are not intended as conjectures.  However, if they were to occur in this way, it may be possible to state with more confidence that the key impacts are more likely to be aesthetic than fishing or crop damage.  The experts could then focus their discussion in more detail on the specific forms of the visibility changes and implications for social value.

 

One drawback of this approach is that the conclusions resulting from such an assessment would not be documentable in a way that would ensure that experts’ conclusions would be “reproducible” by others.  Rather, SAMI would have to accept that the report conclusions would be the unique product and combined views of the particular set of experts that it selects to perform this work. 

 

Another potential drawback is that qualitative discussion might carry less weight in a decision process than numerical ones, no matter how imprecise.  At the same time, masses of quantitative results can fail to be effective if not carefully summarized and interpreted.  The qualitative assessment that this approach would generate could be thought of as the insights and interpretations that even numerical results require to be effective.  Thus, if the writing is properly targeted, it is possible that a purely qualitative consensus could be just as important to decision making as quantitative results.  The Socioeconomic Assessment Workgroup must make its own judgment as to whether it prefers a comprehensive but more qualitative approach or a narrower but more quantitative analysis selected from the options described in the first part of this report.



[1]                 Chestnut, L.D., and R. D. Rowe.  1990.  Preservation Values for Visibility Protection at the National Parks.  U.S. Environmental Protection Agency (#CR-813-686).  Research Triangle Park, North Carolina.

[2]               Rae, D. A.  1983.  “The Value to Visitors of Improving Visibility at Mesa Verde and Great Smoky National Parks.”  In  R. D. Rowe and L. G. Chestnut (eds.) Managing Air Quality and Scenic Resources at National Parks and Wilderness Areas.  Westview Press.  Boulder, Colorado.