Chapter 6: Measurement and Analysis of Benefits and Costs

1
Chapter 6 Measurement and Analysis of Benefits
and Costs

• Steven C. Hackett
• Humboldt State University

2
Measurement and Analysis of Benefits and Costs

• Chapter outline
• Conceptual analysis of benefit/cost estimation
  and its shortcomings
• Efficiency, PDV, and benefit/cost analysis
• Coase Theorem and efficiency in the absence of
  regulation

3
Measurement and Analysis of Benefits and Costs

• Chapter outline, continued
• Benefit/cost analysis and federal
  health/safety/environmental/ regulation
• Measuring benefits
• Quantitative risk assessment
• Contingent valuation method
• Travel cost method
• Hedonic regression method

4
Measurement and Analysis of Benefits and Costs

• Chapter outline, continued
• Measuring costs
• Direct compliance costs
• Indirect costs
• To productivity
• Reflected in market concentration

5
Measurement and Analysis of Benefits and Costs

• In this chapter, benefits refer to the
  benefits associated with additional environmental
  or natural resource preservation, conservation,
  or restoration.
• Likewise costs refer to the costs of
  additional environmental or natural resource
  preservation, conservation, or restoration.

6
Measurement and Analysis of Benefits and Costs

• Costs are usually relatively easy to estimate in
  dollar terms. Examples
• The additional cost of producing diesel engines
  that comply with more stringent particulate
  matter regulations.
• The additional costs and foregone revenues
  associated with certified sustainable timber
  harvest methods.
• The reduced commercial fishing revenues due to
  more stringent fishing regulations

7
Measurement and Analysis of Benefits and Costs

• Benefits are more difficult to estimate in
  dollar terms. Examples
• The improvements in human health associated with
  more stringent particulate matter regulations.
• The watershed benefits associated with certified
  sustainable timber harvest methods.
• The ecological and future gains to stocks
  associated with more stringent fishing regulations

8
Measurement and Analysis of Benefits and Costs

• To do benefit/cost analysis, the apples and
  oranges of benefits and costs must be placed on
  a common metric.
• Note that individuals make these same sort of
  comparisons every day when they compare the
  pleasure of a restaurant meal, for example,
  against the cost of the meal. We can aggregate
  these expenditures to estimate the benefits
  created by restaurant meals.

9
Measurement and Analysis of Benefits and Costs

• Recall from utilitarianism that the challenge is
  to find a common metric that serves as an
  indicator of social utility (benefits) and
  disutility (costs).
• One common metric is money and the monetization
  of benefits and costs.
• Another common metric is voting in democratic
  systems.

10
Measurement and Analysis of Benefits and Costs

• Benefit/cost analysis usually uses money as a
  measure of utility, and thus monetizing benefits
  and costs is an important aspect of such an
• analysis.

11
Measurement and Analysis of Benefits and Costs

• In the current environmental policy debate,
  benefit/cost analysis has become a highly
  charged, controversial issue. Some wish to
  increase the use of benefit/cost analysis in
  order to enhance the efficiency of government
  regulation.
• It can be argued, however, that benefit/cost
  analysis is inappropriate as the single deciding
  policy factor in many circumstances where its use
  is proposed.

12
Limitations of benefit/cost analysis

• Commensurability Using cost/benefit analysis as
  the single deciding factor in setting policy
  assumes implicitly that the values of all objects
  and states of affairs are commensurable, meaning
  that they can be ranked based on a single
  characteristic of value such as money or utility.
  
• What elements of an environmental choice may not
  be commensurable with one another?

Complexity and Interdependence Social and
ecological systems may be too complex and
interdependent to quantify comprehensively
through benefit/cost analysis, and it may not be
possible to adequately measure impacts when only
an element of a larger ecosystem is under
analysis.
13
Limitations of benefit/cost analysis

• Value of Human Lives Some of the benefits of
  environmental improvements include the reduced
  loss of human life. What are the policy
  implications of placing an infinite value on a
  life? Of measuring the value of a life based on
  earnings capacity?

Future vs. Current Generations What discount
rate is appropriate when bringing future impacts
into present discounted value? Will future
generations value things the same way we do? If
not, then how can we bring their values and
preferences into policy debates today that will
affect them?
14
Limitations of benefit/cost analysis

• Marginal Utility of Money When we monetize
  benefits and costs without regard to who receives
  them, we are implicitly assuming that a dollar
  generates the same incremental gain in pleasure
  or marginal utility to all people. Is this
  usually true?

While it is clear that monetization and
benefit/cost analysis capture at best only parts
of the total impact of a policy, and so should
not be considered a sole guide to policy, data on
benefits and costs can be informative and
valuable. If economic benefits arent estimated,
will policy makers assume they do not exist?
15
Efficiency

• The Pareto criterion is considered to be nearly
  impossible to satisfy in actual policy analysis,
  and so Kaldor-Hicks is the usual efficiency
  criterion used.

While the usual method of performing benefit/cost
analysis is to maximize the present discounted
value (PDV) of net monetary benefits (as
described below), an alternative method is to
select policies that generate the greatest amount
of monetary benefit for each dollar of cost,
called the benefit/cost ratio. The ratio method
tends to favor smaller projects, while the net
monetary benefit method tends to favor larger
projects. In the presentation that follows, we
will assume that the Kaldor-Hicks efficiency
criterion is applied to the PDV of net benefits.
16
Maximizing Net Present Discounted Value (PDV)
Dynamic Efficiency Revisited

• To calculate the PDV of net benefits, we must
  first estimate the flow of benefits and costs
  from various project alternatives for each year
  into the future (usually with a finite project
  time horizon). Then we choose an appropriate
  discount rate (the rate of interest charged if
  you loan money for a year rather than use that
  money yourself) and compute the PDV of the net
  benefits for each year into the future.

17
Maximizing Net Present Discounted Value (PDV)
Dynamic Efficiency Revisited

• As we learned in chapter 5, we can calculate the
  PDV of total net benefit (TNB) flows as follows

PDVTNB Si (TBi TC i)/(1 r)i , i0, n
Where TC total cost in a given time period, TB
total benefit in a given time period, r
discount rate, and i indexes periods from the
current. If emissions level z maximizes TNB, then
z
Note TNBmax ? (MB-MC)dz
0
18
An Illustrative Example of Benefit/Cost Analysis
 
 

 

19
An Illustrative Example of Benefit/Cost Analysis
 
 
MC
marginal benefits costs
MB
Quantity of environmental improvement
z
total net benefits
TNB
Quantity of environmental improvement
z

 

20
The Coase Theorem

• It has been pointed out that we may not need
  regulations to achieve the efficient level of
  pollution control described above. In particular,
  economist Ronald Coase argued that the efficient
  outcome may also be realized if we assign and
  enforce property rights and allow people to
  resolve pollution disputes through negotiation.

21
The Coase Theorem

• Suppose that citizens have a property right to
  clean air. If firms wanted to pollute, they would
  have to pay the citizens.
• Suppose firms have a property right to use the
  air as a waste sink. Then citizens would have to
  pay firms to stop pollution.

22
The Coase Theorem

• The Coase theorem is based on a very simple and
  intuitive argument. Suppose that environmental
  protection or enhancement benefits one group of
  people and imposes costs on another. If the
  benefits of environmental protection exceed the
  cost, then the positive total net benefits from
  cleanup can be thought of as a pie to be divided
  between members of society. The size of the pie
  is maximized at the efficient level of pollution
  control, such as the 50 percent level of control
  in the table in one of the previous slides...

23
The Coase Theorem

• What Coase and other economists found interesting
  is that the same cleanup outcome will occur
  regardless of who owns the right to control the
  environmental improvement.
• What differs is who pays.

24
The Coase Theorem

• The Coase Theorem will hold under the following
  conditions
• It is feasible and appropriate to assign
  property rights to aspects of the environment
  such as clean air, water, and earth
• There are positive net benefits from
  environmental improvement
• The transaction costs of coordinating people and
  conducting the negotiation process are low
• There is no free rider problem in gathering
  payment funds from a group of stakeholders
• Any agreement that is reached is legally
  enforceable

25
The Coase Theorem

• Why might transactions costs be high? How would
  high transaction costs limit the effectiveness of
  Coasian solutions to pollution problems?
• Why might there be a free-rider problem, and how
  would free-riding limit the effectiveness of
  Coasian solutions to pollution problems?

• These conditions make the Coase Theorem very
  difficult to satisfy in situations beyond small
  and localized settings involving small numbers of
  people.

26
Operationalizing Benefit/Cost Analysis in U.S.
Environmental Policy
Since Executive Order 12044, issued by President
Jimmy Carter in 1978, federal regulations such as
those for protecting health, safety, and the
environment are required to be cost-effective,
and federal agencies must quantify the benefits
and costs for the various regulations that they
administer.

• Most key elements of environmental legislation,
  however, such as the Clean Air Act and the Clean
  Water Act, set standards based on feasibility
  rather than on passing a benefit/cost test.

27
Operationalizing Benefit/Cost Analysis in U.S.
Environmental Policy
The Safe Drinking Water Act (SDWA), reauthorized
in 1996, now requires the Environmental
Protection Agency (EPA) to utilize benefit/cost
analysis for new regulations. While the SDWA
requires the EPA to publish these benefit/cost
analyses, it does not bind the EPA to reject
regulations based on their failure to pass a
benefit/cost test.

• The Clean Air Act (CAA) amendments of 1990
  required the EPA to conduct a benefit/cost
  analysis of the original CAA of 1970. Each 1 of
  compliance cost to the economy is estimated to
  have generated over 44 in benefits. It is
  interesting to point out that the original CAA
  was widely seen as being more costly than
  necessary.

28
Operationalizing Benefit/Cost Analysis in U.S.
Environmental Policy

• In early 2001 the U.S. Supreme Court ruled that
  the U.S. EPA did not have to consider compliance
  costs in promulgating clean air regulation under
  the aegis of the Clean Air Act (CAA), because the
  language of the CAA does not require the EPA to
  consider compliance costs in promulgating
  regulation. The CAA law only requires that there
  be a public health benefit. The case had to do
  with the EPA setting stringent new standards for
  ozone and particulate matter emissions.

29
Measuring Benefits

• Sometimes a portion of the benefits of an
  environmental improvement can be directly
  estimated from price. Example Access fees at
  parks (recreational benefits) or reduced health
  care expenses and lost work days (human health
  benefits).
• Other times, however, there is no direct
  dollar-denominated benefit, and environmental
  economists have developed special methods for
  estimating these non-market benefits.

30
Measuring the Health and Ecological Benefits of
Regulation Quantitative Risk Assessment (QRA)
and the Value of a Statistical Life

• So far in this course we have assumed that
  supply, demand, costs, benefits, etc. are
  deterministic (not subject to unforeseen changes)
  and known by all in advance.

31
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• Many environmentally damaging practices (such as
  pollution) generate harms to human health that
  are recognizable in overall populations, but have
  uncertain effects on any single individual. A
  natural example is the emission of toxic
  pollution into air or water, which elevates the
  risk of a person contracting cancer, emphysema,
  or various reproductive disorders. In this case
  one way to view the harm is the elevated
  likelihood of a person becoming ill or dying.

32
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• Thus a benefit of environmental protection is the
  reduction in the probability of a person
  experiencing specific adverse health effects.
• Health risk assessment provides a basis for
  quantifying the benefits of different regulatory
  options.
• Likewise, ecological risk assessment is used to
  evaluate the potential adverse effects that human
  activities have on the plants and animals that
  make up ecosystems.

33
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• The EPA's National Center for Environmental
  Assessment (NCEA) describes the following
  elements of the risk assessment paradigm

Hazard identification refers to identifying the
health problems caused by the pollutant. In the
case of human health risk assessment, hazard
identification uses both animal and human studies
to establish the likelihood that a pollutant will
generate harm to human health.
34
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• NCEA risk assessment paradigm

Exposure assessment involves an estimation of the
quantity of the pollutant that people breathe,
drink, absorb through the skin, or are otherwise
exposed to in a period of time. Exposure
assessment also includes an estimate of how many
people are exposed. In the case of air pollution,
for example, exposure assessment includes
measuring the quantity of air emissions from a
particular source, modeling how the pollutant is
transported and dispersed, estimating how many
people are exposed at various distances from the
emission source, and estimating the quantity of
the pollutant breathed by people who are exposed.
35
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• NCEA risk assessment paradigm

The dose-response relationship for a specific
pollutant or human activity describes the
association between exposure and the observed
response (health or ecological effect). In the
case of human health risk assessment, the
dose-response relationship is an estimate of how
different levels of exposure to a pollutant
change the likelihood and severity of health
effects. Just as in the hazard identification,
scientists use results of animal and human
studies to establish dose-response relationships.
36
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• NCEA risk assessment paradigm

The final step, risk characterization, is
presented in different ways to illustrate how
individuals or populations in human or ecological
communities may be affected.
Risk assessments play a direct role in the
formulation and economic assessment of
environmental policy. For example, section 108 of
the Clean Air Act (CAA) directs the Administrator
of the EPA to list pollutants that may reasonably
be anticipated to endanger public health or
welfare, and to issue air quality criteria for
them
37
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• The dominant benefit identified in benefit/cost
  analysis of the Clean Air Act (1970-90) was
  reduced premature mortality due to reductions in
  particulate matter, which contributed 16.6
  trillion of the estimated mean benefits of 22.2
  trillion (in constant 1990 dollars), or
  approximately 75 percent of the total economic
  benefit.

38
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• But how do researchers go from risk assessment of
  a pollutant such as particulate matter to the
  economic value of premature mortality prevented
  by regulation?

One method is the value-of-statistical-life (VSL)
approach. Warning Controversial
39
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life

• Economic estimates of the value of a statistical
  premature death usually use information on (i)
  people's willingness to pay for a reduction in
  the probability of premature death, or (ii)
  willingness to accept higher pay in return for
  higher risk.

• Stated preference approach Survey people on
  willingness-to-pay to avoid a given risk.
• Revealed preference approach Observed wage-risk
  relationships in actual jobs.

40
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• Suppose that a wage-risk study estimates that
  when the annual risk of premature death on the
  job increases by 0.0001 (1 in 10,000), workers
  receive an annual wage premium of 700 per year
  as compensation for this added risk. Assume that
  all other work characteristics are held constant.
  If we assume that those workers are fully
  informed and the labor market is competitive,
  then what is the implied VSL?

41
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• Concept
• Wage premium VSL (increased probability of
  death).
• With a bit of algebraic rearranging we get
• VSL (wage premium) (increased probability of
  death).

Plug in our value for increased probability of
death (0.0001) and a wage premium (700). What do
you get?
42
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• Based on analysis of 26 wage-risk and
  hypothetical willingness-to-pay studies, the EPA
  (1997) estimated a mean value of a statistical
  premature death avoided to be 4.8 million (in
  constant 1990 dollars).

43
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• What are some problems or issues associated with
  VSL estimation?

• Heres one How can we ethically assign a dollar
  value to human life?
• If we dont, then on what basis can a widow(er)
  sue for economic damages due to wrongful death?
  Is there no economic dimension to saving
  statistical lives?
• If we do, then what are we saying about high
  wage-earners lives vs. low wage earners lives?

44
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• VSL and high vs. low wage earners lives

Bowland and Beghin (1998) offers a good example
of VSL differences across rich and poor
countries. They estimated that the value of a
statistical life saved in Santiago Chile due to
reduced air pollution is approximately 600,000,
which is only about 12.5 percent of the 4.8
million value of an American statistical life
used by the EPA (1997). The logic of benefit/cost
analysis might then suggest locating hazardous
life-threatening industrial activity and toxic
wastes in the poorest regions of the world, which
many would consider an unacceptable example of
environmental injustice.
45
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• Other issues associated with VSL estimation

While most of the 26 VSL studies involved the
value of risks to middle-aged working people,
those who die prematurely from particulate matter
are more likely to be aged and past their working
years. Moreover, job-related risks are more
likely to be borne voluntarily, and involve the
risk of sudden and catastrophic death, while
pollution-related risks are borne involuntarily
and involve the risk of longer periods of disease
and suffering.
46
Quantitative Risk Assessment (QRA) and the Value
of a Statistical Life (VSL)

• While there are many problems with the VSL
  approach, simply ignoring the economic cost of
  premature death and leaving it out of
  benefit/cost analysis leads to a substantial
  underestimate of the benefits of environmental
  conservation and restoration.
• An alternative approach taken by Tengs et al.
  (1995) is to evaluate the cost of regulatory
  intervention per statistical life-year saved by
  the intervention. This type of analysis allows
  policymakers to allocate regulatory resources to
  those interventions that generate the most
  statistical life-years saved per dollar of
  intervention.

47
Estimating Non-Market Environmental Benefits

• Categories of non-market environmental benefits

Use Value
Nonuse Value
48
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

Contingent valuation uses a stated preference
approach, which means that survey research
methods are used to ask people to state
hypothetically how much they would be willing to
pay for a given environmental improvement.
49
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

CVM has been around for a while, but the method
was improved substantially after the Exxon Valdez
disaster, when NOAA assembled a panel of leading
economists to assess CVM as a reliable method for
damage assessment.
50
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

• Steps in CVM analysis
• Clearly identify scope of study.
• Focus group to suggest range of valuations,
  wording, etc.
• Development of survey instrument.

51
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

• CVM questionnaire design
• Fair and clear descriptions and images.
• Neutral wording.
• Questions on use and nonuse values, demographics,
  socioeconomics, etc.
• Referendum-style willingness-to-pay question.

52
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

• CVM survey method
• Random sampling of appropriate population.
• Repeated random sampling for different WTP values
  (e.g., 10, 20, 40, 60, 80).
• Many survey researchers recommend methods
  standardized by D. A. Dilman in his 1978 book
  Mail and Telephone Surveys The Total Design
  Method.

53
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

Hypothesis Inverse relationship between
percentage of yes responses to the WTP
question, and the dollar WTP value. If so, you
get a downward-sloping resource demand curve (
WTP on y axis, yes on x axis).
54
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

CVM is controversial because it is based on
hypothetical WTP values (stated preferences)
rather than actual spending (revealed
preferences).
And
55
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

CVM is controversial because it is the ONLY
economic method for estimating non-use values
(existence, option, bequest), AND
These nonuse values tend to be much larger than
active use values. Why do you think that is the
case?
56
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

A key methodological problem with some CVM
studies is embedding bias.
Embedding bias occurs when, for example,
respondents indicate that the value of saving a
500 acre area is worth just as much as saving the
500 acres PLUS an additional contiguous 50,000
acres.
57
Estimating Non-Market Environmental Benefits

• Contingent Valuation Method (CVM)

The CVM has become a political issue. Since it is
the only way to estimate nonuse values, it is an
important valuation method. See the Elwah River
Ecosystem Restoration Environmental Impact
Statement (June 1995) for an interesting
application of the CVM.
58
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

Money and time spent traveling to a recreational
area provides revealed preference information
regarding use values. TCM does NOT provide
information on nonuse values.
59
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

Single-site model Used to estimate the number of
trips taken by a person over a year (or season of
use) as a function of the cost incurred to get to
the site, as well as personal characteristics
(age, sex, income, experience level) and travel
cost to alternative sites. These explanatory
variables are usually gathered using survey
research methods.
60
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

Steps in a single-site TCM study 1. Define the
boundaries of the site to be studied. 2. Define
the recreational uses and season of use. 3.
Develop a sampling strategy for surveying
recreational users.
61
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

Steps in a single-site TCM study 4. Specify a
demand estimation model, including all the
variables to be used in that model. 5. Address
the issue of multiple purpose recreational
trips. 6. Design and conduct the survey.
62
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

Steps in a single-site TCM study 7. Measure
trip cost. 8. Estimate site demand using trip
cost and data from the survey. 9. Calculate the
access value of the site.
63
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

A travel cost study may focus on certain types of
recreational uses, or researchers may be able to
aggregate similar classes of recreational use. A
key decision regarding sampling strategy is
whether to conduct on-site surveys of visitors,
or a random survey of the overall population.
64
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

On-site surveys have the advantage of being
better targeted, requiring fewer survey
solicitations to get the desired sample size. On
the other hand, on-site surveys omit from the
sample people who take zero trips during the time
period under study, which impairs the estimation
of the vertical intercept of the site demand
curve. Likewise, an on-site survey strategy will
over-sample frequent visitors (endogenous
stratification).
65
Estimating Non-Market Environmental Benefits

• Travel Cost Method (TCM)

Off-site surveys of the overall population can
avoid these problems, but a very large number of
surveys must be sent out in order to receive back
a sufficiently large sample of completed
questionnaires from recreationalists who visit
the site. Moreover, off-site surveys require the
researcher to make some assumptions about the
geographical scope of the market containing
site visitors.
66
Travel Cost Method (TCM)
The researcher will develop a demand model that
explains the number of site visits as a function
of the visitors trip cost price, as well as
income, trip cost to substitute sites, and other
visitor characteristics that might influence
their frequency of visits.
67
Travel Cost Method (TCM)
A further complication in using the TCM has to do
with whether trip cost was expended for the sole
purpose of visiting the recreational site under
study, or whether the trip encompassed multiple
purposes including other sites or visiting
friends or family along the way.
68
Travel Cost Method (TCM)
Single vs. multi-purpose travel If the nature of
the study is to focus on day-use trips, then one
can usually assume that the trip cost is
single-purpose in nature. In contrast, if the
recreational use identified in the second step
involves one or more overnight stays, the
researcher should include the question of
single-purpose or multiple-purpose travel in the
questionnaire.
69
Travel Cost Method (TCM)
Single vs. multi-purpose travel The researcher
can then choose whether to drop multiple-purpose
trips from the dataset, or develop a more
complicated demand model that explicitly
incorporates multiple purposes for travel.
70
TCM Questionnaire Design

• In addition to introductory material, the TCM
  questionnaire will include questions for the
  variables identified in step four, as well as the
  number of site visits over the time period in
  question and detailed questions on travel cost
  and other aspects of the most recent visit, which
  are presumably easiest for the subject to recall.

71
TCM Trip Costs

• Trip cost in the TCM includes the cost of travel,
  opportunity cost of time, and other direct
  expenditures required for a site visit. Travel
  cost via automobile would be measured as the
  product of round-trip distance multiplied by the
  average cost per mile from an authoritative
  source such as the U.S. Department of
  Transportation.

72
TCM Trip Costs

• The opportunity cost of a visitors time, a
  component of trip cost, is difficult to estimate,
  and represents a large portion of the economic
  value estimated from the TCM.
• Many researchers make the assumption that some
  portion of time in transit and at the site
  involves the opportunity cost of foregone
  earnings from work, though this is obviously
  flawed for retired people, students, the
  unemployed, and those with paid vacation time.

73
TCM Trip Costs

• It is difficult to apportion part of the capital
  cost of recreational gear and equipment owned by
  the recreationalist to a particular trip. While
  rental rates on equipment can be used to estimate
  equipment and gear costs on a per-trip basis,
  many researchers omit these costs altogether.

74
TCM Site Demand Function

• Once the data are tabulated from returned
  surveys, the researcher uses econometric
  techniques to estimate a site demand function

75
TCM Site Demand Example

• Example
• Site Visits 80,000 0.65INC 40TCSUB
  130TCTARGET
• where Site Visits is the annual number of visits
  to the target area under study, INC per-capita
  personal income, TCSUB travel cost to a
  substitute destination, and TCTARGET travel
  cost to the target area.

76
TCM Site Demand Example

• Suppose that the mean values of the independent
  variables are
• INC 40,000
• TCSUB 500
• TC 100
• Then Site Visits 80,000 0.6540,000 40500
  130100 113,000

77
TCM Site Demand Example

• To estimate a site demand function, we
  incrementally increase TC price holding all
  other independent variables constant
• TC 200, Site Visits 100,000
• TC 300, Site Visits 87,000
• TC 400, Site Visits 74,000
• TC 500, Site Visits 61,000

78
TCM Site Demand Example

• We can now plot these data to get a site demand
  curve

969.23
Site demand curve
Consumer surplus 49.11 million
100
Site Visits
113,000
79
Estimating Non-Market Environmental Benefits

• Hedonic Regression Method (HRM)

HRM is based on the idea that non-market
environmental qualities (views, clean air, short
commute times, safe and friendly neighborhoods,
nearby recreational opportunities) are bundled
with location-specific goods and services
(houses, apartments, jobs) that are traded in
markets.
Thus the value of environmental qualities should
be embedded in home prices, salaries, etc.
80
Estimating Non-Market Environmental Benefits

• Hedonic Regression Method (HRM)

• Examples
• Homes with ocean views are far more expensive
  than similar homes without views.
• The value of a nicely maintained urban park is
  reflected in part in the higher value of homes
  within walking distance of the park.
• Many people are willing to accept lower salaries
  in return for living where there is high quality
  of life and a healthy environment.

81
Estimating Non-Market Environmental Benefits

• Hedonic Regression Method (HRM)

HRM requires that the researcher understand how
to perform regression analysis, an aspect of
econometrics and statistics that estimates
quantitative relationships. Illustrative example
Value of home 50,000 10SQFT - 3AGE -
50CRIMERATE 10,000VIEW 50AIRQUAL -
1,000MILESFROMPARK
82
Estimating Non-Market Environmental Benefits

• Hedonic Regression Method (HRM)

Value of home 50,000 10SQFT - 3AGE -
50CRIMERATE 10,000VIEW 50AIRQUAL -
1,000MILESFROMPARK
Assuming that the regression is done properly and
the model is appropriate and the coefficients are
significant, the above regression equation states
that, all else equal, each additional mile from a
park reduces a homes value by 1,000.
We could sum the additional value of homes near
an urban park to estimate the non-market value
of the park.
83
Estimating Non-Market Environmental Benefits

• Benefits Transfer

Instead of estimating value from scratch, the
benefits transfer approach suggests that the
researcher can instead apply per-unit values from
a study of a similar resource.
Example Value of a chinook salmon on the Klamath
can probably be transferred if one is studying
salmon issues on the nearby Eel or the Mad river.
84
Estimating Costs

• The cost of environmental regulation can be
  divided into direct and indirect costs.

Direct compliance costs include pollution
abatement and expenditures by firms, consumers,
and government, as well as opportunity costs that
can be attributed directly to regulation.
85
Direct Compliance Costs

• Firms bear direct compliance costs for such
  things as additional
• capital equipment needed to meet regulatory
  standards (as well as technology expenses such as
  RD).
• specialized inputs (e.g., low sulfur coal)
  needed to meet regulatory standards.
• personnel to meet regulatory standards.
• Additional planning expenses to meet regulatory
  standards (e.g., THPs, EIS/EAs).
• Opportunity costs (foregone revenues, e.g. when
  logging is prevented in riparian zones).

86
Direct Compliance Costs
Consumers bear direct compliance costs for things
like as vehicle inspections, emissions control
equipment on their cars, any added expenses from
less toxic cleansers and solvents, and water
conservation devices. Local, state, and federal
governments bear direct compliance costs for
activities such as the drafting of regulations,
and the monitoring, oversight, and enforcement of
those regulations.
87
Direct Compliance Costs

• Until 1995 The U.S. Department of Commerce
  reported on the cost of pollution abatement and
  control expenditures (PACE) (omitting opportunity
  costs) in its Survey of Current Business. PACE is
  divided into spending on
• pollution abatement (about 90 of total PACE).
• government regulation and monitoring, and
  research and development (the latter two are
  about 10 percent of total PACE).

88
Direct Compliance Costs
In current (unadjusted) dollars, 1993 PACE was
estimated to be 109 billion. Since
inflation-adjusted PACE spending has been growing
faster than real gross domestic product (GDP),
the share of GDP coming from PACE spending
increased and was estimated to be 1.8 percent in
1993. This represents a slight increase from the
1.7 percent of GDP that was estimated for PACE in
1987.
89
Direct Compliance Costs
The private business sector accounts for
two-thirds of the total spending on the pollution
abatement element of total PACE.
90
Indirect Compliance Costs
Indirect costs occur as feedback effects from
environmental regulation. For example,
regulation that raise firm's marginal costs will
cause higher market prices and thus adversely
affect consumers and change the composition of
goods and services produced in the economy.
91
Indirect Compliance Costs
Indirect cost of compliance on economic growth
Resources allocated to purchase pollution
abatement and control equipment may have
otherwise gone to investment in
productivity-enhancing innovation, yielding an
opportunity cost of a slower rate of economic
growth (assumes economic growth is unambiguously
good).
92
Indirect Compliance Costs
Indirect cost of compliance on economic growth,
continued
The EPA (Ben/Cost Anal. Of CAA 70-90) reports
that compliance with the Clean Air Act had the
greatest sectoral impact on large energy
producers and consumers, particularly those
sectors that relied most heavily on consumption
of fossil fuels. A key aggregate impact of the
CAA was an estimated one twentieth of one percent
per year reduction in economic growth due to
CAA-mandated investment in capital PACE reducing
the level of investment available for capital
formation.
93
Indirect Compliance Costs
Indirect cost of compliance on market structure,
competition, and pricing
With regard to microeconomic indirect costs, one
of these has to do with increased industry
concentration (a few large firms rather than many
small ones) caused by increases in fixed cost due
to environmental regulation.
94
Indirect Compliance Costs
Numerical example of the indirect cost of
compliance on market structure, competition, and
pricing
Suppose that there are currently 10 refineries in
a state. Fixed costs for a refinery are
40,000,000 per year. If the average variable
cost per gallon of producing gasoline is 1.00,
and each gallon is sold for 1.80, then how many
gallons must a refinery sell each year to break
even?
Breakeven Q/year TFC/(P-AVC)
40,000,000/(0.80) 50 million gallons/year. If
the demand by gasoline stations in the state each
year is roughly 500 million gallons/year, then
the market can support all 10 refineries at
breakeven.
95
Indirect Compliance Costs
Numerical example of the indirect cost of
compliance on market structure, competition, and
pricing
Now suppose that new environmental regulations
raise annualized fixed costs to 80,000,000 per
year. If all else remains equal, what is each
refineries' breakeven output/year, and how many
refineries can the market support?
Breakeven Q/year TFC/(P-AVC)
80,000,000/(0.80) 100 million gallons/year.
If the demand by gasoline stations is still 500
million gallons/year, then now the market can
only support 5 refineries, and the others must
close.
96
Indirect Compliance Costs
Numerical example of the indirect cost of
compliance on market structure, competition, and
pricing
With fewer refineries the risk of collusion and
exploitation of consumers (via higher prices)
rises. This increased risk of higher prices due
to there being fewer firms is an indirect cost of
regulation.
97
Excel Exercise
Label a new Excel worksheet in your Econ 523 file
TCM. Consider the following simplified fitted
TCM linear regression equation VISITS 50,000
0.5INC 50TCSUB 100TCTARGET, where VISITS
is the annual number of visits to the target area
under study, INC per-capita personal income,
TCSUB travel cost to a substitute destination,
and TCTARGET travel cost to the target area.
Suppose that the mean value of INC is 50,000,
the mean value of TCSUB is 500, and the mean
value of TCTARGET is 200. Enter all this
information into a spreadsheet.
98
Excel Exercise
a. What does the fitted TCM regression forecast
as the annual number of visits to the target area
when the independent variables are evaluated at
their mean value? Briefly explain in words. b.
Create a table with two columns, one labeled
additional travel cost and one labeled
estimated number of visits. Continue to use the
mean values for INC and TCSUB in the regression
equation, but now progressively increase the mean
value of TCTARGET in 10 increments. This will
tell us the marginal effect of an increase in the
travel cost price on the estimated number of
visits. Record (TCTARGET increment) and the
associated forecast number of visits in each row
of the table. Stop adding increments of 10 to
the mean value of travel cost when the forecasted
number of visits reaches zero.
99
Excel Exercise
c. Plot these data in a diagram where the y
axis is labeled travel cost and the x axis is
labeled estimated number of visits. Label the
curve as the resource demand curve. It should
be downward-sloping and linear. d. Use geometry
to calculate the area under this resource demand
curve, and above the 200 average travel cost
price line. This figure represents a simplified
estimate of the net economic recreational use
value (consumer surplus) derived from the target
area. Provide a brief narrative economic
interpretation of your findings. Given that the
mean value of TCTARGET can be interpreted as the
average actual price paid for the recreational
experience, explain why the area under the
resource demand curve represents a simplified
estimate of the consumer surplus derived from
recreational visitation.