Economics 1100

Document Sample
Economics 1100 Powered By Docstoc
					Ch. 3 Valuing the Environment:

   Benefit-Cost Analysis has always been an important
    tool in natural resource economics
     • but the Exxon Valdez oil tanker fiasco on the Bligh
       Reef in Prince William Sound, Alaska (March 24,
       1989) heightened awareness among the general
       public about the techniques
     • $2.1 billion spent on cleanup
     • $303 million to compensate fishermen’s losses
     • $900 million settlement (paid over 10 years)
     • Jan. 2004: U.S. District Court for Alaska awarded
       $4.5 billion in punitive damages
     • U.S. Supreme Court 8/27/2008 lowered punitive
       damages to $507.5 million. See
     •  3
      Why Value the Environment?
   If you don’t place some value on the
    environment because you believe its
    intrinsic value is infinite:
     o The value that may be used is $0!!!
     oBenefit/cost analysis is becoming more
       widely used in many disciplines.
     oWe still need to consider nonmarket
               Valuing Benefits
   Assessing the magnitude of pollution damage:
    – 1) Identify affected categories
       » Two methods:
            1) animal experiments but extrapolation to
             humans is difficult
            2) Epidemiology- statistically estimate
             relations between health and pollution
             levels in various localities. Must control for
             variety of factors
    – 2) Estimate the physical relationship between
      emissions and damage caused
    – 3) Estimate response by affected parties to
    – 4) Monetary value of damages                        5
                Types of Values
   Stock vs. flow values:
     – Stock value is the present value of the future stream
       of flow values
     – Total economic value: 3 components
        » 1) use value- hunting, fishing, photography,
          viewing, etc.
        » 2) option value- preserve future availability to use
        » 3) nonuse value- derive satisfaction knowing a
          resource exists
     – Total willingness to pay (TWP)
     – TWP = use value + option value + nonuse value
      Classifying Valuation Methods
 Table3.1
 Methods      Observed behavior         Hypothetical
   Direct     Mkt. Price                Contingent valuation
               Simulated Market

   Indirect   Travel cost               Attribute-based
               Hedonic Property Values   models
              Hedonic wage values       Conjoint Analysis
              Avoidance expenditures    Choice experiments
                                        Contingent ranking

     Valuing Nonmarketed Goods
   Four primary indirect observed behavior methods:
     – 1) Travel cost model (Clawson-Knetch Model)
     – an indirect, observed behavior method
        » it is nearly free to go to Yosemite National Park
          (only $20/week for a family, compared to $1,400 for
          a week at Six Flags over Texas.)
        » but people reveal their value of Yosemite by
              1)Monetary expenses: food, lodging,
              2)Value of time (opportunity cost of lost
        » now let’s look at “Saving Tuolumne” article &
          travel cost model used therein
Valuing Nonmarketed Goods
   Four primary indirect observed behavior methods:
    – 2) Hedonic (pleasure seeking) models
       » use statistical models to determine how much
         someone is willing to pay for clean air, reduced
         probability of death (more safety)
       » homes: in cleaner air areas sell for more money
       » workers: paid more for riskier jobs, ceteris paribus
    – 3) Avoidance expenditures:
       » e.g. Air bags or smoke detectors: How much will
         you pay to reduce probability of death by say
         0.0001? $500? Implies value of your life is at least
         $5,000,000 = $500/.0001
     Valuing Nonmarketed Goods
   4) Contingent Valuation Model (CVM)
        » survey individuals, ask them how much they would
          pay to prevent some environmental damage, e.g.,
          save the whale, African elephant, Trinity River
        » use statistical models to determine how much
          someone is willing to pay
        » CVM became very important after Exxon Valdez oil
          spill in Prince William Sound in Alaska
        » Criticisms: Possible Biases: People will say what
          they think you want them to say in a survey, or lie
        » Surveys must be carefully constructed
             It will raise your taxes, are you still willing to
              pay $X?

     Valuing Nonmarketed Goods
•   Indirect hypothetical methods
    • Conjoint analysis: similar to contingent
      valuation, but no willingness to pay is
      stated. Respondents choose between
      alternate states of the world.
    • See Table 3.3

    • Contingent ranking: rank order
      hypothetical situations differing in terms
      of environmental amenities
 4 types of potential survey biases
1) Strategic bias
2) information bias
3) starting point bias
4) hypothetical bias

Using Geographic Information Systems
    (GIS) for Economics Valuation

 –Spatial relationships are
 –Hedonic property value models
  can used spatial relations
   »E.g. air quality, open space

       Valuing Human Life
  –How much will you pay to
    reduce probability of death by
    say 0.0001? $500? Implies
    value of your life is at least
    $5,000,000 = $500/.0001
See Table 3.4 for cost of saving a
Are these efficient regulations?
“The Contingent Valuation Debate: Why
      Economists Should Care”
            Paul Portney
    Journal of Economic Perspectives,
    Vol. 8, No. 4- Fall 1994, pp. 3-17.

 ContingentValuation: willingness to
 pay revealed by respondents is
 contingent on the constructed or
 simulated market presented in the
 1) Origins of the Contingent Valuation
 2) Describing the Methodology
 3) Moving to the Policy Arena
 4) The NOAA Panel
 5) The Importance of the C.V. debate
   Natural Resource Damage Assessment
   New Regulations
   Putting Theory into Practice       16
    1) Origins of the C.V. Method
   1st reference to C. V.: Ciriacy-Wantrup (1947) - benefits of
    preventing soil erosion
   Davis (1963) implemented first contingent valuation survey
    to elicit values of recreational area. He compared them with
    “travel cost” method.
   Harold Hotelling first suggested to the National Park Service
    in 1947 that the price someone “paid” to enter the park
    varied according to their travel cost.
   John Krutilla published “Conservation Reconsidered,” in
    1967, arguably the most influential paper ever written in this
    area. He suggested that many environmental assets had
    “existence values,” e.g., scenic wonders, genetic diversity.
    Often referred to as “nonuse values.”
   But, not until late 1980s did C.V. receive any notoriety. 17
 2) Describing the Methodology
 Main   elements:
 1) Survey must describe the program
 2) Survey must contain a mechanism for
  eliciting the value. E.g. “Your tax bill will
  increase $X under this new program. Will
  you vote “yes” or “no” for the program?
 3) Survey collects information on
  socioeconomic characteristics of respondents

        3) Moving to Policy Arena
 Why has the controversy entered real world?
 3 reasons
     1) Comprehensive Environmental Response, Compensation
      and Liability Act of 1980 (CERCLA), more commonly
      known as “Superfund.” Gave gov’t right to sue for dmages
      to natural resources for which they were trustees. Dept. of
      Interior (DOI) was ordered to write regulations and
      eventually courts required DOI to give equal weight to use
      and nonuse values.
     2) Exxon Valdez oil tanker spilled 11 million gallons on the
      Bligh Reef in Prince William Sound, Alaska (3/24 1989)
     3) Oil Pollution Act of 1990 ordered Dept. of Commerce,
      through National Oceanic and Atmospheric Administration
      (NOAA) to write regulations governing damage assessment.
          4) The NOAA Panel
 NOAA’s general counsel asked Nobel laureates
  Kenneth Arrow and Robert Solow to chair a panel
  of experts to advise NOAA on the following
 “Is the contingent valuation method capable of
  providing estimates of lost nonuse or existence
  values that are reliable enough to be used in natural
  resource damage assessments?”
 Report was published in the Federal Register on
  January 15, 1993.
 It made both sides unhappy.

         4) The NOAA Panel
 Opponents    were upset with “bottom line”
    CV “can produce estimates reliable enough
     to be starting point of a judicial process of
     damage assessment, including lost passive-
     use values.”
 Proponents of C.V. were upset with the
  stringency of the guidelines for C.V.

          4) The NOAA Panel
 Seven  of the most important guidelines:
  – 1) Use personal interviews; second best is
    telephone; third best is mail surveys.
  – 2) Should elicit willingness to pay to
    prevent a future incident rather than
    minimum compensation required for an
    incident that has already occurred.
  – 3) Use the referendum format: “Would you
    vote for program X if it raised your taxes
    Y dollars?”                               22
          4) The NOAA Panel
 Seven   of the most important guidelines:
  – 4) Scenario must accurately and understandably
    describe the expected effects of the program.
  – 5) Must contain reminders that willingness to pay
    for the program would reduce the amount they
    would have available to spend on other things.
  – 6) Must remind them that substitutes are
  – 7) Must include follow up questions to ensure
    that respondents understood the choice they
    made and the reasons for the choice.

   5) The Importance of the
  Contingent Valuation Debate
 Natural   Resource Damage Assessments:
   C.V.  M. will likely influence future damage
   for example, a CVM conducted for the state of
    Alaska in the wake of the accident estimated the
    losses at nearly $3 billion (Carson, et al., 1992)
 New   Regulations
   most significant applications of CVM will
    involve benefit and cost estimates of proposed
    regulations under Superfund and other laws

   5) The Importance of the
  Contingent Valuation Debate
 New  Regulations
   existence values are not necessarily unique
    to environmental policy
   business community tends to oppose use of
   but businesses could use CVM to show
    that persons derive utility from knowing
    that persons have jobs (which might be
    destroyed due to a new regulation).
    5) The Importance of the
   Contingent Valuation Debate
 Putting Theory into Practice
  – the scrutiny of CVM has led some
    economists to think more deeply about
    cognitive processes, rationality, and the
    nature of preference for all goods, public
    or private. We may develop an improved
    theory of preference and choice.
  – CVM is a new technique and may face
    many of the criticisms similarly faced by
    experimental economics                     26
 Valuing the Environment Through
       Contingent Valuation
             W. Michael Hanemann
       Journal of Economic Perspectives,
       Vol. 8, No. 4- Fall 1994, pp. 19-43.

 Conducting   Reliable Surveys
  – serious surveys avoid convenience sampling such
    as stopping people in the street; they employ
    statistically based probability sampling
  – only study where people were stopped for a few
    minutes in a mall was one performed for Exxon
      Conducting Reliable Surveys
 Until mid 1980s, most contingent valuation
  surveys used some version of an open-ended
  question, like “What is the most you would be
  willing to pay for…?”
 Since then, most major CV studies have used
  closed-ended questions like “If it cost $x, would
  you be willing to pay this amount?” or “If it cost
  $x, would you vote for this?” Different people
  are confronted with different dollar amounts.
  Plotting the proportion off “yes” responses
  against the dollar amount traces out the
  cumulative distribution function of willingness-
  to pay.                                        28
        Conducting Reliable Surveys
 The   NOAA panel considered a debriefing
  session essential to determine if the person
  really understood the questions.
 Using the median response is usually better than
  the mean, which is very sensitive to responses
  of the highest bidders.
 Diamond and Hausman select studies from the
  C.V. literature that violate most of the precepts.
  None uses in-person interviews; many are self-
  administered; most use open-ended questions;
  none is cast as voting; many ask questions with
  little detail.                                 29
                Objections to Surveys
 Four common objections to surveys:
 1) Surveys are vulnerable to response effects –
       Small changes in wording can change survey response; need
        to use rigorous cognitive techniques to test for differences
   2) The survey process creates the values
     economists & psychologists have often assumed that
      consumers use a “top-down” or “stored-rule” decision process
     but now evidence indicates that all cognition is a constructive
      process; general principle: people are cognitive misers: they
      tend to resolve problems of reasoning and choice in the
      simplest way possible.
     For non-habituated and complex consumer choices, people
      often make “bottom-up” decisions; that is they make a
      decision rule at the moment they need to use it. Real issue is
      whether preferences have a stable construct.               30
              Objections to Surveys
 Four common objections to surveys:
 3) Ordinary people are ill-trained for valuing the
    but ordinary people are allowed to vote for many
     complex issues.
 4) Survey responses can’t be verified
    3 ways to validate CV results:
      1) replication
      2) comparison with estimates from other sources
      3) comparison with actual behavior where this is
    when CV measures direct use values, we can
     compare with estimates obtained through indirect 31
               Objections to Surveys
   4) Survey responses can’t be verified
     – Knetsch and Davis (1966) conducted the first test,
       comparing CV and travel demand estimates of
       willingness-to pay for recreation the Maine woods.
       The difference was less than 3 %.
     – Over 80 studies that compare C.V. and indirect
       methods. Results are often fairly close: overall, the
       CV estimates are slightly lower than revealed
       preference estimates and highly correlated w/ them.
     – Ideal validation is direct testing of contingent
       valuation predictions against actual behavior. About
       10 such tests exists. Diamond and Hausman mention
       only five of these. The ones not mentioned yield 32
       results quite favorable to C.V.
                 Objections to Surveys
   4) Survey responses can’t be verified
    – 5 studies showing validity of C.V.
        » 1) Bohm (1972): willingness to pay to see TV program in Stockholm
        » 2) Bishop and Heberlein (1990): Wisconsin deer hunting permits:
          willingness to pay was $31 in real sale versus $35 in hypothetical sale.
        » 3) Dickie, Fisher, and Gerking (1987) boxes of strawberries door-to-
        » 4) Carson, Hanemann and Mitchell (1986) water quality bond election
          in California in 1985. Predicted 70-75% yes, actual was 73%.
        » 5) Cummings, Harrison and Rutstrom (1993) emphasized the
          importance of “reason to say no”
        » Diamond and Hausman focus on studies that showed significant
          difference in CV and actual; most had flaws such as not asking to vote
          for higher taxes, using an environ. group for hypothetical treatment,
          comparing hypothetical phone responses to actual mail responses
          (phone response rates are much higher than mail).

         C.V. and Economic Theory
 Critics of C.V. question the validity of the variables in a
  utility function. But according to Gary Becker (Nobel
  laureate) economists generally assume that “individuals
  maximize welfare as they conceive it.”
 Two criticisms raised by Diamond and Hausman
  concern income elasticity of willingness-to-pay and
  “embedding effect”
    income elasticity estimates are generally consistent
     with demand for state and local government services
     in the U.S. (0.3 to 0.6) and with charitable giving (0.4
     to 0.8).
           C.V. and Economic Theory
   “Embedding effect” criticism: means several different
    things. Generally the misconception is that with CV
    you get the same willingness to pay (WTP) if you value
    one lake, two lakes, or ten lakes. Three notions:
    – 1) “scope effect”: WTP varies inadequately with changes in
      the scale or scope of the item being valued
    – 2) “sequencing effect”
    – 3) “subadditivity effect:”
    – scope effect: most studies show that scope does matter. Two
      exceptions. Desvouges et al. (1992) in shopping mall survey
      elicited WTP for preventing the deaths of migratory
      waterfowl. 2,000, 20,000, and 200,000 out of 85 million birds
      might die each year from exposure to waste-oil holding ponds
        C.V. and Economic Theory
– How much should willingness to pay vary with scope?
– Diamond and Hausman assert that economic theory
  requires it to increase more than proportionately with
  the number of bird deaths. Let qF = q0-qR+qS, where
  q0 is original bird population, qR is the number at risk,
  and qS is the number saved. Diamond assumes that
  people should care only about the ultimate number of
  birds, qF. He assumes that preferences are
  quasiconcave in q0. The two assumptions together
  imply quasiconvexity in qR, which is what makes the
  elasticity of WTP with respect to qR greater than unity.
  His conclusion depends heavily on the assumption of
  perfect substitution between q0, qR, and qS.           36
   Contingent Valuation: Is Some
  Number Better than No Number?
    Peter A. Diamond and Jerry A. Hausman
       Journal of Economic Perspectives,
       Vol. 8, No. 4- Fall 1994, pp. 45-64.

 Discusses  how to judge the content in C.V.
  surveys together with evidence from surveys
  that have been done. Tests for consistency
  indicate that C.V. responses are not
  consistent with economic theory.
 Main flaw: “embedding effect.”              37
     Judging Surveys of Willingness to
           Pay for Public Goods
   A number of bases exist for forming judgments about whether
    respondents are answering the right question and whether the
    response is roughly correct.
     – 1) Incredibility test
     – 2) verbal protocol analysis shows that
        » people have difficulty forming their preferences
     – 3) patterns of WTP responses across individuals
        » income elasticities are too low compared to charitable
           giving results
     – 4) patterns of WTP responses across surveys
        » people would pay more to preserve visibility in Grand
           Canyon when it was only question, as compared to when it
           was third question.
       Judging Surveys of Willingness to
             Pay for Public Goods
   4) patterns of WTP responses across surveys
         » embedding effect: people would pay the same to save one,
           two or three stands of timber. Neither income or substitution
           effect can plausibly explain the embedding effect.
         » This seems to confirm the “warm glow” hypothesis.
           Individuals are primarily expressing support for the
           environment that does not vary much with small changes in
           the precise environmental change being described.
         » “Adding up test”: willingness to pay for X and Y (together)
           should be the same as the sum of the willingness to pay for X
           and the WTP for Y, having been given X. Schulze et al.
           (1993) found WTP for complete cleanup was $72.46 while
           WTP for partial cleanup was $72.02. Thus the responses39
           must have been dominated by a warm glow effect.
   Issues in Benefit Estimation
 Primary   versus secondary effects

 Tangible   versus intangible benefits

    Approaches to Cost Estimation
   The Survey Approach

   The Engineering Approach

   Combined Approach

          Treatment of Risk
 Expected   present value of net benefits
  for policy j =
 EPVNB = PiPVNBij (sum over i), j =
 Arrow and Lind argue that risk
  neutrality is appropriate. Total cost of
  risk bearing is small for public
 Discount rate can have a major impact
  on net benefits of a project.
   Choosing the Discount Rate
 Discount  rate can have a major impact
  on net benefits of a project.
 Critical appraisal
  US. Army Corps of Engineers water projects’
   ex ante benefit estimates are too high. We will
   say more on this in water chapter.

    Cost-Effectiveness Analysis
 When    it is difficult to construct a benefit-cost
  estimate, cost-effectiveness can provide
  guidance. For a given environmental goal,
  choose the lowest cost option of obtaining it.
 Second Equimarginal principle (the cost -
  effectiveness equimarginal principle): least
  cost of achieving an environmental target will
  have been achieved when the marginal costs
  of all possible means of achievement are
    E.g. pollution control from various sources

            Impact Analysis
 When   it is difficult to construct a benefit-
  cost estimate or a cost-effectiveness
  estimate, use an impact analysis.
 However, it only lists all the impacts. No
  dollar values are given. Difficult to
  compare impacts.
 Current impact analysis are more
  sophisticated than predecessors and may
  contain B/C analysis and cost
  effectiveness analysis                      45
The End


Shared By: