Applications of Benefit Cost Analysis

Download Report

Transcript Applications of Benefit Cost Analysis

Applications of Benefit-Cost/
Cost-Effectiveness Analysis
1. Tuolumne River preservation
2. Lead in drinking water
3. Habitat Protection
“Saving the Tuolumne”
Dam proposed for hydroelectric power
generation.
The “tension”: valuable electricity vs. loss in
environmental amenities.
Benefits: hydroelectric power, some
recreation.
Costs: environmental, rafting, fishing, hiking,
other recreation.
Question: Should the dam be built?
Influential analysis by economist, Stavins.
Tuolumne: background
Originates in Yosemite Nat’l Park
Flows west 158 miles, 30 miles free-flow
Many RTE species rely on river
Historic significance
World-class rafting: 15,000 trips in 1982
Recreation: 35,000 user-days annually
The Tuolumne: A nice place
Hydroelectric power generation
River’s steep canyon walls ideal for power
generation
“Tuolumne River Preservation Trust”
lobbied for protection under Wild & Scenic
1983: existing hydro captured 90% water
Municipal, agricultural, hydroelectric
Rapid growth of region would require
more water & more power
New hydroelectric projects
2 proposed hydro projects:
Clavey River, Wards Ferry
3 year study on Wild & Scenic stalled
FERC (Fed. Energy Reg. Comm.) from
assessing feasibility of hydro projects.
April 1983, FERC granted permit to
study feasibility of Clavey-Wards Ferry
Project (CWF).
Clavey-Wards Ferry project
2 new dams & reservoirs, 5 mile
diversion tunnel
Jawbone Dam 175’ high
Wards Ferry Dam 450’ high
Generate 980 gigawatt-hours annually
Annual water supply of 12,000 AF
Increased recreational opportunities
Cost: $860 million (1995 dollars)
The opposition
Historical context: John Muir & Sierra Club
lost Hetch Hetchy Valley fight.
Dams would damage
Fishing, rafting, wildlife populations, wild
character.
Recreational opps created are minimal
Cheaper alternative sources of energy
Economic evaluation
EDF economists to evaluate costs and
benefits, including environmental costs
Traditionally, environmental losses only
measured qualitatively. Difficult to
compare with quantified $ Benefits.
Stavins: “Rather than looking at it from a narrow
financial perspective, we believed we could look at it
from a broader social perspective by trying to
internalize some of the environmental externalities”.
Differences in the CBA’s
Stavins’ CBA:
Used data from original project proposal
Included environmental externalities (mostly in
lost rafting and fishing opportunities).
Took dynamic approach – evaluated costs and
benefits over entire life of project (50 year
“planning horizon”), r=10.72%
• 10.72% = 40 year bond rate for district
The costs and benefits
Benefits: $188 million annually
Electricity benefits: $184.2 million
Water yield: $3.4 million
Social Costs: $214 million annually
Internal project costs: $134 million
Lost recreation: $80 million
C (214) > B (188)
Tuolumne River: prologue
Clavey-Wards Ferry project dams were not
built….partly due to formal CBA.
Intense lobbying forced the political decision
to forbid project.
Pete Wilson was senator.
Stavins said: “[Wilson] couldn’t say ‘I did it
because I love wild rivers and I don’t like
electricity’, but he could do it by holding up
the study and saying, ‘look, I changed my
vote for solid economic reasons.’”
“Lead in drinking water”
Should the EPA control lead
contamination of drinking water?
Should water utilities be responsible for
the quality of water at the tap?
Would benefits of such a program
outweigh costs?
Economic analysis at EPA formed basis
for adoption of this rule.
Background
Lead in drinking water is byproduct of
corrosion in public water systems
Water leaves treatment plant lead-free,
lead leaches into water from pipes.
Factors associated with risk:
Corrosivity of pipe material
Length of time water sits in pipe
Lead in plumbing
Water temperature (hotter -> more lead)
Primary issues
Evidence of lead-related health effects
even from low exposure
Tendency of lead to contaminate water
in the house
Decreasing corrosivity of water, also
reap extra economic benefits by
reducing damage to plumbing.
Scientific & analytical problems
No baseline data on lead levels in tap water
High variability in lead levels in tap water
Corrosion control is system specific
Uncertainty over reliability of corrosion
control treatment
Corrosion control treatment may change
water quality and require further treatment.
Approach
Stakeholders: 44% of U.S. population.
2 regulatory approaches:
Define a single water quality standard at
the tap or at the distribution center, OR
Establish corrosion treatment
requirements.
Compare costs and benefits for each
regulator approach
Estimating costs [1 of 2]
1. Source water treatment: for systems with
high lead in water entering dist’n system.
880 water systems, $90 million/yr.
2. Corrosion control treatment: either (1)
adjust pH, (2) water stabilization, or (3)
chemical corrosion inhibitors [engineering
judgement] $220 million/yr.
3. Lead pipe replacement: 26% of public water
systems have lead pipes; usually best to
increase corrosion treatment, $80-370
million/yr.
Estimating costs [2 of 2]
4. Public education: inform consumers
about risks $30 million/yr.
5. State implementation: $40 million/yr.
6. Monitoring: (1) source water, (2)
corrosion, (3) lead pipe replacement,
$40 million/yr.
Total costs: $500-$800 million/yr.
Benefits: children’s health
Avoided medical costs from lead-related
blood disorders: $70,000/yr.
Avoided costs to compensate for leadinduced cognitive damage ($4,600 per
lost IQ point) $900 million/yr.
Offset compensatory education $2
million/yr.
Total: $900 million/yr.
Benefits: adult health
Avoided hypertension, $399 million/yr ($628
per case).
Avoided heart attacks, $818 million/yr ($1
million per event).
Avoided strokes, $609 million/yr ($1 million
per event).
Avoided deaths, $1.6 billion/yr ($2.5 million
per death).
Total: $3.4 billion/yr.
Total (all health): $4.3 billion/yr.
Key uncertainties & sensitivity
Current lead level in drinking water
Efficacy of corrosion treatment
Likelihood of decreased lead in blood
Precise link between lead exposure and
cognitive damage.
Sensitivity Analysis:
Adjusting parameters leads to a range of
costs and benefits.
Summary of costs & benefits
Costs:
$500-$800 million/yr.
NPV = $4 - $7 billion
Benefits:
$4.3 billion/yr.
NPV = $30 - $70 billion
Benefits outweigh costs by ~ 10:1
Reflections on analysis
CBA played prominent role in regulation
Very stringent rule was adopted by EPA
Widespread EPA/public support
Quantitative analysis more likely to
have impact if:
Credibly done and
Done early in process
Ando et al: Species Distributions, Land
Values, and Efficient Conservation
Basic Question: are we spending our
species conservation $ wisely?
Habitat protection often focuses on
biologically rich land
Focusing on biologically rich land results
in fewer acres of habitat to protect
species
Cost-effectiveness Analysis
Goal
Provide habitat to a fixed number of species
No issue of how many species to protect
Compare two approaches
Acquire cheapest land to provide protection
Acquire smallest amount of land to provide
protection
Why is this an interesting question?
Approach
Conduct analysis at county level in US
Use average ag land value for price of land
Use database of species location by county
(endangered or proposed endangered)
Assume if land acquired in county where
species lives  species is protected
Results
Site minimizing vs. cost minimizing
Locations for 453 species
Blue: cost-min only
Yellow: site-min only
Green: both
Cost-minimizing Problem
 c x
j
j J
j
Subject to
x


j
1
For all iεI
j Ni
where J = {j j = 1, ... , n} is the index set of candidate reserve sites,
I = {i i = 1, ... , m} is the index set of species to be covered, Ni is
the subset of J that contain species i, cj is the loss associated with
selecting site j, and xj = 1 if site j is selected and 0 otherwise.
Conclusions
Cost minimizing much more efficient that
site minimizing
Total cost savings of about 80%
Result similar to:
Santa Clara River Group Project
“Ecological Linkages” Group Project
Mini-Group Project Hints
Try to explain the problem & setup to
another person.
Solve it without Excel.
Computers are dumb – they can only do
what we ask them to do.
What is our objective? What are we
choosing in order to meet it? What are
the constraints?
Multicriteria Analysis:
The Concept of an Efficient Frontier
LBV Prob
Efficient Frontier
Attainable Points
Frog Prob
Excel needs 3 things:
1. An “objective” function cell
1.
The thing Excel is trying to maximize (the
probability of survival)
2. A “policy” cell or block of cells
1.
The thing Excel changes in order to maximize
the objective (amount of each site selected).
3. Constraints
1.
Things that “bound” the problem (Xi>0, Xi<100,
C <20,000,000)