AN ANALYSIS OF THE RESIDENTIAL PREFERENCES FOR GREEN POWER-THE ROLE OF BIOENERGY

1
AN ANALYSIS OF THE RESIDENTIAL PREFERENCES FOR
GREEN POWER-THE ROLE OF BIOENERGY

• Kim Jensen, Jamey Menard, Burt English, and Paul
  JakusProfessor, Research Associate, and
  Professor, Agricultural Economics, University of
  Tennessee, Associate Professor, Economics, Utah
  State University

Study funded in part by a grant from the USDA
National Research Initiative Program.
2
Bioenergy
Introduction

• Potential to expand industrial consumption of
  agricultural commodities, adding rural jobs and
  increasing economic activity in rural regions
• Uses renewable resources such as fast growing
  agricultural crops and trees or forest products
  wastes to produce electricity
• Not emission free, but compared with coal,
  significantly lower sulfur emissions
• Considered carbon neutral

3
Considerations
Introduction

• Hydroelectric, wind, and photovoltaic do not
  produce CO2 or SO2 emissions
• Hydroelectric power faces environmental barriers
  related to construction of dams
• Wind machines can be noisy and have significant
  impacts on the landscape
• Photovoltaic costs are relatively high

4
Costs of Renewables
Introduction

• Guey-Lee renewable sources, utilities to
  non-utilities, cents/kWh
• 6.86 conventional hydro
• 11.77 landfill gas
• 11.64 wind
• 15.80 solar
• 9.67 wood/wood waste
• 6.27 municipal solid waste and landfills
• 12.31 other biomass
• Other estimates from biomass-fired plants
• 9 cents per kWh (Department of Energy, Energy
  Efficiency and Renewable Energy Network)
• 6.4 to 11.3 cents per kWh (Oak Ridge National
  Laboratory)

5
Study Purpose
Introduction

• Residential electricity consumers willingness to
  pay (WTP) for electricity from bioenergy and
  other renewable sources
• Expands on prior research by dividing bioenergy
  into two sources bioenergy from agricultural
  crops and bioenergy from forest products wastes.
  Other sources examined include solar, wind, and
  landfill wastes
• WTP is compared across sources
• The effects of demographics, such as income and
  education, on willingness to pay are also examined

6
Prior Studies
Introduction

• WTP more for electricity from renewable sources
  ranges from 30 to 93 (Farhar Farhar and Coburn
  Farhar and Houston Rowlands et al. Tarnai and
  Moore Zarnikau)
• Actual customer participation 4 or less (Swezey
  and Bird)

7
Prior Studies
Introduction

• Farhar-69 percent placed Wind in their top
  three choices, only 26 percent placed Biomass
  in their top three choices
• 93 percent somewhat or strongly favored solar
  power, while 64 percent and 59 percent somewhat
  or strongly favored landfill gas and forest
  waste, respectively
• 53 percent would be willing to pay at least 4 a
  month more for electricity generated from
  biomass. In contrast, 65 percent said they would
  be willing to pay 6 per month more for wind
  power
• Farhar and Coburn-Colorado homeowners
  preferences-1.5 percent listed biomass as their
  top choice, while 33 percent listed solar cells
  as their top choice

8
Survey
Study Methods

• A survey was conducted by mail in Spring/Summer
  of 2003. Prior to the field survey, a pretest
  survey of 50 randomly selected residents was
  conducted
• A sample of 3,000 Tennessee residents was
  randomly drawn. A survey, cover letter, and
  information sheet about the renewable energy
  sources under study were mailed to individuals in
  the sample

9
Survey
Study Methods

• Sections
• Support for and willingness to pay some positive
  amount for energy from renewable sources
• Consumers willingness to pay for renewable
  energy from several sources, including solar,
  wind, landfill wastes, bioenergy from fast
  growing crops, and bioenergy from forest products
  wastes
• Socioeconomics and demographics, such as age,
  education, income

10
Survey
Study Methods

• Participants are asked to treat the hypothetical
  scenario as realistically as possible and they
  are reminded of their budget constraint (Kotchen
  and Reiling Cummings and Taylor)
• By allowing respondents to express support for
  renewable energy without requiring a price
  premium, bias associated with yea saying,
  perceived pressure to provide a socially
  responsible answer, may be minimized (Blamey et
  al.)

11
Survey
Study Methods

• Information sheet comparing land use, emissions,
  and other environmental impacts across specified
  renewable energy sources and coal
• Sample evenly divided among five premium levels
  for a 150kWh block of green power to be purchased
  on the respondents monthly electric bill (1.65,
  3.75, 4.50, 6.00, and 13.00)
• Premium levels, block of electricity sold
  hypothetically are based on data from existing
  green power programs
• Referendum format-respondents asked to indicate
  whether they would be willing to purchase the
  block of power at the specified premium level

12
Economic Model
Study Methods

• Possible outcomes
• not willing to pay any premium
• would pay some nonzero premium less than the
  suggested premium
• would be willing to pay at least the suggested
  premium

13
Economic model
Study Methods
Spike model helps account for large spike or
responses at 0 (not willing to pay anything or
willing to pay some amount less than the premium
provided) (Kriström)

• Probability will pay the premium
• 1-(1/1 exp(adX ßPrem))
• Probability will pay something, but less than the
  premium
• (1/1 exp(adX ßPrem))-(1/(1expadX))
• Probability that will not pay any
• 1/1exp(adX)
• Prempremium, Xdemographics, etc., a,d, ß are
  parameters to be estimated
• WTPln1exp(adX)/ß

14
Results

• A total of 421 responded to the survey
• 38.05 percent were willing to pay something
  more for renewable energy

15
Results
Estimated Models
16
Estimated Models
Results

• Premium was significant in both models
• Income 25,000 or less was significant and
  negative in the models
• Income from 60,001 to 75,000 was significant
  and positive in both models
• College education and contribution of time or
  money to an environmental organization had
  positive influences willingness to pay
• The coefficient on county population was positive
  and significant
• Other variables, such as age, gender, recycling,
  and having had a home energy audit, were not
  significant in any of the models and were omitted

17
WTP Estimates Across Profiles
Results

• WTP estimates calculated at sample means and for
  two profiles
• The first profile is income 25,000 or less, not
  college educated, not a contributor to an
  environmental organization, and living in a
  county with 100,000 population.
• The second profile is income 60,001 to 75,000,
  college educated, contributor to an environmental
  organization, and living in a county with a
  population of 600,000.

18
WTP Estimates Across Profiles
Results
19
Results





5 7 9 11 13 15 17 19
Wind 11.03 15.48
19.94 CL Mean
CU
Solar 9.39 12.71 16.03 CL
Mean CU
Landfill Wastes 7.35 9.75 12.14 CL
Mean CU

• No difference in WTP
• between bioenergy sources or between bioenergy
  and landfill wastes
• WTP for energy from solar or windgt WTP for
  bioenergy

Crops 5.77 7.19 8.62 CL Mean CU
Forest 5.47 6.87 8.28 CL Mean CU
Estimated WTP and 95 Confidence Intervals Across
Energy Sources
20
Reasons for Not Paying More
Results

• Most who would not pay more, did support
  electricity from renewable sources, but they were
  not willing to pay any more. Only about 7
  percent of the respondents did not support the
  concept of electricity from renewable energy
• Reasons why not willing to pay more for energy
  from specified sources
• Wind-visual appearance of the windmills and
  concerns about bird migration/deaths
• Solar-disposal of the solar cells
• Landfill wastes-air emissions from burning
• Bioenergy from crops-environmental impacts of
  agriculture and displacement of acreage for food
• Bioenergy from forest products wastes-deforestatio
  n and concerns air emissions from burning

21
Conclusions

• Percentage of residential electricity consumers
  who are willing to pay premiums for electricity
  is much lower than found in prior studies, at 38
  percent compared with estimates as high as 90
  percent
• Somewhat lower preference for electricity from
  crops or forest wastes than for electricity from
  solar or wind sources
• However, no statistical difference between WTP
  for bioenergy and energy from landfill wastes

22
Conclusions

• About a 5-6 per month gap in WTP between solar
  and bioenergy sources (about .03-.04 per kWh)
• About an 8-9 per month difference in WTP for
  wind compared with bioenergy sources (about
  .05-.06 per kWh)
• WTP estimates are compared with estimated costs
  of generation from prior research (Guey-Lee
  Department of Energy, Energy Efficiency and
  Renewable Energy Network Oak Ridge National
  Laboratory), gaps between WTP and costs appear to
  be greatest for solar and bioenergy sources

23
Conclusions

• Income and education levels, contribution to
  environmental organizations, and urbanization
  influence willingness to pay- suggest potential
  for target marketing of electricity from
  renewable sources
• Study confined geographically to one state,
  capabilities to examine effects of geographic
  location were limited
• Future research should examine WTP across regions
  of the United States
• Future research might also examine how investment
  in local green power projects versus purchases
  off green power markets affect willingness to pay
  for bioenergy