Buying Green

1
Buying Green
By Will Bradshaw, PhD Candidate in Urban Studies
and Planning Massachusetts Institute of
Technology Advisors Lynn Fisher, William
Shutkin, Henry Pollakowski
2
Presentation Outline

• What is green development
• An unsettling observation
• What green development research tells us
• What developers need to know
• An alternative plan for research
• Planning and policy applications
• An Empirical Test

3

• What is Green Development

An Evolution of Green

• From Environment to Triple E
• McHarg and a model of ecological and cultural
  fitness.
• New Institutions and Frameworks
• United States Green Building Council and LEED
• Municipal Rating Systems
• Other institutional actors RMI, New Ecology,
  CMPBS

4

• An Unsettling Observation

Where are the developers?
According to USGBC website in June 2004.
5

• What Green Development Research Tells Us

Public versus Private Benefits

• Public Benefits Reduced need for power
  generation, reduced pollution, less traffic
  congestion, improved public health, conservation
  of open land, etc.
• Private Benefits Three common claims.
• Operating benefits of green building are large
• Total development cost increases from green
  building are small
• In efficient markets, green buildings should be
  worth more
• Case for public benefits has been strong. Private
  benefit claims are based on the possible and not
  the expected. Never tested empirically in real
  estate markets.

6

• What Green Development Research Tells Us

Green Development Research

• Honor Roll Associated with making the business
  case relies on highly successful examples of
  extraordinary buildings.
• Problem Extraordinary versus ordinary
• Still-Life Describes the state of the world
  related to green development practice. Gives a
  sense of policy obstacles and models.
• Problem Still-life research doesnt help build
  a project
• Saved by Green Moral argument around green
  building
• Problem Most effectively reaches people who are
  already interested, and can push away others.

7

• What Developers Need to Know

Thought Experiment I

• You are developer with option to purchase house
  lots how many do you want and how much are you
  willing to pay?
• To answer this question you need answers to
  several others
• How much are houses like yours selling for in the
  area?
• How much do those houses cost to build and sell?
• How many people will need such a house when you
  expect to be done with construction? And how many
  other competing houses will be available in the
  market?
• You will pay up to the difference between the
  expected value of each house and the expected
  cost to develop them. You will buy what you think
  can reasonably be absorbed given your
  competition.

8

• What Developers Need to Know

What does this mean for green?

• Developer concerns revolve around ordinary or
  average behavior in a given market place.
• Place and product are important.
• Mean performance over many buildings provides
  more information than extraordinary performance
  of exceptional buildings.
• Green building claims about private benefits miss
  these points.
• Green buildings operate at significant savings
• Green buildings cost the same or slightly more to
  build
• Green buildings are worth more.
• No one has empirically tested these hypotheses
  for an average building in a real estate market.

9

• An Alternative Plan for Research

A new recipe for research

• Identify real estate market areas with a large
  number of green starts in a single product type
  (Austin, TX Denver, CO Energy Star Homes in
  various regions, Chicago, IL office market?)
• Use hedonic price models to determine if green
  real estate products sell or rent at a premium
• Use utility and other performance data to test
  whether or not green real estate products can
  operate for less
• Use hedonic modeling to determine if there is a
  cost premium associated with green building in
  that area and in that product type
• Draw conclusions about relative value of green
  buildings versus conventional buildings, product
  by product and location by location.

10

• An Alternative Plan for Research

Strengths and Weaknesses
11

• Policy and Planning Applications

Need for analysis in policy and planning

• Power of market To what extent can developer
  and homeowner interests drive a more sustainable
  development industry?
• Efficient use of public funds - No good evidence
  for scale of benefits or costs
• Externalities versus private gain - Are improved
  public benefits and improved private benefits
  correlated?

12

• An Empirical Test What can we investigate

• Austin has a lot of data on green and
  conventional homes
• Willing to share electric and water bills
• No reliable secondary source for construction
  cost data
• Can partially test two of three hypotheses about
  green
• Green buildings trade at a premium
• Green buildings operate at lower costs
• Can begin to make argument about efficiency of
  market for green single-family homes in Austin

13

• An Empirical Test Building a Dataset

ABoR 16,973
TCAD 38,928
WCAD 32,563
Green 3,120
Utility 4,070
Geo-code Addresses with GIS software to find
unique location
ABoR 11,359
TCAD 33,919
WCAD 10,359
Green 3,002
Utility 4,061
14

• An Empirical Test Building a Dataset

ABoR 9,910
Travis 4,491
All 5,715
Green 573
Green 490
Green 503
All Utility 2,420
ABoR Utility 3,594
Travis Utility 2,376
Green 425
Green 484
Green 412
15

• An Empirical Test Map of All Homes

16

• An Empirical Test Map of Homes with Utility Data

17

• An Empirical Test Summary Stats for Green Homes

Important Mean Values
Price 265,000 Sf 2,469 Lot size 8,844
sf Distance 46,550 ft
18

• An Empirical Test Summary Stats for
  Conventional Homes

Important Mean Values
Price 288,517 Sf 2,594 Lot size 15,534
sf Distance 66,196 ft
19

• An Empirical Test Description of Variables

20

• An Empirical Test Regression Results

Highlighting Results

• Green rating has no effect on price but
• Square footage is a very important determinant of
  price.
• Houses are more expensive close to the center of
  town.

21

• An Empirical Test Are green homes more
  electrically efficient?

Highlighting Results

• Green homes seem to use less electricity
• Bad model, but implies some things about
  relationship
• Estimated savings add up to 1.4 of house value
  over 50 year life.

22

• An Empirical Test Are green homes more water
  efficient?

Highlighting Results

• Green homes are not more water efficient.
• Still a bad model

23

• An Empirical Test Problems in Results

• Endogeneity Better house or different behavior?
• Do I have a good proxy?
• Size distribution of green homes has two peaks
• Potential sampling bias in two forms
• Lag in utility data

24

• An Empirical Test Plans to Continue

• Dummy for small house/big house
• Add Energy Variable
• Build more robust utility use model
• Look at utility use without lag
• Generally revisit theory
• Your ideas?