DRAFT—Please do not circulate
Do Urban Energy Initiatives Reduce Cities’ Carbon Footprints?
Judy Layzer
Associate Professor of Environmental Policy
Department of Urban Studies & Planning, MIT
The world is urbanizing. Half of the world’s 6.6 billion people live in urban areas.
More than 75 percent of the citizens of affluent countries live in cities and their
suburbs, and the developing world—where almost all of the population growth
this century will occur—is urbanizing rapidly as well. As a result, planners have
begun to focus on cities as critical battlegrounds in the fight to become more
sustainable. As New York Times journalist Andrew Revkin has observed, in
many ways cities are “where the [sustainability] battle will be won or lost.”
For some, urban sustainability presents a daunting prospect: cities
consume 75 percent of the world’s resources and have dramatically altered the
biogeochemical cycles of the regions where they are located. Yet despite the
inherent challenges, hundreds of cities in the U.S. and around the world have
embraced sustainability planning. Urban sustainability initiatives take a variety of
forms, but most focus first and foremost on reducing the city’s carbon footprint.
For example, more than 825 U.S. mayors have signed the Mayors’ Climate
Agreement, which commits them to pursuing the Kyoto Protocol goal of reducing
carbon dioxide (CO2) emissions 7 percent below 1990 levels by 2012. To
achieve this goal, cities have instituted a variety of policies and practices. Some
cities, such as Boston and Los Angeles, require that every new building meet
LEED certification.1 Others, such as New York City, have pledged to plant
millions of new trees, noting that soil and vegetation not only mitigate the urbanheat-island effect but also absorb CO2. Dozens of cities are retrofitting their
municipal buildings to conserve energy, and converting their bus and truck fleets
to low-carbon fuels.
Many observers believe that such efforts, although valuable symbolically
and as a political stimulus to the federal government, are unlikely to succeed in
actually reducing Americans’ carbon footprint. They point out that cities have
little wiggle room to pursue non-economic policies of any kind, since they
compete in a global political economy in which both capital and labor are mobile.
So, in theory at least, municipal officials will relentlessly pursue economic
development in order to retain capital and high-value taxpayers, and their
willingness to undertake sustainability initiatives that constrain economic growth
will be severely limited. Skeptics also point out that cities do not control many of
the most fundamental factors that affect people’s energy consumption, such as
the price of fossil fuels, federal highway funding mechanisms, subsidies for
depleting natural capital, and the content of international trade policies.
Urban sustainability advocates respond, however, that most construction,
land-use, and transportation decisions in the U.S. are made locally, and those
decisions are central to reducing citizens’ demand for energy. Ample research
1
The LEED certification program provides a national benchmark for the design, construction and
operation of high-performance green buildings.
1
DRAFT—Please do not circulate
suggests that well-designed city—one that is dense and compact and has good
transit options—is an extremely energy-efficient way to organize people.2 In this
regard, Manhattan is an ecotopia: 82 percent of Manhattanites get to work by
public transit, by bicycle, or on foot. Many also live in apartments, which are
smaller and share walls, thereby minimizing their individual heating and cooling
requirements. When people live in dense, mixed-use communities, they can use
distributed power generation, which is more than twice as efficient as central
power production. Compact cities also have fewer infrastructure requirements
because electric, communication, water, and sewage lines are shorter and
require less energy. (Sprawling development, by contrast, not only promotes fuel
consumption in a variety of ways; it also destroys forest- and farmland, both of
which absorb carbon.) Municipal governments are, themselves, large consumers
of energy, so their purchasing and construction policies can have a substantial
impact on an area’s carbon footprint. Moreover, say urban sustainability
advocates, if properly designed, energy-efficiency initiatives are economic
development policies because they not only save money but have the potential to
create jobs in retrofitting, technological innovation, and alternative energy
installation.
In light of this ongoing debate, this research seeks to answer several
questions:
The Proposed Research
To answer the questions raised above, I will conduct a mixed-method evaluation.
The first phase will focus on U.S. cities; subsequent phases will compare U.S.
cities with others around the world, from Latin America to Europe to Asia. The
quantitative portion of the analysis will involve (1) collecting detailed information
on the energy-efficiency policies of the 100 largest U.S. cities, (2) devising
measures to assess the behavior (outputs) and carbon emissions (outcomes) of
those cities, (3) collecting baseline data on both behavior and carbon emissions,
as well as data on changes over time, for each of those measures, and (4)
2
A variety of empirical studies support this claim. For example, researchers have found that
households in buildings with five or more units consume only 38 percent of the energy of
households in single-family homes because of smaller units and shared walls. See Marilyn A.
Brown, Frank Southworth, and Therese Stovall, “Toward a Climate-Friendly Built Environment”
(Washington: Pew Center on Global Climate Change, 2005). CO 2 emissions are about 25
percent higher at a suburban density of four homes per acre than at an urban density of 20
homes per acre. See Patrick Mazza, “Transportation and Global Warming Solutions,” Climate
Solutions Issue Briefing (May 2004), 1-4. A study that compared two households that were
identical in every respect except for neighborhood density found that the household in the less
dense neighborhood (1,000 fewer housing units per square mile) drove almost 1,200 more miles
and consumed 65 more gallons of fuel per year. See Thomas F. Golob and David Brownstone,
“The Impact of Residential Density on Vehicle Usage and Energy Consumption” (March 31,
2008); available at http://repositories.cdlib.org/itsirvine/wps/WPS05_01. The most comprehensive
study of the topic to date concluded that “Despite housing two-thirds of the nation’s population
and three-quarters of its economic activity, the nation’s 100 largest metro areas emitted just 56
percent of U.S. carbon emissions from highway transportation and residential buildings in 2005.”
See Marilyn A. Brown, Frank Southworth, and Andrea Sarzynski, “Shrinking the Carbon Footprint
of Metropolitan America” (Washington, D.C.: Brookings Institution, May 2008), 15.
2
DRAFT—Please do not circulate
collecting data on the non-policy factors that are likely to affect urban areas’
carbon emissions. (Our data collection will build on the partial-carbon-footprint
datasets constructed by Brookings Institution researchers, as well as on carbonemission estimates done by Vulcan project researchers at Purdue University and
the carbon registry at ICLEI. Nevertheless, the outputs/outcomes data collection
will be labor-intensive; it will, however, yield a valuable database that will allow
for a variety of important analyses.3) To analyze the data, we will develop
hypotheses about the effectiveness of various policies and practices based on
the literature on urban sustainability and interviews with policy experts. We will
then run a series of statistical tests aimed at evaluating those hypotheses and
identifying correlations among particular policies and significant carbon-emission
reductions.
The qualitative analysis will consist of a systematic comparison of
matched pairs of cities that are alike on most dimensions but differ in their
approaches to energy efficiency. The rationale for complementing a quantitative
assessment with a more detailed, qualitative analysis is to elucidate causal
mechanisms—that is, to discern which attributes of a policy or practice caused
the observed behavioral changes, and how they did so. Prior to conducting the
case studies we will establish a set of hypothetical causal pathways and posit the
observable implications of those hypotheses. We will also specify in advance the
standards against which performance will be evaluated. For example, one widely
held hypothesis is that painting bicycle lanes on major city streets will lead to an
increase in bicycle commuting because such lanes make riders feel safer and
hence more willing to commute by bike. We will collect information not only on
the extent and quality of bike lanes and the change in bicycle commuting but also
on policymakers’ and riders’ perceptions of whether and how such lanes have
influenced (or fail to affect) behavior. To substantiate our conclusions about the
effectiveness of bike lanes, we will also take into account other factors, such as
the price of gasoline, that might influence ridership.
The research for and writing up of this project will take several years. The
main task for Year 1 is a massive literature review and data collection effort.
Under my direction, three research assistants (two graduate students and one
undergraduate) will assemble the available literature on the effectiveness of
urban energy-efficiency initiatives. Using on-line searches and telephone
interviews, they will gather information on what the largest 100 U.S. cities are
doing vis-à-vis energy conservation and alternative fuels, and categorize cities
according to the nature and extent of their energy planning and implementation.
They will also collect data on energy-consumption outputs and outcomes,
including information on how cities are evaluating their own energy-efficiency
efforts. Finally, they will create a database to manage that information. During
3
According to the authors of a newly released Brookings Institution study on the carbon footprints
of U.S. metropolitan areas “before researchers can appropriately study the impact of proposed
federal policy changes—or even the experiences from state and local efforts—the nation needs a
consistent set of emissions data for multiple periods and at a level of resolution and scale that
can be tied to the activities, land uses, and the infrastructure networks of metropolitan areas.”
See Marilyn A. Brown et al, “Shrinking the Carbon Footprint of Metropolitan America,” 13.
3
DRAFT—Please do not circulate
Year 2, the research team will begin the quantitative analysis. We will also select
cities for in-depth study based on information gathered in the initial data
collection, matching pairs of similar cities for comparison. The on-site data
collection will involve extensive in-person interviews and observations.
This project complements other work going on at MIT, particularly in the
School of Architecture and Planning, aimed at reducing the environmental impact
of cities. For example, architect John Fernandez is developing an urban
metabolism tool to help policymakers comprehend the flows of materials through
their cities. Led by William Mitchell, researchers in the media lab are developing
a City Car—a stackable electric vehicle that promotes more efficient urban
mobility. Lecturer and researcher Harvey Michaels is investigating how market
barriers impede the adoption of energy-efficient technologies. Elsewhere at MIT,
researchers are developing a variety of alternative fuels and energy-efficiency
technologies. The proposed project is unique, however, in its focus on
evaluating the impact of policies, practices, and technology on energyconsumption behavior at the city scale. Since, as many researchers have noted,
there is an efficiency paradox—that is, increases in efficiency often lead to
increases in consumption—it is critical to document the circumstances under
which energy-efficiency measures actually reduce overall fossil-fuel
consumption.
4