Energy Policy 49 (2012) 774–777

Contents lists available at SciVerse ScienceDirect

Energy Policy

0301-42

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/enpol

Communication

Cities reducing their greenhouse gas emissions

Christopher Kennedy n, Stephanie Demoullin, Eugene Mohareb

Department of Civil Engineering, University of Toronto, 35 St. George Street, Toronto, Canada, M5S 1A4

a r t i c l e i n f o

Article history:

Received 2 November 2011

Accepted 17 July 2012Available online 9 August 2012

Keywords:

Cities

Climate change

Greenhouse gas inventory

15/$ - see front matter & 2012 Elsevier Ltd. A

x.doi.org/10.1016/j.enpol.2012.07.030

esponding author. Tel.: þ416 978 5978.

ail address: christopher.kennedy@utoronto.ca

a b s t r a c t

The study asks how well are cities doing in reducing their greenhouse gas emissions. Data from six

cities with repeat GHG emission inventories for the period 2004–2009 is examined: Berlin, Boston,

Greater Toronto, London, New York City and Seattle. All of the cities are reducing their per capita GHG

emissions, primarily through changes to stationary combustion. On average the cities are reducing per

capita emissions by 0.27 t CO2e/capita per year; this is about the same average rate as the cities nation

states, although the cities are reducing emissions faster in percentage terms.

& 2012 Elsevier Ltd. All rights reserved.

1. Introduction

For several reasons cities are increasingly seen as criticalactors in efforts to counteract global climate change (Bulkeleyand Betsill, 2003; Kennedy et al., 2009, 2010; Dodman, 2009;Sovacool and Brown, 2010; Hoornweg et al., 2010, 2011). Themajority of global greenhouse gas (GHG) emissions can beattributed to the production and consumption activities of cities.Meanwhile, with high concentrations of people and infrastruc-ture, cities are particularly vulnerable to the impacts of climatechange. Given also their wealth and potential creativity, citiesshould be leaders in mitigation of GHG emissions, but just howwell are they doing? We find that several major cities arereducing their per capita GHG emissions, primarily throughchanges to stationary combustion; and in some cases the citiesare leading their nation states.

2. Methodology

To assess their progress in reducing GHG emissions, weexamined data from six cities for which repeat GHG emissioninventories have been conducted (Fig. 1). The six cities: Berlin,Boston, Greater Toronto Area (GTA; comprised of several munici-palities), London, New York City (NYC) and Seattle were studiedbecause they have open access inventories, with detailed sectoralinformation (Berlin Brandenburg Department for Statistics, 2010;City of Boston, 2009; Greater London Authority, 2010; Dickinsonand Desai, 2010; Seattle Office of Sustainability & Environment,2008; Greening Greater Toronto, 2011). Inventory methodologies

ll rights reserved.

(C. Kennedy).

differ between cities with respect to the inclusion of scope 3(World Resources Institute, 2004) out-of-boundary emissions,including sources such as: aviation; marine; exported waste;and imported food, materials and other goods. This is a challengewhen comparing cities. To overcome differences in accountingmethodologies used by cities, we reformatted the inventoriesusing the UNEP, UN-HABITAT, World Bank (2010) standard forreporting GHG emissions from cities. The inventories are includedin the supplementary data.

We primarily looked at changes in emissions over the periodfrom 2004 to 2009, where there is greater availability of emis-sions data for repeated years. We sought cities which had at leasttwo, but ideally more, recent inventories (since 2004). In search-ing for inventories, we found other cities having repeat inven-tories, e.g., Barcelona, Chicago, Hong Kong, Philadelphia,Singapore, Stockholm, Tokyo and Vancouver, but these lackedsufficient, recent (post 2004) data. In many of these other cities,per capita GHG emissions were found to be declining. This is alsothe case for the six cities we study, although for Boston andSeattle total GHG emissions have risen with population, while forthe other four total emissions have declined despite populationgrowth (Table 1).

3. Changes by sector

The sector for which emissions reductions were most consis-tently and substantially reported was stationary combustion. Thissector includes combustion of fuels for direct use in residential,commercial and industrial buildings or facilities; and consump-tion-based emissions from electricity use, regardless of whethergeneration is inside or outside of the city. Stationary combustionaccounts for the largest share of emissions in four of the cities;

C. Kennedy et al. / Energy Policy 49 (2012) 774–777 775

in Seattle and GTA (for later years) transportation contributesmore GHG emissions.

Five of the cities have curbed emissions from stationarycombustion, mainly through efforts to reduce the carbon intensityof source fuels for electricity generation and heating. e.g., shiftingfrom coal or oil to natural gas or renewables. Annual percentagechanges range from �6.60% for Berlin to þ1.12% for Boston(Table 1). Berlin used to rely on lignite and hard coal for aboutone-third of its stationary energy requirements, but the hard coalhas been aggressively reduced. The use of mineral oils has alsodropped by one-third since 1990 (see supplementary data), withnatural gas primarily used as a replacement. Boston pursued asimilar strategy in 2006; electricity generation was switched fromoil to natural gas when the price of natural gas fell. The associateddecrease in carbon intensity caused a 12% reduction in GHGemissions from electricity consumed in the city. The followingyear, however, Boston switched back to oil-generated electricity,as the price of oil became more competitive again. Fuel switchinghas been a strategy for most of the cities. In the sole case ofLondon there was evidence that emissions reductions wereachieved more by diminishing consumption than a change ofenergy sources (i.e., London was the only city where reducedemissions were achieved by a reduction in activity level, ratherthan emissions factor). London was also the only city where GHGsfrom commercial and institutional buildings have shrunk too. Inall other cases the lower carbon intensity of the energy supply hasnot been sufficient to offset increased demands. For the otherthree cities, reductions in emissions from stationary combustion

Fig. 1. Changes in annual per capita greenhouse gas emissions for the six cities.

Table 1Annual percentage changes in absolute values of annual GHG emissions for the six cit

Berlin 2004–2007 Boston 2006–2009 GTA 2

ENERGY �5.70% 0.76% �2.45

(a) Stationary combustion �6.60% 1.12% �4.21

(b) Mobile combustion �3.07% �0.18% �0.59

(c) Fugitive sources – –

INDUSTRIAL PROCESSES 2.80% – �3.69

AFOLU – 1.80%

WASTE �4.41% 1.80%

TOTAL �5.07% 0.63% �2.42Population 0.15% 1.80% 2.51%

have been achieved through greening of the electricity supply.Carbon-intensive generation plants using coal and oil have beenreplaced by less Carbon-intensive ones relying on natural gas,cogeneration, nuclear or renewable energy sources. The reduc-tions in GHG emissions from electricity are substantial: 46% (over4 years) for the GTA; 32% (over 4 years) for NYC; and 27% (over3 years) for Seattle.

Five of the six cities have also reduced total emissions formobile combustion, although the magnitudes are smaller inpercentage terms compared to stationary combustion. The great-est reductions are again observed for Berlin at just over 3% peryear. Unfortunately, no explicit vehicle mileage data is reportedfor Berlin, nor London, where perhaps the western extension ofthe congestion charging zone, in February 2007, may have had aneffect. Generally the data reported by the cities is not detailedenough to determine whether changes are due to increasingvehicle efficiency, decreased travel, or switching to lower pollut-ing modes; any of these mechanisms could be linked to rising oilprices.

NYC and GTA have seen their emissions from trucks drop by14% and 11% respectively (both over 4 years). This could berelated to the recession that started around September 2008, asthe latest year of reporting for both is 2009. NYC’s decrease mayalso be due in part to switching of waste transport from trucks torail. Emissions from buses were only specifically reported for twocities, decreasing for NYC and increasing for Seattle. NYC replacedapproximately one-quarter of its diesel-powered buses withbiodiesel and compressed natural gas vehicles, reducing emissionby 15% over 4 years.

Reporting of aviation emissions is an area of contention forurban GHG inventories. Following the UNEP/UN-HABITAT/WorldBank standard we have included GHG emissions for all fuelsloaded onto planes in the GTA and NYC. Note that NYC does notinclude these emissions in its total, but provides them as asubsidiary item. London only includes fuels used for domesticflights in its inventory; these are but a small fraction of fuelsloaded at airports within London, which were equivalent to 3.1 tCO2e/cap in 2005, (Kennedy et al., 2009), i.e., adding almost 50%to London’s total emissions. Seattle’s emissions are based on fuelsused in grounds equipment and by jets during landing and take-off at King County International Airport; and fuel use on domesticflights out of Sea-Tac International Airport, weighted by thepercentage of passengers travelling from Seattle. No aviationemissions are reported for Boston and Berlin. There is clearlyvariation in how emissions are reported, nevertheless wherethere is data changes in emissions varied from �1.1% (over4 years) for NYC to þ16.0% (over 4 years) for the GTA.

Only three of the cities record industrial process emissions intheir inventories; this is likely because major industrial activities,e.g., cement, steel or chemical manufacturing, are absent, orpossibly because data is unavailable. In Seattle and GTA emissionshave fallen at rates of 4.1%/yr and 3.7%/yr respectively; in both

ies, 2004–2009.

005–2009 London 2004–2008 NYC 2005–2009 Seattle 2005–2008

% �0.25% �2.73% 0.80%

% �0.18% �4.13% �0.67%

% �0.50% �0.15% 1.47%

– �0.43% –

% – – �4.07%

– – –

�1.00% �0.50% �4.13%

% �0.90% �2.65% 0.13%0.72% 1.37% 1.21%

C. Kennedy et al. / Energy Policy 49 (2012) 774–777776

cases industrial process emissions are primarily from cementplants. For Berlin, the emissions are mainly from mineral extrac-tion and other extraction and processing activities. Table 1 showsthese emissions have slightly increased between 2004 and 2007,but this is a mild fluctuation around a longer term downwardtrend; industrial process emissions for Berlin are now about 40%of 1990 values (see supplementary data).

Changes in reported emissions from waste range from �4.4%for Boston to þ1.8% for GTA (Table 1), but these numbers needsome interpretation. Seattle (�4.1%) only reports on emissionsfrom closed landfills located within city boundaries, i.e., Seattledoes not include emissions from the waste it exports. Theemissions for GTA include those from exported waste; the regionhas made progress in diverting residential waste, but less atten-tion is paid to commercial waste, which is approximately doublein quantity. The reporting on waste emissions for other cities isalso mixed. Berlin excludes waste from its inventory and Londonis missing residential waste. In NYC, landfill gas emissions,including exported waste, have decreased by 2% (over 4 years)through improvements to the methane capture process. This haspartly been offset though by a rise in N2O emissions fromwastewater handling.

4. Comparison with nation states

Our analysis shows that the cities have generally been redu-cing their GHG emissions, but were they performing any betterthan nations in this respect? Whether a result of deliberate policyor as a consequence of economic change, all of the four nations inwhich the cities are hosted have also decreased their emissions. Inorder to account for differences in population, a comparisonbetween the cities and nations is made using per capita emissions(Table 2). There are still, however, some effects of size that persist.Even though the inventory method for cities broadly derives fromthe IPCC guidelines for nations, there are differences in whichsectors are typically predominant. National inventories recordhigher per capita values because they include emissions fromsectors such as agriculture, heavy industry (only seen in a fewcities), and inter-city transportation, which are usually missingfrom urban inventories. Of course, much of these emissions maybe associated with urban consumer demands, e.g., for food,construction materials or produced goods; and when such indir-ect emissions are attributed to cities their per capita emissionscan be similar to their host nation (Ramaswami et al., 2008;Hillman and Ramaswami, 2010). These indirect emissions canalso be reported under the UN/World Bank tables, but have notyet become mainstream reporting for most cities.

Table 2Comparison of changes in annual per capita GHG emissions for cities and nations.

2004 2005 2006 2007 2008 2009 Rate of cha

Nations

CANADA 23.16 22.62 21.99 22.74 22.03 – �0.28

GTA – 10.73 – – 9.61 8.81

GERMANY – 12.28 12.38 12.11 12.04 – �0.08

Berlin 6.94 6.47 6.58 5.86 –

UK 11.03 10.90 10.75 10.53 10.26 – �0.19

London 6.39 – 6.42 – 5.98 –

USA 21.77 21.75 20.43 20.60 19.77 – �0.50

Boston – – 12.65 13.32 12.75 12.23

NYC – 9.39 8.49 7.89 8.24 7.96

Seattle – 11.97 – – 11.47 –

Average �0.26

The four nations in which the six cities are located have allexperienced reductions in per capita emissions over the period2004/2005 to 2008. The greatest per capita reductions have beenachieved by the US (0.50 t CO2e/cap/yr) and Canada (0.28 t CO2e/cap/yr), but these two are starting from the highest levels. Ratesof reduction for Germany (0.08 t CO2e/cap/yr) and the UK (0.19 tCO2e/cap/yr) are smaller. The stationary combustion sectoraccounts for close to 100% of the GHG reductions in the US; andabout 50% in Canada and Germany. Overall, the average rate ofreduction for the four nations (0.26 t CO2e/cap/yr) is about thesame as that for the six cities (0.27 t CO2e/cap/yr).

For only two cases, GTA and Berlin, the reduction rate of percapita emissions for a city is higher than for its host nation.Berlin’s reduction rate of 0.36 t CO2e/cap/yr, second highest withNew York City, is substantially greater than that of Germany. Also,while NYC’s reduction rate is lower than that of the USA, it ishigher than those of the three other nations.

To correct for differences in the contents of national and cityinventories, discussed above, it is perhaps fairer to make compar-ison in terms of annualized percentage changes in per capitaemissions. By this measure, three of the cities GTA, NYC andBerlin are making faster progress in reducing emissions than theirnation states; and London’s rate of reduction (1.6%/yr) is almostequal to that of the UK (1.7%/yr). The other two cities, Boston(1.1%) and Seattle (1.4%), are notably behind. On average, thepercentage change in per capita emissions for the six cities (2.9%)is greater than that for the nations (1.5%).

5. Conclusions and policy implications

This study has drawn upon a small, limited data set of only sixcities that have conducted recent, repeat GHG inventories, and sobroad generalization of findings to other cities cannot be made.Moreover, our study has not examined the extent to which cities’reductions in GHG emissions can be attributed to climate changepolicy, versus energy efficiency or energy security policy; thereare often co-benefits. Nevertheless, two findings of policy rele-vance are apparent from the study: (i) reductions in GHG emis-sions by cities need to be seen in a multi-level governancecontext; and (ii) despite recent progress, further improvementsin city GHG inventorying procedures are warranted.

The six cities studied here are reducing their per capita GHGemissions, and on average are doing so faster in percentage termsthan their nation states, but care has to be taken in attributing thesuccess of the cities. Some of the most substantial decreases inemissions have arisen due to changes in sources of electricitygeneration; and in most cases this falls under the control ofhigher level governments (NYC is perhaps an exception).

nge (t CO2e/cap/yr) Percentage change in per capita emissions (%/yr)

Cities Nations Cities

�1.2

�0.48 �4.5

�0.7

�0.36 �5.

�1.7

�0.10 �1.6

�2.3

�0.14 �1.1

�0.36 �3.8

�0.17 �1.4

�0.27 �1.5 �2.9

C. Kennedy et al. / Energy Policy 49 (2012) 774–777 777

Moreover, the more modest reductions in transportation emis-sions achieved by most of the cities could be due to changingconsumer preferences towards more efficient vehicles. Perhapsonly London can claim to have made notable changes to trans-portation with its congestion pricing. Cities cannot take all thecredit for reducing their emissions; they work within a multi-level governance context (Corfee-Morlot et al., 2009).

This examination of six city inventories also shows that someGHG emissions are not getting counted. Seattle does not includeemissions from its exported waste; London is missing residentialwaste; and Berlin is missing waste emissions altogether. Thediffering approaches to aviation emissions suggest a misunder-standing of the importance of airports to urban economies. Somecities do not include aviation emissions, perhaps because theyoccur outside of their territory. Others include emissions fromfuels loaded for domestic flights only, to be consistent with theUNFCCC framework for nations. If, however, cities recognize theirrole as the control points and gateways between nations in aglobalized economy (Friedmann, 1986; Sassen, 1991; Kennedy,2011) then emissions from international flights leaving city air-ports should be included. Then of course, there are the GHGemissions embodied in the food, construction materials, producedgoods, etc., which are consumed cities, but not counted ininventories. In short, cities are reducing their GHG emissions,but are not counting all sources.

Acknowledgment

This research was supported by the Natural Sciences andEngineering Research Council of Canada.

Appendix A. Supplementary information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.enpol.2012.07.030.

References

Berlin Brandenburg Department for Statistics, 2010. Energy and CO2 Balance inBerlin 2007. Report of Statistics, Berlin.

Bulkeley, H., Betsill, M.M., 2003. Cities and Climate Change: Urban Sustainabilityand Global Environmental Governance. Routledge, New York.

City of Boston, 2009. Boston Community Greenhouse Gas Inventories. City of

Boston Climate Action Plan, Boston.Corfee-Morlot, J., Kamal-Chaoui, L., Donovan, M.G., Cochran, I., Robert, A., Teasdale, P.-J.,

2009. Cities, Climate Change and Multilevel Governance. OECD EnvironmentalWorking Papers N1 14, OECD publishing, Paris.

Dickinson, J., Desai, R., 2010. Inventory of New York City Greenhouse GasEmissions. City of New York. Mayor’s Office of Long-Term Planning andSustainability, New York.

Dodman, D., 2009. Blaming cities for climate change? An analysis of urbangreenhouse gas emissions inventories. Environment and Urbanization 21 (1),

185–201.Friedmann, J., 1986. The world city hypothesis. Development and Change 17,

69–83.Greater London Authority, 2010. London Energy and Greenhouse Gas Inventory

(LEGGI) 2008. Folders: Data Analysis, London.Greening Greater Toronto, 2011. Living City Report Card. /http://thelivingcity.org/

lcrc/LivingCityReportCard_web_r1.pdfS.Hillman, T., Ramaswami, A., 2010. Greenhouse gas emission footprints and energy

use benchmarks for eight US cities. Environmental Science and Technology 44,1902–1910.

Hoornweg, D., Bhada, P., Freire, M., Trejos, C.L., Sugar, L., 2010. Cities and ClimateChange: An Urgent Agenda. World Bank, Washington, DC.

Hoornweg, D., Sugar, L., Trejos-Gomez, C.L., 2011. Cities and greenhouse gasemissions: moving forward. Environment and Urbanization 23 (1), 207–227.

Kennedy, C.A., 2011. The Evolution of Great World Cities: Urban Wealth andEconomic Growth. University of Toronto Press, Toronto.

Kennedy, C., Steinberger, J., Gasson, B., Hansen, Y., Hillman, T., Havranek, M.,

Pataki, D., Phdungsilp, A., Ramaswami, A., Mendez, G.V., 2009. Greenhouse gasemissions from global cities. Environmental Science and Technology 43 (19),

7297–7309.Kennedy, C., Steinberger, J., Gasson, B., Hansen, Y., Hillman, T., Havranek, M.,

Pataki, D., Phdungsilp, A., Ramaswami, A., Mendez, G.V., 2010. Methodologyfor inventorying greenhouse gas emissions from global cities. Energy Policy 38(9), 4828–4837.

Ramaswami, A., Hillman, T., Janson, B., Reiner, M., Thomas, G., 2008. A demand-centered, hybrid life-cycle methodology for city-scale greenhouse gas inven-

tories. Environmental Science and Technology 42 (17), 6455–6461.Sassen, S., 1991. The Global City. Princeton University Press, NJ.Seattle Office of Sustainability & Environment, 2008. Seattle Community GHG

Inventory, Seattle.Sovacool, B.K., Brown, M.A., 2010. Twelve metropolitan carbon footprints: a

preliminary comparative global assessment. Energy Policy 38 (9), 4856–4869.UNEP, UN-HABITAT, World Bank, 2010. International Standard for Reporting

Greenhouse Gas Emissions for Cities.World Resources Institute/ World Business Council for Sustainable Development,

2004. The Greenhouse Gas Protocol—A Corporate Accounting and ReportingStandard.