Reports
Carbon Monitoring for Action (CARMA): Climate Activism Built on Specious Data
Due Diligence Report on CARMA's Data Methodology
This due diligence report shows that the popular climate management initiative Carbon Monitoring for
Action (CARMA) run by a prominent Washington DC based think tank, the Center for Global
Development (CGD,) utilizes erroneous and questionable CO2 emissions data to rank power plants and
motivate public activism. The report concludes that CARMA’s data are unfit for policy and business
decisions. Our conclusions are derived from comparative analysis of CARMA and United States
Environmental Protection Agency’s (USEPA) data at the level of each power plant, and it shows that more
than 90% of CARMA’s estimates DO NOT match with the US Government’s data. These differences are
large and irreconcilable. We discovered that CARMA’s estimates of carbon intensity—a measure of how
much CO2 is released for every unit of electricity generated—are systematically biased upwards compared
to the USEPA’s estimates for the year 2000 and 2007. This finding is going to stoke speculations about
biased analyses, but we believe that it is simply a case of bad number crunching. CGD’s top management
was first informed about CARMA’s data quality issues on December 4, 2007, but even after five months
there is no comprehensive disclaimer on CARMA’s or CGD’s websites.
Our findings indicate that the overall data and the analytical architecture of CARMA are flawed.
We arrived at this conclusion by checking the following: (1) the precision of CARMA’s ranking, (2) the
extent of the numerical differences between CARMA’s and USEPA’s annual CO2 estimates, (3) the lack of
correlation between CARMA’s and USEPA’s carbon intensity values, (4) the predictable pattern of error
in CARMA’s annual CO2 estimates, and (5) the logical and quantitative inconsistencies in CARMA’s
next decade predictions of CO2 emissions. This report concludes that CARMA’s statistical methodology for
estimating CO2 emissions of power plants is incompatible with the protocols for CO2 monitoring and
verification recommended by the US Government, IPCC and the European Commission. Because climate
management requires considerable coordination across countries, CARMA’s conflicting methodology, data
and results can upset the fledgling progress towards international consensus.
Through this report, we want to amplify three key policy messages. First, climate management actions need
a disciplined data based analysis, and the public and the media should be made fully aware of the data
quality pitfalls. Second, monitoring and verification of CO2 emissions is likely to remain a considerable
challenge, but statistical shortcuts are not a good substitute for the real measured data obtained from proper
on site monitoring and verification as required by the established protocols of various international and
national agencies. And finally, the surging popularity of climate issues has created a tempting environment
for public recognition but the focus should remain on objective and cost effective solutions and policies. As
this report shows, CARMA/CGD has ignored these considerations. And most fundamental of all, climate
campaigns simply can not be built on erroneous and questionable data because it would only undermine
the chances of policy success and hurt our collective interest in managing the challenges of global warming.