The American Economic Association's registry for randomized controlled trials
Clean Development Mechanism
Last registered on March 05, 2015
View Trial History
Clean Development Mechanism
Initial registration date
March 05, 2015
March 05, 2015 1:22 PM EST
Poverty Action Lab
Contact Primary Investigator
Other Primary Investigator(s)
Additional Trial Information
Environment & Energy
Policymakers frequently favor energy-efficiency improvements as a near-term means of carbon emissions abatement. Reports of the Intergovernmental Panel on Climate Change (IPCC) have long stressed the importance of energy efficiency in any climate change mitigation strategy (Edenhofer et al. 2011), and the head of the UN Climate Change Secretariat recently hailed energy efficiency as "the most promising means to reduce greenhouse gases in the short term” (Doyle 2007). This favored position is based on the poorly tested idea that energy-efficiency investments are a low-cost or even no-cost form of abatement, as energy savings make such investments profitable for firms.
The study tests this idea rigorously by conducting a randomized controlled trial of industrial energy audits in India, a fast-growing developing country whose future emissions will be important for global climate change. The project has been carried out among small and medium sized, energy-intensive industrial plants in the state of Gujarat; their technology choices and energy use were tracked against a comparable group of control firms. The study measures the relation between engineering projections for energy savings and actually achieved savings. It also tests two leading economic hypotheses for why industry may not adopt technologies that appear privately profitable: information market failures – based on asymmetries or undersupply – and skill constraints that inhibit technology adoption.
Kalra, Raunak and Nicholas Ryan. 2015. "Clean Development Mechanism." AEA RCT Registry. March 05.
Sponsors & Partners
In the energy audit intervention, half of a total sample of interested factories received energy audits, during which auditors suggested investments to improve the efficiency of energy use and prioritize such investments by their expected economic return.
In the energy manager intervention, half of the sample of audited plants was randomly chosen to receive energy managers – skilled engineers – who stayed on in the plant part-time for approximately three months to implement audit recommendations. These energy managers liaised with service providers, oversaw equipment installation, and trained plant staff on new technology.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
The first intervention tests the pervasive hypothesis that two types of informational market failures prevent the adoption of efficient technology: 1) asymmetric information between firms and service providers may deter adoption of efficient technologies, and 2) information about efficiency may be undersupplied in the market because it is a public good. The energy audit intervention overcomes these obstacles by providing information about energy efficiency, specific to each plant and free of cost.
The second intervention tests the relation between skilled labor and technology adoption. If plants are skill-constrained, then those provided energy managers should adopt a broader set of technologies and save more energy than those provided audits alone.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
The sample set is composed of 435 small and medium sized, textile and chemical industries in the industrial clusters of Ahmedabad, Ankleshwar and Surat in Gujarat, India. Of these 435 industries, 218 are randomly assigned into the control group and 217 into the treatment group that are audited by energy consultants. Half of these 217 industries ie. 109 are randomly assigned to the second implementation phase of implementation by energy engineers.
Experimental Design Details
Randomization done in office by a computer
Textile and Chemical Industrial Units
Was the treatment clustered?
Sample size: planned number of clusters
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
217 industries control, 218 industries treatment
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Approval Date
IRB Approval Number
Post Trial Information
Is the intervention completed?
Is data collection complete?
Is public data available?
Reports and Papers