Energy Ventures Analysis (EVA), widely recognized for their biased views and industry-influenced reports, just released a new assessment of the Clean Power Plan with the National Mining Association (NMA) yesterday. EVA also conducted an analysis for NMA last year on the proposed Clean Power Plan, claiming that the nation's first-ever carbon pollution standards for power plants would raise electricity bills and lead to economic destruction. It doesn't take much to identify some major faults discrediting the 2014 analysis. In response to recent criticism of the CPP highlighting last year's study, we'll look at some of the most glaring inadequacies.
Here's an overview of the shortcomings we can identify:
EVA's deficient analysis of the proposed Clean Power Plan standards is biased and implausible
EVA's November 2014 paper assessing the impacts of the proposed Clean Power Plan suffers from the same lack of transparency as the new report does, but it contained some details that made it possible for us to identify these inadequacies:
(1) The study is missing a Reference Case forecast.
(2) Faulty, outdated assumptions and artificial constraints on resource types severely limit the integrity of the analysis.
(3) EVA neglects to examine cases containing different ranges of assumptions for variables like gas prices and renewable deployment.
(4) EVA assumes gas demand, and corresponding prices, increase in tandem across all sectors and demand centers, but this is not how the markets work.
(5) EVA neglects to account for the value of the economic, health and environmental benefits of reducing harmful carbon pollution from power plants, and inflates its cost estimates by accumulating them over the period.
1. The study is missing a Reference Case forecast.
The EVA analysis lacks the fundamental basis necessary for analyzing the impacts of a policy: a reference case or base case forecast that excludes assumptions for the policy in question. This is essential to being able to measure the changes in the policy case compared with what would have happened without the policy. EVA simply compared its projections to 2012, citing it as the "EPA base year." This is indeed the data vintage the EPA relied on in establishing the standards, but in its analysis of the policy, it measures the cost and price impacts of the proposed standards from the Reference Case forecast. EVA has distorted comparisons by benchmarking to 2012 and has invalidated its own analysis by neglecting to establish a base case forecast. This completely conflates the impacts of the policy with what would happen anyways under business as usual.
2. Faulty, outdated assumptions and artificial constraints on resource types severely limit the integrity of the analysis
EVA does not disclose many of the critical assumptions it developed for both the 2014 and 2015 analyses. In the 2014 paper, EVA calls them "proprietary," but goes on to provide some clues that indicate that these data points are inconsistent with current forecasts. For example, EVA describes a "proprietary econometric multiple regression model" it employed to derive its initial demand forecast, and offers that the compound annual growth rate (CAGR) of electricity demand between 2020 and 2030 was 0.9% in its analysis. Compared with the 0.8% in the same period in the EPA's 2014 Reference Case, and -0.1% in the EPA's cases analyzing Clean Power Plan cases, EVA concludes that its demand forecast is similar to the EPA's with the exception of the demand reduction the EPA calculates for energy efficiency. This is misleading, however, because EVA uses an even more moderate energy efficiency forecast than the EPA does, which results in a demand forecast in the EVA "Clean Power Plan case" that is at least 0.5% to 0.6% (in terms of absolute value) greater than the EPA's over 2020 to 2030.
EVA also relies on proprietary methodology for its data and assumptions for natural gas supply and demand, coal supply and demand, and renewable energy growth. Without any transparency around these parts of the analysis, it is a challenge to determine what could be driving the outcomes.
Additionally, EVA limits compliance almost exclusively to natural gas fuel switching by artificially imposing very limiting constraints on clean energy and services like energy efficiency and renewable energy. This extremely limited scenario is far from what is expected to occur under Clean Power Plan compliance, as states have enormous flexibility and autonomy to determine the most efficient and cost-effective pathways for compliance.
3. EVA neglects to examine cases containing different ranges of assumptions for variables like gas prices and renewable deployment.
EVA had the opportunity to enhance its analysis and redeem its flawed assumptions with sensitivity runs to demonstrate the range of outcomes given more practical assumptions for energy efficiency and renewables, as well as the static characterization of natural gas markets. Instead, EVA failed to study these effects with respect to any of its assumptions. The result is that EVA's analysis is incomplete at best, as it offers a limited (and unrealistic) view of the available possibilities for Clean Power Plan compliance.
4. EVA assumes gas prices increase in tandem across all sectors and demand centers, but this is not how it would work in the real world.
One of the biggest flaws in the November 2014 EVA analysis is how it chooses to misrepresent supply, demand, and resulting prices for natural gas. It claimed that three of the largest natural gas demand centers would grow in tandem without demonstrating the effects that the changes have on these and other demand centers, including the power sector. For example, EVA assumes that the U.S. will significantly increase its LNG exports - this would drive up natural gas prices, presumably lowering demand from the power sector, but EVA's model does not attempt to capture these dynamics. Essentially, EVA artificially drives up natural gas prices without including the dynamics of how the markets would respond - demand for natural gas is held constant. EVA thus fails to capture the effects of trends in the various demand centers for natural gas, leading to a presupposed result and an inadequate characterization of how the policy would affect natural gas markets and prices.
5. EVA neglects to account for the value of economic, health and environmental benefits of reducing harmful carbon pollution from power plants, and inflates its cost estimates by accumulating them over the entire compliance period.
EVA does not address the economic value of the public health and climate benefits of reducing power plant pollution. In its analysis on the proposed rule, the EPA estimated $49 - $84 billion in net benefits in 2030. EVA has left out a critical piece in analyzing the impacts of a policy - how the benefits compare with the costs of the program.
The EPA estimated that the incremental compliance cost for the electric power sector (the amount the power sector would invest to implement the program) of the proposed Clean Power Plan would be between $7.3 and $8.8 billion in 2030. EVA does not present a direct comparison in its analysis, instead focusing on various other costs across multiple sectors. What is more, EVA's calculated costs are not limited to implementation of the proposed Clean Power Plan. They also include the effects of other environmental standards (including MATS) and the "market effects" that were forced into the model. EVA does not describe in detail the calculation of the costs reported in the study, and worse, erroneously compares them directly to the cost assessment in the EPA's analysis.
It's clear from these analytical errors that with this analysis, and others like it, EVA is spreading skewed projections based on arbitrarily fabricated assumptions. EVA shows us time and time again how not to conduct an analysis. Any time you come across shocking suggestions about the Clean Power Plan costing the sector hundreds of billions ($284 billion in the proposed rule, $214 billion in the final rule) check on the source. If it leads you back to EVA, you'll know that none of the claims are remotely credible. And, if it is not EVA and is NERA, we have offered our thoughts on that study too.
Thanks to my colleagues, Kevin Steinberger and Amanda Levin for the contributions to this post.