(Reuters) - “Clean” coal is a rich vein for American investors to mine, thanks to a lucrative subsidy offered by the U.S. government. For many producers of the fuel, the path to profit leads through a laboratory at the University of North Dakota.
The school’s Energy and Environmental Research Center reported earning about $5 million in fiscal 2015-16 performing laboratory tests that qualify clean-coal producers for the subsidy. On any given day, EERC technicians take a sample of up to one ton of the coal from a producer and burn it in a miniature boiler to determine whether it reduces a specific pollutant enough to make the grade.
A stamp of approval from EERC, or a handful of other labs serving the industry, unlocks a tax credit worth more than $7 a ton to producers and their investors. The subsidy, enjoyed by more than 50 clean-coal operations stretching from West Virginia to Wyoming, costs U.S. taxpayers about $1 billion annually.
Technically speaking, EERC’s services aren’t necessary to win the credit. Producers simply need to prove that their product, also known as refined coal, is cutting their emissions of targeted pollutants – a 20 percent cut in nitrogen oxide output plus a 40 percent reduction in mercury. To do that, they can submit the comprehensive data that’s measured around-the-clock at America’s smokestacks – in real-world conditions, instead of a lab – and reported regularly to the Environmental Protection Agency.
But almost all investors choose instead to pay laboratories like EERC for testing that lasts one day, according to industry executives and disclosures by refined coal producers to environmental regulators.
The reason: The lab results almost always show a bigger cut in pollution than the real-world data. And the reductions demonstrated in the labs often do not translate to actual pollution reductions at power plants, according to a Reuters analysis of EPA data, interviews with utility executives, and disclosures by power plants.
The testing regimen casts doubt on whether the subsidy works as intended, said Ron Sahu, an environmental engineer who has consulted with utilities, the EPA and the U.S. Justice Department on power plant emissions.
“That’s a huge leap to give a tax credit based on results from a highly idealized lab test,” he said.
John Harju, the EERC’s vice president of strategic partnerships, said the tests are valid and objective measures of performance.
“We do the tests, and the results are what they are,” he said.
Refined coal appears effective in reducing mercury emissions at actual power plants. But it shows little sign of reducing emissions of nitrogen oxide, or NOx, a key element in creating smog and acid rain and a primary target of the subsidy program, according to a Reuters analysis of EPA data, interviews with industry officials, and regulatory disclosures.
Between 2009, before utilities began burning refined coal, and 2017, the last year for which full-year data is available, NOx pollution rates declined 19 percent among power plants burning refined coal, compared to 29 percent at plants burning raw coal, according to the Reuters analysis of EPA emissions data. Over the same period, 22 of the 56 U.S. utilities that burned refined coal actually recorded higher NOx emissions rates. Only 18 of them recorded a cut of 20 percent or more.
“It’s one of those things that should come with a label that says, ‘Mileage may vary in the field,’” said Martin Hopper, general manager of the MSR Public Power Agency in Modesto, California, which co-owns New Mexico’s San Juan Generating Station.
The Internal Revenue Service, which administers the tax credit program, agreed to accept lab results from subsidy applicants a decade ago. That move came after investors argued to U.S. Treasury and IRS officials in 2007 that lab-controlled tests make it easier to isolate how refined coal performs, eliminating complicating factors at play in a full-sized boiler, according to David Lowman, a partner at law firm Hunton Andrews Kurth LLP, who represented the industry in the negotiations.
The IRS declined to comment on its decision. “When issuing guidance, the IRS and Treasury invite comments concerning our requirements and, based on this, updates can be made to better administer the law,” the U.S. tax agency said in a statement.
FAILING REAL-WORLD TESTS
Over the past decade, the IRS has handed out billions of dollars of refined coal subsidies to companies on the basis of test results that show they are producing a cleaner-burning form of coal.
The semi-annual lab tests often don’t reflect the reality at power plants. That’s because plants generally do not operate under optimal conditions for reducing NOx while burning refined coal, according utility officials and power plant disclosures to environmental regulators.
A typical power plant’s NOx emission rate often fluctuates as operators adjust the flow of air to the combustion chamber and other equipment to meet demands on the facility to produce electricity. Coal that burns hot and fast during wide-open air flow, for example – a setting used when demand surges – will produce more NOx pollution than coal burning at lower temperatures with lower air flow.
In 2010, the IRS said it altered its testing requirements in a way that allowed labs to use different air control settings in the two different tests needed to compare refined coal to raw coal, according to a January 2011 article by two Ernst & Young executives in The Tax Adviser magazine. Although the agency’s reasoning for the change was unclear, the article said the practical effect would be to make the tests easier to pass.
The IRS and EERC declined to comment on the policy change. The EERC said in a statement to Reuters that it has used substantially reduced airflows while burning refined coal, resulting in lower NOx emissions for the same energy output.
The price of emissions credits under the U.S. cap and trade program can also influence a power plant’s rate of pollution. Coal plants that purchase the credits are allowed to emit more pollution, and in recent years it has been cheaper to buy credits than to run pollution-control equipment at peak effectiveness.
As a result, a plant burning refined coal can see its NOx levels spike if pollution controls are turned down or off, regardless of the type of coal it burns, EPA data shows.
“Refined coal is just one of many factors that influence real-time NOx levels,” said Erin Culbert, a spokeswoman for Duke Energy, the largest U.S. utility.
And when utilities decide to burn refined coal, tax credit investors – often Wall Street firms – have no control over how utility owners run their coal plants.
Several utilities have told environmental regulators that they tried to demonstrate pollution reductions big enough to qualify for the subsidy at actual plants, but failed, according to their disclosures.
During a trial run using refined coal at the Stanton Energy Center in Orlando, Florida, in 2010, NOx emission rates increased by 10 percent compared to raw coal, instead of declining. Two other trial runs at Stanton that year proved inconclusive, according to a copy of the results viewed by Reuters. The plant ultimately decided not to burn refined coal, according to regulatory filings.
Stanton declined to comment for this story.
Coal plants achieve most of their reduction in smog pollution through investments in other technology, including equipment that works like the catalytic converter in a car or truck to convert NOx into benign nitrogen and water. Or they cut pollutants by simply switching from coal to cleaner natural gas, or by using a mix of both fuels.
The Brunner Island power plant in Pennsylvania, for example, reduced its NOx emission rate by 63 percent in 2017 compared to 2016, thanks to new natural gas burners and piping, said Todd Martin, a spokesman for Talen Energy, owner of the coal plant.
Some utilities have conceded that they can’t replicate the lab test results they use to win the subsidy at their power plants.
St. Louis-based utility Ameren Corp told regulators in 2013 that it was not confident that a 21 percent NOx reduction achieved by refined coal at EERC could be replicated in the field if it burned the product, according to correspondence with the Missouri Department of Natural Resources related to permitting for a facility near St Louis.
Ameren has two plants in the state that have generated refined coal tax credits for Goldman Sachs and Alabama-based Coal Emissions Reduction Technologies LLC, investors in the facilities that produce clean coal ultimately burned by Ameren.
Ameren declined to comment on refined coal’s ability to reduce NOx. The Alabama investment group did not return phone calls seeking comment.
The two plants burned 6.7 million tons of refined coal in 2017, according to EIA. At the 2018 tax credit amount of $7.03 per ton, using that much refined coal would yield nearly $47 million in annual tax credits.
Ameren said it uses a number of different strategies to reduce pollution at its plants, including refined coal. “Those strategies have resulted in … significant reductions in (sulfur dioxide), NOx and mercury emissions,” Ameren said in a statement.
Whelan Energy Center in Hastings, Nebraska, also burns refined coal, but mainly as a way to cut mercury emissions and to benefit from the subsidy. Plant officials never considered using clean coal for reducing smog pollution because they don’t believe it’s effective, said Marty Stange, environmental supervisor at Hastings Utilities, an owner of the plant.
“We never really looked at it for NOx reduction,” he said.
The laboratory testing is a much safer bet for companies to qualify for the refined coal tax credit. It’s unusual for a clean-coal sample to fail an emissions test in a lab setting, said Murray Abbott, president of Chem-Mod LLC, the leading supplier of chemicals used for refined coal’s emission reductions. Murray said plants usually can’t reproduce the cuts his company’s chemicals achieve in lab results because utilities typically run their plants with power production and costs in mind, not just emissions reductions.
“It’s tough to show the same level of emission reductions at full scale,” he said.
The EERC lab in Grand Forks is a popular destination for refined coal tax-credit investors, according to contract disclosures by the school.
Utility DTE, for example, paid EERC $581,000 to conduct an undisclosed number of refined coal emissions tests at its North Dakota lab in fiscal 2015-2016, according to EERC contract disclosures. DTE’s refined coal operations generated $103 million in tax credits that year, according to DTE’s annual report.
DTE declined to comment on the testing of its coal at EERC labs.
EERC was well placed to become the refined coal industry’s laboratory for testing after the subsidy was adopted in 2004. EERC was officially founded in 1951 as the Robertson Lignite Research Laboratory, a federal facility under the U.S. Bureau of Mines that tested the qualities of different types of coal.
The $5 million in revenue EERC generated in the 2016-16 fiscal year amounted to 14 percent of its total grants and contracts, according to University of North Dakota budget disclosures.
“It’s a fair chunk of work around here,” said the EERC’s Harju.
Highlighting EERC’s central role, Republican Senator John Hoeven of North Dakota referred to its lab testing when he introduced legislation in February to extend the refined coal tax credit after it expires in 2021.
“This would not only benefit the coal-generating power plants in North Dakota that use refined coal, it will also support the work of the EERC, which provides services to facilities around the nation to verify that the refined coal meets the standards required to claim the credit,” Hoeven said in a press release.
And in April, U.S. Energy Secretary Rick Perry named Harju to the National Coal Council, a federal advisory board for formulating coal policy.
EERC’s small test boilers are designed to mimic the performance of actual power plants, lab officials say. Its combustion test facility has an output capacity of less than 1 megawatt per hour, compared to at least several hundred megawatts per hour at a typical coal plant. The lab uses several devices to simulate pollution control equipment in place at a utility, such as scrubbers.
Asked whether the boiler settings are fine-tuned to improve refined coal’s environmental performance, Harju said he was not an expert in such technical details. EERC declined to comment on how often its clients fail to pass its laboratory tests, but Chem-Mod’s Abbott said such failures are rare.
EERC officials declined to provide further technical specifics about its testing process.
“We’re just an independent validator of performance,” Harju said. “It’s an arm’s-length effort.”
Reporting by Tim McLaughlin; Editing by Richard Valdmanis, Janet Roberts and Brian Thevenot