Three Studies Confirm Shale Gas Is Not Worse Than Coal

For most of this year the enormous potential of shale gas has been clouded by controversy over its alleged climate impact. This began with the draft and later the leaked pre-publication version of a paper from a Cornell professor suggesting that the greenhouse gas emissions from gas were no better than those from coal and might even be worse. When I examined Dr. Howarth's analysis in two postings last December and this April I found that his methodology and assumptions were sufficiently flawed to undermine his conclusions. However, I also recognized the informal nature of my assessment and suggested the need for further scrutiny of this issue by organizations with more resources. That has now taken place, though I claim no credit for it. Within the last month three separate teams have issued reports bearing on this question, and not one of them validates Dr. Howarth's findings against shale gas.

The first of these studies comes from IHS Cambridge Energy Research Associates, addressing not just Dr. Howarth's paper, but also the EPA's estimates of methane leakage that were a key input for its calculations of greenhouse gas emissions from shale gas. Although a skeptic might find reasons to dismiss a study from a consultancy with a large energy industry clientele, the other two studies have connections to groups with unimpeachable environmental/sustainability credentials. One is a collaboration between Worldwatch Institute and Deutsche Bank, while the other paper, published in Environmental Research Letters, is from a team at Carnegie Mellon University with financial support from the Sierra Club. I encourage you to read them, but here are the highlights:

The Carnegie Mellon team focused on shale gas from the vast Marcellus formation underlying several eastern states. (See Friday's posting for some perspective of the scale of this resource.) They found that while the current techniques for developing and completing a Marcellus shale gas well do result in higher methane emissions than from conventional gas wells, the extra methane only increases lifecycle emissions from well to burner tip by 3% on average. This is the case because, "The life cycle emissions are dominated by combustion that accounts for 74% of the total emissions." As a result, when burned in a combined cycle power plant to generate electricity, shale gas results in emissions per kilowatt-hour (kWh) that are 20-50% lower than those from coal, depending on equipment and sources. This is the crucial comparison that Howarth's paper gave short shrift. They also compared shale gas emissions to those from LNG, which we'd now be importing in large quantities had shale gas development not ramped up as it did a few years ago. The Mellon team found shale gas and LNG roughly comparable, with both emitting around a quarter less CO2 equivalent per BTU than diesel fuel. That suggests that shale gas isn't just a lower-emitting fuel for power generation, but also for transportation. Finally, they looked at the possibility of shale gas wells being fractured multiple times, rather than just once during their production life, and found that it would take more than 25 fracturing events to negate gas's advantage over coal.

The Worldwatch/Deutsche Bank study considered both top-down and bottom-up views of shale gas emissions, including that of Howarth. They looked at the average US natural gas supply including current proportions of shale gas and found that the emissions from gas-fired power plants beat coal-fired plants by an average of 47%, even with the EPA's higher figures for methane venting during gas production. They also found that among bottom-up assessments of shale gas emissions, including the one from Carnegie Mellon and another from the DOE's National Energy Technology Laboratory, Howarth's results appear to be an outlier, and that shale gas is materially lower than coal in lifecycle emissions for power generation. And while their analysis was performed using the standard 100-year global warming potential for methane of 25 times CO2, they considered sensitivities ranging up to a GWP of 105:1, at which extreme gas still performed better than coal.

It's probably too much to hope that these independent studies will alleviate all of the concerns that have been raised about the greenhouse gas emissions from shale gas, which will only improve as technology and standards progress. (The studies also highlighted both the need and potential to reduce methane emissions from shale gas development, in order to minimize the extra greenhouse gas contribution, irrespective of any comparison to other fuels.) I also get that with the current mood in much of this country, claims for the game-changing energy potential of shale gas must sound too good to be true, without some fatal flaw. Yet everything I see indicates that the problems associated with shale gas development are all manageable, and that while it isn't a panacea, it does represent an extraordinary opportunity for the US from an economic, energy security and environmental perspective. It's time to recognize this as the tremendous gift that it is.

How Small Is That Revised Marcellus Estimate?

I see in The Hill that some critics of shale gas drilling are pointing to a revised estimate of the shale gas resources in the Marcellus formation as evidence that there's not enough gas to justify any risk from hydraulic fracturing. Earlier this week the US Geological Survey updated its previous estimate to 84 trillion cubic feet (TCF) of natural gas, a figure substantially less than the estimate of 410 TCF from the Department of Energy. Now, I'd have thought that even without doing the math on this, 84 TCF would still sound like a heck of a lot of gas, even if trillions have become the new billions in another context.

Perhaps cubic feet of gas don't convey quite the same degree of familiarity as barrels of oil, which most people can visualize, so it might be useful to think of this gas in its oil-equivalent terms. Using standard conversion factors, that 84 TCF in the Marcellus equates to roughly 14.5 billion barrels of oil. For comparison, that's half again as big as the original estimate of 9.6 billion barrels for the Prudhoe Bay field on the Alaskan North Slope. (FYI, Prudhoe Bay had produced a cumulative 11.5 billion barrels as of the end of 2007 and was still estimated to have a few billion barrels to go.) In that light, does anyone still want to argue that the Marcellus resource is inconsequential?

Why Haven't Gas Prices Fallen More?

With the US economy stuck in the doldrums, weakening the demand for oil and its products, and with the fall of at least portions of Tripoli foreshadowing the eventual return of Libyan oil exports to the market, it must seem puzzling that US gasoline prices haven't dropped farther in the last few weeks. As of Monday, the national average price for unleaded regular stood at $3.58 per gallon, only 3% lower than a month ago, when crude oil was just shy of $100 per barrel, compared to around $84 today. On Monday's evening news, CBS ran a segment attempting to explain this apparent disconnect. Unfortunately, they over-simplified the main explanation with a graphic showing cheaper domestic crude oil mixing with higher-priced imported oil. The "A" answer to this question is simpler but not well-understood, even though its elements have been fairly widely reported: Americans are simply looking at the wrong crude oil price, out of long habit. When you compare current gasoline prices and more representative crude oil prices, there isn't much of a disconnect about which to grumble.

The source of this confusion is the price of West Texas Intermediate crude oil (WTI), which for three decades has been the most watched and widely traded oil price in the world, and the basis of what most people mean when they talk about the price of "oil." In fact, there are numerous distinct grades of oil, each with its own price reflecting quality, location and availability. However, until recently most of these prices were based on the price of WTI, plus or minus a relatively narrow band of premiums or discounts, so using WTI as a barometer of all oil prices didn't cause much confusion or inaccuracy. The emergence of a pronounced and lengthy supply bottleneck at the Cushing, OK delivery location for the WTI futures contract has exploded this convenient set of relationships and assumptions.

Because more oil has been going into tankage at Cushing than was leaving those tanks over the last year or so, the price of WTI--itself a category, rather than a single stream of oil--has become massively depressed relative other types of crude oil, not just imported oil but also oil in other locations in the US that aren't affected by the bottleneck. Consider some important examples. While oil produced in Kansas, New Mexico and Oklahoma is all cheaper due to the Cushing effect, Louisiana Light Sweet, which historically traded within a dollar of WTI, is now worth nearly $20/bbl more, putting it much closer to the price of UK Brent crude--the best current gauge of global oil prices--than to WTI. Meanwhile, Bloomberg reports Alaskan North Slope crude (ANS) for delivery on the West Coast at nearly $107/bbl, or $24 over WTI. That's surprising, considering that ANS is heavier and higher in sulfur than WTI, and thus requires more processing. Just as remarkably, California heavy crude at Midway-Sunset is quoted at more than $10/bbl above WTI, when based on history and quality I would expect to see a discount of at least that magnitude. In other words, for now at least, the price of WTI is simply no longer representative of the crude that many US refineries are processing, from either foreign or domestic sources.

When you compare the wholesale price of gasoline from US refineries near the East, West and Gulf coasts to the cost of their crude inputs at around $100 or more, the difference of $15-17/bbl isn't historically unusual. Meanwhile, refineries in the middle of the country have recently been experiencing much stronger margins. This disparity is evident in the second quarter earnings reported by various US refining companies. East coast refiner Sunoco, which hasn't benefited much from cheap WTI, reported a net loss for the quarter, while Valero, with a bigger and more geographically dispersed refining system that includes facilities processing large quantities of WTI-related crude, saw refining segment earnings increase by 39% compared to the second quarter of 2010. The Cushing effect was even more pronounced for the recently merged HollyFrontier Corp., which apparently runs little crude that isn't priced near WTI and saw second-quarter net income almost triple versus 2Q2010. Even after that extra profit margin, gas prices in Tulsa, OK are currently as low as $3.30/gal., or about 15 cents per gallon less than the national average after adjusting for differences in state gas taxes.

Gasoline prices are determined by more than just crude oil prices, though in the long run the two must move together, because the latter represents the largest component of the cost of the former. At least until the bottleneck in Cushing is resolved by new pipeline capacity to the Gulf Coast, one option for which was just canceled, we will need to look beyond our old reliable WTI price indicator in order to compare gasoline and crude prices on a representative basis. I've been paying a lot more attention to the Brent market, and the Wall St. Journal still publishes daily prices for Louisiana Light Sweet and ANS. When and if those indices drop significantly, then it will be time to start looking for a commensurate drop in retail gasoline prices at the pump.

Oil Sands Anxiety Is Overblown

As I was catching up on a two-week backlog of news after my vacation, I ran across a New York Times editorial with the promising title of "Tar Sands and the Carbon Numbers." Thinking that perhaps the Times might have woken up to the necessity of comparing the lifecycle emissions from oil sands to those from other crude oils, I was disappointed to find its editors perpetuating the common misunderstanding concerning these emissions when viewed only from an oil-production perspective. That's a shame, because it results in the scape-goating of Canadian producers and pipeline companies while conveniently avoiding the soul-searching that ought to accompany a clear understanding that, whether we're talking about oil sands or conventional oil imported from any other source, the vast majority of the lifecycle emissions will occur here, when the products into which these oils will be refined are consumed. It is also condescending toward the sovereign responsibility of our NAFTA neighbor for managing their national emissions under the Kyoto Protocol, which they ratified but we didn't.

The pending State Department review of the proposed Keystone XL pipeline project linking Alberta's oil and oil-sands projects to Gulf Coast refineries has become a hot-button issue for US environmental groups. Producing oil from oil sands, which were more commonly called tar sands until that became a term of disparagement, certainly involves more environmental consequences than most--though not all--conventional crude oils. Since US groups haven't been very successful targeting the oil sands projects in Alberta, where they contribute significantly to Canada's oil output and overall economy, the export pipeline has become a target of convenience. From my perspective, the angst about pipeline safety and acidic bitumen is mainly a red herring; the oil industry routinely handles other crude oils of similar sulfur levels and acidity, usually by adjusting the metallurgy of the pipes and vessels involved. The real issue here is greenhouse gas emissions, which the Times and most other critics of oil sands narrowly compare to those from producing conventional oils.

The Environment Canada report cited by the Times indicates that oil sands production and upgrading result in emissions about 70% higher per barrel than for the production of Canada's average conventional oil. That's in the range of other estimates I've seen. However, what the Times fails to mention is that such "upstream" emissions only account for a fraction of the total lifecycle emissions attributable to any oil. By far the biggest portion--even for oil sands--comes from the combustion of petroleum products by end-users.

So if at least 70% of the emissions from oil sands crude occur in the US, rather than Canada, and if the lifecycle (well-to-wheels) emissions from oil sands only average around 15% higher than for the average US refinery's crude slate, while emitting little or no more than some commonly imported crude oils from other countries, are the XL pipeline's opponents exaggerating its impact? I believe they are, unless they're also willing to take on imports of consumer goods and other products from higher-emitting countries like China. That would be difficult to justify to the World Trade Organization, considering that the US doesn't have a statutory limit on its own greenhouse gas emissions. It might also put us in an awkward position with regard to our exports to countries that have adopted strict emissions reduction targets.

Meanwhile we shouldn't forget that under UN agreements it is Canada that bears responsibility for the extra emissions that oil sands generate in Alberta. The Environment Canada report indicates that oil sands are likely to contribute 11.7% of Canada's GHG emissions by 2020, up from 6.7% in 2005, when Canada's share of global GHG emissions stood at less than 2%. The expected increase in oil sands output would account for essentially all of the projected 7% rise in Canada's emissions over that interval, an amount equivalent to 0.1% of current global emissions. The means by which Canada could address those incremental emissions include improved technology, offsetting cuts in other sectors, emissions trading and offsets purchased from other countries, or the Canadian government could simply choose to restrict oil sands output. Whatever path they choose, we have plenty of our own emissions to consider without going into a tizzy over a Canadian sector that currently emits roughly as much as US livestock waste management.

Trying to control the emissions from oil sands by blocking this pipeline is a perfect illustration of the difficulty of attempting to tackle a complex global environmental problem by focusing on isolated measures that only bear indirectly on the outcomes that matter. The weakness of the Times' argument is reflected in the following sentence, referring to Canada's policies: "The United States can't do much about that, but it can stop the Keystone XL pipeline." The implication seems to be that we would be better off if Canada exported its oil sands to developing Asia, their next best market, relieving us of any associated guilt, even if it made no actual difference in global emissions. I hope that when the State Department decides this matter, it gives appropriate weight to the fact that, other than fuel economy improvements in the US car fleet, our energy ties with Canada represent the single most effective energy security measure undertaken by this country since the oil crises of the 1970s.

US Renewables Need A Fallback Plan

When I described some of the energy implications of the debt limit crisis last month, the most serious ones were associated with a default by the US government in the event the debt ceiling wasn't extended. That risk has been resolved, for now. But that doesn't mean that everything looks rosy, especially for renewables. Renewable energy technologies and projects are far more dependent on government assistance and policies than conventional energy. The fate of a wide range of federal energy incentives looks highly uncertain, and the impact of that uncertainty is matched by doubts about the health of the US economy and its growth prospects. With the pace of growth already slowing in some renewable energy sectors, any manufacturers or project developers that aren't thinking seriously about how they would manage without federal incentives could be setting themselves up to become roadkill.

Understanding why requires taking a closer look at the debt ceiling bill that Congress passed in the context of the federal budget baseline--never mind that the US Congress has not enacted a budget in more than two years. In April the Congressional Budget Office (CBO) published its assessment of what the economy would look like under the budget submitted by President Obama in February, as well as under the laws already on the books. The latter comprises the "March CBO Baseline" that was mentioned frequently during the debt limit talks and that formed the basis for comparing different proposals. (See Table 1-5 of the CBO report.) Without factoring in this week's debt limit agreement, the CBO projected a cumulative deficit for fiscal years 2012-21 of $6.7 trillion. That figure is important for several reasons.

First, it serves as a reminder that even after the $917 billion of cuts agreed up front and the $1.2-1.5 trillion of future cuts to be determined later this year, the US debt would still grow by more than $4 trillion over the next decade, mainly through increases in mandatory, or non-discretionary spending--entitlements and other untouchables. That won't change even under the deal done by the Senate and House this week; all of its pre-programmed cuts are to discretionary spending, the category into which most federal spending on renewable energy would fall.

But even that $4 trillion figure looks optimistic. As I understand it the CBO baseline assumes that next January 1 all of the Bush-era tax cuts will expire on schedule, resulting in substantial increases in taxes on both ordinary income and dividend income. And that's not just for those earning more than $200,000 per year, or whatever the threshold of "wealthy" is determined to be; it's for everyone. Nor would the Alternative Minimum Tax, which has been biting a growing number of middle class families every year, be indexed as proposed. It also assumes that the Social Security payroll tax will revert to its normal level of 6.2%, up from this year's 4.2%. Barring a dramatic improvement in the economy between now and the end of the year, it seems unlikely that all of those tax increases will be allowed to take effect. That means that the government's revenue through 2021 is likely to be significantly lower than the CBO forecast, because both growth and tax rates are likely to be lower. That translates into bigger deficits and more pressure for deficit reduction.

So the environment for continued support for renewables will be one in which the government's projected deficits continue as far as the eye can see, even after painful cuts, while its ability to continue borrowing on that scale looks suspect. With the main focus of budget cuts falling on the category that includes cash support for renewables, how likely is it that the Congress would extend the Treasury renewable energy cash grant program when it expires on December 31, 2011, or add new appropriations for the Department of Energy's Loan Guarantee Program? And if the Congressional super-committee's proposals include tax reform that would eliminate many "tax expenditures"--tax credits and deductions--then a host of programs such as the solar investment tax credit, the wind, biomass and geothermal energy production tax credit, various biofuel tax credits, and the electric vehicle purchase tax credit, could end up on the cutting block. In the coming scramble to avoid the budget knife, renewables will be competing with better-established programs with broader and more influential constituencies.

It has always been a risky proposition to build companies and industries, the economics of which depended on substantial government subsidies. Some folks could be on the verge of finding out just how risky. If we go down that path, it will probably also result in awkward questions being asked about some of the decisions made by the stewards of these government programs. They should be; I've never understood what kind of due diligence could have resulted in hundreds of millions of dollars in grants or "loans" going to to clean energy and automotive startups with minimal track records, when private investors weren't willing to bet on those risks at that scale. From a national energy policy and strategy perspective, our focus should not be on saving individual companies--no TARP for renewables, I suspect--but on preserving key capabilities essential to ensuring a long-term competitive US position in the global clean energy market.

What would that entail? First, as government funding for renewables becomes constrained it should be focused on R&D at the expense of deployment. Not only would the available money go much farther, but it would also create more options for the future. The next step should be to ensure that whatever the government does spend on deployment should go to projects that are close to being viable without help, or in the case of the military that enhance combat capabilities. That means, for example, focusing solar development assistance on sunny places like the southwest--preferably in proximity to existing transmission infrastructure--and putting an end to paying people to install utility and rooftop solar in places that receive less than about 5 "peak sun hours" (kWh/m2) per day, on average. Again, the money would go farther, and we'd be shoring up nearly viable operations, instead of trying to command the tide not to overwhelm the marginal ones. And finally, as I suggested last week, a greater emphasis on exports to developing country markets, where energy demand is growing at impressive rates and where renewables are becoming increasingly popular, would increase export earnings and employment while participating in volume-related unit cost reductions. And looking beyond renewable energy, the US government has a bird's nest on the ground in the form of the potential lease bid and royalty income from the substantial oil and gas resources that have been placed off limits for various reasons. Tapping those looks like a much smarter source of revenue--not to mention job creation--than selling off the Strategic Petroleum Reserve bit by bit.

If that sounds like a recipe for putting the US cleantech industry on life support after years of robust government-supported growth, then that's consistent with the severity of the fallback plan that could become necessary. The need for this would depend on the priorities set by the special Congressional deficit reduction committee established by the debt ceiling bill, and by the Congress as a whole, along with the subsequent efforts that will be necessary to prevent our long-term debt from growing beyond our ability to service it. Nor would it be quite the starvation diet it might appear, as long as states kept their renewable portfolio standards in place. This isn't a scenario the cleantech industry would willingly choose, but it's one that it can't ignore.

The Next Big CAFE Loophole

The great pitfall of government policies, no matter how well-intended they might be, is their inevitable unintended consequences. When those are truly surprising, it's hard to attach much blame to the legislators or regulators involved. However, that degree of indulgence shouldn't apply when the unintended consequences are as obvious as the ones inherent in the new fuel economy regulations that were announced with such fanfare last week. After all, an earlier generation of CAFE standards gave rise to what might just be the classic unintended consequence of recent times: the "SUV loophole" that fed a 20-plus-year SUV fad and dug the nation's oil consumption hole much deeper than it needed to be, affecting oil prices, trade deficits and energy security. Now regulators are proposing the creation of a similar loophole for electric vehicles.

I'm not surprised that the coverage I have read on the latest CAFE debate didn't remind the public of the ongoing consequences of treating pick-up trucks and delivery vehicles differently than passenger cars when the first CAFE standards were established in the 1970s. (That loophole was mostly closed just a few years ago.) Who could have guessed that a provision intended to help small businesses would blow up, because an entire generation embraced deluxe versions of such vehicles as their primary transportation--by the tens of millions--undermining the purpose of the CAFE standards to reduce gasoline demand? When I looked at this several years ago, I estimated that SUVs had increased US gasoline consumption by over 400,000 barrels per day, or roughly 5% of total demand, equivalent to the energy contribution of around 10 billion gallons per year of ethanol.

In this case the problem starts with the evolution of Corporate Average Fuel Economy standards from a tool intended solely to improve US energy security by reducing the consumption of petroleum products in transportation, to one encompassing the greenhouse gas emissions implicated in climate change. Although there are important overlaps between these two goals--keeping a chorus of pundits employed touting them--they are not identical in operation or effect. Consider the specifics of the new CAFE proposal.

The "supplemental notice of intent" from the National Highway Traffic Safety Agency (NHTSA) of the Department of Energy, the body that along with the EPA designs and enforces the CAFE standard, spells out the special treatment accorded EVs in the rules that will be forthcoming. It states that EPA intends to give manufacturers multiple credit for each EV, plug-in hybrid (PHEV) and fuel cell vehicle they sell, starting at a multiplier of 2.0 for EVs and fuel cells and declining to 1.5 by 2021, as if these cars somehow canceled the emissions of more than one vehicle. They also intend to treat EVs and the electric portion of PHEVs as having zero emissions, regardless of how the power they use is generated. So in order to meet the tough greenhouse gas standards that accompany the 54.5 mpg CAFE standard, carmakers will have every incentive to produce as many EVs they can. Unfortunately, it's not obvious that this will reduce emissions in the real world, except in the rare instances when EVs recharge exclusively from renewable or nuclear power, which provide only 30% of our electricity mix today, up from 28% in 2005.

One needn't assume that EVs might be recharged using only coal-fired power to see that they aren't always a big improvement, emissions-wise, over non-plug-in Prius-type hybrids or clean diesels. Using the average US grid CO2 emissions of around 1.3 lb/kWh, a Nissan Leaf getting 3 miles per kWh is responsible for the emission of roughly 200 grams of CO2 per mile traveled. By comparison, a 2011 Prius with its 50 mpg EPA average emits around 196 g/mi. A more rigorous comparison would require a full well-to-wheels lifecycle assessment, but that is precisely what the new CAFE rules eschew in the interest of leaning on the scales to help today's preferred vehicle technology.

Subject to further refinement, this back-of-the-envelope analysis suggests that skewing the new CAFE regulations in favor of EVs isn't going to do much to reduce greenhouse gas emissions. Its main advantage is in reducing oil consumption, since less than 1% of our electricity is generated from oil. But if we only cared about oil and not emissions, producing gasoline from domestic coal--in the same manner as a sizeable fraction of South Africa's fuel supply--would be equally effective at backing out oil imports. Meanwhile, a gallon of gasoline saved by an advanced internal combustion engine with stop-start technology and other low-cost efficiency features would be worth exactly as much as a gallon saved by an EV, while costing dramatically less. That's especially true when you factor in the $7,500/car EV tax credit, which I can't help thinking will be a prime target when the joint Congressional committee on deficit reduction established by the debt limit bill passed by the House of Representatives last night and by the Senate just a few minutes ago sets up shop this fall.

The unintended consequence that is easily envisioned from this special treatment of EVs is a massive over-investment in a particular and still very expensive vehicle technology, at the expense of other, less costly and more cost-effective technologies. I certainly accept that EVs represent a major long-term trend in cars, but I don't believe that their development requires fiddling with the CAFE rules in this way. Nor is it obvious that US manufacturers enjoy any particular competitive advantage in producing EVs, which depend on ingredients such as rare earths for which we are even more import-dependent than for oil. If saving oil and emissions is what we really care about, then we are entitled to expect that new fuel economy regulations would focus squarely on those outcomes, without being diverted by the industrial policy fad of the moment. Perhaps this will be one of the topics taken up by the House Oversight and Government Reform Committee of the Congress as it investigates the new CAFE rules.