Splitting rock vs. splitting atoms: What shale gas means for nuclear power

The natural gas boom

Shale gas is nothing new. Geologists have long known that hundreds of trillions of cubic feet of natural gas are trapped in dense underground rock. The first commercial natural gas well drilled in the United States—in Fredonia, New York, in 1821—was in shale (NETL, 2011). But it turned out to be difficult to produce large quantities of shale gas economically. By 1995, US shale gas production was barely 500 million daily cubic feet, and, as late as 2001, it totaled less than a billion cubic feet a day (Sutton, 2011). The US natural gas market was more than 50 times that size.

Indeed, the first years of the twenty-first century were marked by increasingly intense worries that rising US demand for natural gas would soon outstrip falling domestic supply and that the United States would become dependent on imports of liquefied natural gas. That, security strategists feared, would entail big political and strategic risks (Victor et al., 2007). When Russia twice sparked crises in Europe by cutting off natural gas supplies, US policy makers noticed. They did not relish the prospect of becoming potential targets for energy-backed blackmail.

But something else was happening that would change the face of the US energy scene. Devon Energy, already a successful oil and gas development firm, seized on an innovation in natural gas in the early 2000s. Working in the Barnett shale of Texas, it combined two technologies that had long been used separately in the oil and gas industry—horizontal drilling and hydraulic fracturing—to stunning effect, massively boosting production from shale gas resources and turning vast expanses once thought uneconomic into ripe development targets (Whitsitt, 2011). Helped along by high natural gas prices, drillers perfected the technique and reduced costs.

When the financial crisis hit in 2008, and prices rapidly fell, production didn’t cease; instead, it continued to grow. By 2010, annual shale gas production had reached nearly five trillion cubic feet, almost a quarter of all US natural gas output, and more than 10 times its level a mere decade earlier (EIA, 2011b). By April 2011, on the back of ever-growing production, US natural gas prices had fallen below $2 for a thousand cubic feet for the first time in decades, down from a peak of more than $15 barely five years earlier (EIA, 2012a). Analysts were forecasting abundance. The odds of large liquefied natural gas imports had been slashed.

Booming gas production quickly reverberated through energy markets. Natural gas was surging at the expense of coal. In 2011, gas-fired power delivered more than a billion megawatt-hours of electricity for the first time (EIA, 2012b). By early 2012, natural gas was providing 70 percent as much electricity as coal—up from barely more than 40 percent just a year before.

The implications for nuclear power were less immediate but just as stark. At the natural gas prices that prevailed as of mid-April 2012, the all-in cost of electricity from a new gas-fired power plant (including capital costs, operating and maintenance expenses, and fuel) was a scant 3 cents a kilowatt-hour (NETL, 2007; author’s calculations). ¹ Few expect gas to remain this cheap, with futures markets projecting a return to prices around $4.50 for a thousand cubic feet by 2015. Still, even at that higher price, a kilowatt-hour of electricity from a new natural gas power plant would cost only about 5 cents.

Projections for the cost of electricity from new nuclear power plants don’t even come close. A 2009 study by two MIT scholars, based on a survey of recent construction experience, pegged the cost of electricity from new nuclear power plants at 8.4 cents for a kilowatt-hour (Du and Parsons, 2009). (With heroic assumptions about the future availability of cheap capital, that figure dropped to 6.6 cents.) Even the Nuclear Energy Institute, the industry lobby, pegs the price of electricity from new nuclear power plants at more than 11 cents per kilowatt-hour, and expects it to remain above 7 cents through 2025 (NEI, 2011). With cheap gas, new US nuclear energy seems to be pretty much dead.

What about consequences beyond the United States? So far the boom in natural gas production has been an exclusively American story. As a result, its impact on nuclear power has been limited globally. Shale gas has not been produced at scale elsewhere in the world. And, since transporting natural gas overseas is an expensive business—it can cost between $4 and $5 to move a thousand cubic feet of natural gas from the United States to Europe or Asia—cheap gas in the United States won’t automatically mean rock-bottom gas abroad, too. In addition, little export of liquefied natural gas from the United States is currently allowed by law; new large-scale gas exports, which are even more politically controversial than coal exports (because they threaten to undermine cheap electricity prices), would need to be approved before US and overseas markets would be fully linked.

But there is reason to believe that overseas markets could eventually enjoy some of the benefits of the shale gas boom. Advanced Resources International, a respected consultancy, has estimated that the sheer volume of Chinese shale gas resources exceeds that of US resources, and Argentinean resources are close (EIA, 2011b). Other countries—including South Africa, Brazil, Poland, and France—are also believed to have significant stores. Yet cheaply producing these resources is a separate question. Public opposition has shut down shale gas development in France. The recent nationalization of YPF, the dominant oil and gas company in Argentina, will likely scare away foreign investment and, with it, the technology needed to develop unconventional gas. China may have the strongest prospects for large-scale development—yet too little is known about its shale resources to determine whether it can produce gas at prices anywhere close to those seen in the United States. For the time being, the nuclear power companies outside North America may be safe from the threat of a burgeoning natural gas industry.

Questions and controversies looming

Even in the United States, though, not everyone is bullish about shale gas. Some doubt the figures that have been bandied about regarding the size of US resources. Others question the claimed costs of production. And many, fearing contamination of water supplies and despoiling of local environments, oppose shale gas production outright. Any of these could, in principle, send natural gas prices back up, making nuclear competitive. So could strong demand for shale gas from new markets, like natural gas cars.

Estimates of the size of US shale resources are extremely uncertain. In 2009, the Ground Water Protection Council and ALL Consulting stunned observers with the release of their estimate that a whopping 262 trillion cubic feet of natural gas was trapped in US shale (2009). (Annual US consumption is about one-tenth of that, and the shale resources came on top of large conventional ones that were already known.) Two years after that, the Energy Department’s Energy Information Administration (EIA) estimated a massive 827 trillion cubic feet of natural gas, only to drop it back to 482 trillion cubic feet in early 2012 (Urbina, 2012). Private analysts have produced even more varied guesses. Absent more drilling experience, particularly away from the most attractive deposits, resolving outstanding disagreements will be tough.

That is compounded by uncertainty about how much gas a given well will ultimately recover. Shale gas is a young business, but developers expect a well to produce for decades. Long-term production projections thus rely heavily on theory, and there are intense debates over where that theory points. Some expect production to flatten out at low levels but to then continue for many years; others expect it to decline steeply without end. It will likely be many years before this battle is resolved decisively. In the meantime, uncertainty about ultimate well productivity is tantamount to uncertainty about the cost of producing a given amount of fuel.

That all leaves a big question: How much do these differences matter? In 2011, facing questions over natural gas resources and production costs, the EIA took a careful look at five cases (EIA, 2011a). Their best guess, based on moderate-sized resources and moderate drilling costs, saw natural gas prices rise to about $6 for a thousand cubic feet by 2025 and to $7 by 2035. Bigger resources (boosted by 50 percent) meant 2025 prices near $5, and better productivity pushed those down even further, to barely more than $4. Of course, when EIA analysts slashed estimated resources in half, projected prices rose, hitting $7 by 2025. The most extreme case, which featured not only smaller resources but doubled drilling costs, saw prices eventually top $8.

Because of their speculative nature and the lack of experience with shale gas, these sorts of estimates should be taken with a grain of salt. Nonetheless, most of the numbers have something important in common: They look ugly for nuclear power. Even $7 natural gas, one of the worst-case outcomes, translates into new gas-fired power at about 7 cents a kilowatt-hour. Nuclear would have a tough time beating that, at least for the next decade or so, except with the most optimistic assumptions possible about its cost.

There is one way, though, that the cost of natural gas could rise further: fierce public opposition to production, which could keep shale gas in the ground. Indeed, fracking for natural gas has become the focus of immense controversy over the past couple years, particularly in states like New York and Pennsylvania that aren’t used to large-scale oil and gas development. Environmental worries, aside from climate change, span three main areas: People worry about threats to clean water, about impacts to local communities from large-scale industrial activity, and, most recently, about small earthquakes that appear to be triggered by fracking-related activity.

All of these concerns mix real problems (like safe treatment of toxic fluids that flow back to the surface after fracking, damage to local roads, and induced seismicity from poorly placed disposal wells) with manufactured ones (like the possibility that fracking fluids will seep into water supplies through cracks in shale). Unless companies and public policy address both types of concerns effectively, local opposition to development could become a big barrier to extraction, particularly in the Northeast. Gas industry advocates who insist on blaming “public misunderstanding” for many of the concerns (seemingly intent on replicating the worst mistakes of the nuclear industry) are missing a critical point: The public ultimately decides whether or not gas gets produced. If industry and its allies stubbornly resist steps that could build public confidence—like mandatory transparency rules and some sort of inclusion of shale gas development under the Safe Drinking Water Act—on the grounds that the fears in question aren’t justified, they may find their access to reserves cut off. This would be far more costly than spending small sums to comply with new regulations. The net result would be bad for the United States—but, by driving up the price of natural gas, it would be good for nuclear power.

Climate concerns

The biggest unknown of all, though, is the impact of concern about climate change. Natural gas is cleaner than coal—switching from coal to gas lowers carbon dioxide emissions from power generation by 50 percent or more—but it is still far from emissions free. As a result, a permanent transition from coal-based electricity to conventional combustion of natural gas would be inconsistent with stabilization of global greenhouse gas concentrations at anything even approaching safe levels (Wigley, 2011). Models of US climate policy confirm this calculation. Any policy that would ultimately slash US emissions—whether driven by a cap-and-trade system, a carbon tax, or a so-called “clean energy standard” (which would mandate that an increasing fraction of US electricity come from clean sources)—would penalize natural gas but not nuclear power, putting nuclear on far firmer competitive footing by around 2030, though probably not before.

The easiest way to break this down is by looking at the impact of a notional carbon tax; most other broad-based climate policies are functionally equivalent, at least in their impact on electricity prices. Generating a kilowatt-hour of electricity from natural gas using a modern power plant entails about 440 grams of carbon dioxide emissions, including emissions produced in extracting and transporting the gas (Spath and Mann, 2000). Modeling of the ill-fated American Power Act, introduced by Senators John Kerry and Joseph Lieberman in 2010, suggested that it might have resulted in utilities paying a penalty of $30 to $60 for each ton of carbon dioxide emissions by 2020, equivalent to about a penny or two for each kilowatt-hour of power produced from natural gas. That probably wouldn’t be enough to tip the balance toward nuclear, unless gas turned out to be more expensive than expected. By 2030, though, the penalty would rise to between $50 and $100—and the upper end of that would hit natural gas with an extra charge of nearly 5 cents a kilowatt-hour. That could eliminate the cost advantage of cheap natural gas over nuclear power.

What about renewable energy sources? That’s largely a separate question. A price on carbon would boost the prospects of renewables, relative to both coal and natural gas, just as it does for nuclear power. Whether that would be decisive depends on how the costs of renewable power evolve and on whether intermittency challenges can be solved. But the relative merits of gas and nuclear don’t depend on the fate of renewables.

In any case, two big wild cards remain for natural gas. During the past two years, a small group of scholars has raised concerns that natural gas operations are leaking copious amounts of methane, a potent greenhouse gas (Howarth et al., 2011). (Natural gas itself is mostly methane.) Each molecule of methane traps far more heat than a molecule of carbon dioxide does, which means that small leaks can have big impacts on how much heat the atmosphere retains. On the flip side, methane has a much shorter atmospheric lifetime than carbon dioxide, which limits its ultimate ability to warm the planet.

The authors who first raised big public alarms about methane leaks claimed that the potency of methane outweighed its relatively short lifetime, rendering gas worse for climate change than coal. Subsequent scholarship, though, has found major flaws in the original estimates of how much methane is actually leaking (Cathles et al., 2012; Hultman et al., 2011). Other work has shown that even substantial methane leaks wouldn’t ultimately render gas worse for climate change than coal (Alvarez et al., 2012).

Several projects are currently underway that should improve our knowledge of how much methane is actually leaking from natural gas systems. But, because methane is removed from the atmosphere on far shorter time scales than those that matter most for climate stabilization (decades versus centuries), the results are unlikely to change the basic belief that natural gas is a big improvement on coal. Moreover, if much bigger amounts of methane are leaking than most people believe is the case, simple and inexpensive technical fixes (like pipeline repairs and so-called “green completions” that use special equipment to maximize the capture of gas from newly drilled wells instead of venting or flaring it) would be able to fix that. Nuclear will probably continue to face an uphill battle against natural gas, at least for the next decade or two, even in a carbon-constrained world.

And hard-to-predict technological progress could bolster the position of gas even further into the future. Experts have long talked about the possibility of capturing carbon dioxide emissions from coal and burying them underground, an approach known as carbon capture and sequestration, or CCS. The same approach can be applied to natural gas. The cost of modifying a gas-fired power plant to use CCS is actually much lower than the cost of doing it with a coal-fired power plant (NETL, 2007). ² The problem is that plants equipped with CCS consume extra fuel—and, until recently, that looked like a deal-killer for expensive natural gas. The prospect of cheap gas, though, could change that.

Since the United States doesn’t have any experience with full-scale CCS, costs for power plants that use it are enormously uncertain. But informed estimates are possible. Combining a 2007 National Energy Technology Laboratory (NETL) estimate for the cost of gas-fired power that includes carbon capture and sequestration with an assumption of $5-per-thousand-cubic-feet natural gas yields near-zero-carbon gas-fired power at 8 cents a kilowatt-hour, a tough target for nuclear power to beat. But boosting the price of gas to $7 for a thousand cubic feet jacks up the cost of gas-plus-CCS to more than 10 cents a kilowatt-hour. Whether natural gas with CCS will challenge nuclear even in a carbon-constrained world thus depends both on the price of gas and on how quickly carbon capture and sequestration matures. Abundant shale gas, though, makes it difficult to dismiss the possibility.

Yet more surprises may still lie ahead. The massive shifts in the US energy scene over the past few years ought to humble policy makers, investors, and analysts alike. Natural gas looks like it will be dominant over nuclear power, at least in the United States, for the next decade or two. But if there’s any lesson from the shale gas revolution, it’s that previously unimpeachable assumptions sometimes don’t last.