Everything you know about America's shale gas "miracle" is wrong.
I have already shown that we do not have a 100-year supply of natural gas, and that gas production is not profitable at today's prices. I also noted that the U.S. Energy Information Administration recently slashed its resource estimate by 42 percent.
But now there's even more bad news: U.S. gas production appears to have hit a production ceiling, and is actually declining in major areas.
The startling revelation comes from a new paper published today by Houston-based petroleum geologist and energy sector consultant Arthur Berman.
Berman reached this conclusion by compiling his own production history of U.S. shale gas from a massive data set licensed from data provider HPDI. His well-by-well analysis found that total U.S. gas production has been on an "undulating plateau" since the beginning of 2009, and showed declines in some areas in 2011.
Source: Arthur Berman
This stands in stark contrast to recent data provided by the EIA, which shows shale gas production rising steadily for the past two years, and well into the future.
Source: EIA, AEO2012
The EIA's forecast is bullish because it's mainly a view of demand, without great regard for supply limits. But their historical supply data differs for a reason that will be no surprise to experienced observers: the data is bad.
The EIA gets its data on shale gas production by sampling the reports of major operators, then applying a formula to estimate how much gas is actually being produced, according to Berman. This may explain why they only have official monthly historical production data for the two years (unofficially, three) of 2008 and 2009, and only annual data for 2010 and 2011. This has been a big red flag to me in my recent work on shale gas, accustomed as I am to EIA's far more detailed and up-to-date monthly and weekly data on oil, and has made it nearly impossible to verify the claim that we've had "booming" gas production over the past two years. Data is also available directly from the states, but some states have flawed reporting processes, the granularity and reporting frequency varies (as low as every six months, in the case of Pennsylvania), and ultimately the data isn't available in a usable format. It's also inaccurate and incomplete, as one Pittsburgh newspaper recently found out.
Berman reached the same conclusion, noting in his paper that "the data that EIA makes available does not have sufficient resolution to evaluate individual plays or states." So he had to build his own database.
An unprofitable treadmill
One reason for the recent slowdown in production growth is that "unconventional" shale gas wells have to make up for the decline of conventional gas wells, which has accelerated from 23 percent per year in 2001 to 32 percent per year today. The U.S. now needs to replace 22 billion cubic feet per day (Bcf/d) of production each year just to maintain flat supply. Currently, all shale gas plays together produce around 19 Bcf/d.
The shift to unconventional gas has put us on a production treadmill: We have to keep drilling like mad to maintain output because unconventional wells are far less productive and shorter-lived than conventional gas wells. Berman observes that an average gas well in Texas in 2010 produces one-fifth as much gas as an average conventional gas well did in 1972. In 1972, 23,000 gas wells produced 7.5 trillion cubic feet in Texas; in 2010, it took 102,000 wells to produce 6.4 trillion cubic feet.
Another reason was that the spurt of production created a gas glut and drove prices far below the level of profitability. Data from a January, 2012 presentation by the CEO of gas operator Range Resources showed that gas needs to sell for at least $4 per million BTU in order for operators to turn a profit.
Source: Jonathan Callahan, The Oil Drum. Data from Range Resources.
Berman is certain that the $4 threshold applies to new drilling on existing plays only; after accounting for land leasing, overhead and debt service, the threshold would be much higher. In any case, we can see that production flattened out when prices fell below $4 at the beginning of 2009.
Source: Arthur Berman. Data from Natural Gas Intelligence.
A gas price below $3 spells real trouble for operators, and flagging production is but the first effect. The next is debt: According to analysis by ARC Financial Research, the 34 top U.S. publicly traded shale gas producers are currently carrying a combined $10 billion quarterly cash flow deficit. And finally, there will the destruction of forward supply, as new development grinds down. Financing further development with debt in this environment will be extremely difficult, and eventually even the joint-venture sugar daddies that have sustained operators over the past few months will get cold feet. Without a reversal in price, gas production is guaranteed to decline.
The gas gold rush is over
Indeed, Berman concludes that "the gold rush is over at least for now with the less commercial shale plays." Within the major producing areas of the U.S., which account for 75 percent of production, all except Louisiana have been either flat or declining in recent years. Overall, he sees evidence that 80 percent of existing U.S. shale gas plays are already approaching peak production. Rig counts have been falling, and major operators such as Chesapeake Energy and ConocoPhilips have announced slowdowns in drilling in the last month.
The two major plays that do not show evidence of peaking yet are the newer ones: the Marcellus Shale in Pennsylvania and the Haynesville Shale in Louisiana. To see the influence of these two plays on overall production, compare the first chart below, which shows production from all shale plays, to the second, which removes production from those two plays:
Source: Arthur Berman
Source: Chart by Chris Nelder, from Arthur Berman's worksheets
The Haynesville surpassed the Barnett Shale in Texas last year as the top-producing shale play in the U.S., but it may be reaching a production plateau now. Worse, Berman's analysis finds that despite its impressive production, the Haynesville is among the least economic of the shale plays, requiring gas prices above $7.00 per thousand cubic feet to sustain new drilling profitably, and nearly $9.00 per thousand cubic feet after accounting for leasing and other costs. (One thousand cubic feet is roughly equivalent to one million BTU.)
A word of caution is in order here: A one-year decline in production in an unprofitable environment is not proof that shale gas has "peaked." It's certainly possible that renewed drilling could bring higher production when gas prices rise again. The operative question in that case is when. If gas prices recover within the next year or two, it will be relatively easy to bring new wells online rapidly. But if gas prices languish for longer than that, the most productive "core" areas of the plays could become exhausted because the wells deplete so quickly. Without sustained new drilling to replace their production, by the time producers begin drilling again in the remaining, less productive prospects, an air pocket could form in the supply line.
Disinformation and diffusion theory
Berman admits that it's strange for his bottom-up analysis to produce results that are so wildly divergent from the claims of the operators and the data offered by the EIA. "I ask myself: Where could we be wrong?" he explained. "We've looked at the individual wells and it looks like they'll produce less gas than the operators say, so where could we be wrong? Likewise on cost: There are no retained earnings, so how could they be saying they're profitable?"
Having scrutinized the financial reports of operators, Berman concludes that operators are being honest with the SEC, because if they aren't, somebody will go to jail. But then they're telling a very different story to the public, and to investors, particularly regarding their costs. This isn't necessarily nefarious; it's really just a way of working around the natural risks associated with new resource development. They're playing for the future, not for immediate profitability. Early wildcatters gambled on debt-fueled drilling with the hope that they'd be able to hold the leases long enough to see prices rise again and put them nicely in the black, or flip them at a profit to someone who could. And the profit picture is substantial: according to the Range Resources presentation, when gas is $6, they'll be realizing a 135 percent internal rate of return.
"I think these companies realize—clearly—that the U.S. is moving toward a gas economy," Berman observes. "The natural gas industry has been very successful at screwing up the coal industry. . . a huge part of the demand is from the power generation business. The President now thinks, incorrectly, that we've got 100 years of natural gas. [Operators think] 'If we can just get all this land held, drilled, etc., then in a couple of years when the price recovers we're going to make a fortune'. . . and they're right!"
I am inclined to agree. My own analysis suggests that gas is trouncing coal in the power generation sector. I am also strongly against exporting LNG, because it will increase domestic costs across the board, another point on which Berman and I agree. "If they go through with the permits to export LNG, then that's gonna seal it," he remarked. "All you have to do is commit to 20-year contracts to ship a few bcf per day. . . I fear what's really going to happen is that we're going to have to start importing LNG."
Ultimately, we have to ask why there seems to be such an enormous disconnect between the reality of the production and reserve data, and the wild-eyed claims of operators and politicians. Berman's answer is blunt: "We're in a weird place where it's not in anybody's vested interest to say that things aren't wonderful," he said, and went on to relate a few stories of his encounters with politicians. They admitted to him, straight-up, that they can't tell the public the truth about energy issues like gas reserves and peak oil because nobody wants to hear it, and they'll just wind up getting voted out of office.
"This gets back to basic diffusion theory," Berman muses, "where only 5 percent of people base their decisions on information, while the other 95 percent make decisions on what everybody else thinks."
That sounds right to me. It benefits everyone involved to tell happy lies, and benefits no one to own up to the current reality. That is true for everyone from the operators right on up to the President.
Perhaps in the end—like government—we'll simply get the energy policy we deserve.
Photo: Sign outside a Portland, Oregon service station, 1973. From the U.S. National Archives (usnationalarchives/Flickr)