Shale Gas Magical Thinking And The Reality of Low Gas Prices

Enthusiasts believe that shale gas is simultaneously cheap, abundant and profitable thus defying all rules of business and economics. That is magical thinking.

The recently released EIA Annual Energy Outlook 2016 sparkles with pixie dust as it forecasts almost unlimited gas supply at low prices out to 2040 and beyond. Exuberant press reports herald a new era of LNG exports that will change the geopolitical balance of the world and make America great again.

But U.S. shale gas production is declining because of low prices and shale gas companies are in deep financial trouble because in the real world, price and cost matter.

That is not magical.

First Quarter 2016 Financial Performance

The financial performance of shale gas-weighted E&P companies in the first quarter of 2016 was a disaster.

Chesapeake Energy, the biggest shale gas producer in the world, had negative cash from operations. That means that oil and gas sales didn’t even cover operating costs much less capital expenditures like drilling and completion.

Other shale gas-weighted companies including Anadarko, Comstock and Petroquest also had negative cash from operations. Goodrich and Sandridge are in bankruptcy and Exco and Halcon will soon follow. Ultra, Forest, Quicksilver, Swift and Talisman were lost in action last year.

On average, surviving companies out-spent cash flow by two-to-one both in 2015 and 2016 but many normally strong companies greatly increased negative cash flow this year (Figure 1).

Q4 2015 Sampled E&Ps Shale Gas CE-CF
Figure 1. First quarter 2016 and full-year 2015 shale gas E&P company capex-to-cash flow ratios. Source: Google Finance and Labyrinth Consulting Services, Inc.

Devon Energy has been cash-flow neutral through much of the shale gas revolution but disturbingly increased capex-to-cash flow 5-fold in the first quarter of 2016. Similarly, Southwestern Energy has had an excellent record of near-cash flow neutrality but doubled its negative cash flow in 2016.

The debt side of first quarter earnings is far more disturbing. The average debt-to-cash flow ratio for shale gas companies increased almost 4-fold to more than 7, up from less than 2 in 2015 (Figure 2).

Q4 2015 Sampled E&Ps D-CF
Figure 2. First quarter 2016 and full-year 2015 shale gas E&P company debt-to-cash flow ratios. Source: Google Finance and Labyrinth Consulting Services, Inc.

Devon’s debt-to-cash flow was more than 21 and Southwestern’s, more than 17. Gas prices below $3 cannot be sustained without damaging the balance sheets and income statements of even well-managed companies.

Debt-to-cash flow is a critical determinant of risk from a bank’s perspective because it measures how many years it would take to pay off debt if 100% of cash from operations were used for this purpose. This means that it would take these companies an average of 7 years to pay down their total debt using all cash from operating activities.

The energy industry average from 1992-2012 was 1.53 and 2.0 was a standard threshold for banks to call loans based on debt-covenant agreements. That threshold increased in recent years to about 4 but 7 years to pay off debt is clearly beyond reasonable bank exposure risk.

Low Gas Prices and Declining Production

Shale gas is the principal support for all U.S. gas production since conventional gas is in terminal decline. U.S. dry gas production has declined almost 1 Bcf per day since September 2015 largely because of low gas prices (Figure 3).

Dry Gas Prod & HH Price
Figure 3. U.S. dry gas production and Henry Hub price. Source: EIA May 2016 STEO and Labyrinth Consulting Services, Inc.

Henry Hub gas prices have fallen for the last 2 years from more than $6/mmBtu in January 2014 to $2 today and prices have been below $3/mmBtu since early 2015. A similar gas-price decline occurred from June 2011 to April 2012 (Figure 3). Then, dry gas production fell when prices dropped below $3/mmBtu.

$3 is well below the break-even gas price for any operator in any play. Even in the Marcellus–the most commercially attractive shale gas play–break-even prices are more than $3 (Table 1).

Marcellus Operator-EUR Comparison 20 March 2016
Table 1. Marcellus break-even gas prices. COG: Cabot, CHK: Chesapeake. Source: Drilling Info and Labyrinth Consulting Services, Inc.

Shale gas production has fallen 0.83 Bcf/d since February 2016 (Figure 4).

Shale Gas_MASTER Prod
Figure 4. Shale gas production. Source: EIA Natural Gas Weekly and Labyrinth Consulting Services, Inc.

All plays have declined from their respective peaks except the Utica Shale. Marcellus production accounts for more than a third (-0.36 Bcf/d) of shale gas decline in 2016. There is certainly no shortage of supply in that play but low prices and related delays in pipeline commitments have taken their toll on production.

There are no longer any horizontal rigs drilling in the Barnett or Fayetteville, plays that were supposed to help provide the U.S. with 100 years of gas supply . That is the intersection of magical thinking and low gas prices.

Higher Gas Prices Are Likely

Lower gas production along with increased consumption and exports spell higher gas prices later in 2016 and in 2017. Latest data from EIA corroborate the impending late 2016 supply deficit that I wrote about last month (Figure 5).

STEO_JAN 2016 Natural Gas Supply Balance
Figure 5. U.S. dry gas supply balance and forecast. Source: EIA May 2016 STEO and Labyrinth Consulting Services, Inc.

A supply deficit does not mean that there won’t be enough gas but will require more extensive withdrawals from inventory and that will move prices higher. During the last supply deficit in 2013 and through much of 2014, Henry Hub spot prices increased from $2 at the peak of the previous surplus to more than $6 per mmBtu and averaged $4.05.

Comparative inventory (C.I.) is determined by comparing current stocks with a moving average of stocks over the past 5 years. There is a strong negative correlation between C.I. and natural gas price (Figure 6).

Natural Gas Comparative Inventory vs. Henry Hub Spot Price
Figure 6. Natural gas comparative inventory vs. Henry Hub price. Source: EIA and Labyrinth Consulting Services, Inc.

The same June 2011-April 2012 price decline shown in Figure 5 correlates with a strong increase in C.I. in Figure 6. In February 2012, C.I. turned around abruptly and prices responded quickly.

Similarly, the February 2014-March 2016 price decline in Figure 5 correlates with a C.I. increase in Figure 6. That build has slowed in recent weeks and C.I. will probably begin falling as production continues to flatten and decline.

During the period of C.I. surplus from October 2011-March 2013, gas prices averaged less than $3 just as they have during the present period of C.I. surplus since February 2015. I expect prices to move above $3 as the winter heating season begins. A possible temporary price drop in September would be consistent with previous periods when ample winter storage levels are reached after the U.S. Labor Day (J.M.Bodell, personal communication).

Shale Gas Magical Thinking: Price and Cost Matter

Shale gas made sense in the first decade of this century when real gas prices averaged almost $7/mmBtu (Figure 7). That was because there was a supply deficit as conventional production declined before shale gas supply increased to replace it.

CPI-Adjusted U.S. Natural Gas Price 1976-2016
Figure 7. CPI-adjusted U.S. natural gas prices, 1976-2016 (April 2016 U.S. dollars. Source: EIA, U.S. Department of Labor Statistics and Labyrinth Consulting Services, Inc.

Since 2009, however, prices have averaged only $3.81 and that is less than the break-even price for core areas of any play except the Marcellus (Table 2).

Marcellus-Utica-Woodford Break-Even Prices May 2016
Table 2. Shale gas break-even gas price summary. Source: Drilling Info and Labyrinth Consulting Services, Inc.

Shale gas enthusiasts have embraced point-forward economics that ignore many important non-capital costs of doing business. That is the difference between the break-even prices in Table 2 and lower estimates found in many analyst reports.

The EIA magically forecasts that shale gas production will increase from almost 40 Bcfd in 2016 to almost 70 Bcfd by 2030 at $5 (2015 dollars) gas prices; it will increase to almost 80 Bcfd by 2040 at prices below $5 per mmBtu.

The prices in Table 2 are for the core areas of the plays–much higher prices will be necessary to produce the marginal areas needed to support supply after core areas are fully developed. Although I respect EIA’s work and do not hold them to a very high standard on long-term forecasts, this view of the future of shale gas is not helpful.

Falling gas prices have exposed the delusion of shale gas magical thinking. Production growth was funded by debt. Capital in search of yield continued to flow and over-production pushed prices below $2 by the end of 2015.

The wreckage is clear from disastrous first quarter financial data and falling production. The Barnett and Fayetteville plays that were supposed to last 100 years are dead at current prices. The Haynesville will probably follow soon enough.

Capital may continue to flow to shale gas companies but most of it will be used to repair balance sheets. Prices will gradually increase and financially stronger companies with core positions in the Marcellus and Utica plays will survive. Many companies will not.

The U.S. has perhaps a decade of gas supply at about $6 and considerably more at higher prices. By the time prices reach those levels, the folly of export will be apparent.

Figure 8. Magical thinking. Source:



  • Frank Cassidy

    So instead of making a treasure for the more difficult times ahead the US is burning his oil and gas at light speed selling it at a fraction of the cost of the same oil and gas which it will have to import from… some faboulous country which will have enough oil for the insatiable american consumer… a few years from now. This is essentially the UK on steroids. It will a BIG geopolitcal and economical event.

    • Arthur Berman


      To take some liberty with Churchill’s famous reference, “Capitalism is the worst form of economy except for all those other forms that have been tried from time to time…”


  • Nice Comments Art. As a geo-appraiser in the Fayetteville (actually in AR and OK), the rig count fell to zero in January. A friend drilled the one and only well in the state in February I believe. That was is far S. Arkansas. The rig was run by people all retirement age. The young hands have moved on to other jobs. The rig remains on the location as the operator may attempt to back up and kick off a horizontal leg in the Smackover. But they may wait until fall. And the rig isn’t going anywhere soon. We had 4 rigs running in August of last year when a speaker from the O & G Commission spoke before the Arkansas Chapter of the Natl Assoc. of Royalty Owners.

    I believe that the decline will only accelerate and once it gets in the low flow stage it may stabilize somewhat but it will take a long time to “ramp back up”. These rigs are down for the count and unlikely to be easily restored to operational states. Also, once the rigs are mothballed, it is slowly rebalancing the rig count to the work and that means rig rates will rise, perhaps well ahead of product prices

    • Arthur Berman


      Field operations are unlikely to recover quickly, as you correctly observe. Most analysts have no experience in the oil and gas business so they think that once price increases, everything will go back to the way it was. The same lack of industry experience among analysts has produced the phony economics, reserves and supply life for shale gas and tight oil.

      Thanks for your comments,


  • Donkey Kong

    Art Berman for president!

    Keep up the great work Art!

  • Vince

    Thanks for the post. I haven’t been following the sector for long. It really seems that the market has been slow to come to the conclusion that they won’t get bailed out by OPEC. Seems like that was a hope for over a year and finally in January the shale industry figured it out. I think natural gas should benefit, when US shale producers cut back. But like I said, I’m not an oil and gas guy.

  • striebs

    Art ,

    As an Englishman , Frank Cassidy’s comments above about the U.S. attitude towards responsible use of gas resources being like the “UK on steroids” rings some bells .

    Thanks for providing insight into the reason behind high oil prices and the crucial role of cheap debt leading to over production .

    I just sold my shares in ASX listed LNG Ltd which has an LNG export project going through the FERC process at the moment .

    When the penny starts to drop for the U.S. administration , one might expect them first to go into denial .

    Have you seen any evidence yet that people in the Govt are starting to question the 100 year of abundant cheap natural gas hypothesis ?

    • Arthur Berman


      Thanks for your comments.

      Government is oblivious to energy and, therefore, ignorant about it except as it may affect their re-election prospects. This year, energy is almost totally off the radar as an election issue because gasoline is cheap. The EIA headlines celebrate year-over-year record highs for natural gas production–since the declines didn’t begin until September, the good news will continue until the end of summer. The first LNG cargoes have shipped (never mind that they went to Brazil) so that is confirmation bias at work. Americans cling to good news until the bad news can no longer be denied–part of our “can do” attitude and irrational faith in technology.

      All the best,


  • great news to hear of this steady decline in gas production and gas company profits; gives me hope that one day, I shall not see Shale Activity in my countryside….and county…..Susquehanna County, Pa….

    • Arthur Berman


      Don’t hold your breath for no activity in Susquehanna County–that’s the best area of the Marcellus where break-even prices are less than $4/mmBtu.

      All the best,


  • Alex


    Many thanks for your fantastic analysis, in this article and others. I have struggled to calculate breakeven costs on my own. Can I trouble you to explain a little of the methodology that you use to get to your Marcellus BE average cost, for example? How recent is your data?

    I have listened to the Q1 calls from all the big players. By and large they seem to be forecasting declines in 2016, so even the biggest cheerleaders seem to be changing their tune and accepting reality. My back of the paper calcs suggest we could lose a few bcf/day in shale production this year. Then there is likely to be lost production from associated, and legacy conventional is losing a few bcf/day anyway. On top of that net imports look like they will drop a few bcf/day as well. So overall that is a major hit on production capacity. Given what you and others have said about the difficulties with ramping up once prices rise, presumably 2017 could well see another few bcf/day go offline, even with a moderate price run this winter. Which would all suggest prices for winter 2017 (assuming normal temps) are going to have to be WAY higher to first stop declines, let alone see production ramp up. Is that crazy thinking on my part?

    I sat next to the CEO of a gas company recently and he tried to sell me the view that, after this wave of bankruptcies, the PE funds and big O&G players with $100BB+ cash on the sidelines will swoop in and pick everything up for pennies on the dollar, and then it’s back to the races without the debt burden. And having written off sunk costs, production costs would be lower than those you have suggested, so all is good below $3. I see his point, but struggle to believe that scenario plays out until prices are much higher. For those companies that have recently come out of bankruptcy, is there any evidence of production ramping up quickly post restructuring? I assume existing wells continue to produce, but can we expect large capex budgets to be immediately funded?

    Thanks in advance for your continued efforts

    • Arthur Berman


      I taught a 3-day workshop late last year in Moscow about the mechanics of the methodology behind the break-even prices that I use. It is well beyond a blog reply. We have updated the plays in the post within the last few months. EURs don’t change in that time frame nor do costs and certainly not across company or play average well performance.

      Figure 5 is my best conservative guess about gas market balance. My post “Natural Gas Price Increase Inevitable in 2016” shows my best guess on gas prices going forward. Your thinking is definitely not crazy!

      The CEO you mention has obviously not been to the field or, if he has, prefers to ignore what that experience showed him. Companies will gladly take whatever capital is on the sidelines but I think they will have to use it to repair balance sheets and pay overhead expenses. Even if they want to start drilling again, the oil-field service industry will need a long time (a year at least?) to recover crews and stacked equipment to respond to anything more than a limited demand for rigs. I hope that this CEO is right but for now, his view seems quite optimistic.

      Thanks for your comments and questions,


  • Heinrich Leopold


    Thank you for your interesting article.

    In figure 3 you have indicated 74.60 bcf/d dry gas production for March 2016. has a daily comment on dry production which just yesterday published a year to date low of just 70.1 bcf/d. This suggests a 6% drop within three months.

    Are these data comparable with your data.


    • Arthur Berman


      It is difficult to compare Bentek weekly data for late May with the monthly EIA estimate for April (what is used in my report). EIA uses Bentek as part of their production estimate. I don’t have Bentek data because it is expensive. My sense is that May production will be less than April but by less than the Bentek data for the week ending May 20 suggests.

      All the best,


  • Bradley Miller


    Great comments. I have been waiting for another natural gas write up. It seems your previous prediction about a deficit occurring sooner rather than later has arrived. Arent we technically already in a natural gas production deficit considering its the slowest injection season since 2012? And the EIA was predicting August I think.


  • Heinrich Leopold

    Art, just published weekly average dry gas production of just 70.8 bcf/d last week, which is 5% below the average for January- March you have indicated in figure 3.

    Weekly US production average at lowest level thus far this year
    Friday, May 27, 2016 – 12:00 PM

    This is quite disruptive and I am wondering what the reason may be as I am comparing the bentekenergy numbers always with other numbers and the numbers did always compare so far.

    What do you think about my explanation, that companies in February/March period did produce more wet gas as they could not sell natural gas liquids during this time? When natural liquids prices recovered in line with oil prices in April/May companies could sell natural gas liquids again and produced therefore more dry gas again at much lower quantities.

  • Alex


    Many companies are producing at negative differentials to WTI/HH. As an example, SWN forecast a discount to HH in the Marcellus of ~$0.56 for 2016. Assuming that differential holds, when you talk about a B/E for SWN in the Marcellus of $4.73, does that mean they actually need HH to be at $4.73 + 0.56 = $5.29, so that they can then get $4.73? Or does your breakeven analysis allow for differential discounts already, so at $4.73 – $0.56 = $4.17 they are B/E in the Marcellus?

    Thanks again for your insights, Alex

    • Arthur Berman


      I use wellhead and not benchmark prices in my economics so those differentials are accounted for as a variable operating cost.

      Thanks for your question,


  • Jeff

    Hello Art-
    Can you outline how the comparative inventory is actually calculated for oil (or gas) for a point on the line chart? On those charts posted for macrovoices 6/2/16 you show WTI (LHS) and CI (RHS). what do the LHS and RHS designate? thank you

    • Arthur Berman


      LHS and RHS mean Left-Hand Scale and Right-Hand Scale, a fairly common convention. Comparative inventory is calculated by taking the ratio of the current week ending inventory divided by the 4-week moving average of the 5 previous years for the same date.

      All the best,


  • Jeff

    Hello Art-
    A follow up on the CI. You mentioned “Comparative inventory is calculated by taking the ratio of the current week ending inventory divided by the 4-week moving average of the 5 previous years for the same date”. But the CI chart in this article has a unit of BCF. A ratio would be “unitless” and would be expressed a as say “1” if the numerator and denominator were the same. Can you explain this further please? Thank you for the RHS/LHS designation.

Leave a Reply

This website uses cookies and asks your personal data to enhance your browsing experience.