As commodity prices fell throughout 2015 and into the early part of this year, industry discussion became increasingly focused on breakeven economics for E&P activities in individual basins. Over time there were refi nements, improving analysis from stand-alone project economics to more detailed models, including land costs, debt service, general and administrative expenses and so on.

Typically, the role of midstream has not captured center stage in this process. Accounting sometimes provides for a single line item, such as “transportation, treating and gathering,” but rarely does the role of midstream loom large as a critical component.

Bernstein Research has taken a distinctive approach, examining how sunk costs related to “gathering, transport and processing” may have become—counterintuitively—a “curse.” According to its research team, the economic advantage of previously sunk midstream costs “alters the industry’s cost structure and, as a consequence, price outlook.” The fi rm’s medium-term natural gas outlook of $3 per thousand cubic feet (Mcf) through 2020 “is predicated on the ‘sunk cost curse’,” it said.

“By the ‘sunk cost curse,’ we refer to the apparent willingness of producers to deliver gas supply below full-cycle marginal costs for various reasons,” observed Bernstein. “We defi ne marginal cost as the price needed for the worst decile (10%) of new production to deliver a full cycle return.”

The current circumstances are seen by Bernstein as an outgrowth of the U.S. shale gas industry having been “over-capitalized.” Easy access to capital markets in prior times resulted in over-investment in not only the upstream sector, but also the midstream.

Prior to the ramp-up in production in the Marcellus, a number of contracts were signed providing for more than 12 billion cubic feet per day (Bcf/d) of natural gas sourced from “second-tier shale gas basins,” according to Bernstein. These contracts, involving production from the Haynesville, Fayetteville, Midcontinent and Rockies, have an average contract expiration of October 2020. This presented a predicament, according to Bernstein. “The rapid development of the Marcellus in the next few years pushed these producers off the cost curve—unless they treated their considerable midstream commitments as sunk costs,” it said.

“Having over-invested in what turned out to be higher-cost basins, producers are now incented to deliver volumes to help cover their fi xed midstream costs.”

The upshot is that, today, operators in high-cost, fl at-todeclining gas basins, such as the Barnett, Fayetteville and Haynesville, “are correctly treating the approximate $1/Mcf of gathering and local transport costs as sunk, effectively lowering the cost curve of the marginal players to levels competitive with higher quality basins,” according to Bernstein. Contributing to the competitive challenge, E&Ps in higher quality basins “require full-cycle investment [returns] on new gathering and takeaway and are punished by transport differentials, for example, in the Marcellus and Utica.”

Bernstein cited fully loaded breakeven economics for the Fayetteville and Haynesville at $3.70/Mcf and $4.10/Mcf, respectively. However, backing out as sunk costs their gathering and transport costs of $0.80/Mcf and $1.20/Mcf, average breakeven economics at the wellhead are put at $2.90/Mcf for both basins. This substantially narrows the gap to fully loaded breakeven costs cited for the Northeast Marcellus at $2.22/Mcf, as well as for both the Southwest Marcellus and Utica at $2.46/Mcf.

What does Bernstein foresee happening as gas demand grows?

Takeaway projects, primarily pipeline reversals, should debottleneck the southwest Marcellus for the next few years, the research house said. However, thereafter, “as demand grows, we believe older shale gas basins will see a return to activity, as half-cycle Fayetteville and Haynesville [wellhead breakeven prices] are cheaper than building a greenfi eld pipeline to bring Marcellus gas out of the Northeast.”

With natural gas prices having gravitated to around 80% of marginal cost since 2009, Bernstein expected this trend would put gas at about $3/Mcf “until fi nding and development costs in the older, built-out shale plays rise signifi cantly, which we believe will not happen until the next decade.”

What if a rebound in associated gas production trails expectations?

Using a long-term oil price deck of $80 per barrel, Bernstein forecast associated gas growing by 5.5 Bcf/d through 2020. With an incremental 4 to 5 Bcf/d of capacity available from the Fayetteville and Haynesville in case of a shortfall, “we see gas prices capped at $3/Mcf for sustained periods.”