The Marcellus Shale Operator Index, Part 1 - Production Efficiency And Pro Forma

 |  Includes: CHK, COG, RRC, SWN, TLM
by: Moshe Ben-Reuven


A Marcellus Index of operators is proposed, involving eight public companies. As of December 2013, these cover 67% of the Marcellus total production and 59% of total wells.

The objective is to provide a gauge for the economic viability of the Marcellus through the performance of the Index members. This includes production efficiency and pro forma.

Results vary significantly among Index operators. However, for each, Marcellus operations from 2010 to 2013 have resulted in a deficit (collectively, $10 billion).

Following in the footsteps of an earlier article ("Marcellus Shale: Through A Glass, Darkly") which provided a view from an altitude of 33 thousand feet, this is a ground-level look at eight prominent operators in the Marcellus. The questions asked here concern the future of the Marcellus shale natural gas resource in terms of production rates, total recovery, well numbers, and profitability. Based on current status and the recorded short performance history since 2009, a series of near-term observations are offered. A serious concern remains the ultimate recovery and life span of the Marcellus natural gas resource.

Our earlier article has predicted strong demand for natural gas in the US, in the face of relatively modest total proved reserves. The recent DOE NETL Report (2012) puts the Gas in Place (GIP) for the Marcellus Appalachian basin shale at 500 TSCF, or 20-150 BSCF per square mile (BSCF = billion standard cubic feet.) The EIA estimate of proved Marcellus reserves (2011) of 32 TSCF seems in proper proportion to the GIP. Whereas the Marcellus shale proved reserves are under 10% of the total, it is a unique and important resource, and its recovery worthy of careful consideration.

Since numerous operators are involved, most of them quite small, it would seem logical to construct an index for the Marcellus, which should serve as significant subgroup to observe, and provide a benchmark. Also, should not be overly tedious to analyze. The following group of eight operators represents, as of December 31, 2013, 67% of the total Marcellus annual production volume, and 59% of the total wells, including vertical and, mostly, horizontal. Of these, five are independents with market capitalization between $10 and $20 billion each, and three are far smaller operations belonging to Very Large Oil companies.

The selected companies are, in no particular order: Range Resources Appalachia, LLC (NYSE:RRC), Cabot Oil and Gas Corp. (NYSE:COG), Chesapeake Appalachia, LLC (NYSE:CHK), Southwestern Energy Production Co. (NYSE:SWN), Talisman Energy USA, Inc. (NYSE:TLM), Anadarko E&P Onshore, LLC (Anadarko Petroleum Corp. (NYSE:APC)), XTO Energy, Inc. (owned by ExxonMobil (NYSE:XOM)), and Shell Western E&P Inc. (Royal Dutch Shell, Plc (RDS-A)). Their respective share, by the end of 2013, of Marcellus production rate (total 3.165 billion MBTU/y), is given in Fig. 1A, and their respective share of Marcellus wells (total 4,903 productive-wells) in Fig. 1B.

Again, as in the previous article, the motivation is to acquire an understanding of the underlying assets, the strengths and weaknesses of the players, and the near-term trends for the coming 5-year period, based on the performance during 2010-2013.

We are looking for ranking criteria in our Marcellus Index members. Which operator is best? Worst? We will do our best to provide the informed reader with an ability to decide. The only tools required are patience (83%) and some high-school algebra (16%).

1. First Criterion: Marcellus Focus.

The question is, what is the relative importance of the Marcellus shale production to each of the select operators? All operators are public companies, traded on the NYSE.

To gauge the direct income from Marcellus shale NG, the PA-DEP semi-annual production reports were used to calculate the operator's monthly production rates for each reporting period during 2010-2013. The resulting values were used to predict monthly sales, assuming 100% of the production was sold at the then-prevailing Henry Hub spot price. Of course, we know that (1) some fraction of the production is never sold, and (2) the price may be well below the prevailing HH, due to pipeline access and other problems. These pro forma sales are shown in Fig. 2. The actual income figures may have been lower. We note Chesapeake on top, way beyond Cabot, second.

Click to enlarge

We have compared the published operators' income statements, per Table 1, and our pro forma calculation of their respective income from NG sales as described above, for 2011-2013. Figure 3 shows the percent of total income represented by Marcellus NG sales for each of the operators during the period of interest. Virtually all show a regular increase of the Marcellus percent of their income from 2011 to 2013.

Click to enlarge

In Fig. 3, it is observed that only two of the operators, Range and Cabot, have most of their income attributed to Marcellus operations. Caution: Due to the caveats listed above regarding Marcellus NG sales, the actual Marcellus income figures (and hence, percent of total income) would tend to be lower.

Click to enlarge

For the remaining operators, the majority of revenue apparently comes from sources other than Marcellus. Note that Anadarko EP Onshore, XTO, and SWEPI are not listed separately, so the percentages in Fig. 3 use the overall income statements of the parents, APC, XOM, and RDS-A in the denominator. Not surprisingly, these amount to rather small Marcellus-based income percent for APC, XOM, and RDS-A.

What is the significance of the first criterion? Since Marcellus represents a challenge from several aspects, one expects those with a significant income fraction coming from the Marcellus to remain committed and continue to invest, while expecting the others to possibly fold if the going got tough. Conversely, if XOM and RDS-A discontinue their respective investments in XTO and SWEPI, it would be nearly painless for the parent companies. For COG or RRC, such a scenario would be a very different situation.

2. The Second Criterion: Production Efficiency.

Since performance is a strong function of the number of new wells started each period, the ratio of production rate to the number of productive wells provides a measure of efficiency. We hasten to point out that differences in the way fracking was carried out (e.g., number of stages) and variations in the relative richness of the shale rock kerogen may contribute to significant differences in the production rate from 2 distinct wells at precisely the same time after start-up. Also, since the production from each well declines strongly within a month or so of starting, two groups with the same number of productive wells may deliver very different volumes of NG if the ratios of new to older wells within each group are different.

Nevertheless, and assuming no single operator has a secret technology advantage and that the decline of numerous old wells is balanced by the initial pulse-like delivery of fewer new wells, we adopt this ratio as a measure. In other words, operators who manage to produce a given volume with fewer wells are better than others who need more wells for the same result.

Figure 4 depicts production rate vs. producing-well number for the selected operators, using the PA-DEP reported data. Note how closely most operators can be correlated by a straight line. The slopes of the lines are the ratio discussed above. The steeper the line, the higher the efficiency. All lines are expected to pass through the origin (0,0) for some obvious reason. Also, as we move outward from the origin on each line (or point set), we are actually progressing time-wise, since nearly all operators to date have experienced only progressive production rates and increased productive-well counts. Clearly, from Fig. 4, Cabot, Chesapeake and the relative newcomer, Southwestern, are leading, with COG at some advantage. The gradients or slopes of the linear regressions are summarized in Table 2, alongside their respective correlation coefficients (R-squared). The listing is in descending order of production efficiency.

Click to enlarge

It is clear that Range, with the largest well count, along with SWEPI and XTO, define the lower bound on the performance scale, namely, deliver their volumes with the largest number of wells. This should matter a lot financially, since the burden of each well is about $6.4 million, as per COG's recent 2013 update. Fracking wells indeed are an expensive proposition. This is before any operating costs.

What is the significance of the second criterion? Financial performance should be strongly coupled with production efficiency. For comparison, COG produces 1.72 million MBTU/year per well, while SWEPI = 0.43, and XTO = 0.32. These are very significant differences between the higher and lower efficiencies, which likely translate into non-profitability for the latter: To produce 1 million MBTU/year, COG requires a capital investment of $3.7 million, while XTO would require $20 million.

Click to enlarge

Yet, take heart, Marcellus operators. A comparison between Marcellus, PA (2010-2013), and Barnett Shale, TX (1993-2013) is provided in Fig. 4B. According to the DOE NETL report, Barnett is far richer per unit surface area: it has 30% to 100% more BSCF/sq. mile than Marcellus. This shows that the Barnett is by far under-utilized, in comparison. The clear winner is Marcellus, in the aggregate, with annual BSCF/well some 6.4 times more efficient. [Note that 1 BSCF = 1.023 million MBTU, so the vertical scale in Figs. 4 and 4B is almost the same.] This confirms the Hotelling (1931) observation that early entrants to a mineral resource recovery are highly inefficient.

3. The Third Criterion: Financial Prospects.

"Annina: 'Monsieur Rick, what kind of man is Captain Renault?' Rick: 'Oh, he's just like any other man, only more so'." Casablanca, 1942, Warner Bros. Obviously, Rick's statement contains a hilarious oxymoron, like "outstanding at being average", or, "clean coal". To further qualify our select Marcellus index members, financial performance comes to mind (which must have been on Rick's mind, too.) Finances have two diverse aspects: (1) the actual Marcellus performance of each operator, and (2) Market sentiment. The latter is taken up in Part 2 of this article.

3.1 Marcellus Index Pro Forma. As mentioned in our earlier Seeking Alpha article, Marcellus NG production is like a high-wire act on a bicycle: you must keep moving, or you will fall down. Operators must keep completing new wells at a given rate, for if they slow down, production rates will stagnate or fall, followed closely by revenues.

Indeed, each and every one of our select operators has consistently increased the number of their Marcellus wells each year between 2010-2013, mostly at a constant rate. A few words about the assumptions made toward the enclosed pro forma results.

3.1.1 Income is calculated from the respective time-wise production correlations of each operator. These correlations yield a monthly volume, e.g., millions MBTU, which are then multiplied by the prevailing Henry Hub Spot price to produce a monthly gross income figure. The caveats were explained above. The calendar 12-month revenue is summed for the annual gross income figure. This process was carried out for each of the select Index operators, between 2010 and 2018. We deal here only with results to end of 2013.

3.1.2 Capital Expenses are calculated for each annual period, using the reported number of new productive wells in the calendar year. We have assumed an expense of $6.4 million per each well. This figure comes from Cabot 2013 Update, and it corresponds to multi-stage fracked, horizontal, multi-well pads. We nevertheless use this cost figure for all operators and all well types, most of which are fracked and horizontal, to apply a fair common denominator. We have applied 100% of the well cost in the year the well was completed, even though a variety of financing would be used, which would be different for distinct operators.

3.1.3 Operating Expenses for each operator are calculated from the annual volume of NG sold. The operating cost, in $/MBTU, is again taken from the COG 2013 Update, which has specified values applied between 2010 and 2013. Thus OPEX would be proportional to the volume of NG produced, as reported by PA-DEP. Note that the report lists volumes, and 1,000 SCF = 1.023 MBTU, where MBTU= 1 Million British Thermal Units.

3.1.4 Earnings Before Interest and Tax (EBIT), in our pro forma, equals Income less the capital and operating expenses, computed for each select operator annually. The foregoing simple pro forma elements pertain to Marcellus operations only. Issues of debt, G&A, other related earnings, hedging offsets income, and other accounting delights were precluded; with our apologies. The results are summarized in Table 3, for the years 2010-2013. It must be emphasized that we have taken consistently the most optimistic approach, in applying the maximal possible income and the smallest possible expenses.

Click to enlarge

3.2 Discussion of Index Results. For more graphic detail, the EBIT block is summarized in Fig. 5. It appears that only 3 of the 8 operators were to break a profit, for the first time, in 2013. These are, Cabot, Chesapeake and Talisman. Virtually each and every member of our Index group has a significant financial deficit carried over from 2010.

The aforesaid accumulated deficit is calculated by summing the respective EBIT figures for each of the operators over the 4-year period of interest. Even if somehow the costs were significantly lower, we would still get a very negative accumulation here. The total accrued 2010-2013 operations profit for each index member is plotted in Fig. 6.

The undisputed leader of the index group in this respect is Range. We may recall RRC, in Fig. 4, as having defined the lower bound on production efficiency. It seems to be collaborated by the largest accumulated Marcellus-related deficit in the group. But, wait a minute, Range is the only member of the Index to report appreciable condensate production from wet shale gas in the Marcellus. Should it not offset this deficit? Indeed, RRC's reported (PA-DEP) annual condensate production during the period 2010-2013 were (in million BBL/year): 0, 0.475, 1.670, and 2.869 respectively. Using as costing 50% of the then-prevailing WTI (Cushing, OK) Spot price ($79.48, $94.88, $94.05, and $97.98 per BBL) we get, for the same period, the RRC condensate income: 0, $23 million, $79 million, and $141 million. Total accumulated condensate credit: $242 million. This should reduce the RRC total Marcellus deficit to about $3.02 billion at the end of 2013. Regrettably, however, no change in its status as deficit champion.

Perhaps not surprisingly, Cabot and Southwestern, which defined the upper bound on production efficiency, have the smallest deficit. Chesapeake, in the meantime, has significantly more wells than COG, its well ratio at the end of 2013 being 620/231 = 2.7, while its deficit ratio in Fig. 6 is 1,532/504 = 3, quite significantly close. From Table 2, COG has 42% higher production efficiency than CHK, in collaboration with the direction (favorable to COG) of the foregoing deficit ratio.

The question to ask at this point is not how could virtually all of our select Marcellus Index operators reach such precarious condition, but how long can they sustain their immersion in the Red Sea? Will Wall Street keep rescuing our heroes? Stay tuned for some further troubling questions in Part 2 of this article.

4. Conclusions.

4.1 The projection of 4 TSCF/y by end of 2014 is dependent on adding 1,262 (25%+) new wells in the Marcellus during this year, thus adding 1 TSCF/y (+33%) production capacity. There is no assurance whatsoever that this is being done. To the contrary, nearly every operator in our Marcellus index has indicated it has expanded its interest in wet shale, or shale oil, or conventional oil production, elsewhere. Presumably reassuring shareholders that they are refocusing on more profitable operations.

4.2 An additional troubling aspect, the Income Statements of each and every one of our 8 Marcellus Index members shows ZERO for R&D. No interest, apparently. At a cost of $6.4 million/well, with production of an uncontrollable initial pulse followed by slower prolonged decline, 5-6 million gallon/well water used for fracking, and radioactive effluent waste, equals complete satisfaction? To those awaiting another miracle to increase proven reserves in the Marcellus, good luck.

4.3 At current NG prices, the Index collectively appears to have all the drivers for a high-CAPEX, high-volume, and low-margin business model, as noted in a recent Oxford Energy Institute study of US pricing of NG; perhaps resembling utilities. Large international oil companies (IOCs) tend to do rather poorly. This is a pity, as the sector needs participants who can weather the initial turbulent times and possibly invest in better technology. Our earlier article inferred a significant increase in NG prices in the near term. This would be good news not only to Marcellus operators, but also to the economy, as well as to the US energy future.

Disclosure: I am long XOM, APC, RDS.A. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.