Bakken Oil Producers: IP30 Data And Well Metrics Update - Economics Of The Average Well

|
Includes: CLR, COP, EOG, EQNR, ERF, HES, HK, MRO, OAS, QEP, SM, WLL, WPX, XOM
by: Karl Francis

Summary

IP30 trends (oil) are shown for 17 Bakken producers.

Additional well metrics and info is provided for major producers.

Early 2017 data are discussed.

Required WTI prices for certain IRR and cash flow levels are specified.

Data for the following companies are shown: Hess (HES), Whiting Petroleum (WLL), Continental Resources (CLR), XTO Energy (XOM), Oasis Petroleum (OAS), ConocoPhillips/Burlington Resources (COP), Qep (QEP), Statoil (STO), EOG (EOG), Marathon Oil (MRO), SM Energy (SM), HRC (HK), Newfield Production, WPX Energy (WPX), Petro Hunt, Enerplus (ERF), and Zavanna.

The article is a follow up to the article published here. That previous article shows the production trends since 2014 until the first half of 2016. The current article focuses on full year 2016 and Q1 2017.

IP30 (oil) means average daily oil production of a well in its peak production month. It's usually expressed in barrel per day (bopd). Oil includes condensate. The IP30 month follows the completion month with a lag of one to several months; so, the correlation between both numbers isn't perfect.

The official NDIC (North Dakota Industrial Commission) database lists 71 operators having drilled horizontal wells in North Dakota's Bakken. The table below shows average IP30 values and the associated number of wells for all active companies which had at least 40 IP30 wells since the beginning of 2014 and 10 wells with Ip30 in 2016. They represent 97% of all horizontal wells.

The numbers are based on sales (runs). On the aggregate level presented her, the difference between runs and production is in the order of 1%. The reason for choosing runs is that the runs data of confidential wells are as up to date as the date of non confidential wells.

Nb

Nb

Nb

Nb

IP30

IP30

IP30

Total

2014

2015

2016

2014

2015

2016

BAKKEN

4485

2058

1606

821

559

594

690

CONOCOPHILLIPS/BURLINGTON RES

287

76

128

82

748

523

548

CONTINENTAL RESOURCES

437

227

154

55

436

555

615

ENERPLUS RESOURCES USA

50

15

18

17

1396

1184

1024

EOG RESOURCES, INC.

181

87

48

46

949

817

760

HESS BAKKEN INVESTMENTS

546

223

233

90

640

633

669

HRC OPERATING, LLC

128

55

39

34

922

1026

934

MARATHON OIL COMPANY

144

68

59

17

558

645

1055

NEWFIELD PRODUCTION

113

54

43

16

536

583

894

OASIS PETROLEUM NORTH

364

194

101

69

429

564

801

PETRO-HUNT, L.L.C.

100

63

19

18

586

507

584

QEP ENERGY COMPANY

209

98

62

49

801

967

838

SM ENERGY COMPANY

99

20

32

47

359

291

263

STATOIL OIL & GAS LP

188

70

88

30

444

461

660

WHITING OIL AND GAS

520

261

199

60

549

652

974

WPX ENERGY WILLISTON

103

49

31

23

780

747

1120

XTO ENERGY INC.

370

128

138

103

515

468

518

ZAVANNA, LLC

42

7

20

15

764

754

713

821 wells had IP30 in 2016, 433 in the first half, 388 in the second half. The first half number is somewhat lower than the corresponding number in the previous article as the now available longer time series of production data have revealed that some of the H1 wells had their real IP30 in later production months. That adjustment has lifted average first half 2016 Bakken IP30 from the previous number of 631 to 638 bopd.

XTO had the highest number of new IP30 wells in 2016, closely followed by Hess and ConocoPhillips/Burlington.

Production decline rates

Average production decline rates of the wells vary quite significantly from company to company and are a lagging indicator because one ore more years of post IP30 production data are required to have meaningful information. That limits the analysis to wells with IP30 until the first half of 2016 currently.

The most important differences between companies show up in the first production year. After the second year, when production is already quite low, it's assumed that year/year decline rate converge, but the decline in the first year defines the level of the tail.

Two key figures characterizing first year decline are used:

  • First year IP30 multiple (the total first 12 month production starting with the IP30 month expressed as a multiple of the IP30 month production. That value has a meaningful impact on the economics of the well.
  • First year decline rate (the ratio of 13th month production vs. IP30 month production). That measure is a reasonable indicator of the level of the long tail of the curve - the production level relative to IP30 in the following years.

Both measures are sometimes corresponding (high first year multiple and low first year decline rate), but not always. Some companies use pressure management to flatten the peak. That results in a relatively higher first year IP30 multiple, although production may then show a steep decline in the last months of the first year, resulting in a rather mediocre First year decline rate.

The company specific values are mentioned in the company related discussion below.

For the Bakken average horizontal well with IP30 in 2014, the First year IP30 multiple was 5.23 and the First year decline rate was around 76%. For the details, see here.

The two metrics didn't change significantly for wells with IP30 in 2015, compared to 2014. Horizontal Bakken wells completed in the 2009 to 2013 period had more moderate first year decline values of 65% to 70%. The deterioration can be attributed to tighter spacing in the first place.

Average Bakken IP30 performance

Average IP30 has been relatively flat from the second half of 2014 to the first half of 2016, but made a significant jump in the 2nd half of 2016, lifting the 2016 average to 690 bopd. There are two main reasons for this. The first is the extraordinary IP30 performance jump of a few companies, notably WLL and OAS. The second is the strongly falling contribution of under-performing XTO wells (from 83 new wells with IP30 in the first half to 20 in the second half of the year).

Selected companies

The companies showing the largest number of wells with IP30 in 2016 are discussed in alphabetical order below.

For 3 of the mentioned companies (XTO, COP, and STO), tight oil is only a side show and little information is known about well design.

The mentioned numbers for net acreage owned by the companies are derived from company information. The quantity is no indication of quality. In several cases, the acreage quality (measured in oil production) of a company varies by more than a factor two. Nearly all operators have focused completions in 2016 on their most productive acres.

For several companies, average IP30 performances vary by 10% to 20% from quarter to quarter. Even half year numbers display a significant volatility. That's the result of the relatively low number of wells. The productivity of one SU (Space Unit) can have a significant impact on the company's averages. "Wells" means wells with IP30 in the period.

ConocoPhillips/Burlington

COP is the second largest player by acreage in the Bakken with about 620k net acres. It had 82 new wells with IP30 in 2016. That was less than in 2015, but more than in 2014.

COP wells had their peak IP30 in the first half of 2014 (800 bopd) and had then stabilized in the 550 bopd area. Judged by the decline rates, the later wells seem to have been less pushed than the 2014 wells, with emphasis on steady performance. First year decline has been around 70%. Late in 2016, COP has changed designs, using much more proppant/ft, with good results showing up in 2017.

Continental Resources

Continental is the largest Bakken player with approx. 800k net acres. The company touts it high quality acreage, but most of it is outside the core fields. Continental had 56 wells with IP30 in 2016, one quarter of the 2014 number and just 18 wells in the 2nd half of 2016. With the announced increase of proppant per feet to the 700 lbs to 1200 lbs range, more fracking fluid and more stages per well, coupled with high grading, IP30 performance has increased significantly in the second half of 2016, reaching the Bakken average. CLR used to have higher IP30 multiples and below average well decline rates in the past, compensating for lower IP30 numbers. It remains to be seen how the decline profile of the recent more pushed wells, together with more aggressive flow backs and lifting, will impact the well decline profile.

EOG

EOG has pioneered enhanced completion in the Bakken. The company started to use high levels of sand and frack-water in 2012 and generalized that method in 2013, routinely applying 1000 to 1500 lbs/ft of sand. More than 2000 lbs/ft have been tested in 2013, but apparently didn't show better well economics.

Nevertheless, well productivity is trending down since early 2014. From an average of 1200 bopd in the first half of 2014, IP30 has fallen to 760 bopd in 2016. In 2015, EOG mentioned that for its 2015 Q3 wells, it had switched to high intensity fracking (more fracks per stage), with a reduced frack radius around the bore, to reduce interference. That's like placing a bigger tap on a smaller reservoir. These wells show very steep decline rate between month 12 and 18. Later wells show more moderate decline trends. Overall, the decline trend of all 181 EOG wells with IP30 since 2014 has been worse than the Bakken average. After 18 months, production was at 15% of IP30 and after 2 years below 11%. That's only 75% of the Bakken average well.

EOG is now clearly focused on its non Bakken areas and seems to work off its Bakken inventory of 230k acres.

EOG's said that its 2016 D&C costs were below $ 5M, despite enhanced completions. That is the lowest reported D&C cost in the Bakken. It may be the result of vertical integration and ignoring cost of capital of the production factors. By contrast, D&C costs of EOG's Delaware wells, typically only with 5000 ft laterals, are given with $ 7.8M, a $ 3M difference.

Hess

Hess has 570k net acres in the Bakken. It has produced the highest number of Wells with IP30 since 2014, in sum close to 550 Wells. In 2016, it had 90 wells with IP30, the second highest number of all Bakken producers after XTO.

IP30 performance of its wells had been relatively flat since 2014, oscillation in a range of 600 bpd to 650 bpd until the middle of 2016. The company had said to apply an optimized and standardized completion design, which can be qualified as high intensity fracking (50 stages) but only using a relatively low amount of sand (550 lbs/ft). The company said that it was not targeting the highest IP30 numbers but the best well economics. In the 2 nd half of 2016, Hess has started to increase proppant per feet to the 1000 lbs level. For 2016 IP30 was 5 % above its long term average but remains slightly below the increased Bakken average.

The company's well production decline metrics have constantly been the worst of all the major companies observed here, with first year decline rates around 83% and a first year IP30 multiple around 4.5. Hess wells are tight spaced.

Taking into account the steep decline rate, production of a Hess' average well was Bakken average in the past, but has underperformed the Bakken average in 2016 by 10% to 20%.

According to Hess, its D&C costs are $1 M lower than the costs of competitors (except EOG), for a similar enhanced completion design.

HRC

The IP30 of HRC wells has oscillated in the 900 to 1000 bopd range since 2014. In 2016, the number of new wells was around 60% of the 2014 level. First year IP30 multiples have been 10% below the Bakken average. Recently, HRC sold most of its Bakken assets and is now focusing on other plays.

Oasis Petroleum

Oasis has around 500k net acres in the Bakken and is the purest Bakken play. In 2014, the IP30 of Oasis' wells was significantly below Bakken average. Performance has since increased significantly and has made a very impressive jump in the 2 nd half of 2016, with IP30 above 1000 bopd.

One reason for OAS' increased IP30 performance over the years has to do with improving acreage quality through land lease purchases. Regarding well design, for the first half of 2016 OAS said that all of its new wells were benefiting from enhanced completions. Oasis had tested two variants, one with 1000 lbs/ft sand and the other, called slick-water well, with 450 lbs/ft sand+ ceramics and a very high level of frack-water. The slick water well was apparently the preferred design then. But in the second half of 2016, the high proppant variant, adding $ 1M to D&C costs became preferred and has contributed to the rise in IP30 values.

Another factor explaining the IP30 jump was the concentration of drilling in the company's core acreage. According to OAS, 22% of its location inventory is core. With the projected drilling rates for 2018, the core capacity would last until 2021-2022. After that, OAS would have to move to the "extended core" locations, with average well performance being only 54% of core well performance. The wells of the third category, "fairway", will show only 37% of core well performance (all derived from company presentations, assuming a similar shape of the type curves and the same well design and high proppant volumes per ft)

On the negative side, decline rates have worsened. First year IP30 were 10% below the Bakken average. First half 2016 wells indicate that the first y/y decline rates continue to deteriorate. The IP30 push and maybe tighter spacing seems to come at a price.

Qep

Qep is a relative small player in the Bakken with 116k net acres. Relative to the size of its acreage, Qep has maintained a rather high completion level. Qep had 49 wells with IP30 in 2016 - roughly half the number of 2014 wells.

The quality of Qep's acreage is above average. IP30 reached a peak in the first half of 2015 with over 1000 bopd trough the application of enhanced completion techniques but has continued to fall back since than by 20%. Y/Y decline numbers are 10% worse than Bakken average. Like EOG, Qep seems to work off its Bakken inventory.

Whiting Petroleum

Whiting is the 6th largest player by acreage in the Bakken with 450k net acres, before the recently announced sale of top acreage, to reduce debt.

Whiting's wells have shown Bakken average IP30 performance until the middle of 2015, when enhanced completions with high proppant use was embraced. IP30 of later wells has risen strongly, reaching more than 1000 bopd in the second half of 2016, after displaying some weakness in the middle of the year. The IP30 increase in the last years has been accompanied by a slightly steeper decline rate than the Bakken average but not as steep as that of OAS wells. WLL's D&C costs are similar to OAS' well costs and close to the Bakken average.

SM Energy

The IP30 performance of SM wells went from bad to very bad in the 2014 to 2016 period, falling below 300 bopd in 2016. Apparently unfazed, completion volumes did not decline, with the number of IP30 wells even reaching a new high in the 3th quarter of 2016.

SM has tried to sell its Bakken assets but, not surprisingly, there was no taker. The 3rd quarter of 2016 seems to have been the last period with completions and new wells with IP30.

Statoil

Statoil has 250k net acres in the Bakken. With 30 wells in 2016, one third of the 2014 number, Statoil fell out of the top 10 in 2016. The IP30 performance of Statoil's wells has been significantly below average and relatively flat in 2014 and 2015, but show a jump in 2016, reaching 750 bopd in the 2nd half of 2016. The company has apparently employed enhanced completions for these wells. The former low well performance was associated with significantly below average decline rates in the first and second production years until 2016, compensating for the low IP30 numbers. Preliminary numbers for new wells in 2016 indicate that decline rates are now closer to Bakken average.

XTO

XTO has close to 500k net acres in the Bakken, competing with OAS for 4 th place. IP30 of XTO's wells was below the Bakken average in the last 3 years. In the 2nd half of 2016, IP30 dropped to the abysmal 300 bopd area. XTO's level of activity didn't react to the low oil price until the middle of 2016, but then dropped like a rock.

Well decline rates and first year IP30 multiples of XTO's 2015 wells have been a few % better than the Bakken average, suggesting that well performance and space density wasn't really pushed.

Outlook for 2017

The mid- to top performers (IP30 wise) are now all using enhanced completion techniques with proppant/ft typically in the 1000 to 1200 lbs/ft range. That's lower than the proppant intensity of Permian wells, but the benefits of using even higher proppant volumes may not be as high as in the Permian.

ND-Bakken completions have been around 60 per month in the first months of 2017. From the NDIC Director's cut information, one can derive that the average rig drills 16 wells per year. With the rig count having increased to 50-55, monthly completions should increase to the 80 level later in the year, assuming a constant DUC inventory. That would mean total 2017 completions close to 900 wells, an approximate 15 % increase over 2016 (in 2016, more wells had IP30 than were completed).

Earlier in the year, CLR, COP, HES, MRO, OAS and WLL had announced drilling/completion programs implying a total increase in net completions in the order of 150 wells above their 2016 activity levels. Recently, several companies have reduced their targets.

On the other side, the probable absence of new SM wells and the reduced activity level of XTO could result in a reduced contribution from these companies by up to 100 wells, compared to 2016.

Early 2017 IP30 numbers

The following remarks are based on Q1 2017 IP30 production data.

The North Dakota Bakken total number is below 200 new wells. Bakken average IP30 has moved comfortably above the 700 bopd level, maybe reaching 750 bopd in the first half of 2017. The main reason seems to be the strongly reduced contribution from under-performers like SM and XTO, compared to 2016.

Company specific IP30 data are based on small numbers of wells and tend to jump significantly from one observation period to the next. The following remarks should take that into account. The mentioned numbers are subject to moderate revisions, when longer production time series become available.

EOG had the largest number of new wells (30 wells) with average IP30 far worse than in 2016. IP30 was 20% below the Bakken average in Q1 2017. In 2013 until the first half of 2014, EOG wells outperformed the Bakken average by more than 100%. Clearly, the Parshall field seems to get exhausted.

Statoil was the second largest contributor (21 wells). IP30 was 20% below the company's 2016 level and more than 30% below the Bakken average.

HES, CLR and QEP were hovering around the Bakken average, based on 12 to 15 wells each.

XTO (14 wells) has recovered somewhat from the abysmal numbers of the 2nd half of 2016, but with an IP30 around 450 bopd, well economics can expected to be awful.

OAS, WLL and WPX's new wells continue to show IP30 around 1000 bopd, based on 8 to 10 wells per operator.

Well economics - the average Bakken well

The well economics are expressed as the required WTI price to achieve a certain internal rate of return, or IRR, on equity. Input assumptions and calculation methods are detailed here. With the now available data, the results for the full cost case are as follows

2016

Required WTI for the average Bakken well in 2016 (IP30 = 690 bopd)

IRR = 0% (that's close to US GAAP break even): $ 56

IRR = 8.8% (that's the cost of equity in real terms according to Damodaram): $ 69

IRR = 15% (that used to be the targeted return for energy investments outside shale):$ 77

2017 estimates

For the 2017 average Bakken well, with an assumed IP30 of 750 bopd (920 boepd), average D&C costs increasing to $ 7.5 M because of rising completion costs and a higher ratio of enhanced completions, a first year IP30 multiple of 5.15, a slight increase in the real cost of equity to 9%, other things being equal, the required WTI prices would be as follows

IRR = 0%: $ 53

IRR = 9%: $ 68

IRR = 15%: $ 76

The required WTI prices in 2017 are close to the 2016 numbers. The higher IP30 number is offset by higher D&C costs and a slightly deteriorating first year IP30 multiple. The EUR of that average 2017 well is 637k boe.

Required WTI price for cash flow financing - Bakken well with IP30 of 750 bopd

Many companies say that they will now operate within their cash flow generated from production. But at $ 50 WTI, reinvesting all free cash flow (no payout to equity owners) is not enough to maintain the added first year production of a new well. Based on the assumed metrics of the average 2017 Bakken well and reinvesting all free cash flows from op., the production level resulting from the new well investment would fall to 40% of the first year production level after 10 years with WTI at $ 50/b.

A WTI price of $ 64/b is required to maintain the annual first year production of a new well on average over the next 10 years by reinvesting all cash flows under the above scenario. The problem is that the currently drilled core acreage gets exhausted sooner or later, requiring an increasing WTI price over time.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

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.