Fatih Birol, Chairman of the International Energy Agency was recently quoted as saying US oil production grew by 864,000 barrels per day, according to weekly estimates, since October 2017 (it was actually only 658,000 barrels per day from November 1, until the first week of February) and that it was astonishing. The comment was made after the EIA reported a weekly increase of 332,0000 barrels per day in the Feb 7, 2018 weekly Petroleum Status Report. The definition of astonishing is “extremely surprising or impressive; amazing.” But having said that, I won’t even get into the monthly production reports and 2018 and 2019 forecasts as they are so far “off” the weekly estimates.
Call me a heretic, but I am beginning to be very skeptical of the weekly estimates. They are too optimistic. As I have mentioned in previous posts, actual production data is not available until the third month after production. Estimates using other data or algorithms are just that, estimates.
Why am I skeptical of this ‘astonishing’ growth? First and foremost, the three-month dq/dt of 658,000 barrels per quarter represents an annual increase of 2.63 million per day, and the ridiculous weekly increase of 332,000 barrels per day in one week represents growth of over 17.2 million barrels per year! Second, and more important, both OECD and US storage numbers decreased over this period. The IEA has confirmed that OECD stocks decreased by 45 million barrels in December. In fact, stocks are approaching the 5-year average. And finally, the TRC has updated Permian production YTD 2017 versus 2016. In 2016, Texas Permian production averaged 1,559,432 barrels per day, which has increased to 1,689,474 bpd year to date as of November, representing an increase of only 130,042 barrels per day (the question is when did much of this increase take place – most likely in the back half of the year).
The only way to ground truth the veracity of the weekly estimates is too look back at actual state production numbers for the months of November, December and January. As this data becomes available, I will submit monthly updates. Rather than reporting on all states, I will only look at North Dakota, Wyoming, Colorado, New Mexico and Texas as all of the major shale oil plays are located in these states.
Maybe I’m just naïve, but I find it amusing how easily taken for granted the market hangs on every weekly production change while discounting the realities of the physical market. Oh wait, this is exactly what the hedge funds want! Volitivity arbitrage for short and long positions.
I have to take my hats off to the US producers making great strides in the success of the Permian, making it the pre-eminent play in the world. As to where the peak rate lies and how fast it takes remains to be seen. Since my last article on the Permian, I’ve looked into where activity is taking place. In New Mexico, activity is focused on Lea and Eddy counties which have a total area of 8,600 square miles. Texas activity is found in five counties, Reeves, Loving, Midland, Upton and Reagan which have a total area of 7,000 square miles. Together these seven counties comprise 20% of the total Permian footprint.
Should these counties comprise the Permian sweet spot, about 46,000 wells will deplete this resource assuming two section units with six wells per unit. That said, based on rig counts from 2011 to 2015, which averaged deployment of 520 rigs, and assuming one-mile laterals, one well per rig, and six wells per section, about 5,200 sections have been developed. In years 2016 and 2017 an average of 320 rigs were deployed drilling two-mile laterals for the most part, exploiting a further 2,560 sections. So, if my math is correct, a total of 7,760 sections have been exploited, leaving about 7,840 sections left, or 23,520 two-mile laterals, or a about a four-year inventory at current rig counts.
Do I expect that producers will find other sweet spots? Probably. Is it entirely possible that I have underestimated the inventory of wells? Probably.
My point is that production estimates are just that. Historical results and capital budgets define the trends, not algorithmic forward estimates.
Randy Evanchuk, P. Eng., has 35 years of experience in the patch. From 2007 until he retired in 2015, Mr. Evanchuk was involved in all phases of of unconventional resource development including;evaluation, economics, production and facilities. As as senior consultant with Murphy’s Holdings, he evaluated their Montney holding as well was as a member of evaluation team. Mr. Evanchuk was the Vice President of new ventures at Daylight Energy where his team was successful in acquiring a substantial Duvernay position. At Seven Generations Energy he was Executive Vice President looking after facilities, marketing, production operations and long range facility and marketing planning