A short blog for a long weekend.
I have been reading a lot in the media lately about how in the context of this downturn, the rig count doesn’t matter, because of a variety of factors with the most prominent being efficiency gains, the fracklog (wells drilled but not completed) and higher initial production.
It is theorized that all these factors contribute to making the rig count irrelevant, the prevailing wisdom being that it is possible to do more with less, production enhancements were always low hanging fruit anyway and an inventory of wells made drilling obsolecent. Ergo the stubborn production numbers (which of course become less stubborn by the week).
At any rate, I thought to myself that it might be worth a little digging. Just for fun. Note that this analysis is oil only and confined to the United States since they have such great data.
First stop on our tour is the context of the actual rig count, which we know has declined a wee bit since January 2015. More precisely, the number of rigs drilling for oil in the US has fallen from 1500 to about 300 or a staggering 80%.
Production during that time went from 8.6 mm bpd up to 9.2 mm bpd in June of 2015 (last peak) before declining to 8.3 mm bpd today. This represents a decline of 4% and 10% respectively, pretty far off from 80%, right?
So clearly, drilling is more efficient so the rig count doesn’t matter.
After all, what is more attractive in the face of a downturn than a more efficient industry, able to eke out gains by rationalizing operations and innovating? Squeezing more molecules out of the ground by being relentlessly inventive? A simple glance at the EIA data shows increasing levels of new well productivity per rig, even as the rig count falls. Pretty awesome.
Interestingly, taking a look at the “big 3” basins – the Bakken in North Dakota, the Permian Basin and the Eagle Ford in Texas and finally adding the Niobrara in Wyoming/Colorado, a different if somewhat muddled picture emerges.
On an aggregate basis, this represents tight oil – at April 30 2016 about 4.8 million barrels per day or more than half of US production. Why just look at tight oil? Because this is what the media wants to talk about. Legacy production? Offshore? Alaska?!?!? Who cares, it’s all shale all the time.
Using just these areas, rather than the 4% reduction from January 2015 and 10% from the peak, the numbers are actually 6% and 7%. What skews these numbers is that while production from the Bakken, Eagle Ford and Niobrara is off an average 19% from the peak in June 2015 and 16% from January 2015, production in the Permian actually increased by 13%. What?
So what gives? Why is the Permian growing? Is this the efficiency story we have all been looking for? And why do the other areas not perform as well?
A simple look at the rig count for each region is informative, as each region saw a sharp decline into 2015 and then a gradual reduction through 2015 until today, except the Permian which actually saw an increase in the middle of last year, coincident with final spike in production.
Also, the Permian basin had by far the largest amount of active rigs (and still does) which means that its fracklog is likely the highest, so it is the most likely area to have the stickiest production numbers.
Also of note, relative to production, the Permian has the lowest decline rate on legacy production of each key region, allowing it to hold up further.
In other words, more rigs, larger fracklog, lower decline rate…
All in all, this may actually serve to bolster a better theory, in that in a downturn, producers will allocate resources to the highest quality land holdings in the highest quality areas in the highest quality basins and lowest cost resource plays in order to extract cash flow. In the case of the United States, this is clearly the Permian.
On the efficiency side, I looked a little closer at one of the more commonly cited EIA production numbers: new well productiivy per rig.
One can’t help but be struck by the data they present that shows significant and ongoing increasing productivity by rig. Pretty cool, but then I looked at the graph to the left and thought to myself how do they even come up with these numbers?
On the surface, they actually seems to come from simply dividing new production coming on stream by the number of rigs at the end of that period. Call me skeptical, but given the time to drill and complete a well and a cratering rig count, surely the only efficiency that this data shows me is a denominator that is half or less what it was a year ago.
Perhaps this is more reflective of the exploitation of the “fracklog” than any phantom increased per rig efficiency.
Speaking of that annoying term, I read recently that the US fracklog at December 2015 was at about 4,000 drilled but uncompleted wells, split relatively evenly amongst the main areas with the Permian having the most and the Bakken declining as shown in the chart to the right. At a typical completion rate of 100 wells per month, actual distribution among the plays and ignoring decline rates, the article postulated that (at various price levels) the most commercial production that can be added from this fracklog in 2016 is between 300 and 450 thousand barrels per day of production based on prices between $30 and $40 WTI.
So, call it upside 450,000 bbls of new production (presumably a third already added given the time elapsed in 2016 so far) and a decline in legacy production of between 150,000 and 250,000 barrels a month and only 300 rigs beavering away in the tight oil regions. Tell me again how efficiency trumps that?
And with the rig count so low and no money for new exploration, how much is really going to get added to that fracklog?
Put another way? Look for production in the US tight oil areas to take another sharp dive. We likely still have another 500,000 to 700,000 barrels a day of decline to see this year (simple math – 1.6 mm annual legacy decline less 450,000 new barrels less current YTD decline of 400,000 leaves 750,000 bpd of decline to come).
So yes, indeed, the game is rigged. It’s rigged towards those areas that are the lowest cost and have the best immediate prospects. It’s rigged to producers who have drilled and uncompleted wells. And it’s rigged to the data crunchers who want the data to tell a particular story.
But to my simple mind? The rig count matters.
Special Contest Alert!
OK – week 3 of the contest. People are coming out of the woodwork
Up for grabs for the winner is a bottle of whiskey (of my choosing, don’t worry, I have excellent taste) and, of course, all the accolades and imagined immortality that comes from choosing the name of a blog with a narrow and targeted distribution.
So far I am working on these suggestions.
- The Weekly Stew
- Crude Observations
- Not as Boring as it Seems
- Your Week in 3 minutes or Less
- Stu’s News
Prices as at May 20, 2016 (May 13, 2016)
- The price of oil ended the week up
- Storage posted a surprise increase
- Production was down again
- The rig count was flat and appears to be bottoming out
- Conitnued production declines in the US are helping keep prices up as are the production shut-ins around the world, but negative noise out of Saudi Arabia and Iran is capping any rally
- Natural gas fell marginally during the week
- WTI Crude: $48.48 ($46.57)
- Nymex Gas: $2.062 ($2.096)
- US/Canadian Dollar: $0.7612 ($ 0.7730)
Highlights
- As at May 13, 2016, US crude oil supplies were at 541.3 million barrels, an increase of 1.3 million barrels from the previous week and 59.1 million barrels ahead of last year.
- The number of days oil supply in storage was 33.6, ahead of last year’s 29.8.
- Production was down again for the week at 8.791 million barrels per day. Production last year at the same time was 9.345 million barrels per day. The decrease in production this week was mixed as a recovery in Alaska deliveries was offset by a drop in lower 48 production.
- Imports were flat during the week
- Refinery inputs were up marginally during the week
- The effect of the Fort McMurray shut downs have yet to be felt in these numbers
- As at May 6, 2016, US natural gas in storage was 2,754 billion cubic feet (Bcf), which is 41% above the 5-year average and about 40% higher than last year’s level, following an implied net injection of 73 Bcf during the report week.
- Overall U.S. natural gas consumption data was up during the week by about 2%
- Oil rig count at May 13 was flat at 318 from the week prior.
- Rig count at January 1, 2015 was 1,482
- Natural gas rigs drilling in the United States was down 2 at 85.
- Rig count at January 1, 2015 was 328
- The collapse in the US rig count is likely over with most of the damage already being done
- As of May 16, with break up in full force, the Canadian rig count was at 36 (5% utilization), 25 Alberta (5%), 6 BC (8%), 5 Saskatchewan (4%), 0 Manitoba (0%)). Utilization for the same period last year was about 10%.
- US split of Oil vs Gas rigs is 80%/20%, in Canada the split is 20%/80%
- Offshore rig count was at 24
- Offshore rig count at January 1, 2015 was 55
Drillbits
- Fire crews continue to battle the Fort McMurray wildfire which now covers an area the size of Prince Edward Island or, for our American friends, all of Delaware
- Estimates of production shut in are about 1.2 million barrels, and although some restarts are slowly occuring (Imperial Oil Kearl River), it will be weeks before production returns to previous levels
- A work camp with 655 units was destoryed by the fire and others were threatened
- Canada’s National Energy Board (NEB) gave conditional approval to Kinder Morgans Trans Mountain Pipeline twinning project. The project, which is expected to come into service in 2019 will triple the capacity of the existing line and allow oilsands crude to be shipped to verseas markets. There are still hurdles to overcome, however this is an important first step. The project needs to meet 157 conditions imposed by the regulator and still requires apporval by the Federal Cabinet
- Oilfield service companies FMC Technologies of Texas and the French company Technip agreed to merge this week. The combined TechnipFMC would have a market value of $13 billion, and with 2015 combined revenues of $20 billion
- Drumpf Watch – Better yet, Watch Drumpf! This is courtesy of my 10 year old. Watch it, c’est bon.
