Commentary – Does Moore’s Law Apply to Rigs?
“How much more productive can these new wells get?” I asked my host who had kindly invited me on a field trip to some of western Canada’s most prolific oil fields.
Looking down the aisle of bobbing pump jacks, seven in a row on one side of the immaculate gravel pad, the veteran oil executive replied, “We can get up to 1,000 barrels a day out of some of the new ones, but that’s not a limit; we’re improving the economics and productivity with each new well.”
Impressive I thought, subconsciously nodding my head in sync with the leading pump jack.
“Back when I was in field exploration,” I said sounding like a grey-haired guy, “we used to high-five if a new well put out 100 barrels a day.”
The stats certainly show that the industry’s ability to pull oil out of the ground from parts of North America has recently improved by an order of magnitude. In other words, ‘rig productivity’ – the amount of new oil production that an average rig can bring on in one month of drilling– has increased 10-fold in the last 6 years.
Both of us stood marvelling at the operation in front of us, tacitly thinking the same questions. What are the limits to this remarkable energy megatrend: 1,000? 2,000? 5,000 barrels per day per well?
“All this appears amazing,” I said, “like some sort of Moore’s Law for guys in hard hats and overalls.”
Moore’s Law – the observation that the number of transistors on a computer chip doubles about every two years – is the gold standard for benchmarking innovation. But at the same time I explained to my host that I’m always cautious about being duped by exuberant technology, trends and numbers. For example, we’ve all subscribed to “high-speed” Internet services that claim dozens of Megabits per second, but grinds to a fraction of the advertised rate when half the neighbourhood is binge-watching Netflix. “The notion of ‘rig productivity’ has to be taken with caution,” I noted. “We can’t assume that the best posted performance in the field is the norm for all wells.”
My thoughts turn to the US Energy Information Agency (EIA), which every month publishes a report on rig productivity http://www.eia.gov/petroleum/drilling/.
July data from the Eagle Ford play in Texas – one of the most prolific in North America – shows that rig productivity in the area is now over 1,000 B/d per rig, similar to the Canadian play area I’m visiting. The rate of innovation has been impressive; between 2007 and 2014, when oil prices were above $100/B, rig productivity was increasing by 110 B/d per rig, every year.
But a curious thing is noticeable at the end of 2014: Rig productivity suddenly kicked up in all the US plays. The Eagle Ford was particularly impressive, going from 550 to 1,100 B/d per rig. I did a double take. Really? A doubling every 18 months? That’s better than Moore’s law!
The well pad I’m standing on is a testament to ongoing innovation, but there is a statistical distortion at play. Starting in late 2014, the severe downturn in oil prices forced the industry to park three-quarters of their rigs and “high-grade” their inventory of prospects. Producers focused on only their best rocks, drilling with only the most efficient rigs. All the low productivity stuff was culled out of the statistical sampling, skewing the average productivity numbers much higher.
It’s only an estimate, but the average productivity in the Eagle Ford using a wider spectrum of rocks is probably at least 30% less, in the 700 B/d per rig range. Higher oil prices are needed to attract rigs to those lesser quality locations. The much-vaunted Permian only averages about 350, despite showing 500 B/d per rig in the EIA data.
But top-end numbers like 700 B/d per rig should still be a cold-shower wake up call to high-cost oil companies, or to the champions of rival energy systems trying to supplant oil. Notwithstanding the high-grading of the rig sample to premium “sweet spots”, the average productivity in the best North American plays is still improving by about 100 B/d per rig per year, with several years of running room left. It’s not exponential like Moore’s Law, but the pace of innovation is on par with many trends in the tech world.
“We’ve got some of the best acreage and best practices in Western Canada,” said the CEO, looking into the distance and panning his hand across the company’s leased land.
“Yeah, it’s amazing, by the numbers you’re as good as or better than some of the best in North America,” I replied. “But ‘best’ also reminds me to consider that other areas are not as good.”