Pompe à filtre à sable : black friday – choisir – promo – Abrupt Reversal Of Shale Oil’s Fortunes Points To A Radical Reset Of Oil Prices

Last Tuesday, the Energy Information Administration (EIA) predicted that shale oil production in February will hit an all-time record.

One thing it didn’t mention:

I left out the predictions for January and February, and the latest actual is December.

Yes, shale oil output is growing. But the rate of growth more than halved in a year. The last time growth dropped anywhere near that fast was in 2015.

Something changed, and there is no sign that whatever did change is going away.

What changed?

Well, a week ago, WorldOil reported that Halliburton (NYSE:HAL) and Schlumberger (NYSE:SLB) said:

« U.S. shale oil fracking has already peaked and is in a period of sustained contraction. »

« Halliburton cut 22% of its frack fleet last year, Schlumberger has already cut its fleet in half; it has no intention of bringing that equipment back into service. »

You would have thought that Halliburton and Schlumberger, the two biggest pumpers in U.S. shale oil, might have a better idea of what’s going on than EIA or Rystad Energy that have been predicting a rosy future for shale oil.

They talk to the customers. They know what’s going to happen in six months’ time, and hiring a frac spread is not like ordering a cab from Uber. EIA and Rystad just know what happened last month.

Why shale peaked

For shale oil, the biggest single cost is the frac, which typically takes about three days, and then you get the oil. No frac, no oil.

The number of frac spreads working is a measure of how aggressively shale operators are going after oil.

In 2014, it took seven months before the penny dropped that the oil price might stay down, i.e. before the frac spread count started to drop.

That was perhaps because in October 2014, Goldman Sachs, which at the time Reuters called the most bearish of all the analysts they polled, predicted an average of $85 in Q1 2015.

The stampede only started once the price dropped below $60 WTI. Then there was a pause after it climbed back from $45 to $60, which suggests then the average break-even was about $60 WTI.

So anyone who was active before the price bust was likely making a spectacular profit, which could explain why operators piled in with such enthusiasm, and why, after the writing was on the wall, so many hung on for too long.

But most of that profit was on paper because the expensive new equipment that powered the first boom was bought on credit.

When the price dropped below $40 (with the helpful encouragement of OPEC), the liquidations started. Those who survived, and the big boys like Halliburton and Schlumberger, were able to buy hardly used frac spreads and the like for pennies.

But the survivors all had debt, or in the case of the big boys, balance sheets to square away, so when the price finally edged above $50 WTI, they started going back in, putting the idled equipment to work, for pennies.

No doubt, heavily pressurized by their bankers and bond holders to at least make enough money to pay interest; maybe pay a little principal, but forget about paying dividends.

So then there was the second shale boom.

There is an opinion making the rounds that not much money was made by the operators in the second boom. All they managed to do was pay down debts taken on during the first boom to some extent and survive. Many didn’t.

Both Halliburton and Schlumberger took hefty write-downs on that episode.

Looking to now:

At first sight, the recent sharp drop in working frac spreads looks extremely odd. Did something change?

On the last two occasions that the price was in the range of $55 to $60 WTI, the frac spread count was a high of 400. Now it’s 300, and the trajectory looks decidedly down.

In the first half of 2017, when WTI was hovering around $50, operators started jumping in with both feet. This time it’s been toying with $60 and they are jumping out.

Make up your minds guys! Supposedly you have magic new technology and Rystad Energy says your break-even is less than $40. Don’t forget, you have debts to service and investors to look after. Go to work!!

But perhaps operators and pumpers have caught up with their debts and so they are no longer being pressurized to work for break-even plus a razor-thin margin to cover the debt service?

If you have debt around your neck, and your customer says, « I see you been stacked for six months, let us do you a favor and help you out », you roll over and let them scratch your tummy.

More to the point. Your banker compels you to take whatever silly price is on offer, so long as you make an operating profit, and forget about the cost of wear and tear on your equipment, and don’t even think about paying dividends.

If you are debt free, it costs you almost nothing to stack, or big enough to talk back to bankers, you hit the customer with an outrageous price. But of course never so high they can’t afford it.

That’s called Parasite Economics. It’s what Halliburton just did with its frac spreads.

Perhaps also some frac spreads are worn out?

A typical spread does 30-to-40 fracs a year. Some have been working flat out for more than five years. Major overhauls of the 50,000 HP power-packs which are the heart of a spread do not come cheap. Also the sand eats up the pump-impellers and housings, and cavitation and surges blow things apart.

What’s certain is no one is forking out $30 to $60 million per spread to buy new maybe because they can’t get credit anymore, and since they haven’t been paying dividends, they lost touch with equity investors, or maybe like Halliburton and Schlumberger, they saw the writing on the wall. Maybe because operators can’t pay enough to give a decent return on a new rig.

Whatever the reason, no one is buying new, and the ones that are left are falling apart.

Initial Production

The critical benchmark in shale is initial production (InPro). That’s how much new oil flowed in the first 30 days after you cracked open the tap. It’s reported by EIA in its monthly Drilling Productivity Report (DPR), albeit somewhat confusingly. You have to do some arithmetic, which may be why the journalists don’t track it.

The total increase or decrease in production capacity in a month is InPro minus legacy loss. Legacy loss is how much the existing wells declined. That’s about 6% of total production per month.

EIA publishes those numbers on the third Tuesday of the month, so now we know what happened in December.

InPro minus legacy loss is a measure of the capacity to produce oil. It is a better benchmark for the trend than month-on-month changes in production, which bounces around:

Long term, the average of the jagged line is the same as the smooth line. Notice the trend. Neither of the lines are going up.

Looking to the future, legacy loss is highly predictable from total production.

InPro next month can be reasonably accurately predicted from the frac spread count, which is logical. The best-fit algorithm for that over the past year is:

Where FS means numbers of frac spreads this month, and FSL3 means the numbers three months back. The RSquare on the model plotted against actual is 92%, and the standard error is 2.8 % of the estimate.

That’s a pretty good model albeit on not much data. Interestingly InPro depends more on what spread count was three months ago than this month, which suggests that for many completions, it takes two months from the frac until the oil starts flowing.

Primary Vision tweets the frac spread count every week, so you can make a pretty accurate estimate of what IP will be from that, and you can be two to three months ahead of when EIA reports InPro.

This is how that looked:

There was a small change in the direction of the trend in InPro in December. That could be because in the week ending 20th December, 28 spreads were pulled presumably for the Christmas break. 28 came back over the past two weeks. But the 29 that were pulled after Christmas might not be coming back. Perhaps some spreads are more productive than others, and it’s the weaker ones that get pulled?

If you believe what Halliburton and Schlumberger are saying, the frac spread count is not likely to go anywhere but down in 2020.

Looking at the same story from another angle:

That’s the line of frac spreads from the second chart superimposed on the first.

Notice how the number of frac spreads is a leading indicator of total production. Notice the dive in frac spreads from October. That’s what is hitting production right now. Notice what’s about to happen in February.

The future

This is a scenario for what if from now the frac spread count bucks the trend and stays at 300 units for the next 10 months:

The predictions of future production for the scenario « what if frac spreads stay the same » add up to an annual drop in production of 200,000 barrels per day for full-year 2020.

Oil Prices

The oil market is currently priced on the premise that the predictions of EIA and Rystad Energy of plus 900,000 barrels per day in 2020 are about right.

So what if it goes down 200,000?

That’s a difference of 1.1 million barrels per day; talk about a missed expectation if that happens!

What would happen if OPEC announced tomorrow it would take that much more oil off the table? Would there still be a glut? Might the price go up?

In an excellent piece on WorldOil (referenced above), David Wethe wrote:

« That view from Halliburton Co. and Schlumberger Ltd. signals an eventual deceleration in U.S. oil production, which is currently at record highs. Slower output growth would have global ramifications, given additional American barrels are forecast to account for most of the increase in worldwide supply this year. »

« Global ramifications » quite so. If you believe Halliburton and me starting about now!

Putting that into simple language, if the assumptions the markets are relying on turn out to be wrong by 1.1 million barrels per day, or if you believe me, probably more, there could be a spike.

At $75 Brent, operators still won’t be able to get credit to buy new frac spreads until the price stabilizes there for six months.

Spreads take nine months to a year to make. So if Brent goes to $75 next month, say, it won’t be until June 2021 that the new spreads start to be deployed.

Whether the reason operators are pulling and stacking frac spreads is that now they can afford to wait for better prices (like Halliburton), or because, as Primary Vision seems to be saying, frac spreads are falling apart, or both or even something else, the prospects of a change in direction seem very slim.

Personally, I can’t think of any reason why operators might start jumping into frac all those DUCs, with the same enthusiasm they had in 2017 until Brent hits $75.

When we will see $75 will depend not so much on when the current glut gets used up, but more likely it will be when the market wakes up to the reality that change is in the air. In 2014, there was a small glut, but the fear that shale oil would lead to an oversupply was what triggered the massive reset.

A reset the other way, if it happens, might well be equally dramatic.

If glut turns to the supply crunch Schlumberger was warning of a year ago, the price could go even higher than $75.

So could there be a spike, like to $90 Brent in 2020?

With oil prices tanking because of the China virus, and inventory builds, and X-Y-and-Z, that seems a frivolous question.

For now, the prevailing wisdom is that U.S. shale oil is fine, the headlines say production is breaking records, and EIA, Rystad Energy, and IEA, all highly regarded in the industry, are saying the growth in 2020 will be 900,000 bbl/day – steady as she goes.

Last November Reuters published a poll they did of 42 analysts and economists, which predicted that Brent would average $65 in 2020 and range between $60 and $70. And there would be no fireworks!

Here’s a thought:

Perhaps the consensus of the 42 experts might be wrong? After all, even experts can get it wrong from time to time.

For example:

In December 2013, Reuters published a poll of 27 analysts which predicted Brent in 2014 would average $100, and wait for it, there would be no fireworks.

In October 2014, after Brent had dropped to $90, Goldman Sachs, one of the analysts polled by Reuters, revised their prediction from $100, down to an average of $85 for Q1 2015.

Reuters reported that then Goldman was « the most bearish of any of the major financial institutions. »

Goldman got its second prediction wrong too.

The Q1 2015 average came in at $54.

Brent bottomed a year later at $27.

No one talks about how come so many experts got it so wrong.

To this day, there isn’t any consensus on why what happened did happen suddenly out of the blue like as if a black swan pounced.

The November 2019 Reuters poll of 42 analysts was presumably of the same analysts they had polled in 2013, plus some extra for safety in numbers. Presumably all using the same tried and trusted methodology they used in 2013?

If this time they are out by a factor of two, like last time, except the other way: 2 x $65 = $130

I submit that over the past four years, the price of oil was dictated by the price shale operators were forced by banks and bond holders to sell it for.

That’s not the normal way the oil market works. Normally the price is dictated by what customers can afford and not go excessively into debt, because they can add more value doing things with the oil. Parasite economics.

In October 2014, when Goldman Sachs put out its prediction for $85, one lonely analyst of a mildly Austrian persuasion predicted that oil would bust down to $60 and then stay down for four years.

If OPEC had not decided to drive the price down to $27 to kill shale, that might have been what would have happened, so the prediction was half right.

The one-eyed man perhaps? The same lonely analyst now says, at some point, probably within the next year, Brent will go up over $90 and stay there for a minimum of three years, and it could indeed spike to $130.

That prediction has nothing to do with frac spreads. It’s based on an estimate of what the customers can afford and observation of the dynamics of bubble bust laid out by Ludwig von Mises and Bob Farrell.

But it is 100% consistent with the story about frac spreads.

If that turns out to be right, then finally offshore E&P stocks like RIG might be let out of their misery.

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.