#42. The petroleum switchback

Before I became a strategist back in 2009, I had spent more years than I care to remember working as an oil sector analyst, required to supply a continuous narrative on developments in petroleum markets. So the events of recent weeks, which have seen crude prices in free-fall, haven’t put me under the kind of pressure that I became accustomed to whenever the oil markets took a dramatic turn for the better or the worse.

Even so, I feel a need to provide some analysis here, not least because I have read so much nonsense on the subject of late, particularly where medium- and longer-term prospects are concerned. It was ever thus – a sharp fall in prices inevitably triggers predictions of long-run disaster for the industry, just as surely as a spike provokes dire warnings of sustained disaster for everybody else.

Let me cut to the chase. The recent crash in crude prices is eminently rational, and should indeed have happened quite a long time ago. No political sub-text is required to explain this slump, which is not to say that there are no such sub-texts as, pretty clearly, there are.

Second, things are likely to get worse – a lot worse – before they get better, and markets are likely to be far more glutted with oil by March than they are now.

Third, though, what we are witnessing is not the dawn of an age of cheap energy.

To understand why the slump in oil prices was not just inevitable but long overdue, take a look at the table below, which is based on OPEC data and is expressed in millions of barrels per day (mmb/d).

Oil supply and demand custom

Comparing 2014 with 2011, demand for oil has increased by 2.9 mmb/d, a number exceeded by the combined total of non-OPEC production (+3.4mmb/d) and non-conventional output from OPEC countries (+0.5 mmb/d). The overall effect has been to squeeze the market’s need for OPEC crude downwards by about 1.0 mmb/d, from 30.3 mmb/d in 2011 to around 29.3 mmb/d this year.

Moreover, this number could worsen further next year, under the combined impact of economic weakness and further increases in non-OPEC supply. Since much of what is shown in 2014 – and, specifically, the probably oversupply of close to 1 mmb/d – was eminently predictable, the only real surprise about the recent slump in oil prices is that it didn’t happen earlier.

If you understand how oil markets work, you can only anticipate further pressure over the coming three months. Essentially, there are two peaks in oil demand – the winter heating season, and the summer motoring (and flying) peak. These are at their most pronounced in January and August, respectively.

These are the periods of peak demand for refined products. The peaks in the purchasing of crude oil, however, anticipate these consumer peaks, because oil has to be shipped, refined and distributed before it reaches the customer.

Thus, any crude purchased after November isn’t going to reach the end-user until after the winter peak, which is why crude purchasing reaches its high-point in November. Thereafter, crude purchasing diminishes until it begins its run-up to the second (June) peak.

Hence, industry purchasing drops sharply between November and March, and this swing factor can easily be of the order of 5 mmb/d. This is why, come March, markets are likely to be glutted with oil. You can see how this works on this chart:

Crude purchasing

Of course, OPEC is always able, theoretically anyway, to shore up prices by reducing its production, something that on this occasion the cartel, guided by Saudi, has been conspicuously unwilling to do.

Some have tried to impute political meaning to this reluctance. Is the intention to bankrupt Iran, or Russia, or maybe both? This seems somewhat unlikely. Low oil prices will indeed inflict severe damage on the Russian economy, but, and as the forex markets have been telling us for months, Russia was already well on the way to economic catastrophe even without an oil price crash. As for Iran, the suffering inflicted by low oil prices will be severe, but it is by no means obvious that this will have a particularly acute impact on the policies of the Iranian government.

Rather, the unwillingness of OPEC (meaning, in effect, Saudi and its closest allies) to cut production results from two, essentially economic calculations. First, Saudi sees no reason why it should cut its export volumes and revenues to benefit its competitors.

Second, Saudi has probably calculated that a period of severely depressed prices can deal a mortal blow at competing producers, principally in the unconventional sectors of shale oil and tar sands production.

And this, of course, brings us to what used to be called “the sixty-four thousand dollar question” – what is the outlook for shale oil production?

This is clearly critical because, within the 3.4 mmb/d increase in non-OPEC production since 2011, no less than 2.8 mmb/d is attributable to increased shale oil output in the US.

According to the Energy Information Administration (EIA), shale output is scheduled to increase significantly further between 2014 (4.07 mmb/d) and 2021 (4.80 mmb/d) before entering a very gentle decline. With demand growth likely to be sluggish, this is clearly an unappetising prospect for Saudi, and one that they would dearly like to stop in its tracks.

Fans of shales seem to believe that low oil prices are unlikely to inflict significant damage to production, arguing (amongst other things) that improving technologies will drive costs downwards. Others, including me, believe that shale production is going to peak a lot sooner than the EIA believes, and that the subsequent decline will be very much sharper.

I am on record as believing that the economics of shales are far from robust, mainly because output from each well deteriorates very rapidly indeed, declining by 60%, or even more, within the first year of production.

This being so, the only explanation for the continued increase in shale oil output lies in an extraordinary rate of drilling, which has enabled output from new wells to more than offset declines from existing sources, but has also inflicted negative cash flows on the industry, even with oil prices in excess of $100/b. This is the “drilling treadmill”, which forces producers to keep drilling in order to stave off a slump in production.

If I’m right about the drilling treadmill, US shale oil production was already poised to enter a precipitate decline from 2018-19. This suggests that OPEC’s unwillingness to shore up oil prices could bring that downturn significantly nearer, not by causing the shut-in of existing wells but, rather, by blocking the flow of funds that keeps the drilling treadmill turning. We should have a pretty clear idea about the outlook for shales within a year from now.

Meanwhile, there is one factor that we should not lose sight of, and that is the cost-mix of the global oil production portfolio. That cost-mix continues to rise, and is the best long-run indicator of the outlook for prices. Output from big, low-cost fields continues to decline, and the cost of replacement production continues to increase.

The marginal cost of replacement barrels is now far higher than the market price of crude oil, which can have only one implication on any long- or even medium-term assessment of the outlook.

25 thoughts on “#42. The petroleum switchback

  1. Tim,

    It’s refreshing to see some commentary that isn’t centred on how Saudi wants to discipline OPEC members, crush Russia or bankrupt US shale. The simple truth is that last time they tried to defend a price floor they simply managed to ruin the economy for 20 years, and they’d rather not do that again.

    I agree with you on shale (especially the Red Queen treadmill), but the other big factor in all this is how much older conventional production is taken offline in places like the north sea. There’s plenty of old infrastructure out there which has been kept alive by $100+ oil but which is probably now slipping back into loss-making territory. I’ve seen estimates that there might be as much as 8mbpd globally which is currently in this category, with this fall having the potential to take maybe half of that out. My strong suspicion is that, if shale takes a hit as well, supply will quite significantly undercorrect, and then we’ll see how much spare capacity OPEC has really got. I think a price spike by some time in 2016 is on the cards.

    In the shale patch, most operators are saying that they’re going to target the best sweet spots more aggressively. I have no reason to disbelieve them, and suspect that we might even see an increase in average well productivity this year, and possibly continued though slowed growth. However once those sweet spots are drilled out there’s only the much less productive acreage to drill. I suspect that this move might serve to increase how high shale’s ultimate production peak is, but also to bring that peak forward in time and increase the steepness of the ultimate decline. Shale producers aren’t interested in sustainable flow, just production growth. Recipe for disaster if ever there was one.

    I remember reading an old article on the oildrum, about the peaking of whale oil. Around the time of the peak, the price began to strongly oscillate in an unpedictable and nonlinear manner. Makes you wonder.

    • Sam

      Thank you. When I worked as a full-time analyst, I never ceased to be amazed by some of the nonsense in the media – I used to call it “the black gold syndrome”

      You are right about Saudi. Way back, I was invited to write a strategy paper for OPEC. My conclusion was that very low prices could indeed bankrupt higher-cost competitors, and stimulate demand, but neither would happen quickly – so could OPEC members afford umpteen years of ultra-low prices? Conversely, cutting production to shore up prices would only subsidise competitors whilst undermining demand, and volumes would remain low for umpteen years. So they should compromise on a medium-low price, which they did.

      I mention this because low prices do not kill competitor production on opex grounds. For instance, current North Sea fields (with sunk capital) remain in the black, cash flow-wise, at prices far lower than here. But what IS impacted is investment in new production. Existing fields’ output declines by 7-8% annually, so, if you’re Saudi, that’s what you can reckon on if you keep prices low enough to stop new investment.

      Shale, of course, is quite different, because decline rates are so much faster. A few months ago I saw a paper by a retired senior US military officer predicting sharp shale decline after 2017, and that was based on $100 oil, with sweet-spots becoming exhausted. Until now, shale growth has been propelled by investors who, believing the “Saudi America” spin, have put up huge amounts of capital. If that spin story is sunk, the capital will dry up, stopping the treadmill – again, not overnight, but much quicker than in conventional plays like the North Sea. Incidentally, taking both opex and the NPV of capex, much of Mr Salmond’s North Sea “wealth” would cost $100/b or more to produce.

      As for shale investors, it’s a case of “believe the hype and sell on to someone who believes it even more than you do”. In other words, the “greater fool theory” of investment…….

    • I think there’s a chance you might be surprised by how much legacy conventional declines. I know that BP are reviewing almost all their legacy fields in the Norwegian north sea, have to imagine other operators will be doing the same. These are fields which would have been decomissioned maybe 4 or 5 years ago at a $60 barrel. Similarly, the Forties field has had its decline slowed significantly thanks to a heroic drilling programme by Apache going after some very marginal targets. That’s going to stop pretty sharply. Though the UK north sea has declined so much that frankly there’s not all that much left to lose. Not great for Salmond!

      Recent data I’ve seen from the Bakken is showing that % water cut might be increasing. A potential sign that sweet spots are getting drilled out, or wells are interfering because they’re spaced too closely.

    • Or is it “understand that shale is all hype, get in as early as possible, then unload on idiots who actually believe the 100 year story”

    • Hi Tim, I was wondering if you had a link to the report that predicts a sharp shale decline after 2017. I would be interested in looking at the data.

  2. Hi Tim

    Another factor that seems to get overlooked is the level of mal-investment and high debt in the shale industry. All mining in the US only accounts for 2.6% of GDP of which oil and gas is 1.7%. However in 2013 30% of US CAPEX was in the oil and gas industries. Of all junk bonds issued in 2013, Oil and Gas companies made up nearly 16% of the total. Those figures don’t really square with the GDP ratio suggesting heavy mal-investment. Not least because companies were spending $1.30 in input cost for every dollar in revenue. Rates on junk bonds have doubled since June from 5 – 10%. One needs to factor in debt repayments to see how many shale companies are really viable. many were drowning in debt with oil over $100 so this must be killing them. Basically these companies cannot afford to cut production, indeed they need to produce more to even start to meet their debt obligations but we also find that shale companies are massively cutting back on CAPEX for 2015. Many seem to be aiming at a 50% reduction which will wipe 15% off total US CAPEX next year. All of this is going to have a serious impact on the US economy when multipliers kick in.

    it used to be banks which were too big to fail. All global central bank policy seems to have done is rendered everything too big to fail and require endless QE propping up the whole inverted pyramid scheme of things.



    • Thanks. Two points occur to me.

      First, industry generally isn’t investing enough – and capital formation is difficult when real interest rates are negative.

      Second, though, the capital cost of shales is ludicrously high. Investors, believing the “Saudi America” hype, have entrusted vast capital to shale. Big mistake, as they will soon discover.

      A lot of America’s energy sector debt was junk even before the oil price slumped. With oil prices lower there has to be very significant default risk (just as there also is in Russia).

      Lastly, some recent figures suggest that, without the shale boom, US economic growth and job growth over the last five years would have been minimal.

    • Agreed

      I think a lot of people are under estimating the impact this is going to have. The only out is for the FED to prop up the junk bond market through secret proxy purchases.

    • Another point. Charles Hugh Smith and I are putting together a new book “Open Source Government: Toward a Community Operating System”.

      If you are interested in the concept and contributing it would be good to get you on board.



  3. Tim – the numbers I see from Reuters are as follows:

    12-1st half 89.68128717
    12-2nd half 89.83083083
    13-1st half 89.61196729
    13-2nd half 90.69364283
    14-1st half 91.10622417

    Are these correct? Because they indicate stagnation rather than a glut.

    The peaking of shale is clearly a huge issue but conventional oil peaked in 2005 and shale is, as T Boone Pickens said in an interview yesterday ‘the only thing between us and $170 oil’

    Total global production is of course what matters and will we not hit total peak well before shale peaks i.e. if shale cannot offset continuing declines in conventional, then the game is over long before shale peaks.

    • Not sure about what those figures are?

      The current “glut” is caused by shale, which in turn reflects huge investment, promoted, I believe, by hype. But the fierce decline rates of shale are bound to assert themselves, just as soon as investment declines and/or the best well sites have been used up.

    • Tim,

      I believe those figures are for oil production. Specifically they look like they’re the “all liquids” category, which includes crude + condensate, NGLS, biofuels and refinery gain.

      Since some of those categories don’t actually contribute any energy to society as refinery gains indicates volumetric gain from cracking longer to shorter chain hydrocarbons, and arguably represents a net energy loss and biofuels have at best an EROI of 1, it’s kind of a difficult number to glean much information from. Refinery gain has been increasing over the last years as we switch from lighter to heavier crudes, such as Venezuelan and Canadian tar sands.

      There’s an interesting discussion on modelling decline in shale production here: http://peakoilbarrel.com/oil-field-models-decline-rates-convolution/

      I’m currently playing around with some models myself, trying to see what things might end up looking like over the next year.

    • Sam

      Thank you – all of this is very interesting indeed, as is the link. Moreover, the timing could not have been better, as my first project for the coming year is a new version of the Surplus Energy Economics Database, “SEEDS 2015”.

      The source of my global supply and demand numbers is OPEC’s monthly report. When presenting these, I treat processing gains as a deduction from demand, rather than as a source of supply. The analysis of US liquids output is sourced from the EIA, of course.

      I take the point about shale decline – in one sense it’s like any oil province, where the decline rate of individual wells is moderated by the drilling of new wells. But there are various reasons why drilling might cease – lack of equity investment capital, lack of loan capital (it seems to me that shale co bonds were already close to junk even before the crude price slumped), and exhaustion of best drilling locations. The decline rate per field would still be a bit lower than the single well rate, but would nevertheless be severe, far more so than the EIA (with its 2022 peak) seems to assume. A senior but retired officer in the US military wrote a paper about shale, showing a sharp decline from 2018, and I find his arguments convincing. This clearly matters to the military – if the US really is “Saudi America”, the military could scale back its Middle East presence, and I see no sign of that.

    • Tim,

      I’ve been looking at some of the dynamics based on what we might expect to see from the production in terms of possible drilling dynamics over the next year. Unfortunately convolution is a pig of a process to get to work in excel. The assumptions that I’m currently using are:

      1) All wells drilled are the average bakken well, which is approximated by a hperbolic decline curve and has an EUR in the 350k barrel range with an IP of around 1000 bpd.
      2) All wells produce for 10 years and are then shut in.

      #1 is close to reality, though average well profile may improve as drilling shifts to high quality areas to make up for low price. #2 is maybe a bit dodgy, because some wells might end up getting shut in sooner if opex is too high, and that’s real possibility considering all the wastewater and oil have to be moved around by truck, which costs a fair bit.

      If say drilling declines by 50% this year and holds roughly that level for say another few years then you lose about 20% of production pretty quickly, and then enter a longer slower decline period over subsequent years. However there’s a very sharp inflection point at the peak, which would probably knock people pretty hard.

      If drilling were to stop altogether then you get a disastrous period of maybe 2 years where 50% of production declines, then the fat tails of the wells kick in and things are smoother over the last decade.

      I’m going to try and put something more comprehensive together, but suffice to say the final output is strongly dependant on what your wells drilled function ends up looking like. However given that shale has been the sole engine of global liquids growth since 2005, a 50% decline in drilling might bring about peak oil by 2016 (assuming nothing else can pick up the slack and people are more reluctant to invest in shale if the bubble pops nastily).

  4. Pingback: Here’s to interesting times in 2015 « Simple Living in Suffolk

  5. Pingback: #43. Oil – where next? | surplusenergyeconomics

Comments are closed.