Shale watch… Lucky or smart?

” I have spent my whole professional life as an international economist thinking and writing about economic geography, without being aware of it”

Paul Krugman

From a great article in Businessweek:

today the Permian is so important to Chevron’s fortunes that saving 10 seconds each time drill pipes are linked can translate into millions of dollars in cost savings across Chevron’s huge operations there…

“We were ready to let others spend their money and work through those 5,000 ways how not to make a light bulb before the opportunity to make money really presented itself,” he says. “The decision to hold the Permian Basin was very deliberate.”

On a day in which the EIA upped its well productivity numbers for US shale:


It’s all about productivity. Which is why, without an industrial cluster, I can’t see this Saudi dream working for many years…

Oil prices, speculators, and supply expansion…

An article from the FT here touches on an issue that has been discussed since there was an oil market:

Who trades oil is changing, however. Investors who bother little with details such as inventories and pipeline flows are replacing dwindling ranks of specialist commodities hedge funds. The shift could alter the way prices are formed…
Then who is driving oil positions higher? Newly prominent oil speculators are not necessarily reacting to news about supply and demand or utterances from Riyadh. Instead, they may be buying and selling oil based on moves in currencies, interest rates or the price of oil itself.

Namely, are speculators affecting the price of oil? You can see from the graph above that the exponential growth of Brent Ice futures contracts, which is cash settled and does not require physical product delivery, bears no relationship to the relative steady increase in the demand for oil. Some demand for these futures clearly reflects increasingly sophisticated financial risk management techniques, but some clearly represents purely speculative capital trading on price moves (often with large amounts of leverage).

There has been an entire industry in trying to ascertain the economic effects of speculators in oil markets. The IMF view is that they have no effect, but reputable economists at institutions such as the St Louis Fed disagree. A good summary is here.

My own (simplified) view, that accords to a well researched positions, is that speculators affect the volatility of the oil price but the not the final price over the long run.  Basic economic logic alone should dictate that if there is an increased amount of capital being invested in an asset class it will cause the price to rise, but when combined with leverage it adds huge volatility (quite simply if you have borrowed money to buy something and the price drops you tend to liquidate quickly to minimise loses). Which is why you see such huge swings in oil investment positions with a clear procyclical bias:

The big long.png

But the major point for those involved in service industries to my mind is that this is part of the explanation why there is not a linear relationship between the oil price and demand for oil field services. Directors at E&P companies make decisions about the long term price but ultimately the market for physically delivered product is more important when investing in production infrastructure, despite the large trading arms of the supermajors,  because they obviously do have deliver in the physical form eventually. They also benefit from miscalculating demand on the upisde through rising prices and a higher ROE on invested capital, so although they give up some amount of market share it’s a fairly small downside for erring on the side of caution.

Too many models that forecast the demand for oilfield services work are based on the forecast oil price rather than physical volume required. Too many management teams in offshore are using a rebound in the oil price as “proof” the “market” will eventually recover in demand terms when it is clear there is no linear relationship. As shale becomes the swing production method of choice offshore demand in particular should be relatively easy to forecast because in the new environment it will be supplying a baseload of physically demanded production while short-term changes in demand are managed by tight-oil. If someone in oil services tells you their business model is fine because the price of oil will rise I would suggest examining things a lot more carefully.

The shale productivity revolution in context…

“[W]e can see the computer age everywhere but in the productivity statistics”

Robert Solow


A great article in the FT today (behind paywall) on the boom in shale being driven by productivity increases (a near facsimile earlier version appears here). Readers of this blog will notice a consistent theme, as Krugman said, “productivity may not be everything, but in the long run it’s nearly everything”. The importance of this is that the US is turning shale into a manufacturing industry, small incremental improvements day-in, day-out, that cummulatively dramatically lower overall per unit costs:

In 2015, shale oil producers on average used 3,300 tons of sand per well, according to Petronerds, a consultancy. By last year, that had almost doubled to 6,100 tons per well. Delivering that much sand to the well site can require 250 truck movements. Other techniques for shale production have also been refined to increase the amount of oil that can be extracted. Modern rigs can drill faster, further, and more accurately than their predecessors. The process of hydraulic fracturing is being split up into more “stages”, allowing effort to be focused more precisely on oil-bearing rocks.

Innovations using the latest computing and communications technology, including remote operations, are also starting to be used more widely. Schlumberger, the oilfield services group, says that in 2014, 13 per cent of jobs it worked on at US onshore wells were supported by technical experts watching from its Houston campus. By 2017, that was up to 31 per cent.

This coincided with the IEA forecasting that the US will become the world’s largest oil producer by 2023 (graph above). [It is well worth having a look at the 13 slides at the bottom of the page of the IEA link]. Investment remains depressed in all but tight oil and the comment at the bottom of this slide regarding offshore is telling:

IEA Upstream Spending.png

The IEA is worried:

that despite falling costs, additional investment will be needed to spur supply growth after 2020. The oil industry has yet to recover from an unprecedented two-year drop in investment in 2015-2016, and the IEA sees little-to-no increase in upstream spending outside of the United States in 2017 and 2018.

“The United States is set to put its stamp on global oil markets for the next five years,” said Dr. Fatih Birol, the IEA’s Executive Director. “But as we’ve highlighted repeatedly, the weak global investment picture remains a source of concern. More investments will be needed to make up for declining oil fields – the world needs to replace 3 mb/d of declines each year, the equivalent of the North Sea – while also meeting robust demand growth.”

The real questions here revolve around how much capacity is being replaced annually, and it is simply not true that 3mb/d are not being replaced at the moment. The producutivity improvements in shale above are part of the solution. Other questions are what sort of price increases would crimp demand? etc. There appears to be no change in investor expectations that they want E&P companies, certainly large ones, to reduce debt and increase shareholder payouts, and therefore capital projects will remain subdued. There is also a strong feeling in the investment community that reserves can be run down.

Without wishing to sound like a broken record the “Demand Fairy” isn’t saving anyone in offshore. Offshore needs to re-engineer it’s business model to compete. The IEA is clear that this weakness will be felt from 2020 onwards, so even if you accept their reasoning, that is a long time to keep burning OpEx if your business model cannot even breakeven in the current climate.

I view shale as a technological revolution and believe that no economy is better suited to maximising its potential. Perez defines the major economic technical revolutions since 1770 into five categories, and the US is dominant in the last three:

Tech Rev 1770 -2000s.png

Revolution is an overused word. But according to Perez’s definition, that I agree with, the shale industry is a Technological Revolution (TR):

Tech Rev Definition.png

This to me is the most interesting part of economic history, because while there are nuances the broad economic development of industrial patterns are really well understood. A classic article here compares the development of the computer industry to electric dynamo. Like shale it is a story of how US capital markets funded ambitious companies, vast economies of scale, manufacturing efficiency gains, and the slow initial diffusion of producitivity gains (think tight-oil 2007).

How long can this positive productivity feedback loop, where innovations throughout the system positively affect other inputs, continue? A long time I suspect. Shale may not be subject to the same volume effects as the PC industry but it still makes an interesting comparison (Allen):

Computers deflator.png

(For those who can’t remember logarithmic maths from school all the left hand bar of the top graph means is “this is a really big number… so big we have a formula to make it shorter”. And the bottom graph just means that even though the price dropped really quickly a lot of new features were added as well). This is a hallmark of the US economy and a manufacturing industry based on constant productivity improvements.

An earlier, and slightly different technical revolution, can be seen with the invention of Corliss Steam Engine,  which allowed America to break free from the constraints of water power in the 19th century (Rosenberg and Trajtenberg). Like shale this was an energy revolution, one that changed the structure of the US economy and allowed manufacturing and urbanization to begin in earnest. Corliss Steam Engine.png

Everytime I write about shale I want to write something about the incredible economic period of the 1930s: How US mass production techniques, a revolution in both managerial skill and capital formation, led to the creation of the economy that created Victory Ships and transformed the Ford factory at Willow Run (YouTube watch this, seriously) into the manufacturer of the B-24 (“the Liberator“), innovations that arguably changed the course of the WWII, and ultimately the post-war global economy. There are surprisingly (and disappointingly) few journal and web references to this, and all are about the mathematics of productivity, when this is really a story to be told at the company level. However, economic development is path dependent and these processes and learnings today are being applied in the shale basins every day, even if unwittingly. (For a broader read about what an incredible period the 1930s was in microeconomic terms this the best I have turned up so far).

So although I don’t understand the individual impact of every innovation listed on this slide I understand where they came from and the process driving this:

Nabors Rig.png

Once upon a time a PSV went for the low-to-mid USD 20ks a day. At the time of writing the BHGE rig count hit 800, the revolution therefore continues…

Shale productivity and swing production…

“We have just started to get into the manufacturing and harvest mode of the shale revolution,”

Concho Chief Executive Officer Tim Leach

I keep going on about productivity because as Krugman says, it’s not everything, but it’s nearly everything. This article from Bloomberg highlights how Encana is pushing the technical boundaries of shale further with a new Cube process (pictured above).

Basically, you do a whole pile of layers at the same time. I have no idea if it will be successful or not, but these companies will learn a lot from this and not all this knowledge will be wasted even if the process itself is a failure. Economists talk of dynamic capabilities, where companies build a knowledge base, in a path dependent manner,  by trying to do new things; and this is a classic case. This article from 1994 is as relevant now to this as it was to the semiconductor industry which led to the original observations (remember Moore’s Law).

Which is why Encana can show this:

Encana Productivity

On a per dollar basis no one in offshore is showing those sort of gains. A large part of the cost reductions have come as equity in assets has been wiped out through oversupply. This is a genuine ‘learning-by-doing’ productivity gain. I have consistently said that once the US economy starts to turn this into a manufacturing industry there will be a consistent reduction in per unit costs, it is the raison d’etre of American capitalism (which to be clear I don’t think is an amazingly astute observation).

I mentioned the other day I believe shale production was a genuine swing producer and looked for some data on this. And I found it:

The simulations show that the U.S. supply response is much larger now due to the shale revolution. Given a price rise to $80 per barrel, U.S. oil production could rise by 0.5 million barrels per day in 6 months, 1.2 million in 1 year, 2 million in 2 years, and 3 million in 5 years. Nonetheless, it takes many months before a substantial portion of the full supply response is online, longer than the 30 to 90 days typically associated with the role of “swing producer” such as Saudi Arabia.

The thing is this research used well data from 2010-2015, so if you look at the productivity gains over the last two years then shale is even more responsive. A comparison with offshore, given cycle times and unit costs are so different, would be interesting, but methodologically challenging I think. The point is shale is undeniably a swing producer and with a much lower capital commitment (i.e. low CapEx/ high OpEx) than the industry has had before and gives E&P companies an optionality that offshore doesn’t (but not the high flow/low lift costs advanatges of offshore).

This is where the US economy is so good. Look at this data from Precision Drilling Corp:

PDC Productivity

These cummulative gains are huge. Day-in, day-out, this is an economy, and in this case an energy production system, focused on driving down unit costs. I get you can run into limits, but we appear to be someway off them here. As I say, I think it takes a brave person to bet against this, and from an investors point-of-view such productivity gains are enormously attractive.

The case against shale…

FT Commodities Note has a opinion today by Robert McNally today making the case against shale acting as a swing producer in the oil market. The argument goes against everything I think and have written about here, and is contrary to the argument Spencer Dale makes.

I don’t agree with many points. For example:

US shale production is comprised of many dozens of highly idiosyncratic public and private companies, each competing with each other to maximise reserves and production. Shale’s shorter cycle ebb and flow can stabilise prices, but only coincidentally and depending on prevailing, broader market fundamentals…
True swing production is a very different animal: swing producers comprise a relatively small number of government-sanctioned entities controlling the bulk of low-cost wells that collude under a policy mandate to stabilise oil prices. Historical examples include the Texas Railroad Commission and other oil states, Seven Sisters, and Opec. Swing producers subordinate profit maximisation to price stability and bear the costs of holding idle or spare production capacity to contend with disruptions and unexpected demand surges.

McNally’s definition of swing production is an organised price cap not swing production based on excess capacity/capital that would not be profitable in its own right.. It is a cartel to cap prices irrespective of profit. Current shale producers seek to make a profit as the price of oil rises. The shale industry has shown it can be a swing producer, and do so with a sole profit maximisation, because it has a different economic model where latent capacity can be maintained without the expense that traditionally large projects entailed? The tradeoff is a much higher operating cost instead of a higher initial capital cost.

I think McNally misses the whole point: that an entire industry has been formed out of market demand for swing producers, and that they don’t require a subsidy to compete. This industry seems to be relatively adept at adding to rig supply or reducing it as prices change (as the above grpahic makes clear). We have heard a large number of reasons why shale will fail: :”Permania” land values, US capital markets shut down, technology limits etc., and shale is still here and going strong with E&P majors committing vast amounts of CapEx to it.

McNally makes the case that shale is too small and slow to fully plug the gap left by conventional projects. I think he misses the point of the declining importance of reserves, and also the unwillingness of future US administrations to give up a degree of energy security the country has developed. However, McNally is a respected thinker on energy so if you want the bear case for shale, and the bull case for high oil prices, read it.


The age of abundance… Schlumberger exits seismic…

“Geophysical measurement, survey design and seismic operations have been an essential part of Schlumberger and our R&E [research and engineering] efforts for more than 30 years,” Kibsgaard said, pointing out the company’s unique position in terms of intellectual property and its engineering and manufacturing capabilities. But as the downturn enters a sixth year for the seismic data acquisition business, “the present outlook provides no line of sight for the market recovery.”


Spencer Dale and Bassam Fattouh published an excellent article at Oxford Economics on Peak Oil Demand and Long-Term prices this week (where the headline graph comes from and is a summary of various forecasters). The core of the article is that we have entered an ‘age of abundance’ in oil and it discusses some of the consequences of this:

the real significance of peak oil demand is that it signals a shift in paradigm from an age of (perceived) scarcity to an age of abundance. The conventional wisdom that dominated oil market behaviour over the past few decades, based around the notion of peak oil ‘supply’ and the belief that oil would become increasingly scarce and valuable over time, has been debunked.

Over the past 35 years or so, for every barrel of oil consumed, two have been added to estimates of Proved Oil Reserves.  In its recent Outlook, BP estimated that based on known oil resources and using only today’s technology, enough oil could be produced to meet the world’s entire demand for oil out to 2050, more than twice over! And future oil discoveries and improvements in technology are likely to only increase that abundance. The world isn’t going to run out of oil. Rather, it seems increasingly likely that significant amounts of recoverable oil will never be extracted.

Co-incidentally, or perhaps thinking abstractly causatively, Schlumberger also announced this week that it was divesting it’s interests in seismic.

Historically one of the key drivers of offshore (and onshore) demand at the front-end has been the need to keep reserve replacement ratio (“RRR”) above 100% for supermajors. If you want an idea of how key this is read the first few chapters of Steve Coll’s outstanding book ‘Private Empire: ExxonMobil and American Power‘ (and while you’re at it do yourself a favour and grab a copy of ‘Ghost Wars’). In an age of scarcity knowing that you had a stock of a scarce resource was an important differentiating factor and gave shareholders comfort that they were paying for a company with a future.

Now the RRR of the E&P majors is declining:

Reserve Replacement Ratio Majors Since 2014

Eventually it will tautologically lead to increased exploration activity in the future. For offshore a key question is how much and in what time scale? Clearly Schlumberger think it is someway off happening. In an age of abundance reserves assume considerably less importance.

And indeed investors don’t seem to value it as much either:

Rohan Murphy, energy analyst at Allianz Global Investors, which holds shares in Shell, BP, Total and Statoil, sees a reserve life of eight to 10 years as “quite a healthy level”.

“I don’t think these companies should have a reserve life much above eight to 10 years, especially when we are trying to get to grips with what oil demand will be in 10 years from now.”

From an E&P company these reserves will not be worth as much in the future, and potentially with technology changes will be cheaper in real terms to extract. So what investors are saying (already) is don’t pull forward this expenditure now, wait… And that is a completely different dynamic to the one prior to 2014 where first oil drove all decisions.

Not only are reserves going nowhere but neither is demand as Dale and Fattouh point out. The demand side of the market is likely to be relatively stable for decades, it’s on the supply side where the action is occurring. But it is hard to overplay the significance of this, an 8 year RRR would be a ~30% drop from historical levels and see a massive reduction in rig work and associated activity. Schlumerger is changing its business model accordingly. Slowly working through these reserves down to an 8 year period would fundamentally change models used to forecast rig demand used in exploratory wells for example.

This is playing out exactly as Spencer Dale predicted in 2015 where shale can act as a “kink” in the supply curve. Marginal production gives E&P companies more flexibility and lowers CapEx and offshore commitment. It makes uncomfortable reading for offshore as the logical assumption, from a senior figure who obviously has a large influence in how an E&P major acts, is that short-term price movements will be met with an increase in investment in tight oil. It might not be as profitable per barrel, but it can adjust quickly to short term movements, and isn’t as risky given the cycle times and upfront commitment offshore requires.

The call on offshore requires a vast number of assumptions in models, which as the graph at the top of this post shows can lead to a multiplicity of reasonabe assumed outcomes contains vast room for deviation. A few percentage points in extending current fields, a few percentage points increase in productivity from existing wells, and a small drop in forecast demand (either from efficiency, substitution, or a global recession) and the amount required from offshore drops signficantly. E&P companies are underpinning a huge amount of base demand on offshore but those few percentage points in each assumption that could change they appear to be backing themselves to supply with tight oil. The future for offshore will not be a mirror of the past.

Schlumberger exiting seismic is a small example of how the offshore industry will delverage from a position of being overcapitalised as an industry relative to the amount of work, and it is a clear industrial sign that “the age of abundance” isn’t some ethereal academic concept. Schlumberger also took a huge writedown on the asset value associated with its seismic business which again places a numerical value on an economic theory. If you go long on seismic companies at this point you might want to ask yourself what you know that Schlumberger management, actively reweighting their corporate investment profile to tight oil, know that you don’t?

The scrapping of older tonnage, which has begun in earnest in the rig industry, has been delayed for longer in the offshore industry as banks resist the economic reality of depreciating tonnage by delaying principal payments. Such a situation cannot continue forever. The Seacor/COSCO deal this week, which sees vessels taking delayed delivery in the future is a sign of industry weakness not recovery. Declining values highlight that these assets no longer offer access to a scarce resource but something now viewed as an abundant commodity.

I’d have to say the more I look at presentations where the core of the logic is a return to 2013/14, after a suitable gap for the market to return to “normality”, the less convinced I am.


The drawdown on offshore… My views on shale and offshore…

A friend asked me this week what my view on the graph above was (courtesy of Sparebank1  Markets) given my views on shale? I have been pretty consistent here that shale isn’t going to displace offshore, but it doesn’t need to in order to have a major impact on the economics offshore: It just has to take demand at the margin.

There are many graphs floating around like the one above. The Seadrill restructuring presentation contains this:

Seadrill offshore gap.png

Oceaneering had this graph recently:

Oceaneering Deepwater Demand.png

Ocean Rig has this one (while I agree with the headline the logic and data supporting this don’t make sense to me as the sanctioning replacement ratio has been historically over 100% for meaningful periods so a drawdown as firms pay down debt in a low price environment is logical):

Ocean Rig Industry Demand.png

You get the idea.

Schlumberger recently put the scale of shale production into perspective:

2017 Oil Supply

So to be clear: all tight oil had to do was add ~5% to the global supply and it has turned the entire offshore industry into a financial mess. Now it isn’t a strict causation, offshore has suffered a severe financial bubble based on oversupply as well as a demand crisis driven by the speed of change in the shale industry, but still it shows how finely balanced things were.

And indeed if you look at all the stories of hope and recovery that aim to recreate the world cast in a 2013 shadow they all profess unrealistically high utilisation and day rates at their core. The reason is obvious: the industry from 2007 was built on very high day rate levels and utilisation figues and any small change in those realities, given the very high fixed cost base, causes financial chaos.

A few percentage points in utilisation and day rates is all it takes to massively swing profitability in such a high fixed cost industry. Economic change happens at the margins.

Look at this chart from HugeStadSea which sums up the dominant thinking in the market (Q2 2017):

HSS Market Assumptions

Back to a cheeky 90% utilisation from 60%. Nice… what could go wrong?

Seadrill has the same:

Seadrill RP day rates

Everyone has the same story. But the problem is unless everyone is at very high utiilisation then day rates won’t pick up as the economic incentive for anyone with idle tonnage is to bid it cheap. Rowan in their latest results stated that they believed the market had to hit 85% utilisation before day rates improved.

So shale has an importance on the economics of offshore far beyond it’s output in the physical market displacing offshore oil as a source of supply. Shale only needs to reduce the utilisation of the offshore fleet by a few percentage points and that fundamentally changes the economics of rig and vessel companies. Seadrill, Solstad, Bibby, Technip DOF, in fact EVERYONE in the industry, is a completely different financial proposition at 51% utilisation compared to 91% with concomittant increases (or decreases) in day rates.

You also get an idea how large the investment in shale has been (Source: Schlumberger) since 2008:

Shale CapEx.png

25% investment since 2008 has gone into tight oil and it has seen productivity improvements like this (although the presentation highlights these rates are slowing):

Shale productivity.png

So I repeat again that that shale will not displace offshore as a source of supply. But it doesn’t need to in order to completely upset the economic structure of the offshore industry by lowering the amount of marginal demand generated where offshore service companies made profits above their fixed costs. By that I mean if your rig/vessel covered its costs and overheads on 85% utilisation and profits came after that (i.e. at 86% utilisation and above you started to be profitable), and the impact of tight oil is such that you only ever get to 85% for ever, then shale will have managed to removed excess profitability from the industry by ensuring a drop in demand at the margin. All shale needs to do is meaningfully alter the global fleet utilisation, and win a significant amount of E&P CapEx share, both goals shale has achieved, to have a massive impact on the offshore industry.

In economic terms this is what looks likely to happen:


The Demand (for offshore services) = the extra unit of revenue firms get for selling (Marginal Revenue) which matches the point where the extra costs of supply (Marginal Cost + Average Total Costs) balance. So yes, there will be more work, and assets may well be busier than they are now, but it could be just enough to keep everyone in the industry cash flow positive but making zero economic profit. I am not saying there won’t be accounting profits, that all firms will all be loss making, and all shares will go to zero, but I am saying firms will find it very hard to earn returns above their cost of capital.

The one prediction I will make is that any business model in this industry that just relies on an increase in day rates and utilisation is doomed unless it has a massive cash pile (because getting there is going to take a very long time) or you are buying assets at distress prices. But most of the distress investors have moved far too early and there is so much money floating around from these funds I think the distress funds are killing the price discovery mechanism in this market.

It is clear that the quantative deployment of capital in large E&P companies will have a significant portion focued on tight oil as well as offshore. A few percentage points, that could go either way in many companies, collectively have a huge impact on the global offshore fleet utilisation (and therefore dayrates) and that is the core impact shale has.