January 13th, 2015
by Andrew Butter
This is an update on a long-term oil price forecasting model developed in early 2008 first posted in April 2009 as "The Impending Mother of All Oil Shocks" which was updated in December 2010 and posted as "Crude Oil Price Ten Year Forecast to 2020".
Reference the current drama, in 2011 the model said:
- Brent at $125 was a bubble
- Bubbles pop
- Whenever the bubble popped the price would halve
In 2011 the model predicted the forthcoming crash would bring Brent down to $67 which was out by 30%...or more - Brent today is $50. That doesn't prove the model is no good, just it needs calibrating, the busting of the 2011-2014 bubble, if that's what it was, provides new data; which is the reason for re-visiting the model.
With regard to semantics, what is called a "bubble" in this analysis is when price departs upwards from what makes sense economically. That encourages what Austrians call, "mal-investment" - i.e. building condos at the height of the housing bubble, or in this case, drilling for expensive oil. A bust is the process of the mal-investors and/or their bankers losing their shirts, or if they are smart persuading governments to bail them out. Figuring out what does, and does not make sense economically (called Other Than Market Value, (OTMV)) is what this analysis is about.
Whether or not $125 Brent in 2011 and $140 Brent in 2008 were bubbles...certainly not many are talking in those terms...the price today is indeed less than half of what it was in 2011, which is pretty much what the analysis in 2010 said would happen. That, along with four other predictions by the model that turned out to have been reasonably correct, and none that were precisely wrong, is offered as evidence the model, or more specifically the valuation opinion of the model driver, Other Than Market Value (OTMV), may in general terms be correct. That said, clearly the model is better at looking long-term, i.e. on average, than for day-trading, i.e. individual data-points.
There could be other explanations for the recent plunge in prices...butterflies flapping wings in Brazil, Black Swans; China doing whatever; "imbalances" between production and consumption...whatever that means - when I went to school that was called "inventory"...and of course there are the conspiracy theories. Yet none of those collective nuggets of wisdom are offered by anyone who said, three years ago, that what did happen just now; was going to happen, which suggests perhaps the idea of bubble/bust may be correct.
By definition, the market does not recognize bubbles; if it did there would not be bubbles. So long as there is a common belief that the market is working efficiently, bubbles persist. The basis of this analysis is first...Big Idea...bubbles actually happen, they really do...remember the housing bubble, the stock-market bubble, and what looks suspiciously like what was a bubble/bust in gold, and perhaps a bubble and one of these days a bust in U.S. Treasuries (watch this space)...and second that markets do work efficiently, in the medium/long term, which is why every bubble must be followed by a bust. The thesis on which this analysis is based is that bubble/bust is zero sum, it creates no long term economic value, therefore in the long-term the Net Present Value (NPV) of cash flow into the marketplace above OTMV (+) plus out, below OTMV (-), must be zero; and that provides a basis for calibrating OTMV.
This analysis says:
2011 to 2014 was a bubble, during that time the world paid about $600 Billion more for oil than what was "fair"... where "fair" is OTMV; which incidentally is about the price H.H. Prince Turki Al Faisal Al Saud, has been saying was fair since 2008. It is not known whether he based his opinion on reading this analysis; probably not, particularly since he has the advantage of being able to talk to his customers. In which case that provides a level of confidence, because if you work out price/value two different ways, and both give a similar answer, that suggests both methods may well be valid.
If that's right, what has to happen next is that at some point in the future the world will pay $600 Billion less for oil than what is "fair", plus the time value of money; because bubble/bust is zero-sum. In effect, the world overpaid for oil between 2011 and 2014, therefore, according to this thesis, at some point in the future they will get their money back on the difference from "fair", with interest. When, is impossible to predict, this analysis says that regardless of "when", how much the refund will be, is not in question; the interest paid on top will depend on when.
Since the bubble (if that's what it was) does appear to have well and truly popped, unless it is re-inflated (always a possibility, but less likely once there has been a good-clean-bust; see below), likely by 2020 all of that $600 Billion, plus interest, will have been "refunded". By that time the "fair" price will be over $100.
- The oil industry will likely have been totally spooked by four-to-five years of $50-$70 oil (Brent), so drilling activity will have fallen through the floor.
- Likely investment in new Big Oil projects that take five-to-ten years before you get any oil, if you are lucky, will have been put on hold. [IF] so spare production capacity in 2020 will be minimal; [IF] so it is very likely a bubble will start again in about 2020....[IF] history repeats, oil could go to $150 in 2022 (that would be a bubble...which would at some point bust down to perhaps $70, depending always on when). If...if...if.
- Rental/charter rates on drilling and related equipment will be through the floor, which is why there is a good argument to say that now is a good time for Shell to start drilling the Arctic, for the Brazilians to start developing sub-salt, and for the Saudi's to start re-drilling/working-over their aged wells, taking advantage of equipment and people for hire at half-price; but that probably won't happen. The danger of everyone stopping drilling is that once prices get back to "pay-back-values", there is a shortage of equipment and people; which means bubbles are more likely.
- There will be a strong downward pressure on interest rates because mispricing of oil alone explains 76% of changes in yield on the U.S. 10-Year since 1970 (P<0.0001*). Run that along with a model with the FED-rate and GDP growth as explanatory variables and you explain 90% with the "oil" component explaining 46%. That, plus the windfall of the refunds helping the current account deficit, may persuade central banks to start the dangerous maneuver of letting steam out of the pressure-cooker to wean their economies from QE-X. It will be tricky to do that without crashing stock-markets or creating a bust in bonds. If they get it wrong they risk serious deflation; printing money to pay budget deficits is inflationary, but printing it to bail-out incompetent bankers is highly deflationary. The successful strategy to save the banks from their collective insanity, over the past seven years, has, with regard to deflation; been countered to some extent by the bubble in oil prices which is/was highly inflationary, evidence the inflation in the 1980's which can be completely explained by oil mispricing (P<0.0001*). This analysis says that "safety-valve" is now removed by the bust in oil prices, [IF] importing inflation is indeed what Martha Stewart calls a "Good Thing", which is debatable.
Wars that affect supply of oil, as did the Iran-Iraq war...many forget that it was that, not a conspiracy by OPEC or even an outbreak of Black Swans that caused the oil-price bubble in the 80's, and it was over-drilling during that time that caused the bust that followed...and/or outbreaks of financial mass-stupidity, for example when bankers and economists got into their heads that "house prices always go up"...are impossible to predict; they are for markets, like the waves generated by hooligans tossing bricks into the into a pond. The analysis also does not say when or if there will be more bubbles or when they will pop, all it does is provide a basis for (a) identifying them....when the price is more than 15% above Other Than Market Value (OTMV), that's a bubble...assuming the estimate of OTMV is correct +/- 10%; and (b) forecasting the size and the dynamics of the busts that inevitably follow.
1: This analysis is a valuation model. To use the terminology of International Valuation Standards (IVS) there are two types of valuation, (A) Market Value (B) Other Than Market Value (OTMV). Market Value is the price you can sell something for to someone dumber than you are. Other Than Market Value is the price they ought to be paying if the market is not in disequilibrium. Bubbles and busts are markets going through a period of disequilibrium.
The central thesis of this model is that if you can understand what is OTMV then you can tell if there is a bubble (or bust), and how big it is. Ben Bernanke and Allan Greenspan both say it's impossible to know if there is a bubble; and everyone says, because they are so wise, that must be true; although I imagine by now they have at least worked out how to recognize a bust in progress, so that's good news. This analysis says it's not impossible to recognize a bubble or to say how big it is.
2: For most markets...housing, oil; stock-markets...bottled water, resort hotels, whatever... you can usually figure OTMV +/- 20% from 10-to-(optimally)-50-years of data by using multivariate regression and/or neural analysis from (a) a proxy for demand, GDP (nominal) in source markets often works well; (b) divided by supply capacity or a proxy for that (for housing and oil), that's not how much was used, or pumped, supply capacity of houses includes empty house even if they are shuttered, plus (c) a function of the yield on Treasuries for assets (oil wells and houses), as the main explanatory variable. To calibrate the model to make it more accurate you need an understanding how bubbles and busts work.
Throw a pebble in a pond and you get a wave. The peak of the wave above the equilibrium is exactly the reciprocal of the trough - but you don't change the water level in the pond, and eventually that's where the water level (prices), settle down. See The Seven Immutable Laws of Bubbles.
Bubbles happen generally for two reasons, (a) Wars (b) Easy/Stupid-money. They create no economic value, just as the pebble does not raise the water level in the pond. All they do is re-distribute wealth, from those who paid too much, to those who paid too little; or from those who got paid too little, to those who got paid too much.
3: It is not possible to predict a timeline for bubbles or busts with this model, that's work in progress; apparently George Soros does it by consulting his chiropractor...seems to work for him, there may be other ways, not found so far, although what people do in the future in reaction to events, is hard to predict. If for example the Saudi's had been persuaded to cut their oil production in half, and cut their own throats in the process, then the bubble could have been re-inflated. But "someone" must take the hit of $600 Billion less oil revenues (than what's "fair"), plus the time value of money, some-time in the future.
The idea about the Saudi's taking 100% of the hit was great idea (except for the Saudi's), sadly they didn't play ball. The debate right now, is who takes under or over fair-share of the hit; usually the Saudi's and other producers with lots of spare capacity, increase production in bubbles (a bit) to make more money, and lower production in busts (a bit) so all the other producers are the ones who pay the refund. Intriguingly this time Saudi is offering to pay their fair-share of the refunds, which is why it's so hard to understand what the other producers are so upset about?
4: Assuming that no-one else steps up to re-inflate the bubble, and as always, that no one starts a war that affects supply, that $600 Billion windfall to buyers of oil, and hit to producers of oil, will happen sooner rather than later. The line drawn on the chart simply copies the dynamics of how the previous bubble popped - extended; and is calibrated in the final iteration by a calculation of NPV from January 2011 to the estimated the time that price will likely re-visit OTMV using an arbitrary discount rate of 5% and assuming there is no "re-inflation" of the popped bubble.
5: January 2020 is a guess based on eyeballing, if prices stay very low in the short-term, the date of the "cleansing" or payment in full of the "refund" (plus interest), will be sooner.
6: The projected line of OTMV is predicated by growth (nominal) of world GDP (GWP) and the growth of world consumption of oil (CW) which historically can be reasonably well modeled by an algorithm with GWP and oil-price mispricing as explanatory variables (P<0.0001*). One reason for the "mistake" in predicting $67 Brent, might be that GWP did in fact drop in 2014, whether or not anyone noticed, and given that's not a super-accurate number +/-2% nominal, that could be a reason.
The range 3% to 5% for GWP growth going forwards is chosen arbitrarily to represent the most likely outliers reading between the lines from what the IMF and the World Bank say, not that any of them make 10-Year forecasts, and not that anyway their forecasts prove very often to have been correct +/- 15% six-months later.
It is possible that growth of world consumption (CW) will be less than would be expected from such a large drop in prices, because some of the countries that subsidize oil consumption (China, India, Saudi Arabia, and...Wait for it, yes U.S.A.); may do the sensible thing and not pass on the savings. It's much easier to sell a price that did not change, than to sell a price increase from removing subsidies; as part of an election-candy strategy.
7: What happens in 2020 or whenever the price gets back to what is "fair" will depend on whether or not there has been investment in new long-term production capacity, during the bust. That is not shale-oil which is short-term, pay-back on those plays is one-to-two years, that production can be turned on, or off at will, which means that once Brent starts going above $90 that production capacity will start to increase, providing some sort of safety valve against a bubble down the road. The big question will be Big Oil, today, anyone who goes to a banker or to the investment committee of a major oil company, with a plan to develop a new set of oil fields that can only give a reasonable return on investment over ten years, if oil is $100 or more, risks serious questions being asked about his/her sanity.
That's why it's quite likely that in 2020 there will be another bubble, because every time prices after a bust; return to OTMV (as they did in 2006; and in 2010 - according to this analysis anyway), the next thing that happened was a bubble.
There are two likely reasons for that, first whenever the price of anything doubles in under two years, which is what typically happens after the "refunds" are all paid (in 2004-2006 and 2009-2011), all of the genius bankers and economists draw a straight line through the dots of the recent past into the future, using the well-proven Nassim Taleb "How to predict the weight of a Turkey" methodology....which was employed by the rating agencies prior to the housing bust in U.S.A.; and so stupid money gets thrown into the ring, which as George Soros has explained over and over until he went blue-in-the-face, is (temporarily) self-perpetuating.
It's hard, perhaps impossible, to prove, but the availability of tons of easy/stupid-money thanks in 2005-2008 to Goldman's God's Workers & Co, and in 2011-2014 thanks to the Federal Reserve, and the fact that there were what looks most probably like two bubbles in oil prices in less than ten years may perhaps not be a complete coincidence.
The other reason is that bubble-and-bust is all about what the Austrians call mal-investment. That can go two ways; during the bubble it's easy to get money for investments that make no sense at OTMV let alone at bust prices, and it's particularly easy when there is easy/stupid-money around.
Most of the shale-oil drilling was done on credit with the production hedged up to pay-back...hedging that price required, in hindsight - a big suitcase full of stupid money....It is highly likely that a good proportion of the equipment used in the shale oil boom will soon be "bank-owned", like all those houses in Detroit owned by Norwegian pension funds thanks to what in retrospect appears to have been a bubble in house prices, although of course, at the time everyone, including two chairmen of the U.S. Federal Reserve, no less, said it was not a bubble...until after it bust.
Another example, since 2011 state-controlled banks in China have been offering finance to speculators who put 1% down to build offshore (jack-up and floating) drilling rigs. Most of the 180 rigs that according to RS PLATOU, a rig broker, will be completed in 2015 and 2016 are currently under various stages of construction in China.
That's nominally a 30% increase or more of drilling capability than is currently provided by the fleet of 550 rigs currently working; but it's more because the new rigs are a lot more productive than the old rigs, conservatively I'd say in 2016 the capacity to drill offshore will have increased by 50%.
At a guess I'd say that if oil stays at around $60 for a year or so - and this analysis says it could be five-years, the amount of offshore drilling for new wells and increasingly re-drilling old wells, will halve because the oil companies, particularly the state-owned ones, will get spooked.
In January 2015 the 550 rigs were pretty-much all working, it's highly unlikely that in January 2016 there will be work for more than 200, so that means a utilization rate of less than 25% on the new fleet which means charter rates will likely be 30% or less of what they were in early 2015.
Will many of the 550 old rigs get scrapped? Unlikely, particularly not the jack-ups, all you do is find a pleasant spot, jack-up, if you like put some extra protection around the splash zone, and wash out the pre-load tanks with fresh water which is a real good idea if it's hot and the residual salt water evaporates to brine, then turn off the lights, and wait. Those things were built to last, just like the old dry-bulk-carriers.
It looks highly probable that anyone who recently invested in a spanking new offshore drilling rig, based on a business model that said house prices will go up forever "oil prices will go up forever", made a mal-investment.
Mal-investment goes the other way, in a bust no one drills, no one invests in maintenance, let alone upgrading; so when the bust ends, often there is not enough spare production capacity to see-off the bubbles created by the silly-money-bubble-blowers or to handle small "shocks" like the Libyan crisis which removed 1% of the world's oil supply, and that was supposedly the reason oil prices bubbled (that wasn't the reason, that was just the story put out by the bubble-blowers).
8: This analysis offers no explanation for why there were bubbles in 2008 and in 2011 through 2014. They were clearly not caused by war, apparently they were not caused by manipulation of oil markets by easy-money, so what was it? A possible reason is that the easy-money was used to finance economic activity that makes no sense when oil prices are at OTMV or lower. Obviously the oil industry is one area, on the consumer side, the availability of finance at low rates may have allowed processes that are marginally profitable at high oil prices, to keep going, for example subsidies that keep the price to the consumer low.
The central thesis for the model is that since 1970 the price of oil has a fundamental* value dictated only by GWP (world GDP nominal) divided by CW (world consumption of oil). That is somewhat counter intuitive, it says if consumption goes up and GDP stays the same, the "value" will go down; according to Adam Smith "price" will go up. But CW or production, is just a proxy for supply capacity, and the "fundamental" (OTMV) is not the spot-price; it's an estimate of what the spot price ought to be, if the market was not going through a period of disequilibrium.
The story so far according to the model:
- In the 1970's oil prices were controlled by the "Seven Sisters" and it was in their interest to keep the price they paid to the owners of the oil, low, and to make their margins transporting it and refining it, it a classic Third-World shakedown.
- OPEC got its act together for the first, and perhaps the last time, and forced prices up
- That caused a "bubble", helped by the fact that demand for oil was going up, thanks in part to many years when it had been sold at a price lower than OTMV, and also that because of low prices, investment in developing new production capacity had been less than optimal. That bubble was about to pop, but then Iran fell apart.
- Then there was the Iran-Iraq war, which affected production, and production capacity of both countries, there was no spare capacity anywhere, even though the Saudi's increased their production to the maximum (swinging to (a) keep prices down in a bubble, or looking at it another way (b) profiteering on the high oil price), except they dropped production during the "Tanker War", some say that was "swinging" to keep prices up, another plausible explanation is that no one could buy insurance to ship oil out of the Arabian Gulf.
- During that time (14-years of bubble), there was a massive investment in new production capacity, mainly offshore in GOM and the North Sea, that started coming on line in a big way in the early 1980's, and was likely the reason the bubble started to pop.
- Then the war ended, and the bubble popped, according to this analysis at it's peak, oil was priced 2xOTMV so the pop to ½ x OTMV is a "calibration point"
- Then there was the Gulf War, production capacity went down, prices went up, causing a "mini-bubble"
- The "mini-bubble" then bust after hostilities died down.
- After twenty-years of oil-prices being about half OTMV (as estimated by this analysis), finally everyone received their "refunds" from over-paying in the 1980's (NPV of that cash flow is about zero, although it's tough to figure because interest rates were bouncing around).
- Then there was a bubble - most likely caused by (a) low oil prices having encouraged wasting oil (b) under-investment in new production capacity for many years, perhaps aided and abetted by financial "engineering".
- The bust at the end of 2008 was precipitated by the financial crisis; this provides a second "calibration" point.
- The bubble of 2011, in retrospect that was probably a continuation of the 2008 bubble, caused in part by everyone having got used to cheap oil, plus a lack of investment in new capacity. The bubble over-stimulated development of new production. The bottom of the "pop", whenever it is clear what that was, will provide a third "calibration" point.
- The analysis says that OTMV ought to be regained in about 2020, by which time OTMV ought to be about $100. At that point there is a risk of a bubble up to perhaps $150 in 2022, depending on (a) how much Big Oil invests in the next few years, taking advantage of low prices for developing oil production capacity, and (b) how fast the world economy grows (nominal).