Bespoke alternative investment solutions for institutional investors

Research papers

Why the next bull market in oil is not so straightforward

Issues to be addressed in this research paper

When the price of oil drops from $115 to under $30 in only 18 months, as Brent oil did between June 2014 and January 2016, and subsequently rises to $45, I can fully understand why a bull market ethos is building. I have certainly been in that camp myself, and still am to a degree. It is not quite so simple, though, which is what this research paper is about. In the following, you will see why I think oil prices can still rise meaningfully, but I will also explain why the lofty days of 2011-14 are not likely to return anytime soon. To wrap it all up, I will suggest which investment strategy we think is the most appropriate, given those expectations.

Continue reading

We publish investment strategies and opportunities in our research papers. This research paper is available to professional investors as part of ARP+ subscription.
Already a subscriber? Login

Contango vs. backwardation

Let’s go back to school for a minute. The relevance of this will become apparent later. Commodity futures curves are said to be either in contango or backwardation. A commodity is in contango when the spot price is lower than the forward price, and in backwardation when the opposite is the case (chart 1).

Chart 1: Anatomy of a typical commodity futures curve
Chart 1: Anatomy of a typical commodity futures curve
Source: Goldman Sachs Global Investment Research, April 2016

This is relevant because the change in the shape of the curve from day to day tells you how much tighter (or softer) the market is turning on the margin. The shape of the forward curve is in many ways more relevant to oil traders than what the price of oil actually is.

As oil supplies have been ample since the summer of 2014, oil prices have been very soft, and the market has been in contango, following several years of back-wardation (chart 2).

Chart 2: 1-month vs. 2–month time-spreads
Chart 2: 1-month vs. 2–month time-spreads
Source: Goldman Sachs Global Investment Research, April 2016

A few more observations re the forward curve:

The front end of the curve tends to be driven by inventories, where high inventories lead to contango (chart 3). Contango is therefore often, but not always, associated with weak economic conditions, whereas robust conditions would usually lead to backwardation.

Chart 3: Inventories vs. Brent time-spread
Chart 3: Inventories vs. Brent time-spread
Source: Goldman Sachs Global Investment Research, April 2016

In the context of that, I note that the high inventories we have enjoyed in the last couple of years have indeed led to contango, but high inventories were not caused by weak economic conditions this time. Rather, they were caused by excessive supplies. Meanwhile, the back-end of the curve is driven mainly by marginal production costs (chart 4).

And now to the important point: Today, with the Brent forward curve being relatively flat, and turning flatter every day, inventories must be tighter than they were only a few months ago. The recent strength in oil prices may have surprised a few, but they didn’t surprise the trading community at all. The changing forward curve told the traders all they needed to know.

Chart 4: WTI 5-year contract price vs. marginal cost of production
Chart 4: WTI 5-year contract price vs. marginal cost of production
Source: Goldman Sachs Global Investment Research, April 2016

Adding to that, ‘financialization’ of commodity markets has probably resulted in exaggerated moves in either direction – and will most likely continue to do so. Take CTAs. When a trend is first established, they line up to establish a position in the commodity in question. For example, when oil prices turned in January, CTAs became massive buyers of oil futures. I am not criticizing CTAs at all – just saying that they have a bigger impact on oil prices than they did 10-20 years ago.

A reminder of our view on oil prices

Our view on oil prices has, for quite some time, been driven by a view that the marginal cost of production effectively sets the downside of a relatively wide trading range, whereas the average production cost of non-conventional oil (deep water drilling, horizontal drilling, shale, etc.) defines the top end of that same trading range. On that basis, long ago, we came to the conclusion that oil should trade in a range from $40-50 on the downside to $70-80 on the upside – at least in the medium term.

That said, I also pointed out that oil prices are more than capable of breaking out of this trading range for shorter periods of time. That has in fact already happened, but oil prices are now trading inside ‘our’ trading range again. Since we last published a research paper on oil back in October 2015, I have received further evidence to suggest that we may have a period of quite unusual activity in front of us.

The bull case

It is hardly a secret anymore that oil prices have been driven lower by over-supplies more so than by a fall in demand, and it is also a fact that much of that over-supply comes from countries outside OPEC. The U.S. as well as some other oil-producing countries probably got a little intoxicated by oil prices staying above $100 for an extended period of time, and chose to ignore simple logic.

As oil prices began to unravel, because much debt had been established to ramp up production in the first place, logic did not apply. Where you would normally expect supplies to drop, production was kept at unsustainably high levels in order to service the debt. The end-result was inevitable.

Now, a few months into 2016, you are finally seeing the supply side changing shape (chart 5). Both in the U.S. and in other non-OPEC countries, production is finally being cut, and the better balance between demand and supply has already had an effect on price.

The over-supply of oil is not expected to entirely disappear this year; however, by the end of 2017, it is expected to have largely gone (chart 6). There is obviously an element of guesstimate in that projection, and the higher oil prices go in the short term, the more likely various producers are not to implement their pre-announced production cuts. After all, we are only dealing with human beings!

Chart 5: Non-OPEC oil production
Chart 5: Non-OPEC oil production
Source: Original source unknown. Chart provided by Frere Hall, April 2016

In other words, the bull case is very much a story about overall supply and demand being better balanced going forward. Supply won’t necessarily fall on a world-wide basis, but the growth rate will drop quite dramatically.

Chart 6: Global oil supply and demand
Chart 6: Global oil supply and demand
Source: JP Morgan Asset Management, April 2016

The bear case

The bear side of the story really has three legs:

  1. Easy access to credit in the U.S.
  2. Iranian production coming back.
  3. Demand problems in China and/or elsewhere.

It has been surprisingly easy for U.S. shale producers, despite oil prices remaining well below production costs, to finance ongoing operations. I obviously don’t know if that will continue but, at least so far, U.S. banks have been amazingly accommodating. If that attitude persists, U.S. shale production could fall much less than anticipated.

Also, when I first wrote about U.S. shale last year, I was informed that average production costs were $70-80 per barrel, and I have no reason to believe that was incorrect. I am now told that U.S. shale producers have dramatically lowered their production costs – from an average of $70-80 a year ago to $55-60 now. If that is correct, shale producers will return much faster than many expect them to.

Iran next. The country’s oil minister has stated that Iran will not even consider reducing its output until it is back at 4 mbd - a production level they last reached in August 2008. Should the Iranians be able to produce that much relatively quickly (which many doubt), global over-supplies will be bigger than ever.

Finally, as far as the bear case is concerned, few talk about demand these days; it is all about the supply side. The problems in China haven’t miraculously gone away; neither have the many challenges that Europe is facing. And the economic upturn in the U.S. is getting a bit long in the tooth. After all, we are now more than seven years into this cyclical upswing.

As far as the U.S. economy is concerned, I also note that it is the only major industrialised country to whom falling oil prices are not a significant net positive. That is a function of the oil sector’s growing importance to the overall economy.

The U.S. operate with something they call severance tax over there, which is a percentage of the value of the oil (and other raw materials) extracted. To an oil state like Alaska, severance taxes are very important - they accounted for 72% of total tax revenues in 2014. The dramatic fall in oil prices since then has led to a corresponding fall in tax revenues in Alaska. Other big oil states like Texas, North Dakota, Wyoming and New Mexico have all seen a significant drop in tax revenues, even if severance taxes are not as important to those states as they are to Alaska.

All this leads me to conclude that the probability of a significant economic slowdown in the U.S. is not as low as claimed by some. It is not a given, but it would be naïve to simply ignore that risk factor, and the re-balancing of demand and supply as portrayed in chart 6 is therefore also associated with some risk.

Has shale reduced the upside?

As is clear from the above, I see positive as well as negative factors lined up ahead. That said, as I pointed out to begin with, I am still bullish on oil prices, so let me make it clear why that is. My bullishness is largely driven by the high probability of a continued fall in supplies - at least over the next 12 months.

Chart 7: U.S. rig count
Chart 7: U.S. rig count
Source: Original source unknown. Chart provided by Frere Hall, April 2016

As far as the U.S. goes, there has always been a close link between the number of rigs deployed and oil production – although there is usually a 12 month lag between the two. The rig count dropped quite dramatically in 2015, and continues to fall in 2016 (chart 7). It is therefore more than likely that U.S. production will continue to fall into 2017.

Outside the U.S., capital expenditures have been cut back substantially in the last couple of years. Oil producers have admittedly been able to maintain their production levels so far despite lower capex, but reality will catch up sooner or later. The countries most aggressively reducing their capex programmes would include Russia, Latin America and China.

Consequently, the most likely path over the next year or two is rising oil prices. That said, I am reducing the upper end of my trading range from $70-80 to $60-80. I hear from well informed sources that supplies coming out of the U.S. will rise quite substantially, should oil prices approach $70.

Some analysts have argued that shale producers will not come back. Their financial strength has weakened so much in the 2014-15 bear market that they don’t think they will ever come back (or so they say).

I don’t believe that for one second. The entrepreneurial spirit that is prevalent in the U.S. combined with an accommodating banking sector will almost certainly bring the shale brigade back at some point, and an increase in shale production will effectively limit the upside potential.

The combination of a mixed macro outlook and rather healthy micro factors apart from the possibility of rising U.S. shale production has lead me to increase the spread at the top end, until we have more clarity as to which of the three factors will prevail.

Over the next 12 months, as inventories continue to fall, the forward curve will likely turn more and more backwardated with the front end of the curve taking its price based on the spread between the spot price and the 12-month price. As the draws get larger, that price could possibly approach $80, but my point is that it will be the spot price that goes there rather than the forward price (which is the one you trade). In that scenario, trading the shape of the curve rather than the absolute price of oil becomes the key to success.

Other investment implications

Provided the curve is relatively stable (with spot prices also creeping up whilst the forward curve is in contango), I would not hesitate to recommend a long-only investment strategy in oil as investors shouldn’t lose money as they ‘roll up’ the curve; but that is not the future we are looking into.

Given the rather mixed macro outlook, we expect the forward curve to alternate between backwardation and contango. Such alternation can turn into a very expensive experience for long-only investors. I therefore believe that a much smarter approach to oil trading would be curve trading and, in order to do that, a long/short strategy is required.

Following on from that, given the relatively high volatility of most commodity long/short managers, I believe a portfolio approach is the best way forward. According to our research, it is indeed possible to create a portfolio of a small number of energy trading strategies, where the portfolio volatility is equity-like in nature but the expected returns much higher.

Such a significant reduction in volatility is best achieved by investing in energy more broadly and not just in oil. We would include coal, carbon trading, natural gas and power trading (electricity) in our universe of energy trading strategies.

Niels C. Jensen

25 April 2016