This week, in the run-up to the 173rd OPEC-meeting (30 November) I thought we should take a look at the oil market. It is divided in two parts, one pertaining to the status of standard metrics of the oil market and the other part to a more difficult but nonetheless important discussion on the future of oil. Now, I am no expert, so I would really appreciate any comments and thoughts you might have on these subjects (and others). With that, let’s dive in!
Chart 1: Oil prices over time
I thought it would be fitting to start off with a chart of oil prices for a long haul (historical data from BP, updated by MB) and it was actually more interesting than I had expected. It is especially interesting to see the massive “thematic” swings over the past few decades, especially since “unconventional hydrocarbons” (shale gas, oil sand etc.) and “electrified” (Electric Vehicles, improved battery tech etc.) themes have also been, or are now, building.
Measures of the supply situation
That said, since I suspect demand for oil is quite stable (not to mention hard to measure) in the short run and probably follows the general economic cycle, it should make sense to continue by having a look at the supply outlook. One oft-used measure is the calendar spreads where I use the difference between 1st and 7th position Brent Futures (and apply a one-sided HP-filter to take away some of the noise).
Chart 2: The calendar spread
We can see that the market comes from a view of ever-increasing supply (I assume the oil demand outlook has been rising in tandem with the economy over the past year or so) reflected in the oil market’s previous deep “contango”1 (when the future price is higher than the spot price). Currently, however, it seems as if oil markets have moved from normal (a small) contango into a backwardation (when the future price is lower than the spot price), indicating that supply has probably become more restrained in relation to demand (why the future spot price is set to rise and, probably more interestingly, there is room for so called roll-yield where investors earn a profit via continued prolonging of their future contracts even if spot prices are still). As can be seen in the following chart, the spot price and the calendar spread do co-vary quite nicely.
1A “small” contango is normal for any non-perishable commodity (and reflects a premium consisting of warehousing costs, opportunity costs etc)
Chart 3: Calendar spreads and spot prices are much the same phenomenon
In addition to looking at the future price curve we also want to take a look at crude and product stocks to check as many boxes as we can on the “rebalancing and recovery trades” currently holding a grip on the oil market.
The first few charts indicate some recent rebalancing of the supply and demand situation, but is this also visible in oil inventories?
Charts 4a-d: Inventories have shrunk, in some cases (gasoline) dramatically
Note: Data on oil consumption and inventories are harder to come by and is published with a strong lag. An alternative is to look at more up to date US data (only).
Even when looking at more timely US data the impression remains, it has been a pronounced draw-down in stock, in particular, motor gasoline, despite refineries running close to peak capacity for the main part of the year (the hurricanes nonetheless ended that pattern).
Charts 5a & b: US Crude oil stocks
In essence, it seems most indicators point to a floor under the oil price and even prospects of continued rises in the oil price. In addition, the above also implies that the crack spread should continue to be on the high side of historical experiences.
Graph 6: US Crack Spread
Margin players – Shale producers
It seems the OPEC has indeed managed to cap oil production, but there are always margin producers (mainly shale, non-OPEC but also other producers of non-conventional hydrocarbons) waiting on the side lines should oil prices rise above their breakeven costs. While there is no way of knowing for sure where breakeven costs for shale producers are, we can make a rough estimate through looking at the correlation patterns of WTI and Rig-counts. Typically, the US rig-count seem to respond to the WTI spot price with a lag of 16-20 weeks, but with shale oil companies being margin producers there must also be a reverse relationship which we might be able to explore.
Graph 7: Correlation is not causation
As it turns out the maximum negative correlation is around the same lag length. However, since I want to get my hand on the intercept (approximating a break-even cost) I perform a rolling regression of the real oil price on the change in rig counts. The completely unscientific rolling regression does in this case help us out with the possibility of a time-varying intercept (time varying break-even costs). That said, how do we choose the length of the regression window? – To be honest, I have no idea, but I figure two-three years should suffice for new technology to spread in the shale industry.
Chart 8: A crude estimate of break-even costs
Indeed, if not in theory, my results are at least anecdotally appealing as they indicate that shale producers’ break-even costs fell from 40 USD/bbl to almost 20 USD/bbl in the early years of the millennia. From then on, they hovered around 30 USD/bbl until oil prices took off again in which more and more (and more “expensive” wells) were utilized. As oil prices collapsed in 2014, and the expensive rigs were shut down, break even costs again plummeted and are currently at 41 USD/bbl (in real terms), which should imply still below 50 USD/bbl in nominal terms; close to what more rigorous studies of shale companies’ financial reports also seem to suggest.
In the graph below, I have plotted the difference between real crude prices and our break-even cost estimate and it seems to work quite well. When the margin is positive, the rig-count increases (and vice versa).
Chart 9: An alchemist’s approach to shale break even costs
Into the big unknown
With that out of the way, let’s look deep into the future of oil markets instead. There are a number of themes currently discussed in oil markets, with electric vehicles and the above mentioned unconventional hydrocarbons being a couple of the factors depressing oil prices. Balancing this is of course the general improvements in economic conditions that are hoped to continue over the coming years. Now, we know that the energy intensity of GDP has been improving over the long haul.
Chart 10: Energy intensity of world GDP
But that is not the same as saying that environmental effects are positive, even when controlling for the higher incomes. We also need to take into consideration what source of energy that is explaining the improvements. If, for example, it was coal that explained developments, the improved energy intensity would amount to naught.
Note: EIA offers this data dating back to the 18th century, but we have not yet included it in the Macrobond database (keep them support tickets coming).
Chart 11: US primary energy consumption by source and sector
We can clearly see that the share of fossil fuels is slowly decreasing and that relatively environmental friendly fuels (i.e., nuclear energy) have increased. Looking at older vintages of EIA data we can see that there was a massive decrease in coal dependence from almost 80% in the beginning of the 20th century to the current levels around 15-20% (in the chart above the increase in coal dependence during the 1970’s is probably a reflection of the two oil crises during that decade). In the right-hand chart, we can discern the gradual increase of electric power generation which partly (but far from entirely) is a reflection of the shrinking residential, commercial and industrial shares. More worrisome perhaps, is the increase in transportation as electric cars are still in their infancy.
Graph 12: Thankfully, EV-adoption is moving fast
Note: The Y-axis is log scale. Electric car (incl plug-in hybrids) graph line starts in 2011. Source: IMF (WP/17/120)
On a more positive note, the adoption of EV’s is clearly moving very fast and as with the introduction of conventional motor vehicles, the usage became more widespread only as prices decreased. In 2015’s PPP-prices the T-Ford cost almost 140 000 USD in 1910, which makes today’s high profiled EV-makers pale in comparison. Thereafter it took the T-ford less than ten years, again in PPP-adjusted terms, to reach today’s average car price (ca 34 000 USD). A swift look at Tesla’s (PPP-adjusted) MSRP’s shows that after a slow start, prices on EV’s are now developing in a similar manner.
Of course, there is so much more to discuss (we haven’t even begun to discuss speculative positioning or fiscal breakeven prices!). Unfortunately, I have to draw a line somewhere and a tentative conclusion from the above is:
- Oil markets are cyclically strong and future prices and stock-levels indicate that prices have at least found a floor. In other words, OPEC did indeed manage to reign in production, while receiving a helping hand from a benign economic cycle
- The (relatively new) margin players, like shale producers, are set to keep any future price increases in check, which is why (in my opinion) the most probable scenario should be range bound prices, with deviations mainly explained by changes to the economic outlook.
Thirdly, and anecdotally, the final steps of electrifying the society (thanks to EV’s, improved battery technology etc.) are well under following on historical patterns. Eventually this should lead to a swifter convergence of oil prices (per energy content) to other energy sources.
Chart 13: Per unit energy, all sources should be created equal
Note: One barrel of crude oil is approximated to contain 5.729 million BTU’s
Disclaimer: We don’t usually have views and opinions about economic and financial states of affairs, (not ones that we express publicly as a company, anyway). We do believe, however, that people can and do appreciate a variety of perspectives. What you’ve just read is the perspective of the author. While we think our writers are very smart, Macrobond Financial does not expressly endorse the views presented here. And, as the old adage goes, you shouldn’t believe everything you read (not without finding the data, performing a few analyses and presenting it in a nice chart). We want to make it clear that we are not offering this information as investment advice. That being said, if you have Macrobond, you can easily check everything that’s mentioned here, and decide for yourself. If you don’t have Macrobond, now you have a great reason to get it.