The relative rate of growth in supply v. demand will ultimately determine stock levels and prices. And the three key predicting agencies, the International Energy Agency (IEA), Energy Information Administration (EIA), and OPEC have different views on what is likely to unfold.
OPEC does not often predict is own production, but in December it forecast it would average 33.2 million barrels per day (mmbd) during 2018. That would far exceed its projected “call on OPEC oil,” which is world demand minus non-OPEC production. For 2018 as a whole, it predicts that figure will be 33.1 mmbd.
That demand for OPEC oil is based on a gain in demand of 1.52 mmbd and a rise in no-OPEC production of 1.15 mmbd. In my view demand is likely to be a bit stronger due to world economic growth. However, the non-OPEC supply number is much too low, given the recent rise in U.S. production of 886,000 b/d from August through November. (December production was down a bit for seasonal reasons.) Furthermore, U.S. production has yet to respond to $60/b. The rise in output last autumn was a response to $50/b.
The EIA has the most aggressive non-OPEC production estimate of a gain of 2.5 mmbd, with 2.0 occurring in the U.S. alone, and the balance in Canada and Brazil. The EIA forecast is based on a gain in crude production of 1.5 mmbd and a rise in other liquids of 500,000 b/d. WTI did not exceed $60 in any month since 2015 until January 2018. And the year-over-year gain in March 2018 is estimated to be 1.29 mmbd. And so the industry’s response to $60/b could very well enable the 1.5 mmbd gain. Continue reading "Updated 2018 Crude Oil Outlook"→
On November 30, 2016, OPEC’s press release announcing the supply target of 32.5 million barrels per day included the following reference to inventories:
“The numbers underscore that the market rebalancing is underway, but the Conference stressed that OECD and non-OECD inventories still stand well above the five-year average. The Conference said it was vital that stock levels were drawn down to normal levels.”
Since the middle of 2017, OPEC has compared the OECD inventories to the five-year average, which had been 2010 to 2015. At some point in 2017, OPEC adjusted the five-year average to include 2011 to 2016. In doing so, it included two-and-a-half years of glutted (not normal) inventory levels. The effect was to make current levels appear to be closer to “normal” levels.
Given that OECD inventories are approaching the elevated five-year average, Saudi Energy Minister Khalid al-Falih has recently questioned that yardstick.
"Do we need to adjust for rising demand and look at forward day cover? How do we deal with non-OECD inventory? (It's) less transparent and reliable,” Falih said. “We have to think of the global market, the center of demand has shifted from OECD to non-OECD.”
Using historical supply-demand data and prices, I found a correlation between stocks and prices over time, but it is far from precise. That makes sense because price behavior is much more complex than using one measurement to define it. Market sentiment and positioning tend to cause prices to overshoot and undershoot equilibrium prices. To paraphrase the Noble Prize-winning economist Robert Shiller, prices are more volatile than the fundamentals imply.
Using monthly data from January 2008 through December 2017 (a full 10-year period), I found a -79% correlation. The Cartesian coordinate graph is depicted below:
I developed a simple linear regression to fit prices, given the inventory level, and graphed the actual prices with fitted prices:
This illustrates how far prices can travel from an equilibrium price, especially in 2008-09. On the other hand, the fitted prices do match up with actual prices over time. And the December 2017 fitted price ($61) is quite close to the actual price ($58).
This historical analysis begs the question, where are prices likely to go in 2018 and 2019? It also serves as a guide for understanding what stock level OPEC+ needs to achieve by withholding supplies.
To answer the first question, I used EIA’s STEO forecast of OECD stocks for 2018 and 2019. The forecast shows stocks bottoming in February, which would correspond to a topping of prices at $63.76, using this methodology. It implies that the $66.66 reached in January is likely to be the peak for 2018 and 2019, with prices dropping back into the lower $40s next year.
I also included EIA’s own price forecast on the graph for comparison. It shows similar expectations for the first half of 2018, but that prices will hold above $55 for the forecast period.
Regarding OPEC’s target, the regression shows that if inventories remain right about where they were at end-December (2.870 billion), the WTI price would remain at $60/b. If it wants $70/b, it needs to get OECD stocks to drop to about 2.800 billion. By the way, the latest 5-year monthly moving average is at 2.830 billion.
This model is very simplistic and does not include the impact of trader positioning and sentiment, which I believe are highly influential to the price. For example, the large drop in prices during the first week of February illustrated that factor. I use my Vertical Risk Management model to assess sentiment for positioning.
The other qualification is that the marginal cost of production and the timing of supply response have changed greatly due to the shale oil revolution. The large inventory of DUCs and much faster response of short-cycle oil has changed the market. For those reasons, lower inventories are required to support the same price. On the other hand, there is much more demand at the same price than compared to five to ten years ago. On balance, those two factors may be doing a good job canceling each other out since my regression using forward cover, instead of stocks, produced a lower correlation.
Disclosure: This contributor does not own any stocks mentioned in this article. This article is the opinion of the contributor themselves. The above is a matter of opinion provided for general information purposes only and is not intended as investment advice. This contributor is not receiving compensation (other than from INO.com) for their opinion.
The crude oil price started the year off strong, as January posted the highest OPEC Reference Basket price ($66.85) since November 2014, the month in which the Saudis decided to wage an oil price war with American shale oil. But the market gave up its 2018 gain during the first week of January, as the Energy Information Administration (EIA) incorporated the huge November production surge into its short-term outlook and weekly time series data. To top it all off, Baker-Hughes reported the most significant one-week gain in its oil-directed drilling rig count.
Whether the market shifts back to bullish sentiment, or whether the bearish sentiment takes control this year, depends mainly on several key assumptions. The central hypothesis is how fast shale oil production will grow this year, and the second is what OPEC production will be, given the on-going risk to Venezuelan output. Based on U.S. production from August through November, the recent lagged response in drilling rigs, and the high prices experienced October through January; I expect that U.S. production will rise faster than either the DOE or OPEC assume in their forecasts.
EIA’s February Outlook
The EIA released its outlook, revising its U.S. crude production estimates much higher. For the year, it now expects crude production to average 10.59 million barrels per day (mmbd) in 2018, and to exit the year at 11.13 mmbd.
The Energy Information Administration (EIA) reported that November U.S. crude oil production averaged 10.038 million barrels per day (mmbd) in November, up 384,000 b/d from October. The monthly product number was just shy of the 10.044 mmbd record set in November 1970. This gain was on top of a 17,000 b/d upward revision for October, making the total rise 401,000 b/d. By comparison, the Saudi production cut was about 460,000 b/d.
About 200,000 b/d of the increase was expected since Hurricane Nate had disrupted production in October by about amount. But about 175,000 b/d of the rise was new production. The bulk of the increase was in Texas, accounting for 114,000 b/d. Production in the mid-west was up 23,000 b/d. Gains were wide-spread among numerous states.
Production has surged by 846,000 b/d from September through November. This increase is far more significant than the one reported by the EIA in its weekly numbers or forecast by the EIA in its monthly STEO. The interpolated weekly figures for November imply a monthly average of 9.667 mmbd, 371,000 b/d lower. And the latest weekly average reported by the EIA was 9.199 mmbd. Clearly, the EIA will need to upwardly revise its weekly model soon, probably in next week’s report.
According to the Energy Information Administration (EIA), U.S. petroleum inventories (excluding SPR) fell by 14.2 million barrels in the week ending December 15, 2017. They stand about 2 million barrels (mmb) higher than the rising, rolling 5-year average and are about 96 mmb lower than a year ago.
Commercial crude stocks fell by 6.5 mmb, and SPR stocks were built by 0.4 mmb last week. Gasoline stocks rose by 1.2 mmb, and distillate stocks gained 0.8 mmb. Primary demand rose by 640,000 b/d to average 19.948 million barrels per day (mmbd).