Crude markets sold off aggressively over the course of the week following the broad equity sell-off triggered by last week’s jump in volatility. The S&P is now down by 10% from its highs, which is the technical definition of a ‘correction’. This was ultimately the trigger for the sell-off in crude, but its not as if there was a lot of resistance given how high oil prices had risen since last summer, and especially at the start of the year, and the seasonal softness in the physical markets. Indeed, many were hoping for a correction. In that sense the market’s reaction to the DOE weekly inventory report may have been somewhat exaggerated. It was not so much the report itself, but rather the timing of the sudden reported surge in US crude oil production that is now causing a broader rethink of the validity of the bull thesis. The fear has been triggered by the risk aversion in broader markets, which has caused potential buyers of crude who had been looking for a correction to pause. The loathing is mostly driven, once again, by the dramatic underperformance of energy equities, which are tracking the sell-off in longer dated crude. At the moment, however, WTI is simply tracking equities.
As if to emphasise the point, the oil market this week completely ignored a very bullish Chinese January crude import number and instead chose to focus on the EIA data which showed a 0.33 mb/d w/w rise in US production led by the Lower-48. This of course came on the heels of the final figures for November 2017, published by the EIA in its Petroleum Supply Monthly (PSM) report last week, which showed a mammoth 0.38 mb/d m/m gain in US output (for more details please see Data review: US output and shale, 31 January 2018).
|Fig 1: WTI vs equity indices, % change||Fig 2: US Risk assets performance ytd|
|Source: Bloomberg, Energy Aspects||Source: Bloomberg, Energy Aspects|
Let’s just start with the obvious. US production did not grow by 0.3 mb/d in one week. People who are now trying to extrapolate from this number would find that US production grows 1.2 mb/d on the month, and this would effectively double US production over the course of the year. Its important to remember that the weekly production number is based on a model, and not actual reporting by producers. It is at best an estimate, and to be honest it’s more of a guesstimate. The methodology is openly explained by the EIA on its website. While this hardly makes it the most satisfying data point, there’s little point complaining about portion size when it’s the only item on the menu.
A bit on the EIA methodology here: the weekly estimates are driven by the EIA’s Short-Term Energy Outlook (STEO) since collecting weekly crude oil production data from operators is not feasible. In fact, the EIA notes that the 12,000 operators in question (spanning 1.2 million existing wells) are challenged to even provide accurate reports on their monthly activity for the EIA-914 survey (which is published in the PSM), which is due 40 days after the end of the month. Alaska’s production is reported by the state (estimated from deliveries to the TAPS pipeline), but figures for both Gulf of Mexico and onshore Lower-48 production are driven by the STEO, which in turn is based on the EIA’s Drilling Productivity Report (DPR), crude prices, rig counts, other well data and the latest EIA-914 survey. In other words, if the latest final monthly figure published in the PSM are substantially higher or lower than expectations, that sets a new baseline for STEO, and then the same growth or decline trajectory is applied to the new base. So, the weekly production growth of 0.33 mb/d was entirely a reflection of the agency’s upward revision to their 2018 STEO production forecast (+0.33 mb/d), based on (among other factors) a higher-than-expected November 2017 output number in the last PSM.
We think the market reaction to the data point says more about the market than it does about fundamentals. After all, it’s not as though the surge in US production has snuck up on anyone: US crude grew by over 0.8 mb/d from July to December 2017. But the reality is that no one cared about US growth if stocks were drawing, which they did impressively in H2 17. The increase in production coinciding with inventories shifting from draws to builds, coupled with the broader risk wobble, is accentuating this correction. Eventually, this will be positive for H2 18 as the positioning in crude will be ‘clean’. Of course, that doesn’t help right now.
Indeed, the problem this week has been that US crude stocks built again, naturally bringing the focus to rising US production. But these builds have been expected seasonally, and importantly, thus far they have been much shallower than normal, and we do not expect them to last for a meaningful period of time. Given the time of the year, the physical market is going to wait for some clarity on refinery turnarounds. And, non-oil specialists are still trying to figure out if this equity ‘correction’ morphs into full blown risk-off. In the meantime, given the reduced appetite by non-oil specialists, the path of least resistance in the very near term may be lower, until something happens to shift the focus from current meagre stock builds to H2 18 stock draws, which we anticipate despite our forecast for y/y US liquids production growth of over 1.5 mb/d. The catalyst may well be the return of Chinese buying, which is unlikely before early March, after the Chinese New Year holidays, but ultimately, given that the fundamentals in H2 18 are in such solid state, it is only a matter of time before oil gets bought. With OPEC still committed to supply management, demand growth accelerating into H2 18, and a global supply deficit still in place, the weakness is temporary.
Meanwhile, energy equities continue to grossly underperform. XOP has failed to rally with crude when its strong, and has exceeded declines when crude has sold. Despite the strong rally in crude prices from the summer, XOP right now is within spitting distance of its all time low. The degree to which the space remains shunned by the broader market is curious. We still view the lack of upward lift in the longer dated contracts as a big reason why there remains lack of faith in the narrative. Even though the prompt crude contract is still up some 20% since the summer, Dec-19 and Dec-20 remain largely flat. The backwardation has managed to attract ‘new buyers’ into the commodity, but a lack of movement in the back end continues to deter new buyers of the equities. Oil’s correlation with equities is accelerating, which to us reinforces that people are still scared, and will want some stability before reengaging. It’s hard to see crude prices rallying if VIX remains elevated.
Brent: The outlook for Brent remains unchanged from last week. Spreads are soft, although they have held up well this week, and the softness is not particularly noteworthy considering the time of the year and that the arb east remains shut. European margins have stabilised (albeit at a low level) and Asian margins have actually rebounded. The prompt flat price weakness was triggered by TARs but has been exacerbated by the macro wobble. If April-May Brent spreads can expire stable without falling into contango, that could serve as a catalyst given April is the peak for global refinery works.
WTI: Crude stocks rose by a modest 1.9 mb w/w to 420.3 mb. This is the second build in a row for US crude inventories, but stocks remain well below 2017 levels, and in the year to date total stocks have drawn compared to a build on a five-year average basis. The weekly build comes despite a huge 0.78 mb/d w/w increase in runs to 16.8 mb/d. We expect crude exports to remain steady at 1.28 mb/d in February before dropping by 0.15 mb/d m/m in March on softer export economics. However, the March dip should be short-lived, with the WTI-MEH forward strip more than compensating for the rally in WTI-Brent from April onwards. Spreads have been softer, and we do see some downside risk to WTI spreads from March onwards as southbound flows from Cushing drop. For more details, please see our Department of Energy data review, 7 February 2018.
|Fig 3: Ytd change in US crude stocks, mb||Fig 4: Ytd change US petroleum stocks, mb|
|Source: EIA, Energy Aspects||Source: EIA, Energy Aspects|
Gasoline: After a strong counter-seasonal decline in last week’s data, US gasoline stocks rose by 3.4 mb w/w to 245.5 mb, pushing stocks back above the five-year average. PADD 1 led the gains, rising by 1.6 mb w/w as imports into the USEC jumped by 0.26 mb/d w/w to a 17-week high of 0.70 mb/d. While the seasonal build in gasoline inventories is much shallower than last year, we still think that what is being stored is more summer grade components. Summer gasoline will need help in the form of refinery outages to outperform.
Distillate: US distillate stocks rose by 3.9 mb w/w to 141.8 mb as runs recovered from the disruptions plaguing USGC refineries in recent weeks. Indeed, runs rose sharply, supported primarily by a 0.56 mb/d w/w rise in USGC throughputs. The build pushed US distillate inventories back above the five-year average for the first time since October 2017. While diesel imports into the USEC dropped by 0.21 mb/d w/w, PADD 1 led gains, with stocks up by 2.2 mb and could remain under pressure in the short term with additional supplies landing this week—0.7 mb of ULSD from Saudi Arabia arrived into NYH earlier this week. However, the influx in overseas supplies combined with the imminent rise in PADD 3 turnarounds has weighed on the arb to ship ULSD between the USGC and NYH, with the route becoming unworkable on paper for the first time since late October 2017 on Tuesday. So PADD 3 resupplies to PADD 1 bear watching.
|Fig 5: Ytd change US gasoline stocks, mb||Fig 6: Ytd change in US distillate stocks, mb|
|Source: EIA, Energy Aspects||Source: EIA, Energy Aspects|