The science behind the polar vortex.(NOAA)
Weather is the perpetual wildcard in the natural gas market, but it’s
been particularly shifty this winter, keeping market participants — and
weather forecasters, for that matter — on their toes. Gas futures prices
started this season at $3.30-plus/MMBtu, but then endured some of the
warmest weather on record (in November and January), including a couple
of polar vortex head fakes over the past month or so — weather forecasts
at times in January started off much colder but ultimately reversed
course. Prompt CME/NYMEX Henry Hub futures prices have seesawed as a
result. Despite the weather setbacks, however, prices have held on in
the $2.40-$2.70/MMBtu range through much of winter and averaged more
than $0.60/MMBtu higher year-on-year in January. And, with an Arctic
blast set to unfurl across the Lower 48 this week, prices last Friday
topped $3/MMBtu again in intraday trading before settling in the
high-$2.80s/MMBtu Friday and Monday. Today, we examine the supply-demand
factors underlying the recent price action, and prospects for sustained
$3/MMBtu gas prices.
In yesterday’s blog, Cold As Ice, we discussed the potential impacts of this week’s polar vortex event on the U.S. propane market (and how it may differ from similar events in the past). Now we turn our attention to what the extreme weather will mean for the U.S. natural gas market.
Well before the storage withdrawal season got underway in November 2020, it was apparent that the gas market was poised to tighten in late 2020. Rig counts were at decade-long lows, Lower-48 gas production was trailing by more than 7 Bcf/d year-on-year heading into the winter season (November through March), and it would take months, if not more than a year, for production volumes to return to their pre-pandemic peaks. At the same time, U.S. LNG exports were set to make a strong comeback after being throttled by anemic international demand and prices for much of 2020. But the storage inventory had exited injection season (April through October) with a 200-Bcf surplus vs. the previous year. So the question was not whether the market would tighten but rather by how much. And that ultimately would come down to weather — still the biggest driver of domestic gas consumption and the #1 unknown, made even dicier by the polar vortex effect in recent years. A cold winter could rapidly wipe out the storage surplus and even lead to a substantial deficit, enough to support prices throughout 2021. On the other hand, another warm winter could easily offset the effects of production losses and LNG exports, leaving a lingering or even larger storage surplus to weigh on prices.
Figure 1. CME/NYMEX Henry Hub Prompt Futures. Source: CME via Bloomberg
Given all that, it’s not surprising that the market has been dancing to the tune of weather forecasts this winter, making double-digit moves day to day, or even intraday, as weather models were revised. What’s actually transpired though has been more complicated and nuanced, particularly when put in perspective with previous years. For example, what otherwise would have been considered normal-to-bullish weather for November was exceptionally bearish for the gas market in November 2020 when compared with November 2019, especially the first half of the month. As a result, the year-on-year storage surplus expanded in November 2020, and the December prompt contract, which had started November near $3.35/MMBtu, quickly fell below $3/MMBtu to as low as $2.59/MMBtu and ended the month in the $2.80s/MMBtu (dashed orange box). Similarly, December 2020 weather was closer to normal, but weather-driven demand still lagged December 2019, which was especially cold. Once again, fundamentals did little to make a dent in the storage surplus, and prompt prices lost more ground, falling to a low near $2.30/MMBtu by the end of 2020 (dashed orange oval).
The opposite occurred in January 2021. It was one of the warmest Januaries on record, but still managed to be bullish relative to January 2020 (which had been very mild), allowing the storage surplus to shrink. The price action was halting, however, as weather forecasts bounced around, hinging on the direction of the wobbly polar vortex split — what happens when the stratospheric air above the North Pole warms and breaks down the polar vortex wind pattern, blasting icy Arctic air farther south in the Northern Hemisphere, either toward Europe and Asia or towards North America, or both. Prompt futures last month pinballed from the $2.50s to $2.70s, back down to the $2.40s, and again up to the $2.70s (dashed green oval) as weather models at least twice forecast, then retracted, expectations for extreme cold.
Now, both a polar vortex event and $3/MMBtu prices are back in play as forecasts show the polar split breaking towards the U.S. this week, bringing freezing, even sub-zero temperatures as far south as Texas. What’s been the net effect of all the weather shuffling so far this winter, and what will this week’s Arctic blast mean for supply-demand fundamentals, storage and, ultimately, prices?
A good way to assess the relative tightness (or looseness) of the gas market is to compare current fundamentals with the previous year. We do that using historical supply-demand data from the daily RBN NATGAS Billboard report. Figure 2 shows the year-on-year changes for each of the supply and demand components for the first three months of winter 2020-21, from November through January. The navy-blue bars indicate an increase vs. the same period last year, while the red bars indicate a decline.
Figure 2. Year-on-Year Changes in Lower-48 Supply-Demand Balance. Source: RBN NATGAS Billboard
Starting with supply to the left in Figure 2, Lower-48 production for November 2020 through January 2021 was down a whopping 3.9 Bcf/d year-on-year, averaging just 91.4 Bcf/d this winter vs. 95.3 Bcf/d during the same period a year earlier. Imports from Canada jumped 1 Bcf/d year-on-year to 5.7 Bcf/d, partially offsetting U.S. production losses, though LNG sendout (pipeline receipts of regasified LNG volumes from imports) slipped 0.2 Bcf/d from last winter to just 200 MMcf/d this time around.
As for demand (middle section of Figure 2), Lower-48 consumption from the primary sectors — power generation, residential and commercial heating (res/comm), and industrial end-users — has been substantially lower too this winter. Power burn, which averaged 27.4 Bcf/d in November through January, has lagged by 1.4 Bcf/d year-on-year. Res/comm, the main driver of demand during winter, has trailed by 2.2 Bcf/d year-on-year, with an average 44.2 Bcf/d this season, down from 46.4 Bcf/d last year. While res/comm was nearly flat year-on-year in December and up nearly 4 Bcf/d year-on-year in January, November 2020 was especially bearish, lagging 2019 levels by nearly 11 Bcf/d. [Note that the res/comm bucket in our model includes miscellaneous volumes, including fuel, pipe loss, and any observed flows that can’t be categorized due to insufficient visibility of what happens to the molecules once they get on an intrastate system or behind a local distribution company (LDC).] Finally, industrial demand also took a hit, averaging 24.4 Bcf/d, down 0.7 Bcf/d from a year earlier.
The only upside in demand has come from exports. Exports to Mexico averaged 5.5 Bcf/d this winter through January, up 0.3 Bcf/d year-on-year as some additional pipeline capacity was completed, particularly on the Mexico side and gas from the Permian moved south to displace LNG imports previously at Mexico’s Manzanillo terminal (see the NATGAS Permian report for details). Then there are LNG exports, which, after months of languishing well below capacity, have surged to record highs as a number of factors — strong winter heating demand in Europe and Asia, a supply shortage in Asia due to Panama Canal delays and vessel shortages, and fast-declining storage inventories in Europe, among others — tightened global LNG supply and sent international gas/LNG prices soaring (some of the logistical constraints have eased in recent weeks; see the LNG Voyager report for details). LNG feedgas deliveries for November-January averaged 10 Bcf/d, up 2.3 Bcf/d year-on-year.
That brings us to the Lower-48 gas market balance (the difference of supply minus demand, shown on the right side of Figure 2). If we net the supply components — production plus imports — total supply averaged 97.3 Bcf/d from November through January, down a net 3.1 Bcf/d. Demand, including exports, totaled 111.5 Bcf/d, down 1.7 Bcf/d as incremental exports were more than offset by declines in domestic consumption. With supply down more than demand, the balance averaged negative 14.2 Bcf/d (97.3 Bcf/d of total supply minus 111.5 Bcf/d of total demand), about 1.4 Bcf/d tighter year-on-year (3.1 Bcf/d less supply plus 1.7 Bcf/d less demand). In other words, production declines have done much of the heavy lifting when it comes to shrinking the storage surplus, with help from LNG exports — if not for incremental exports mitigating the bearish weather effects on domestic consumption, futures prices likely would be closer to $2/MMBtu than $3/MMBtu today.
Now consider what’s transpiring in February. Production remains at a year-on-year deficit. LNG exports remain stout, with nearly all terminals operating at or above full contracted capacity. What’s different this month is that given the potentially prolonged Arctic wave bearing down on the Lower 48 this week, U.S. gas consumption will likely set new records — potentially by a lot. The latest weather forecasts as of Monday morning showed somewhat less extreme temperatures for the current storage week but only because it pushed the coldest days out by a couple of days into the weekend. Moreover, Monday’s early model runs ended up adding another 50 Bcf of demand to the back half of February. On top of that, production is likely to see further declines as freezing temperatures cause wellheads to freeze-off, a phenomenon that disproportionately affects liquids-rich production regions. So if weather and related demand materialize as expected, the winter supply-demand balance will tighten significantly this month compared with last year and we should see the year-on-year storage surplus, which was down to less than 100 Bcf/d as of January 29, not only disappear but flip to more than a 200-Bcf deficit by mid-February. As of Monday morning, our temperature-based storage model was estimating a total withdrawal of over 800 Bcf for weeks ending February 5 through February 26, compared with 518 Bcf in the same period last year and the five-year average of about 440 Bcf. (Our weekly storage outlook is updated daily in the NATGAS Billboard report, based on the latest weather, supply and demand fundamentals.)
This scenario certainly makes $3/MMBtu gas prices more plausible. But for now, the market appears to be just as sensitive to the downside as it has been to the upside. As we saw last week, while prompt futures breached $3/MMBtu in intraday trading last Friday, they ultimately recoiled and settled at $2.86/MMBtu after midday weather runs moderated the length of the deep freeze. And price action largely appeared to stall Monday, with the March contract settling at $2.882/MMBtu, up just 1.9 cents from Friday. Normal temperatures gradually rise through late February and March as winter transitions into spring, so the ability for cold weather to make quite as big an impact on heating demand and the overall supply-demand balance will become handicapped after February, a reality that is likely weighing on traders’ minds.
Looking past winter, it gets even trickier. While production so far this year has been showing year-on-year losses, that could change by May, with volumes potentially showing substantial gains through spring and early summer compared with last year, when producer shut-ins were in effect. Even if shut-ins occur to some degree this year, such as in the Northeast if takeaway constraints develop, they are unlikely to be as severe as last year given the higher commodity prices. If we extend our January production average of 91.5 Bcf/d through summer, that will equate to nearly 3 Bcf/d of incremental production year-on-year through injection season. That means that demand will have to come in at least that much stronger in order to keep the supply-demand balance in line with last year. That may not be so tall an order, since LNG exports experienced a level of cancellations through the 2020 injection season that is also less likely to repeat, at least to that extent. But as production continues its recovery, it will take both full utilization of LNG export capacity and supportive weather to balance the market. And given that the weather wildcard is ever-looming, the market may remain rangebound and highly sensitive to both upside and downside signals.
"Harlem Shuffle" was written by Bob Relf and Earl Nelson and originally released by the soul music duo Bob & Earl as a single in 1963. That release went to #44 on the Billboard Hot 100 Singles chart. It was subsequently re-released in the UK in 1969, when it went to #7 on the UK charts. George Harrison reportedly called it his favorite record of all time.
The Rolling Stones' cover version of the song appeared as the third song on side one of their 20th American studio album, Dirty Work. Released as a single in February 1986, the song went to #5 on the Billboard Hot 100 Singles chart. It was the first cover song that the Stones released as the opening single of a studio album since 1965. Personnel on the record were: Mick Jagger (lead and backing vocals), Keith Richards (guitars, backing vocals), Ronnie Wood (guitars, tenor sax, backing vocals), Bill Wyman (bass), Charlie Watts (drums), Chuck Leavell, Phillip Saisse (keyboards), Dan Collette (trumpet), with backing vocals from Bobby Womack, Jimmy Cliff, Don Covay, Beverly D'Angelo, Kirsty MacColl, Dolette McDonald, Janice Pendarvis, Patti Scialfa, and Tom Waits.
Dirty Work was recorded between April and August 1985 and released in March 1986. Produced by Steve Lilywhite and The Glimmer Twins (Jagger and Richards), the album went to #4 on the Billboard Top 200 Albums chart. It has been certified Platinum by the Recording Industry Association of America. Two singles were released from the album.
The Rolling Stones are an English rock band formed in London in 1962 by Mick Jagger, Keith Richards, Brian Jones, Bill Wyman, and Charlie Watts. Brian Jones left the band one month before his death in 1969. He was replaced by Mick Taylor, who left the band in 1974, to be replaced by Ronnie Wood. Bill Wyman left the band in 1993, and Darryl Jones has been their bassist since that time. The Rolling Stones have released 30 studio albums, 33 live albums, 29 compilation albums, three EPs, and 121 singles. They have won one Billboard Music Award, two Grammy Awards, seven Grammy Hall of Fame Awards, three MTV Video Music Awards, and two World Music Awards. They were inducted into the Rock and Roll Hall of Fame in 1989. The band continues to record and tour.