Last week, natural gas prices faltered for a third consecutive week, sliding 8.2% on the week, including 2.6% on Friday, to close at $1.80/MMBTU. After reaching the lowest prices in over a decade at $1.75/MMBTU on December 17th, the commodity surged by 40% in the next three weeks as long-term forecasts suggested a below-average January, February, and March that would bolster demand, reaching as high as $2.48/MMBTU on January 8. Since then, however, heating demand has remained tepid and the storage surplus versus the five-year average has continued to edge higher, reaching +555 BCF or 28% as of the week ending February 12. With unusually mild temperatures expected through early this week and natural gas production at record highs, the outlook for the commodity has soured and many expect continued weakness as fears of a storage crisis looms.
I agree that the environment remains bearish for natural gas in the near term, but expect that the arrival of cooler temperatures late this week that will last into March along with the imminent arrival of natural gas exports, leads me to believe that natural gas is poised to slow its freefall as it tries to establish a bottom over the next few weeks. While I remain reluctant to go long natural gas at this juncture, I believe that natural gas will make an excellent value trade in the next 1-2 months, but that time is not yet here. This article discusses the short- and medium-term outlook for natural gas as well as my trading strategy.
Last week began with some of the coldest temperatures of the season with multiple stations in the northeast setting record lows of -10F to -25F. However, that cold was rapidly pushed aside as a ridge built into the nation's midsection with temperatures in the 50s and 60s as far north as Chicago, Detroit, and Cleveland by Friday. Natural gas storage withdrawals were well-above average during the first few days of the storage week with withdrawals over -30 BCF/day, but by Tuesday had drifted below the five-year average of 21 BCF/day and by Friday were creeping towards single digits with an estimated draw of just -10 BCF on Friday. Figure 1 below plots daily storage withdrawals for the week of February 13-19.
Figure 1: Projected natural gas storage withdrawals for last week showing the impact of a rapid warming trend. (Source: Data is my own via CelsiusEnergy.)
Overall, for the storage week of February 13-19, my model, which incorporates both temperature and pipeline data, is projecting a -145 BCF storage withdrawal when the EIA releases official numbers on Thursday. This is 1 BCF greater than the five-year average withdrawal of -144 BCF but 75 BCF less than last year's impressive -220 BCF draw.
The week ahead will start off exceptionally mild across nearly the entire nation with some relief expected later in the week as cooler temperatures ease southward. Temperatures over the weekend continued to be unusually warm east of the Rockies with estimated population-weighted mean temperatures of 52F-55F versus the historical average of 43F. We would not expect to see temperatures that warm until the first week of April. This warmth will continue into the early portion of the work-week. Figure 2 below shows projected departure-from-average temperatures Monday morning based on Sunday forecasts showing generally above-average readings east of the Rockies with near-to-below average readings isolated to the Pacific Northwest.
Figure 2: Forecast departure from average temperatures Monday afternoon showing generalized above-average temperatures across most of the nation. (Source: Tropical Tidbits.)
By Friday, the pattern will begin to shift with a trough replacing the ridge across the East leading to cooler temperatures. Figure 3 below shows forecast departure-from-average temperatures Friday morning, showing readings 5-15 degrees below average across the Southeast with above-average temperatures across the West.
Figure 3: Forecast departure from average temperatures Friday afternoon showing temperatures at or below average across most of the nation. (Source: Tropical Tidbits.)
The forecast population-weighted temperature Friday morning looks to be around 43F, 2 degrees below average and 10 degrees cooler than the start of the weekend, driving daily withdrawals to a projected -21 BCF, 1 BCF greater than the 5-year average daily draw of -20 BCF. Unfortunately, it will be too little too late to save the week for the bulls. My model is projecting a very bearish -55 BCF withdrawal for the week ending February 26, 82 BCF less than the five-year average and 174 BCF less than last year's impressive -229 BCF withdrawal. If it verifies, this draw would be the smallest in the past five years.
Looking long-term, computer models have been somewhat inconsistent, but the latest trends (as of mid-day Sunday) suggest the potential for considerably cooler temperatures heading into March for at least a few days. For the week ending March 4, I am projecting a storage withdrawal of -126 BCF, 7 BCF greater than the five-year average withdrawal of -119 BCF but 72 BCF less than last year's -198 BCF withdrawal. Moving further into March, temperatures look to slowly return to near-or-above-average levels. For the week ending March 11, I am projecting a -54 BCF withdrawal, 27 BCF less than the 5-year average but 11 BCF greater than last year's mediocre -43 BCF withdrawal.
In summary, for the 4-week period from February 12 - Mar 11, my model is projecting that -380 BCF will be withdrawn from storage with a projected March 11 inventory level of 2327 BCF. Should this verify, the storage surplus will grow by 101 BCF versus the five-year average to +656 BCF. This data is summarized below in Figure 4.
Figure 4: Projected weekly storage withdrawals for the next 4 weeks showing widening of the storage surplus from +555 BCF to +656 BCF. (Source: Data is my own via Celsius Energy.)
Should these projections verify, the +656 BCF storage surplus will be the largest since the week ending June 1, 2012. 2012 was a perfect storm of natural gas bearishness with rebounding natural gas production following the Great Recession rig cuts of 2008-2009 coupled with a record warm spring. Storage bottomed out that "winter" at just 2369 BCF, more than 800 BCF above the 5-year average. Natural gas prices plunged to $1.90/BCF, the lowest in 8 years. As figure 4 above showed, I expect that 2016 storage will finish below that level with a projected storage level at 2327 BCF on March 11 with 2-3 weeks of storage withdrawals to go.
These projections are based on Sunday PM computer model data and may change slightly in coming days. For those interested, I am now publishing my natural gas model data that was used to generate the above projections online at CelsiusEnergy.blogspot.com. This includes daily and weekly storage projections for the next four weeks updated twice daily as well as temperature, nuclear power outage, wind & solar generation data updated hourly, and natural gas pricing data updated live. There is no paywall, no free trials, just quality data that I use to guide my own natural gas trading.
One additional factor that may help to tighten the natural gas market is the imminent departure of natural gas exports. For Cheniere Energy (NYSEMKT:LNG), the second time may be the charm. The company was initially set to begin exports at their 1.8 BCF/day Sabine Pass facility in early January and had even begun loading natural gas into storage facilities for cooling with an LNG tanker waiting offshore in the Gulf of Mexico to begin loading. Then, on January 14, the company abruptly announced it was delaying exports until "late February or early March" due to instrumentation issues discovered during the final phases of the plant commissioning.
Now, over the past week, pipeline volumes on the Creole Trail Pipeline - Cheniere's dedicated Sabine Pass pipeline - have surged to record highs suggesting that the company is finally prepared to get this show on the road. Figure 5 below shows total volumes on the Creole Trail Pipeline since December.
Figure 5: Creole trail pipeline volumes over the past four months showing surge to record levels in the past 2 weeks suggesting the imminent onset of exports. (Source: Creole Trail.)
As the figure shows, after the "false start" in December when flows climbed to around 0.2 BCF per day, flows during January and February, generally flatlined to <0.05 BCF/day. Over the past week, volumes have again surged, peaking at 1.0 BCF on February 17.
Furthermore, Bloomberg reported last Thursday that the company had secured approval for and had begun loading coolants to begin and maintain the liquefaction process. Platts estimates that Sabine Pass's 17 BCF storage terminals were between 65% and 82% filled at the end of last week and, at current rates, the facility will need another 3-8 days to completely fill storage and begin exports, which could be as early as this week.
When they begin, exports will initially be around 1.5 BCF/day before increasing to 1.8-2.0 BCF/day this summer. At these rates, if supply and demand were otherwise perfectly balanced, it would take 52 weeks to completely wipe out the current storage surplus. Thus, exports alone are unlikely to restore/erase the storage surplus alone and, at least initially, their psychological role may be greater than the fundamental role.
In summary, natural gas will start out the next month decidedly bearish with a 5-year low storage withdrawal expected this week (to be reported by the EIA the following week). Temperatures will cool later this week and into early March with at least one above-average withdrawals likely. Nonetheless, I expect that the storage surplus will continue to edge higher, increasing by 101 BCF to +656 BCF by the second week of March.
Undoubtedly, over the next 2-3 months, comparisons will be made to 2012 when the spring storage minimum reached a record of high 2369 BCF. Current storage levels of 2706 BCF for the week ending February 12 are 55 BCF lower than 2012 at the same time (when storage was 2761 BCF). Should my 4-week projection of an inventory level of 2327 BCF for the week ending March 11 verify, this gap will narrow slightly to 42 BCF.
Investors are heavily discounting natural gas on account of the ballooning surplus. Figure 6 below plots the price of natural gas in 2012 versus the 2016 price.
Figure 6: Natural gas prices in 2012 versus 2016 showing a steep discount for 2016. (Source: Yahoo Finance.)
Despite current inventory levels trailing 2012 levels through the same period, natural gas is trading at a 33% discount to 2012 prices, $1.80/MMBTU vs $2.69/MMBTU. However, the substantial discount is not entirely unfounded. In 2012, natural gas bulls caught a break as record warmth and coal-to-gas switching drove an unprecedented 20 straight weeks of storage surplus contraction that saw the surplus narrow from a peak of +924 BCF in late March to just +262 by September. Even despite this dramatic improvement in fundamentals, season-ending inventory levels in November 2012 still peaked at 3929 BCF, which was a record at the time.
It will be a challenge for 2016 to see the same level of improvement. While early forecasts suggest that this post-El Nino summer may indeed be warmer than average, the odds are slim that we see the same level of warmth as 2012. According to the National Climate Data Center, the June-August 2012 period was the 116th warmest out of 118 years of record with record heat focused on the major population centers of the Midwest and Northeast urban corridor. Secondly, thanks to recent coal plant retirements and a prolonged period of cheap natural gas, coal-to-gas switching has nearly been maxed out and it will be difficult to boost natural gas demand much further by this route. Furthermore, last summer the West Coast saw significant hydro-to-gas switching thanks to the historic drought across the region, by as much as 1-3 BCF/day. Thanks to El Nino, hydroelectric levels have rebounded this winter and any additional coal-to-gas switching we do see this summer may be countered by losses from restored hydroelectric output.
Many investors - including yours truly - have counted on record low rig counts, currently at 101 natural gas rigs according to Baker Hughes eventually leading to a decline in production. According to Platts, this has not occurred and production levels last week remained at record highs above 73 BCF/day. With only 100 active rigs, there is little room to further cut the rig count, even if prices remain suppressed.
The one thing that 2006 may have going for it are natural gas exports. Exports will help tighten the natural gas market as discussed above by up to 14 BCF/week, but this would only narrow the surplus by 350-400 BCF, about half of the season-ending surplus.
Thus, the current natural gas discount, even against 2012 prices, is being driven by the fear that fundamentals do not support a rapid correction this summer as was seen in 2012 even with exports and a decimated rig count. This could lead to storage levels approaching capacity next fall. Current demonstrated working capacity according to the EIA stands at 4350 BCF while design working capacity is 4660 BCF, 519 BCF and 829 BCF above the 5-year average, respectively. Thus, if the natural gas surplus reaches +656 BCF as I project and then tracks the 5-year average for the next 7 months, it will exceed the demonstrated working capacity leading to a potential supply crisis.
It is my opinion that the combination of natural gas exports, above-average temperatures, and flat production will be greater than anticipated and inventory levels will not reach capacity next Fall. I expect that the combination of massive losses in the oil and natural gas rig counts will eventually pay off and production, particularly in the Marcellus Shale, will stabilize and decline. Natural gas exports, while not an immediate fix, will pay dividends long-term, particularly combined with increased domestic demand in the setting of a hot summer. Natural gas is trading at a nearly 50% discount to its 5-year average price of $3.40/MMBTU and has the potential to be one of the best trades in the next 1-2 years as fundamentals correct, and perhaps overcorrect.
However, in the short term, while I would not be surprised to see a bounce in natural gas over the next two weeks due to the forecast of colder temperatures and the arrival of natural gas exports, I expect renewed selling as soon as any bullish stimuli runs its course as we have seen all winter. I expect continued fear-driven trading through at least early April and would not be surprised to see natural gas remain under $2.00/MMBTU until that time. I believe that the turning point may come after storage bottoms out and investors get an early idea of how tight the natural gas market will be this summer after the first 2 or 3 reported injections. This however, in the short term, while I would not be surprised to see a bounce in natural gas over the next two weeks due to the forecast of colder temperatures and the arrival of natural gas exports, I expect renewed selling as soon as any bullish stimuli runs its course as we have seen all winter. I expect continued fear-driven trading through at least early April and would not be surprised to see natural gas remain under $2.00/MMBTU until that time. I believe that the turning point may come after storage bottoms out and investors get an early idea of how tight the natural gas market will be this summer after the first 2 or 3 reported injections. This would also allow time to evaluate the impact of natural gas exports. Should injections start coming in lower-than-expected, I see the potential for a substantial rally. I feel that a good time to consider going long would be just before then, particularly if natural gas trades under $1.75/MMBTU.
While it might be tempting to go long immediately to avoid an earlier rally, I am reluctant to do so due to the pervasive bearish sentiment and large contango. The spread between the March 2016 contract ($1.80/MMBTU) and July 2016 contracts ($2.09/MMBTU) is 16%. Thus, even if natural gas trades flat for the next few months, popular ETFs such as the United Natural Gas Fund (NYSEARCA:UNG) which must close out near-month contracts and rebuy the T+1 contract will suffer price-independent losses roughly equal to the contango. While the contango will narrow and these losses will diminish should natural gas rally, as long as the price remains depressed, a long natural gas position will lose value even if the natural gas futures price stays flat. If and when I do go long, my preferred ETF will be UNG. I am reluctant to purchase a long position in a leveraged natural gas ETF such as the 2x ProShares Ultra Natural Gas ETF (NYSEARCA:BOIL) or the VelocityShares 3x Long Natural Gas ETF (NYSEARCA:UWTI) due to recent volatility with routine daily moves >3% that will favor leverage-induced decay of these products. I am likewise reluctant to sell short the inverse leveraged ETFs, namely the VelocityShares 3x Short Natural Gas ETF (NYSEARCA:DWTI), a strategy that has worked well for me in the past, due to the risk of rapidly accumulating losses should natural gas continue its steep decline.
In short, I believe that natural gas offers excellent long term value at its current decade-lows and that investors are likely too bearish in their concerns of a storage crisis next fall. However, given the short-term likelihood of a growing storage surplus and legitimate concerns about the sector's ability to rapidly reduce glut this summer, I am poised to watch from the sidelines until the fundamentals give some signs of improvement. While I might sacrifice some early gains, I would rather see concrete evidence of a rebalancing of supply and demand than gamble that "this is the time."
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.