From Seeking Alpha (charts did not paste)
Natural Gas Bull Market... Delayed, Not Denied
Aug. 23, 2019 7:20 AM ET|
3 comments
|
About: The United States Natural Gas ETF, LP (UNG), UGAZ, DGAZ
Atlas Research
Atlas Research
Long/short equity, macro, value, growth at reasonable price
(1,202 followers)
Summary
The U.S. natural gas market is currently oversupplied by roughly 1.9 Bcf/d.
Today's oversupply is a direct result of four delayed export projects, originally slated for Q2 2019 or earlier.
Absent these delays, the U.S. gas market would have flipped from surplus to deficit this summer.
As these delayed projects hit the market, producers must supply 5.2 Bcf/d of new gas production to avert a supply crunch in 2020.
It's been a brutal summer for natural gas (UNG) bulls...
With gas prices recently threatening to break below $2 per Mcf -the lowest seasonal level in over two decades - it might seem irrational for anyone to justify a bullish view on the 2019 gas market. So was I wrong to expect gas prices to double in 2019?
Believe it or not, I will happily admit when an investment idea blows up in my face... it's happened before, and I'm quite sure it will happen again. But before issuing a mea culpa, let's first take a step back to review the original bullish thesis for 2019, which was based on two simple factors:
1) a slower rate of gas production growth
2) massive new demand for U.S. gas, mostly from pipeline exports into Mexico and LNG facilities coming online
So let's start by reviewing the latest production trends in the U.S. gas market.
2019 Production Stalling... For Now
So far, the 2019 gas production story has played out as expected. Based on the latest EIA data, May 2019 gas production of 89.9 Bcf/d came in just 1 Bcf/d above the 88.9 Bcf/d produced in December 2018. That's an annualized pace of just over 2 Bcf/d growth- a huge decline from last year's 9.5 Bcf/d year-on-year growth.
The chart below puts 2019's year-to-date gas production in context of the more torrid growth rates registered in late 2017 through 2018, a phenomenon driven less by economics and more by producers fulfilling pipeline commitments (most of which were signed 3-4 years prior, when gas forward curves traded well above current prices):
2019 us natural gas production
Source: EIA
Now before proceeding, the big caveat on this 2019 production profile is that the data above is clearly a few months old. In their more recent data estimates, the EIA indicates that Appalachian gas drillers have grown production further over the recent summer months. That means we can will likely see an uptick in growth when the EIA releases its finalized June - August production data.
But ultimately, even though we don't have the latest official production numbers, the latest production trends will show up indirectly in the supply/demand balance via changes in gas storage, as I'll show in the following section.
Weather-Neutral Gas Storage Indicates ~1.9 Bcf/d Excess Supply
Last April, I showed that the U.S. gas market was oversupplied by roughly 3 Bcf/d. As a brief review, the methodology used in that article assumed a linear relationship between heating degree days (HDDs) and winter gas demand. Said differently, every 1% increase in HDDs during the winter should translate into a roughly 1% increase in natural gas inventory drawdowns. So by normalizing winter gas storage drawdowns by the number of HDDs, you can derive the underlying weather-neutral supply/demand balance.
During the injection season (May through October), you can use the same basic methodology, except normalizing the storage injection numbers with cooling degree data (CDD). In the chart below, you can see that 2019 has been almost exactly average in terms of cumulative CDDs versus the five year average.
2019 Cooling Degree Days
Source: American Gas Association
By the numbers, we've accumulated 921 CDDs from early May through Mid-August, compared with a 5-year average of 926 CDDs over the same time period. In other words, this summer is almost exactly average with respect to weather-related deviations in demand. So to keep the math simple (and because this is a rough approximation) we can assume no adjustment for anomalous weather demand so far this injection season. That means we can directly compare the average daily gas injection volume versus the previous 5-year average to estimate the relative supply/demand balance, as shown in the chart below:
natural gas storage injections
Source: EIA
As show above, we've injected an average of 11.9 Bcf/d per day during this year's injection season. That's 1.9 Bcf/d above the five-year average of 10.0, which implies an average oversupply of roughly 1.9 Bcf/d since May of this year (full disclaimer: this is a rough estimate).
While this 1.9 Bcf/d of excess supply marks a notable improvement from last winter's 3 Bcf/d surplus, going from bad to less bad provided little consolation to the bulls as prices crashed on persistent above-average storage injections this summer.
So what happened, and where do we go from here? For that, let's turn to the demand side of the equation.
Force Majeure Takes 1.5 Bcf/d Summer Demand Off the Market
At this time last year, the massive Valley Crossing – Sur de Texas pipeline project was slated to begin exporting up to 2.6 Bcf/d of Texas gas into Mexico starting in October 2018. For reference, 2.6 Bcf/d exceeds the entire growth in associated gas from the Permian over the last 12 months (based on the EIA's current estimate of 14.6 Bcf/d of Permian gas production this August). So it's no surprise why this huge outlet for U.S. gas formed a key piece of the 2019 bullish thesis that I laid out last year.
But fast forward 10 months later, and this pipeline still hasn't come online. Here's what happened...
As the name implies, this project included two segments: the Valley Crossing pipeline on the U.S. side of the border connecting with the Sur de Texas-Tuxpan (STTP) pipeline on the Mexican side of the border. Canadian pipeline operator Enbridge oversaw construction of the Valley Crossing section of the pipeline, hitting their original project deadline of October 2018. Meanwhile, a 60/40 joint venture between TC Energy (formerly TransCanada) and Sempra Energy’s Mexican subsidiary Ienova, known as Infraestructura Marina del Golfo (IMG), oversaw the construction of the STTP.
Now, if it were simply a matter of construction, the STTP pipeline would have likely also hit its October 2018 deadline. However, progress on STTP was delayed for months on end due to a series of indigenous protests blocking construction progress. After more than six months of delays, IMG announced mechanical completion of the Sur de Texas pipeline on June 11th ... but this only opened the door for a separate set of legal delays between IMG and its intended customer, Mexico's state-owned utility CFE. Let me explain...
You see, the original estimates in the Sur de Texas contract called for $2.1 billion in project capex. But thanks to the project delays, the final cost ballooned to $2.5 billion. Given that these delays were caused by a series of unpredictable indigenous protests, the IMG joint venture invoked a “force majeure” clause in the contract. Effectively, IMG claimed the right to charge CFE for pipeline capacity payments dating back to the original October 2018 start date. CFE fought these claims, bogging down the start-up of pipeline flows further throughout the summer.
Of course, it's in neither party's economic interest to see pipeline flows further bogged down in a legal dispute. So my bet is that's its only a matter of time before some kind of agreement is reached, and volumes start flowing through the pipe. Indeed, just a couple of days ago, we got the first indication of a potential breakthrough in the negotiations...
On August 19th, Mexican President Obrador noted in a morning press release the potential for an imminent resolution between CFE and the pipeline operators. In the event of a resolution, energy research firm Platts Analytics expects that flows through the pipeline could reach as high as 1.5 Bcf/d imminently. Longer-term, we can expect the full 2.6 Bcf/d capacity on this pipeline to fill up, as Mexico’s domestic gas production continues declining while the country continues its switch from coal to gas fired power generation.
Now remember, I just showed that we’ve been oversupplied by roughly 1.9 Bcf/d so far this summer. So this one single project alone - had it not suffered from months of unforeseeable delays - would have wiped out more than 75% of this summer's gas surplus. And this wasn't the only major source of demand destruction this summer...
LNG Delays Cut 2 Bcf/d of Summer Gas Demand
Across-the-board delays in LNG export facilities delivered the final knockout blow to the gas market this summer. Of course, with the benefit hindsight, you might argue that 2019's LNG construction delays were at least modestly more predictable than the indigenous protests and legal wrangling over the STTP project. But that would have ignored the precedent set by Bechtel – the construction contractor who set the bar high with their track record of bringing Cheniere's LNG facilities online ahead of schedule and under budget.
If the other construction contractors in the space performed anything like Bechtel, then the U.S. gas market would have benefited from an additional 2.25 Bcf/d of nameplate LNG demand this summer. The table below came from the EIA's website in March, which detailed the list of upcoming LNG export facilities based on the publicly available project timelines from the project owners (the yellow highlighted rows show the projects scheduled for completion by this June):
Source: EIA
However, as of mid-August, not a single one of these projects had yet reached commercial production, thanks to across the board construction delays at the Cameron, Freeport and the Elba Island LNG facilities. Assuming these facilities would have averaged 90% utilization, that translates into 2 Bcf/d of lost potential demand this summer.
So when you add this 2 Bcf/d of lost LNG demand with the 1.5 Bcf/d of lost pipeline flows into Mexico, basic math shows how - absent project delays - this summer's 1.9 Bcf/d surplus would have flipped into a 1.6 Bcf/d gas deficit. Just to reiterate...
Delayed export projects meant the difference between surplus and shortage in this summer's gas market.
So yes, I will happily admit my failure to predict months of indigenous protests followed by months of legal wrangling over the STTP pipeline, in addition to across the board delays among three separate LNG export facilities. For those who correctly predicted this series of events (whom I've yet to meet), congratulations.
But of course, that's old news today. What about going forward?
LNG Export Demand Delayed... not Denied
On August 19th, news broke that commercial operations had officially started at the Cameron's Train 1 LNG facility. On the same day, a spokesperson for the Freeport project announced the start of sustained LNG production for Train 1, with commercial volumes expected to commence in September.
Meanwhile, on Kinder Morgan's latest earnings call on July 17th, management noted mechanical completion on four of the first ten Elba Island trains, and that the facility was ramping up to full commercial operations. They haven't provided a firm date on their projected commercial start-up, but my bet is the first series of trains hits the market sometime in late Q3 or early Q4.
So within the next couple of weeks, we can likely expect roughly 1.2 - 1.3 Bcf/d of new gas demand for Train 1 of both the Freeport and Cameron LNG projects, plus another 0.2 Bcf/d from Elba Island Trains 1 - 6 within the next few months. This adds to the potential boost of roughly 1.5 Bcf/d when the legal wrangling over STTP wraps up. That's a potential of about 3 Bcf/d of imminent demand in just the next few months.
And if we look out further into 2020, we have additional LNG export demand coming from Freeport Trains 2 - 3, Cameron Trains 2 - 3 and the remaining Elba Island trains coming online. The chart below sums up all of the additional LNG projects we can expect to hit the market in the next 12 months:
LNG Demand
Source: EIA
Assuming a 90% average utilization rate from nameplate capacity, that's a combined 4.2 Bcf/d of new LNG capacity that will bring the total up to 10 Bcf/d by Q3 2020. And as the Valley Crossing - STTP pipeline into Mexico ramps up over time, we can expect that pipeline will eventually reach its full 2.6 Bcf/d of capacity.
Meanwhile, the ongoing switch from coal to gas fired power plants in the U.S. electric grid continues, with an average year-over-year gain of 1 Bcf/d so far this year that shows no signs of stopping anytime soon. So here's the bottom line...
Significant New Production Needed to Avert 2020 Gas Supply Crunch
Putting all the pieces together, here's what we can expect over the next 12 months:
1) 4.2 Bcf/d of new gas demand for LNG from Freeport Trains 1 - 3, Cameron Trains 1 - 3 and Elba Island Trains 1 - 10.
2) A conservative estimate of roughly 2.2 Bcf/d in new gas exports to Mexico, driven largely by the Valley Crossing - STTP pipeline.
3) A conservative estimate of 0.7 Bcf/d of new gas demand for power generation.
That's 7.1 Bcf/d of new gas demand in the next 12 months, even allowing for further project delays among new LNG facilities. Given our starting point today of roughly 1.9 Bcf/d in excess supply, that implies we'll need 5.2 Bcf/d of new gas production over the next 12 months to avert a supply crunch and a bull market in gas prices in 2020.
So that only leaves one logical question to answer regarding the future direction of the U.S. gas market...
Will producers step up to fill the projected supply gap of 5.2 Bcf/d to avert a 2020 supply crunch? Stay tuned for future articles, where I plan to address that very question.
Thanks for reading and good investing,
- Atlas
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.