Page 1 of 1

EIA - Natural Gas Storage Report - Aug 6

Posted: Thu Aug 06, 2020 9:56 am
by dan_s
Working gas in storage was 3,274 Bcf as of Friday, July 31, 2020, according to EIA estimates. This represents a net increase of 33 Bcf from the previous week.
Stocks were 601 Bcf higher than last year at this time and 429 Bcf above the five-year average of 2,845 Bcf.
At 3,274 Bcf, total working gas is within the five-year historical range.

The build was 5 Bcf higher than the 5-year average for the last week of July, probably because all the rain we had that lowered power demand a bit in the Texas and the SE.

For August the 5-year average builds are higher (42 to 66 Bcf), so we should see the delta to the 5-year average go on decline. If so, we might see the October NYMEX contract push over $2.50; it is trading at $2.35 at the time of this post. My forecasts are based on Henry Hub gas averaging $1.75 for Q3, $2.25 for Q4 and $2.50 for 2021.

Updated forecast/valuation models for CRK, EQT and RRC are now on the EPG Website under the Sweet 16 tab. You can download the Excel spreadsheets and see what happens in 2021 if you change the gas price to $3.00/mcf. If HH futures for Q1 push firmly over $3.00, there is significant upside for these three.

Re: EIA - Natural Gas Storage Report - Aug 6

Posted: Thu Aug 06, 2020 11:27 am
by dan_s
Goehring & Rozencwajg
Natural Resource Market Commentary
August 4, 2020

North American natural gas has been in a vicious bear market for 15 years. Surging supply
brought about by the shale gas revolution has resulted in a persistent surplus. Although
demand has also surged over the same period, it has not been able to keep up with the
unrelenting growth in production.

This is all changing as we speak.

Supply has now begun to contract, and the North American natural gas market is about to
swing from long term structural surplus to deficit.

One barrel of oil contains the same energy content as six thousand cubic feet of natural gas
and historically this has anchored together the price of the two fuels. From 2000 to 2005,
the price of oil averaged 7.5 times the price of natural gas -- not far from the energy equivalency.

The relentless supply surge beginning in 2006, combined with the environment
regulations that curtailed a utility’s ability to burn residual fuel oil (the most competitive
fuel for natural gas), caused the energy link between natural gas and oil to break down. After
the warm winter of 2011-2012, natural gas price fell below $2 per mmcf. With oil priced at
$103, the oil-natural gas ratio hit 53:1, -- almost 9 times its energy equivalent. In the 40
years of data that we keep, this is by far the highest (i.e., most bearish) oil-natural gas ratio
ever.

Over the last twelve months, the oil-natural gas ratio has averaged approximately 25:1 –
still far below its energy-equivalency. If our research is correct, we will see the ratio fall
dramatically and may even see it return to its historical six to eight-times ratio.
Excess production
is what caused the link to break and we are now entering into a period of declining
supply.

The price of natural gas peaked in 2005 at over $15 per mmcf, and today stands at $1.65 --
almost 90% below the peak. The fundamental reason for the bear market has been simple:
US natural gas supply surged due to the shales. The initial successes in the Barnett by
Mitchell Energy in the early 2000s was followed by the discovery of the Fayetteville by
Southwestern Energy in 2005, the Haynesville by Chesapeake Energy in 2007, and then the
massive Marcellus field by Range Resources soon after.

After having declined consistently over the previous 10 years, natural gas production eventually
bottomed in 2005 at 49 bcf day. By 2019 US dry gas supply had nearly doubled to 92
bcf per day -- a stunning increase of 4.6% per year. Shale dramatically changed the composition
of the US natural gas supply between 2005 and 2019.

It was not only shale production from the gas fields that contributed to the growth. Led by
surging production from the Permian and Eagle Ford oil shales, so-called associated natural
gas (by-product gas produced from oil wells) grew to be 16% of US gas supply.

By 2019, primary gas production from the Marcellus and Haynesville, along with associated
gas production from the Permian, had grown to almost 45% of US gas production and
over 100% of total natural gas production growth.

Because the shale fields have been such prolific drivers of supply growth, many analysts do
not appreciate that they eventually succumb to the same geological forces affecting conven-
tional gas and oil fields. Production ramps up following an initial discovery, plateaus once
drilling productivity begins to falter, and ultimately declines once the drilling productivity
can no longer overcome the underlying depletion rate of the field.

All Oil & Gas Fields eventually peak and then go on decline

The first two shale gas fields put into production (the Barnett in East Texas and the Fayetteville
in Arkansas) have already ramped up, peaked, and declined in the same sequence of events
experienced by conventional fields. The Barnett ramped up production starting in 2001
while the Fayetteville began its steep ramp up in 2007.

The Barnett and Fayetteville both ultimately peaked in 2012 at 5.2 and 2.9 bcf/d respectively.
Since then, both fields have declined by 60% and 65%, respectively, from their peak
levels and production has entered into terminal decline. Today, neither field has a single
rig drilling for gas.


We used our neural network to analyze both fields and we identified two important data
points that coincided with the beginning of declines in both fields. First, production
declined in both fields once 60% of their total “Tier 1” acreage (as defined by our neural
network) had been drilled up. This coincided with the moment when 50% of the fields’ total
recoverable reserves had been produced.

Below is a map that shows the density of Barnett wells drilled through 2007. We have outlined
what our neural network identified as the best Tier 1 acreage. As you can see, early drilling
in the basin was widely scattered; operators were only just learning where the best areas were
located. Eight years later, things were very different. By 2012, drillers knew exactly where
the best acreage was located and were rapidly drilling it out.

Our neural network suggests that by 2012, nearly 60% of all the best Tier 1 locations in the
Barnett had already been drilled. It now seems that this 60% Tier 1 development threshold
coincides with a stagnation in overall drilling productivity Once 40% of the best wells are
drilled drilling productivity plateaus and begins to slow and once 60% of the best wells have
been developed overall production begins to fall.

The same phenomenon occurred in the Fayetteville shale as well. This map shows all the
Fayetteville wells drilled through 2009 along with our neural network’s estimates of Tier 1
locations. On the right is the same map with wells drilled in 2013. Although it is less pronounced
than the Barnett, the drillers had focused in the best part of the cores here as well.

The two main sources of recent growth have been the Marcellus and Haynesville, but our
models tell us they are going through the same phenomenon. The following maps show how
concentrated drilling has become in the best areas of both basins in 2018 and 2019. We estimate
that over the last five years, Marcellus producers have concentrated their drilling in Tier 1 areas
from 45% of all wells to 60%. In the Haynesville drillers have gone from 53% to 67%.

The five "gassers" in our Sweet 16 and Small-Cap Growth Portfolios are:
> Comstock Resources (CRK), a pure play on the Haynesville Shale
> EQT Corp. (EQT), Marcellus & Utica Shales
> Range Resources (RRC), Marcellus & Utica
> Goodrich Petroleum (GDP), Haynesville
> Gulfport Energy (GPOR), Utica Shale and Central Oklahoma


Our neural networks suggests that 60% of Tier 1 wells have been developed in the Haynesville
and 40% have been developed in the Marcellus. If these plays follow the same path as the
Barnett and Fayetteville, then the Haynesville has entered terminal decline while the Marcellus
is in the process of plateauing. Since the end of 2019, gas production from the Marcellus has
declined by almost 1 bcf per day while Haynesville production has declined by 400 mmcf/day.

The only other field that is likely to continue growing over the next several years is the Utica
in eastern Ohio and western Pennsylvania.
The Utica was developed much later than any other
play: production did not start growing materially until 2014. Using the most recent drilling
data, our neural network estimates that only 22% of its Tier 1 locations have been drilled to
date. Given that significant Tier 1 inventory remains, our models suggest production could
still rise by 50% over the next several years.

In our 3Q2019 letter, we also established the relationship between total recoverable reserves
and peak production. We showed how a shale gas field’s peak production has occurred once
half of its total recoverable reserves have been produced. This relationship is very well established
in conventional basins, but we showed how it held true with the Barnett and Fayetteville
as well. According to this our analysis, the Haynesville has likely produced more than half
its total reserves and as a result has likely peaked. The Marcellus is not far behind. The Utica
once again is the only basin that shows signs of potential growth.


Production of associated gas from the shale oil fields will be challenged as well. Please see our oil
section for an in-depth discussion of the challenges going forward. In the short term, associated gas
is almost guaranteed to decline as drilling in the Permian has fallen by 70% in only three months
.

In total, we believe that shale production will decline by 1.5 bdf/d per year over the next five
years after having averaged an incredibly 4 bcf/d of growth each year over the past decade.