Why U.S. Natural Gas Liquids Production is Surging

Monday, 01/09/2023 Published by: David Braziel RBN Energy rbnenergy.com

This is an excerpt.  To see the graphs mentioned in the article, use this link:  https://rbnenergy.com/when-the-levee-breaks-why-us-natural-gas-liqu...

Since the advent of the Shale Revolution way back in 2008, U.S. production of natural gas liquids from gas processing has grown pretty much non-stop, from an annual average of 1.8 MMb/d 15 years ago to 5.9 MMb/d in 2022 — a 9% compound annual growth rate. Today, NGL production exceeds 6.1 MMb/d and that number might be even higher if the glut of supply wasn’t depressing prices and discouraging the recovery of a lot of ethane. All that production has major implications for domestic pricing, upstream economics, midstream infrastructure, and downstream consumers like petrochemicals, not to mention international markets, which now receive roughly 40% of U.S. output. In today’s RBN blog, we examine what’s causing NGL production to continually increase.

To understand what’s going on with U.S. production of the mixed stream of natural gas liquids collectively known as NGLs (ethane, propane, butane, isobutane and pentanes+), it’s important to recognize the relationship between NGLs and the production of crude oil and natural gas — after all, they all come from the same holes in the ground in hydrocarbon-rich areas like the Permian, Bakken, and Eagle Ford. Because of their common origin, RBN refers to the three commodity streams (crude, gas and NGLs) as the “drillbit hydrocarbons.”

These days, about 80% of drilling in the U.S. is primarily directed at crude oil production, which makes sense because (generally speaking) crude is the most valuable of the drillbit hydrocarbons on a per-Btu basis. Crude doesn't emerge from shale plays on its own, of course — instead, it comes out of the ground mixed with what’s typically referred to as associated gas, a gurgling combination of methane (natural gas), mixed NGLs and various impurities. The composition of this oil/natgas/NGLs stew varies widely, not only between shale basins but within each basin and from well to well — and even within each well over time.

The differences in drillbit-hydrocarbon composition between oil-focused basins, within basins and from well to well is easy to wrap your head around — depending on location, there will be variations in rock and hydrocarbon content within that rock. As for the changes in composition over time at individual oil-focused wells in key shale basins, they tend to result from an increasing gas-to-oil ratio (known as the GOR and calculated as Mcf of gas per barrel of oil). In other words, the output of individual wells and entire shale basins tend to become “gassier” from year to year. As we discussed in the blog Don’t Stop Me Now, the main reason for rising GORs is that gas type curves generally tend to be shallower — meaning they decline less quickly — than oil type curves. Also, additional natural gas and NGLs tend to be captured as gathering and processing infrastructure is built out and restrictions on flaring tighten.

Rising GORs help to explain (1) why an increasing share of U.S. natural gas production is coming from oil-focused shale basins (the Permian being a prime example) and (2) why U.S. crude oil production has stalled at just over 12 MMb/d but natural gas and NGL production continue to grow. They also help to explain why — even as demand for natural gas for LNG exports is rising — that the Haynesville is the only place where gas-directed drilling is increasing significantly, with much smaller gains in the gassier parts of the Rockies and Eagle Ford. (Appalachia, once the star of domestic gas production growth, remains mired in pipeline purgatory.)

As you would expect, the trajectory of GORs differs from basin to basin, depending on a number of factors such as the degree of recent drilling activity (again, new wells tend to be oilier than older ones) and where E&Ps are choosing to drill within the basin. The graphs in Figure 1 show the trends since January 2017. The average GOR in the U.S. had been steadily declining (graph to left in figure 1) — primarily as a result of decreased gas production from legacy fields. The ratio spiked in 2020 when crude prices crashed and a lot of primarily crude production wells were shut-in, but has since restabilized at about 9.4. The major unconventional crude oil basins have notably different trends. As you can see, only the Niobrara has seen a decreasing GOR, from 11.0 to 7.7, over the past five years (light blue line in set of graphs to right), while the GOR has risen from 14.5 to 16.0 in the Anadarko (red line), 1.6 to 2.7 in the Bakken (purple line), 5.1 to 6.0 in the Eagle Ford (orange line), and 3.4 to 3.8 in the all-important Permian (green line). That increase of 0.4 in the Permian’s GOR may not sound all that profound — until you consider that the basin now produces 5.7 MMb/d of crude and that the higher GOR translates to an increase of more than 2 Bcf/d of gross gas — an enormous volume to be sure.

That’s one reason why the Permian continues to require additional gas takeaway capacity even as it has plenty of crude pipeline capacity. In fact, Pioneer Natural Resources CEO Scott Sheffield made headlines last Thursday when he said, among other things, that “the (GORs) in the entire Permian Basin will continue to go up. We're seeing that. You'll see the percent oil drop for all those companies, most likely below 50% over the next 10 years. And the gas itself will get up to about 30 Bcf/d. We're going to need a [new] gas pipeline at least about every 18 months to two years going forward.”

Production in the Permian, like in several other oil-focused shale plays, has another characteristic particularly pertinent to today’s blog topic — namely, the associated gas produced there is richly saturated with NGLs. NGL content in associated gas is often measured in gallons per Mcf of gas, or GPM. Depending on the hydrocarbon mix of the basin in question, each Mcf of associated gas may have anywhere from a couple to several GPM of mixed NGLs entrained within it.

The label GPM is generally applied to wellhead gas — in other words, how many gallons are in an Mcf when it comes out of the ground. However, wellhead GPMs are highly variable and not widely published so few know that figure for sure other than the producers. What is reported — and therefore what we can know — is how much liquid is extracted by the time the resulting natural gas makes it through to the tailgate of the processing plant. That tailgate volume of liquids is less than the total volume of liquids, sometimes by a considerable amount, because some of the ethane is not recovered and is instead left in the gas stream (more on that in a moment). So when we refer to GPMs in this blog, know that we’re discussing tailgate GPMs.

As shown in the chart to the left in Figure 2, the recovered GPM across the U.S. has been rising steadily since 2008, from 1.3 GPM back then to 2.3 GPM today. A rising GOR coupled with a rising GPM means that now, on average, for every barrel of crude oil produced in the U.S., 9.4 Mcf of gross gas is produced and 22 gallons of mixed NGLs (9.4 x 2.3 = 22) — or about half a barrel of NGLs (22 / 42). A major reason for the rising GPM is that new, highly efficient processing infrastructure has enabled a higher percentage of NGLs to be recovered from the gas stream. (In an upcoming blog, we’ll explore how today’s highly efficient gas processing technologies enable a higher percentage of NGL recovery.) The graph to the right in Figure 2 shows that GPMs vary considerably by region. PADD 3 (Gulf Coast; green line), home to the Permian and Eagle Ford, has experienced a steep rise in GPMs through the Shale Era and now has the highest GPM of any PADD, edging out PADD 2 (Midwest; gray line), home of the Bakken, since the mid-2010s.

There are at least another couple of other things to consider when you're looking at NGL production and GPMs. One is that — as we mentioned just above — the GPM number only reflects the NGLs that are separated out by gas processing plants and does not include ethane that, for one reason or another, gets left in the residue natural gas stream. And the degree of recovery differs widely by region. For example, in PADD 5 (West Coast, yellow line), the recovered GPM is kept lower because none of the ethane is recovered — it’s all left in the gas stream. On a similar note, in PADD 1 (East Coast); purple line), the substantial increase in GPM starting in 2013-14 was due to the startup of significant processing in the region, with the first ethane recovery there occurring in December 2013.

Another quirk of all this liquids production that bears mentioning is that in shale plays like the Permian that have experienced a buildout of efficient new processing capacity, the average recovered barrel of mixed NGL production (known as y-grade after it’s been processed out of the gas stream) tends to contain a greater portion of lighter components like ethane and propane than heavies. As shown in Figure 3 below, on average since 2005, the proportion of ethane in the U.S. NGL mix has grown from 38% to 41% (orange layer in right graph) and propane’s share has grown from 29% to 31% (purple layer), while the combined share of heavier purity products has declined from 33% to 28%. That’s going to have an impact on the types and configurations of facilities upstream needed to handle the liquids as well as downstream markets and prices.

And one more thing: As with crude oil and natural gas, as production of NGLs has ramped up during the Shale Era, surplus production has pushed its way into international markets to the point where the U.S. is now a leading exporter of crude, gas and NGLs. Just as noteworthy, while on average in 2022 the U.S. exported about 30% of the crude oil it produces (up from only 4% in 2014) and 20% of its natural gas (up from 6% in 2014), fully 40% of its produced NGLs are now exported (teal layer in graph to far right in Figure 4), with a whopping 60% of propane production sent abroad.

But whereas, prior to falling at the start of 2023, crude and gas prices were elevated last year due to a variety of reasons, including restrained supplies due to capital discipline and the war in Ukraine, the price rise in NGLs has seen more of a decline in comparison (left graph in figure 5), with several key price ratios like the propane-to-crude ratio and ethane-to-gas (at Henry Hub, or HH) ratio sinking very low indeed. As shown in Figure 5 below, the price of a weighted average basket of NGLs (green line in left graph) has fallen compared to the price of crude oil and natural gas. That means that the propane-to-crude ratio is down to around 40% lately while the ethane-to-gas ratio has recently been below 1:1 and only in the last week, risen back above that threshold.

The difference for NGLs is that their production is impacted by the multiplicity of factors we noted above — rising GORs, rising GPMs, and a rising share of lighter NGLs among them — while the demand for NGLs from the petchem sector and other consumers has grown only modestly and has taken a hit as the outlook for the global economy has dampened. 

With NGL prices so low compared to oil and gas, weird things happen. Most notably, when the price of ethane falls below the price of natural gas on a Btu equivalence, it incentivizes processors to reject their ethane, meaning that they leave it entrained in the natural gas stream to capture the higher value.

So how long might the flood of NGLs persist? Well, there is little to make us think that the GOR, GPM and lighter-barrel trends described above won’t continue. And that suggests that, with U.S. demand for NGLs largely satiated, the incremental barrels will overflow into international markets like river water jumping its levee in a flood.