U.S. natural gas supplies should be about 3.1 Bcf/d higher this summer than a year ago, but incremental demand is forecast to be lower, which would keep the markets long through the injection season, Genscape Inc. analysts said Wednesday.
Senior natural gas analyst Rick Margolin shared a microphone with Randall Collum, managing director of supply side analytics, to offer the firm’s domestic summer supply and demand forecast. Overall, production is expected to be 3.3 Bcf/d higher this summer, with some supply diminished by 0.2 Bcf/d of lower imports from Canada. Genscape is forecasting Canada imports to average 4.8 Bcf/d this summer, down from 5 Bcf/d a year ago. On the demand side, 2.4 Bcf/d of incremental growth should come from overall gains in coal plant retirements and hydropower generation, as well as 2.42 Bcf/d of exports to Mexico, a 0.3 Bcf/d year/year increase. That would total 2.7 Bcf/d of demand, "which basically means we continue to remain a long situation this summer," Margolin said.
Total gas demand for the Lower 48 this summer is estimated at about 64 Bcf/d, which is higher than last year. That's about 2.4 Bcf/d more than that demand averaged in 2014, Margolin said. And it's 1 Bcf/d more than in the "remarkably high" demand year 2012.
The injection season "is going to start with more or less normal inventories, and by the end of the summer, there's a very real potential that we have new records for working gas in storage," he said. As of now, Genscape is estimating that when injection season ends Oct. 31, there will be 3,988 Bcf in Lower 48 working gas inventory.
"Certainly, weather is going to be the big driver on whether or not that number is actualized," Margolin said. "A record warm summer would probably push demand higher and leave inventories lower, somewhere in the upper 3.7 Tcf range. However, if we have a very cool summer and demand for gas power generation doesn't manifest itself, there's a very real potential that working gas inventories at the end of the summer crests the 4 Tcf level."
Genscape had forecast that 12 Bcf would be injected for the week ending March 20, a figure that turned out to be right on the money (see related story). The injection was the first this year, a bit earlier than normal, but not too much off the mark, Margolin said. Through the rest of this month, Genscape is estimating that inventories will remain "right about where they are now. Putting this in the context of recent years, this is just shy of the five-year normal…and it is substantially higher from where we entered last winter. There's a pretty decent amount of gas in storage for the start of the season, and that's going to be important to keep in consideration to round out the demand story."
The "big story" this summer is the amount of incremental gas burn expected from the "large swing of coal retirements," Margolin said. “Throughout 2015, we are estimating 27 GW of nameplate capacity on coal units in the United States will be retired. Dissecting that into the summer, we estimate about 18 GW will go off..." Most of the coal plant retirements are in PJM Interconnection, followed by the Southeastern Electric Reliability Council and the Midwest Independent Transmission System Operator.
"If we were to replace those coal units on a 1:1 basis with natural gas, certainly we would get a large number. But what's important to keep in consideration is that the coal units that are going offline have not been very heavily utilized in recent years," Margolin said. At the coal units Genscape is tracking, average utilization in 2014 across the to-be-retired fleet "was just 26% of nameplate capacity. In effect, most of the capacity has already been idled. So this is going to diminish the incremental gas burn gains that could be grasped from these coal units going offline. Nonetheless, we do have retirements that are going offline and some of this is going to get replaced with gas."
Factoring in the utilization rate of the retired units with rates on gas plants in areas surrounding those units, Genscape estimated that the incremental burn from those retirements would generate just under 900 MMcf/d of new gas demand this summer.
"Over the course of the year, because some of those retirements are scheduled to take place beyond the summer strip, the total number gets a little bit higher, to just over 1 Bcf/d," Margolin said. "There are incremental burn opportunities that will occur from the coal retirements, but the number may not be as robust...Nonetheless, demand growth is demand growth."
Other opportunities for increased gas demand this summer would come from the West, where hydro supplies are low and are not expected to show up in force. Genscape calculated that in low hydro years, the West has averaged a burn of about 5.3 Bcf/d. In an average hydro year, the region has been burning about a bit more than 5.1 Bcf/d, which means in a low year, there’s an additional 0.12 Bcf/d more gas demand. That incremental burn, however, only would run through July.
Growth of renewables also is going to reduce what heretofore were gas gains. More than 25 GW of renewable generation is expected to be installed this year, by Genscape estimates. Although most of the additions would not happen until later this year, the summer should see more than 6 GW of wind and solar capacity added into the market. That means that toward the end of the summer, those renewable sources could pressure gas demand to contract.
A growing source of demand for U.S. gas is from Mexico's escalating thirst. Mexico plans to add about 14 GW of gas-fired generation capacity through 2020, mostly in 2016 and 2017, "but in the interim, there also is increasing utilization of existing power infrastructure, Margolin said. "At the same time, Mexico's domestic gas production has remained flat to declining in some areas. With demand growth and production declining, it's become increasingly dependent on U.S. gas, as seen by growth in new cross-border pipelines.
He pointed to the start up of the NET Mexico pipeline, with 2.1 Bcf/d capacity (see Daily GPI, Nov. 17, 2014). The next two years "should see pretty aggressive growth to Mexico." For this summer, Genscape is forecasting that flows to Mexico should average just under 2.5 Bcf/d, which would represent a 0.3 Bcf/d summer/summer increase.
Meanwhile, gas production may be slowing, but it's not changing the supply picture too much, said Collum. Earlier this month he told NGI that Appalachian Basin gas output was on pace to increase even with lower capital spending (see Shale Daily, March 20).
The gas and oil rigs continue to decline in the U.S. onshore, and while the biggest hit is to the oil side, the gas markets are in turn impacted.
Genscape doesn't see a bottom on the rig count until the third quarter, likely in August "and we actually think they will start rebounding from there as the contango in the crude market, prices go up, which will in turn drive the economics up for oil drilling," Collum said. Reductions in service costs "will also will help motivate the rigs to get out there and start drilling again. We actually think in 4Q2015 we'll see a decent little ramp in oil rigs."
Genscape's outlook for U.S. gas production is well below what it was last August, all because of the decline in crude and gas prices. "April 2015 crude is trading about $50 lower than it was back in August. And natural gas is trading almost $1.00 down," Collum said. "The one thing that keeps production propped up even as rigs have already been dropping is the deferred inventory that's already been built up from pad drilling over the last year or so," the drilled but uncompleted wells. "It takes a while before that price actually impacts the production. Right now we're seeing about a five- to six-month lag between a price impact and a rig impact, and then another two- to three-month, sometimes four or five, before you a actually see a production impact on that."
Genscape now is calling for gas production overall to peak in the May-June period. "That's about a seven-month lag between when the rig count peaked at the end of October 2014. For this summer, we're thinking we’re going to see around a 3.3 Bcf/d year-on-year increase. It's slightly lower than last year’s summer increase, where we had a 4.3 Bcf/d year-on-year increase."
Northeast production has begun to slow. Genscape said gas output there has increased only around 1% since December. "At this point last year, we were closer to 9% or so," Collum said. "More than likely, we think production will peak this summer” as the inventory that had built up is eaten away.
A potential risk to the downside are more drilled wells being deferred. Based on exploration and production company announcements to date, Genscape is estimating that there are around 720 wells that have been drilled but intentionally deferred "because of the contango in the oil market."
Collum compared the rig declines today to the last downturn during the 2008-2009 financial crisis. October 2008 was the peak rig count previously and last October was the peak rig count today.
“They followed quite the same path" on a percentage basis. "Last time around, prices rebounded so quickly that the rig count decline ended at about 30 weeks. Now we think the rig count declines will end about 41 weeks, so another two-and-a-half to three months" from the last time this occurred in 2008-2009.
If prices were to rebound over the next month or two, the rig decline could possibly reverse, Collum said. For the week ending March 20, two rigs were added, the first gain in a period of 13-14 weeks when on average almost 75 rigs per week were laid down.
"Lower 48 production has not hit a peak in the last several years. It just keeps going up and up. The reason we think this time is different is because even though in 2009 when Texas production started declining, we had the Haynesville Shale for the next three years pick up the pace. In 2008, operators were spending crazy amounts of money on leases, $30,000 an acre. They were signing two to three year leases, typically they were five to 10 year-type leases, so the producers had to go out and drill those wells. They ramped up production quite a bit because they had to hold those leases by production."
The Haynesville "kept Lower 48 production propped up through 2011, when the Marcellus and Utica were discovered," and since then, gas production has never slowed.
"Now we're starting to see Marcellus and Utica rigs drop," down by more than 30 rigs year/year. "Recently, we started to see some processing plants get delayed in several regions across the country. We think all of this is adding up. There's no new production sources that are going to help keep production from declining at some point this year, we believe."
Even if Appalachia gas output surprises to the upside, "next year we are calling for about 1.5 Bcf/d of production declines. We think absolute upside will still decline about 400-500 MMcf/d year-on-year. There's a high probability we'll have declines next year."
The recording and slides from the Genscape webinar are available here.
Correction: Genscape Inc. is projecting that total natural gas demand for the Lower 48 this summer is estimated at about 64 Bcf/d, which is higher than last year. That's about 2.4 Bcf/d more than that demand averaged in 2014, said senior natural gas analyst Rick Margolin. And it's 1 Bcf/d more than in the "remarkably high" demand year 2012. NGI regrets the errors in the original piece, and they have been corrected.