Natural Gas Markets Headed for A Crash?
G. Allen Brooks, Managing Partner
Parks Paton Hoepfl & Brown
Rigzone
Tuesday, August 18, 2009
Natural gas prices after rallying on surprisingly strong labor market news have retreated in recent days as the prospect of full storage suggests the industry will be forced to curtail production unless demand picks up. At the end of July, natural gas in storage was almost 3.1 trillion cubic feet (Tcf), or about 25% above the 5-year average for volumes at this time of year. Estimates of full storage capacity range from 3.7 Tcf to 4.1 Tcf. At the date of this report from the Energy Information Administration (EIA), there were 10 weeks left to the storage injection season meaning that without a strong pick up in gas demand or a collapse in production, domestic gas producers are facing the eventuality of all having to curtail their production. When that happens, we should expect a meaningful drop in natural gas prices.
This industry-wide predicament was highlighted by Aubrey McClendon, CEO of Chesapeake Energy (CHK-NYSE) on his company's earnings conference call. Mr. McClendon, the poster child for aggressive gas production management during periods of weak gas prices, announced his company was not planning to curtail production since it expected storage to max out and thus they, along with all other producers, would be forced to shut in flowing gas volumes. For the first time, Chesapeake was not about to exhibit discipline in supporting gas prices for the benefit of producers who did not curtail their production. Does this suggest that leaders of the natural gas industry are prepared to ignore production economics to demonstrate a point to their fellow producers?
We have been watching and writing about the travails of the domestic gas business as the collapse in the drilling rig count does not seem to have dented gas production as everyone assumed. Since we last opined on the gas market, the Energy Information Administration (EIA) has released its monthly gas production estimates gleaned from their Form 914 survey of operators. These surveys, reportedly providing the industry with more accurate production data, started in 2005. The only problem is that the data is still dated as the latest monthly estimated production volume figure released was for May, some 60 days old.
The May 914 gas production was 62.84 billion cubic feet per day (Bcf/d), down from the revised April monthly data showing production of 63.35 Bcf/d. Many analysts, gas producing company executives and forecasters jumped on this decline as confirmation the long-anticipated gas production decline was underway. On closer examination, however, we can't be totally sure because there have been a number of other recent months when the initial monthly gas production estimate was revised lower. The initial May production estimate now is virtually identical to the revised December 2008 estimate.
The initial gas production estimate for April was revised down, but only from 63.37 Bcf/d to 63.35 Bcf/d. The revised April production estimate was down from the March revised figure by approximately 200 million cubic feet per day (MMcf/d), but it was essentially flat with the revised February production estimate of 63.58 Bcf/d. Can we take solace in the May production estimate decline? Is the recent monthly revision pattern being reduced a sign that when the May estimate is revised it too will show even lower production?
Since January 2005, there have been 52 revisions to the initial monthly production estimate. One revision showed no change. Of the remaining revisions, 33 were higher than the initial estimate and 18 were lower. Increased estimates were made nearly two-thirds of the time. Admittedly, there were stretches when the revisions were always up, just as there were stretches when they were all lower. At the moment, we appear to be in a period marked by mostly lower revisions, but we can't find any rhyme or reason why historical patterns of revisions shifted from mostly up to down or vice a versa.
Given the data history showing such a strong bias in favor of increased monthly production estimate revisions, we remain skeptical in calling for a further reduction for May's initial estimate.
The other concern we have had about the gas production scenario is the developments in the drilling rig market. We, along with everyone else, watched with horror last fall as the domestic drilling rig business entered a freefall. We and others have wrestled with determining exactly how far down the rig count would go in this market correction and when it might bottom. More recently we have begun focusing on the pace and shape of the rig count's recovery.
One aspect of the drilling industry decline that has been of particular significance for the gas business has been the difference in the type of drilling rigs that were being laid down. This interest has gained significance by the emergence of the gas-shale plays. Data has shown that wells drilled horizontally in these gas-shales have tended to be more prolific than wells drilled vertically. The guiding principal behind the significant initial production volumes coming from gasshale wells has been the successful marriage of horizontal drilling technology with improved formation fracturing capability. Drillers have been able to rapidly drill long lateral well sections in the heart of many of the gas-rich formations. Well stimulation technology has enabled the development of multiple stage fracturing applications within the same well bore. Together these technologies have produced gas wells with initial production volumes multiples of conventionally drilled and completed gas well volumes.
We showed in our last Musings drilling and production data for Fayetteville gas shale wells derived from Southwestern Energy's (SWN-NYSE) financial reports. The data, covering a two-year period since early 2007, showed significant progress in drilling time, drilling performance and well production. The length of time required to drill the wells fell from 20 days to 12 while the lateral distance drilled increased 84% to almost 3,900 feet. At the same time, the 30-day average production rate grew from 1,006 MMcf/d to 2,373 MMcf/d.
Given the growing importance to the nation's production of natural gas from wells drilled horizontally, we examined overall gas production figures versus measures of drilling rig activity. When gas production is paired with gas-oriented drilling rigs, one sees a dramatic fall-off in rigs since last fall with barely any movement in the Form 194 monthly gas production volumes so far this year, based on the initial monthly production estimate.
On the other hand, if we match the same Form 914 gas production volumes against the number of active horizontal rigs, we also are hard pressed to see any impact from the downturn in drilling.
On the other hand, if we plot the percentage of all rigs drilling horizontally, there is a pattern of a steady increase as gas production increased and as it is holding steady now.
While we were wrestling with this data, we came across an interesting article by Arthur Berman published in World Oil and republished by the Association for the Study of Peak Oil in its latest newsletter. In the article, Mr. Berman re-analyzed well production data from the Barnett Shale, the initial stimulus for the gas-shale drilling explosion. He updated the data from the roughly 2,000 horizontal wells that he initially studied two years ago. Based on this new study of well performance, he concluded the following points: there is little correlation between the initial production rate and the well's ultimate recoverable reserves; the life of average well production is shorter than predicted; the volume of commerciallyrecoverable gas has been over-stated; core areas of the play do not provide higher recoverable reserves; recoverable reserves from horizontal wells are no greater than reserves in vertical wells; and average well performance has decreased consistently since 2003 for horizontal wells.
The key finding of the study, and a point that will hearten those who are arguing that the fall-off in gas production is imminent due to the drilling collapse, is that the overall ultimate recoverable reserves from horizontal wells decreased by 30% from Mr. Berman's earlier projection. Additionally, he found that the average ultimate recoverable reserve estimate per well fell from 1.24 Bcf to 0.84 Bcf. These findings reflect the fact that "most wells do not maintain the hyperbolic decline projection indicated from their first months or years of production." What Mr. Berman found was that wells experienced an abrupt negative departure from the hyperbolic decline as early as 12-18 months, but more likely in the fourth of fifth year of the well's production. This conclusion may temper the expectation of a significant fall-off in gas production given the drilling rig cutback.
What Mr. Berman discovered in updating his study was that the decline curves are rapid and more severe than previously thought, so the ultimate amount of reserves recovered from producing wells is less than originally thought. For those who believe in the longterm positive outlook for the oil service industry, this finding is highly supportive. On the other hand, the fact that these gas wells don't fall off their hyperbolic decline curves as quickly as some people have thought (or hoped for) may signal that the domestic land rig count might experience another downturn when natural gas prices hit the wall as gas storage capacity fills to the brim. Or possibly the rig count will experience a much longer and more gradual recovery than currently expected.
Additional findings from Mr. Berman's study point to less favorable production economics from horizontal wells. Mr. Berman focused a part of his study on the wells in the "sweet" spot of the Barnett formation, or those wells located in Tarrant and Johnson counties in and around Ft. Worth, Texas. This area comprises about 9.5 million acres. What he found was that this sweet spot did not produce appreciably higher ultimate recoverable reserves per well than the overall play. Horizontal wells only resulted in about a 31% improvement in reserves recovered compared to their 2.5-times greater cost; not a positive for profitability. This conclusion suggests that many of the claims about low costs associated with the gasshale plays may prove inaccurate. According to Mr. Berman, "If every operator in the Barnett Shale was hedged at a netback gas price of $8/Mcf, only 31% of horizontal wells would break even or make money. At $6/Mcf, only 15% of wells would reach this commercial threshold." Does this analysis suggest that many of the gas producers are deluding themselves about how successful they are progressing in developing gas shales? Will the new gas-shale plays really have better economics? Are producers who suggest they have superior acreage positions and better technology really exceptional? Maybe they are all citizens of Garrison Keillor's Lake Wobegon where everyone is "above average."
Posted by Arthur Caldicott on 27 Aug 2009