The Oil Drum | DrumBeat: August 20, 2008

196 comments on DrumBeat: August 20, 2008

Comments can no longer be added to this story.

Show without comments | PDF version

Gail the Actuary on August 20, 2008 - 9:27am

A couple of days ago, DownSouth brought up the question of a discrepancy between EIA natural gas data for Texas and the corresponding data from the Texas Railroad Commission.

I contacted both the EIA and The Texas Railroad Commission. I also looked at the data myself. My conclusion is that the EIA data is right. The Texas Railroad Commission data, according to both the EIA and RCC, is at this point substantially delayed in processing. The part that is especially missing is production on the last six months for Barnett Shale. There can be production missing for a year or more, however.

According to the e-mail I got from EIA,

Katie [from the Texas Railroad Commission] is correct about Texas Natural Gas production. Texas RRC has had many internal problems collecting their production data in a timely manor in recent years. They have lost staff to higher paying jobs in the oil and gas industry and, as stated below, so many new wells have been drilled in the Barnett Shale that collecting and processing that data in a timely manor has been difficult.

In the past, Texas production was the gold standard. Within 3 to 6 month you had 97-100% of the data. Their first estimate would be within 4-5% of total production. Now, the first monthly estimate may be missing as much as 20% of the production and it takes almost 12 months to reach 99% of the production total.

While the increase in Texas production for the past year seems unlikely, it is indeed happening. The production data we have collected from the operators on the EIA-914 Natural Gas Report backs our estimates. With time, as Texas gets the data processed and people realize the increases from shale are correct, they will know that EIA has made very reliable estimates.

When I graph the monthly Texas Railroad Commission data as currently reported, this is what I get:

Texas natural gas production apart from Barnett Shale has been flat for years. If one only looks at the Barnett Shale data as reported, you get the impression that the Barnett Shale production peaked and has started to decline. The fact of the matter is that there is a serious processing problem at the Texas Railroad Commission with the Barnett Shale data, because of all of the new wells. The paperwork needs to be processed before this production can be entered into the system. There is at this time a huge paperwork backlog.

Once the new Barnett Shale well data is processed, it is likely that the growth pattern in Texas will be more like that shown in the EIA data (with a 15% or 16% year on year growth) than in the Texas RRC data. Nearly all the growth is likely to be from Barnett Shale.

It was good DownSouth brought up the question. Thanks DownSouth! It got us to look into the question, and better understand what the real situation is.

Comments can no longer be added to this story.

westexas on August 20, 2008 - 10:02am

I certainly believe the part about a backlog at the RRC. We are facing incredible delays in getting new drilling permits.

In any case, I assume that the bottom line is that the EIA is just estimating the increase in shale gas production, while the RRC is trying to count the actual production, but as noted, the RRC is behind in the process. However, I would assume that the annual RRC data, which did show an increase in 2007, are fairly accurate. What the annual RRC data show is that we are producing at about two-thirds of our 1972 peak rate, but that it took four times as many wells to bring production up to two-thirds of our peak rate.

However, the shale plays are probably the best thing going for the US Oil & Gas industry, but as we have discussed, we have the "Red Queen" problem. We are replacing a smaller number of higher volume conventional wells with a much larger volume of generally lower production rate and faster declining unconventional wells. The limiting factors become equipment, personnel and infrastructure.

Comments can no longer be added to this story.

Gail the Actuary on August 20, 2008 - 11:04am

I think the EIA is probably doing a reasonable job of estimating the increase in production. The companies probably would not be reporting the higher production on the EIA-914 reports unless it was really there. When a person backs into what the Texas Railroad Commission thinks the ultimate production will really be, it comes out pretty much in line with what the EIA is now estimating. I expect that most of the increase will be in Barnett Shale. There is also a little "other shale gas" (Bossier shale and shale gas from the Toyah, NW field).

This is a link to the previous discussion.

Comments can no longer be added to this story.

Moneyman on August 20, 2008 - 10:57am

When I started in the industry in 2002, my first job as a Production Accountant involved sending in the state production reports. It was a mess back then and from what I hear, it has gotten much worse. At that time, reports were printed, then mailed (hundreds of pages per month), then keyed in manually by the state. Then we usually got questions, which had to be resolved. A reallocation of production from comingled fields could cause many revisions several months later.(although gross would remain the same) After I moved to another position, the state came up with a plan to bring the system out of the stone age and accept electronic submissions. From what I hear, it was poorly executed. All in all, the data is probably pretty good quality after six months or so. I would not be surprised to see long delays in the reporting for new leases, but new wells drilled in unitized fields should see no delay. I'm not familiar with Barnett at all, but it seems likely that new wells are on new RRC leases. Reporting for unitized fields is much simpler....
Although it probably doesn't affect the data quality as much as it does on the RRC side, the turnover in most company production accounting departments is pretty high. Good employees move on to bigger and better things, leaving new hires and whoever is left to ensure data quality.

Comments can no longer be added to this story.

DownSouth on August 20, 2008 - 11:26am

Hi Gail the Actuary,

Thanks for contacting the EIA and the Texas Railroad Commission. Your due dilligence is exemplary, and the response of these two agencies was desperately needed. However, I'm still not convinced.

I think we left it off on Monday's drumbeat with your comment:

I talked to a woman at the Texas Railroad Commission this morning. Their statistics for recent months are understated by some unknown amount because production has been growing so rapidly on Barnett Shale that they have not been able to keep up with processing of various paperwork. Until the paperwork gets into the system, the natural gas production cannot get into the system either.

Because of this, it looks to me as though at least part of the problem is with the Texas Railroad Commission data. I have sent e-mails to both the Texas RRC and EIA. I'll see how much more I can figure out.

I would like to use this comment, plus your follow-up today, as a hook on which to hang a couple more observations.

My comment the other day dealt with mass balance, how things aren't adding up between production, consumption and storage.

Today I'd like to talk about rig count vs. production, production cost vs. production, and natural gas price vs. production.

Facts: Texas rig count and production...

Year      Avg. Rig Count     Gas Prod     Gas Prod
            during Jan.         EIA       Texas RRC
                              (BCFPD)      (BCFPD)

2000            285              na         15.78
2001            429              na         15.99
2002            372              na         15.71
2003            370              na         15.95
2004            459              na         16.56
2005            548             16.26       16.58
2006            664             17.21       17.41
2007            790             18.86       18.58
2008(1st 6 mo)  858             21.01       18.16

References:

Rig count
http://files.shareholder.com/downloads/BHI/391398890x0x223315/61FEC3DE-7...

EIA Production
http://www.eia.doe.gov/oil_gas/natural_gas/data_publications/eia914/gros...

Texas Railroad Commission Production
http://webapps.rrc.state.tx.us/PDQ/home.do;jsessionid=CLGUte61dXzWjmk8kEbB7MT4CRMO48RDUC4VRg9QKk2Pe4ivmQbs!622569757

Observations: The EIA figures are asking us to believe that drilling has suddenly become much more efficient, that is that a slight uptick in rig activity is yielding huge increases in new gas production.

Also, if we take a look at this map of the distribution of rig activity across Texas...

http://gis.bakerhughesdirect.com/RigCounts/default2.aspx

we see only a handful of rigs running in the Bend Arch-Fort Worth Basin (Barnett Shale). Most Texas rigs are running in the Permian Basin, the Gulf Coast region, East Texas and South Texas. (You can drill down on the Baker Huges web site to get a closer view.) By looking at the map, maybe 10% of Texas rigs are currently drilling Barnett Shale, and certainly not more than 20%. This does not support what the lady at the Texas Railroad Commission said about "production has been growing so rapidly on Barnett Shale" being the cause of their work overload.

Facts: Production cost vs. production...

Using Chesapeake Energy as an example, direct from its own financial statements, look what is happening to costs:

                                          Investment in
Quarter          Oper. Costs           Property & Equipment
                 (per MCF)         (per MCF produced during qtr)
Q2-2003            $2.27                     $58.86
Q2-2004             2.60                      63.54
Q2-2005             3.11                     120.80 
Q2-2006             3.90                     116.52
Q2-2007             4.50                     154.00
Q2-2008             4.73                     142.71

Observatons: If indeed drilling for natural gas was becoming more efficient, that is if more gas were being produced for each foot drilled, then one would expect the production costs to be dropping. So far they are not. (By the way I am just using Chesapeake as an example, but I just as easily could have used Devon or any other major natural gas producer. Their costs are all similar.)

Facts: Natural gas prices vs. production:


                Production        Avg. Monthly
Month          EIA Bulletin        Gas Price           
                (Lower 48)        (Henry Hub) 
     
Jan 07            55.66              $6.55           
Feb 07            55.45               7.98              
Mar 07            56.70               7.10            
Apr 07            56.82               7.59  
May 07            57.06               7.63  
Jun 07            57.68               7.36
Jul 07            57.48               6.21      
Aug 07            57.91               6.23
Sep 07            58.17               6.08 
Oct 07            58.32               6.80  
Nov 07            59.69               7.14      
Dec 07            60.38               7.14
Jan 08            60.31               7.98          
Feb 08            61.25               8.55
Mar 08            61.91               9.44
Apr 08            61.62              10.13
May 08            61.82              11.21
Jun 08                               12.69
Jul 08                               11.06

Observations: I don't have a clue what determines natural gas prices, whether it is perception (hype and manipulation) or fundamentals (supply and demand). But if it is the latter, then the huge run up in prices in the first half of this year doesn’t seem to be consistent with a simultaneous balooning in supplies.

Time will certainly tell. But in the meantime all the conflicting information certainly makes for a mental and intellectual challenge and, on my part at least, for a lot of fun.

Comments can no longer be added to this story.

Iconoclast421 on August 20, 2008 - 5:13pm

Observations: The EIA figures are asking us to believe that drilling has suddenly become much more efficient, that is that a slight uptick in rig activity is yielding huge increases in new gas production.

I'm guessing that the numbers from 2007 (2007 790 18.86 18.58) were too high given the market conditions at that time. In other words, there were too many rigs in 2007, 790 when there probably should have been only 740. If you plug 740 in there in place of the 790, it makes a bit more sense.

Observatons: If indeed drilling for natural gas was becoming more efficient, that is if more gas were being produced for each foot drilled, then one would expect the production costs to be dropping. So far they are not.

Q2-2006 3.90
Q2-2007 4.50
Q2-2008 4.73

Technically, going from 4.50 to 4.73 is not a drop, but wouldn't you expect a number far greater than 4.73? It's only 5% higher at a time when costs all across the board were rising at rates much higher than that.

Comments can no longer be added to this story.

DownSouth on August 20, 2008 - 7:40pm

Yes, Iconoclast421, I agree, and I don't see a single one of the four observations I made that is ironclad, that couldn't be explained in a differnt way.

The increase in $/MCF could indeed be due more to an increase in $/foot-drilled than a drop in MCF/foot-drilled.

Likewise, one could argue the March to July run-up in gas prices had nothing to do with supply and demand, but instead was a bubble caused soley by speculation and manipulation. Many in fact do make this argument.

And if we take a look at rig activity, one could surmise that 400 rigs are needed just to maintain production flat. Anything over 400 rigs adds new production. So an increase from 500 rigs to 600 rigs doubles the number drilling for new production (600-400)/(500-400)=2 whereas in absolute numbers it is only a 20% increase. An increase from 500 to 800 quadruples the number drilling for new production: (800-400)/(500-400)=4, whereas in absolute numbers it is only a 60% increase. So a relatively small percentage uptick in overall rig activity could indeed represent a much larger percentage uptick in the number of rigs toiling at increasing gas production.

And I suppose natural gas demand really could have increased by 11% over the last 16 months. If the production figures are correct, that would have to be true, because as memmel pointed out, it hasn't gone into storage.

It's just that when one looks at the whole picture, it seems to me that the Texas Railroad Commission production figures do a lot better job of explaining all the interrelated and interdependendent phenomena than the EIA figures do.

Like I said, time will tell. But for me, the jury's still out.

Comments can no longer be added to this story.

JonFreise on August 21, 2008 - 12:16pm

It was the small rig increase versus the huge production increase that made me wonder about shut in gas. We will see the truth as the year plays out.

We do know that the number of MCF produced per foot drilled is falling from EIA data. But it is not real time data.

Canadian production is down. US increases might be being used to replace falling Canadian gas. Just brainstorming....

Comments can no longer be added to this story.

Gail the Actuary on August 20, 2008 - 5:24pm

You make some good points. I don't think the situation is as problematic as you suggest, though.

I notice from Baker Hughes data (which you also quote), there was an increase of 195 in US drilling rigs in the last year. Of this, Texas received 94, or 48% of the additional rigs. There was also a big shift into horizontal drilling rigs, and I would bet that the shift occurred in Texas as well. Most of the drilling rigs are for gas, so it doesn't seem like we should be too surprised if there was a fairly big increase in gas production in Texas. Clearly someone is interested in drilling in Texas, whether or not it is in the Barnett Shale area.

Regarding the higher price of natural gas in early 2008, at least part of this was related to the higher price of oil.

I agree that costs per MCF are going up. The business is fairly oil intensive, and oil costs have been going up. Long term one would expect the price of natural gas to continue to increase, to keep parity with the price of oil (and perhaps increase even more, since the current price is low relative to oil on a BTU basis). I would imagine this is why companies keep drilling, even with the higher costs. Once the profitability is no longer there, or they cannot get a needed input, like drilling pipes, the increase will stop.

I don't think the supply - use is as out of balance as you indicate. US net imports are way down, so that total natural gas available was up only 4.7% comparing the first five months of 2008 with the first five months of 2007. Consumption increased a little less than that--4.0%. There is some seasonality and storage plays a role, so one wouldn't expect amounts to match up completely.

If someone is looking at supply going forward, they are likely to look at the recent 8.8% increase in US dry gas production. If this percentage increase continues going forward, we are likely to be somewhat oversupplied for our current uses. This assumes that we won't have another decrease in net imports in the future. This reasoning may be part of what is keeping natural gas prices lower now.

Comments can no longer be added to this story.

DownSouth on August 20, 2008 - 7:42pm

Gail the Acturary,

I don't know what the source for the data was for your table entitled "Comparison of Changes in Supply and Use of US Natural Gas".

However, it states that US Dry Gas Production for the first 5 months of 2008 is 8,496,022 (MMCF?). If you divide that by the 152 days in the first five months of 2008, you get an average daily production rate of 55.9 BCFPD (55,900 MMCFPD).

The EIA figures, however, peg average daily production for the Lower 48 for the first five months of 2008 at 61.4 BCFPD.

So here's yet another source that is in disagreement with the EIA figures, and in the same direction as the Texas Railroad Commission figures, to the tune of 5.5 BCFPD. That's not an insignificant difference.

My background is in science and engineering, so all these mushy and inconsistent figures just drive me crazy. As Jacques Barzun wrote in From Dawn to Decadence:

The road to the present was hard and long because the old systems were good. They had consistency and completeness; only at a few points did contrary facts or gaps in explanation threaten their validity. One such fact was the odd behavior of the planets, especially Mars, which at times went backward instead of forward...

The larger picture was this: in the heavens, with Earth at the center, were several huge spheres, one within the next, each made of finer and finer stuff, and all revolving and emitting the "music of the spheres..." Sphere and circle, the two perfect figures, were essential to this perfect movement; it was unconscionable on the part of Mars that it should retrogress. Other irregularities were taken care of by old Ptolemy's epicycles, circular paths around the point where the errant body should be.

It made a very complex structure, and at last the mind rebelled at more and more contortions.

Comments can no longer be added to this story.

Gail the Actuary on August 20, 2008 - 9:57pm

If you look at the EIA data, you will discover that there are several different versions of natural gas production, because natural gas tends to be used/lost as it goes through the system:

-Gross Withdrawals
-Marketed Production
-Dry Production

I notice you are quoting data relating to the 48 state total. This adds another variation, since most EIA data tables are for the US in total. An EIA exhibit that gives most of these is here.

Regarding consumption, there is

-Total Consumption
-Natural Gas Delivered to Consumers

If you are using two different data sources (such as EIA and RRC), and are trying to match actual production amounts, you need to be very careful to match like with like. Otherwise, you are likely to come to erroneous conclusions.

In my exhibit, I was only trying to show year to year percentage changes, so matching was not an issue. What I showed was dry production from here, which is a lower amount. When I originally put the exhibit together, I had an additional line for marketed production. The percentage change in that was 8.8%, which is the same as the percentage change in dry gas production, so I took the line off the exhibit.

I think you are fighting an uphill battle trying to match natural gas numbers between sources, without doing a lot of background work to make certain that the numbers you are using are comparable. It is very easy to make a mistake. At a minimum, you want to go back to the EIA tables and start working from them. Then you at least know what number you are actually working with.

Comments can no longer be added to this story.

ROCKMAN on August 21, 2008 - 8:25am

Gail/DS,

I’ll throw in a couple of points to highlight some additional difficulties trying to reach your worthy goals. First, the TRRC is always behind a good bit in reporting every aspect of Tx production. No one’s fault…they don’t really have as many folks working there as you might guess. But, more importantly, trying to characterize the current state of shale gas development in Tx, as well as anywhere else, is a very moving target. On the question of efficiency: there has been, and continues, a very steep learning curve in completion technology. Production profiles and recovery estimates from wells drilled just 3 or 4 years would look very different than if they were done today. Back then a well might have 2 or 3 fracs pumped into it. Now 12 fracs per well is not uncommon. Not surprisingly, new frac protocols might deliver 5 times the initial rate. Also, as Gail mentioned, the shift to horizontal drilling will make for even greater changes. Just yesterday we were given the go ahead to ramp up our shale gas drilling rig count to 14 for 2009….about a 50% increase over original plans. On the other side of the ledger, these “efficiency” improvements have come with a much higher price tag. Actually diesel prices increase haven't been the big problem. It’s been steel. On 1 Jan we made a projection on casing cost increase by year end. We hit that number 2 months ago. I myself don’t see much of the internal economic data but it’s easy to guess that it’s very fluid. And fracs are essentially priced by the pounds of sand pumped down the hole. A lot more sand in 12 fracs then 3 fracs.

There is a cost/efficiency number out there for wells drilled to day which is quit a bit different than one drilled 3 or 4 years ago. And a well drilled 3 or 4 years from today will probably look a lot different than one drilled today. I wish you the best of luck in figuring out this elusive target.

Comments can no longer be added to this story.

Gail the Actuary on August 21, 2008 - 10:34am

Thanks! Your comment is very helpful.

I think we often forget about improvements in technology as a driver of production increases. When I visited BP's tight gas facility in Wamsutter, one thing I commented on was how much technology had changed in recent years. It would not be surprising if similar changes are taking place on Barnett Shale as well.

Regarding the higher price of steel, I expect there is at least a little tie back to the higher price of oil. When the price of oil rose, so did the price of coal. One of the big costs of coal is transporting it. Another is mining it. Both of these are quite oil-dependent. There was also a shortage of coal. If other fossil fuels had been available in greater quantity at lower price, this might not have been such an issue.

Comments can no longer be added to this story.

ROCKMAN on August 21, 2008 - 12:03pm

Gail,

Certainly energy cost are a factor in the steel price run up but I can't really tell to what extent. But from the position the steel makers have taken it strikes me that competition among buyers is the dominant driving force of the inflation. For quit a while now the steel makers won't even quote a price: if you order 400,000' of a particular casing size they tell you the run will be out in, let's say, 4 months. They won't give you a price for the order. They tell you the cost when the pipe is ready to ship. You can accept that price or pass. It doesn't matter to the mill...they have several buyers right behind waiting to pay the price. Over the last few months I've heard that much of the competition has come from overseas buyers willing to meet the local price plus the additional shipping cost. We're probable in the top 5 casing buyers in the US and are on a strict monthly casing allowance. I can't fault those companies for maximizing their profit. We do the same. But it does add to that complexity of determining the real economic value of a drilling program with the parameters changing so quickly combined with predicting the unpredictable (like NG prices in 2011). As I’ve mentioned before, it’s the rapid pay out of these resource plays which is driving activity. Even if the total recovery were twice the average, if payout took 3 or 4 years you wouldn’t see the drilling levels we have today. Net present value is THE controlling factor. It's the only way to mitigate rising cost and the future pricing risk factor.

I still haven’t been able to devote enough time to the resource play decline model but will keep pushing. Something DS said reminded me of an old analogy for such plays: it’s like seeing you gas gauge running low so you speed up so you can reach your destination before running out of gas. Foolish logic for sure. But imagine a different reaction: you stop on the shoulder of the road, with the engine running, so you can make your remaining fuel last longer. Your engines runs much longer but you make no progress. This silly example is exactly where I see the resource playing public companies are today. The more wells they drill to replace their rapid declining reserve base the more rapidly declining reserves they add to their portfolio. And thus they need to drill even more wells. There will come a point for every public company when they won’t be able to stick with this plan. Could be any one of several reasons: play runs out, NG too low for new drilling, too much competition from other companies splitting the pie, etc. I’m guessing this is one reason we’re seeing companies paying top dollar to tie up as much acreage as possible. Even if all the other factors support the expansion, only he with the gold (acreage) rules.

Comments can no longer be added to this story.

Bryant on August 21, 2008 - 1:17pm

You can accept that price or pass. It doesn't matter to the mill...they have several buyers right behind waiting to pay the price

We are see some of the same things in the hard rock mining business, and it is not restricted to supplies like steel and tires, but services as well; drillers, rigs, lowboy transports etc.

Comments can no longer be added to this story.

SamuM on August 20, 2008 - 3:05pm

Great work Gail! Thanks for this. An example of what we should do in a face of a data discrepancy like this.

Comments can no longer be added to this story.

JonFreise on August 21, 2008 - 12:02pm

Hi Gail,

I am late to the party, but thank you for calling and getting the official story. I think we should keep a close track on how the two databases match in the future.

I find it a bit disturbing that the EIA is claiming this 15% increase when no one actually knows how much gas has been produced. The EIA is working with surveys and models. If the RRC does not know the true production, then no one does. I need to read the EIA press release again, but it did not sound like they were claiming a "forecasted increase of 15%".

Comments can no longer be added to this story.

196 comments on DrumBeat: August 20, 2008

Comments can no longer be added to this story.

Show without comments | PDF version

“If kindness and comfort are, as I suspect, the results of an energy surplus, then, as the supply contracts, we could be expected to start fighting once again like cats in a sack.”

—George Monbiot