Tuesday, September 14, 2010

Petrohawk's Incorrect Rig Count Data

A couple of weeks ago, I read a presentation Petrohawk Energy delivered at the Enercom conference in Denver.  In it, the company claimed that the Haynesville Shale rig count was declining (below: Petrohawk's slide (page 6), my annotation).

I don't see it that way. I thought about commenting at the time but let it go.  I see today that the same chart showed up in Allen Brooks's influential "Musings from the Oil Patch" semi-monthly newsletter, so I feel the need to speak up.

Rig count data is not an exact science.  There are more than a thousand rigs out there and they are moving around all the time.  There is an aspect of corporate chess involved too.  But the reality is that there are X number of rigs out there and Y number of rigs working.  It's black and white, but the data isn't easy to gather.  I've heard from some service companies that Baker Hughes rig count numbers (upon which I depend for my analysis) can run low.  For instance, rigs aren't counted if they are on the move from one place to another.  Looking at the data myself, I also see that there is sometimes a frustrating multi-week lag period. Definitely not an exact science  But I found that my numbers are much higher than Petrohawk's (mine is the red line) for the time period the company shows.

So what gives?  Each week I review the Baker Hughes data for every single well in the region (around 200 right now), and it is pretty clear what is and is not a Haynesville/Bossier well.  Why are my counts on average 49 rigs higher than Petrohawk's?  There are other companies collecting rig data, but I doubt the numbers are this far off.

Producers are definitely in a bind right now.  Because of low commodity prices, their annual capital expenditures are higher than operating cash flows.  That condition is bearable in the short term where a company can generate capital by borrowing and selling equity to generate cash, but it is unsustainable in the long run.  This is becoming a theme among investment analysts (Allen Brooks was making this same point when he used the chart).  Interestingly, I read a chilling comparison regarding the use of (and destruction of) capital between natural gas E&P companies and the airline industry in the Wall Street Journal yesterday.  Not exactly apples to apples, but it's never a good day when you are compared to the airline industry.  

The ball clearly is in the court of producers like Petrohawk to sell their stories.  The best way to get some traction is to show that gas prices will be going up, either through increased demand or decreasing supply/production.  Since rig counts are viewed as a leading indicator of production (at least in a pre-shale world), selling a decline in Haynesville rigs targeting dry gas is a compelling story.  I just don't think the data supports it.  

In the Haynesville Shale, rig counts are pretty much sticking to the plateau around 175 to 180 rigs.  I have seen some weakness lately, but we are only talking about a small handful of rigs.  You can bet that as soon as that figure starts trending downward, I will let you know.  The chart below shows a running count of Haynesville rigs since the beginning of 2010 when I started counting rig-by-rig.  I'll put it on the main page and update it weekly.


Anonymous said...

The Petrohawk numbers are not correct as the values for 5/20 thru 7/8 are overstated and the 8/12 value is too low.

The Petrohawk numbers are for Louisiana only.

Les B

Robert Hutchinson said...

A reader pointed out to me that Petrohawk's "rig count" is actually the LA DNR's "drilling in progress" category.

While this may be true, it is very misleading for HK to present the information this way. HK is too big and smart of a company to make an error like this. It is either sloppy or misleading - I can't say which. But if I can figure it out, I'm sure they can too.

Anonymous said...

First off, let me say that your website here, dedicated to the Haynesville, is by far my favorite. I check it almost daily. Second, a big question floating around the industry right now is when will the Haynesville rig count drop. It is inevitable, but WHEN. I'm compiling research from many sources, including this one (mainly this one), and it would greatly help if you could post a Haynesville Rig Count (Excel Sheet or Graph) dating back as far as you have data. Please let me know what you think about all of this. Thanks again for running such a great and useful site! -JD

Robert Hutchinson said...

JD - email me at haynesvilleplay (at) gmail.com to discuss.