I have been sitting on this post for while pending the publication of an article by Mellisa Davis (of The Street Sweeper). She has now gone public
quoting me - so here is something I have been working on.
High growth subprime companies - a brief stylized history
Until the crisis one of the staples of my life was financial institutions which under provided for inevitable losses and grew really fast. Examples include Metris and Americredit - both of which collapsed, partially recovered and were eventually taken over.
Before the collapse they had less-than-prime lending businesses in credit cards and auto loans respectively. Metris was the more pure example. Another example was the credit card business of Circuit City.
These companies understated losses. That made them seem profitable. That profitability is only temporary because eventually it becomes obvious to even the most stubborn and blinkered management that the loans are not going to pay. When that happens a charge is inevitable - and future profitability winds up at a lower level (more akin to economic reality).
However the companies in question deferred the day of reckoning. They did this a couple of ways. First when someone could not pay their loan they tended to extend them more credit. Metris’s average balance outstanding went above $4000 (or more than double average balances at similar companies). That meant that when the losses (inevitably) came they were bigger.
The second way that they deferred the day of reckoning was just to grow really really fast. After all, if you don’t recognize any losses on new loans, then filling your portfolio with new loans meant your aggregate credit looked OK, even if the old loans were toxic. The denial solution de jour was hypergrowth. The hypergrowth hid problems and also (incidentally) drove stocks to the moon giving management ample opportunity to cash out. In other words the path that made management rich was to make the problem bigger and bigger.
Sometimes before they blew up (and they inevitably blew up) you would get a signal: a quarter with an unexpected “reserve adjustment”. Reserve adjustment being a (belated) admission that the reserves were not adequate. Sometimes there was no signal except insider selling. Mostly even the insider selling was not a good signal because the insiders were always selling.
Lesson from this
There is a lesson I drew from this. Be very wary of fast-growing hyper profitable companies (especially companies in competitive industries) where the earnings are critically dependent on a reserve or variable that has to be estimated and on which the estimate is really a guess. This could be lending or insurance or large contract construction with contract warranties or in this case an oil company. The company in question is Northern Oil and Gas
(Northern): a member of the S&P mid-cap index.
The Bakken shale oil plays
Northern Oil and Gas (and its little sister company Voyager Oil
) own acreage and part shares in oil wells in the Bakken Shale
. The Bakken Shale is one of the hotter properties in North America and is the subject of much promotion including several (sometimes anonymous) stock promotion blogs dedicated entirely to investing in the Bakken (see here
, and here
). Some have go-go names like “Million Dollar Way
” appealing to relatively unsophisticated investors.
That said - the Bakken is a real oil field and it has real players and is producing a lot of oil.
The Bakken is a large (200 thousand square miles) mid depth (typically about 10 thousand feet) and not very wide (typically about 40 feet wide) oil bearing shale deposit. The area overlaps two American States (North Dakota and Montana) and one Canadian Province (Saskatchewan).
The field has been known about for many years because traditional oil fields underlie the Bakken and as the wells have been drilled, small quantities of light sweet crude have been logged. However the technology to extract oil in quantity from the Bakken is relatively recent and involves drilling down 10 thousand feet, kicking a horizontal well down 5-8 thousand feet, putting explosives down the well to crack the rock and then chemicals and water at very high pressure to extensively fracture the rock. And then the oil flows.
This has driven North Dakota oil production extremely well:
If you extrapolate graphs of production (something I don’t think you should do) then North Dakota overtakes Alaska in 2017 to be the major US oil producing state.
The problem with Bakken shale oil
The problem with Bakken shale oil is that the pores in the rock are tight and the rock needs to be fractured to extract it. It is a stylized fact of oil production that the tighter the rock the faster the oil well declines to a trickle - especially after the reservoir has been forcedly fractured.
This makes sense. If the rock is homogenous, quite porous and extends over a large area you would expect an oil well to flow for a very long time (multiple decades). However if the rock flows only a small amount of oil without fracturing (the small amount being the amount near the well bore) then when you fracture it it will flow many small amounts (the small amounts being amounts near the fracture system). And when you have exhausted the oil that happens to be near the fracture system then the flow should flow to a trickle as the rock is not very conducive to flow.
The risk with the Bakken is that the wells decline much faster than anticipated when the well is drilled and faster than anticipated in the accounts.
If this happens the company will have to take a “depletion adjustment” - the analog of my subprime company taking a “reserve adjustment”.
Now, also as an analog of the subprime company if the company, is drilling a whole lot of wells that are declining faster than anticipated, it can hide this through growth. After all, new wells don’t just stop flowing: they need to be old wells before that happens. And you can hide an aggregate decline problem by drilling like crazy. It just winds up being a bigger decline problem when you stop drilling.
And now you see my subprime oil company: it is Northern Oil and Gas and it is a doozy. Market cap is over $1.7 billion and the stock is up from 3 and a bit dollars to almost $30 since the beginning of 2009. It is - through partners - drilling like crazy.
It is a strange company, despite the large market cap it only has 11 staff. It takes minority shares in other companies oil wells. It does not drill anything itself. Every well seems to strike oil (that is what it is like in the Bakken) but because they do not operate the wells they only have limited insight into the decline rates. Shareholders have even less insight into the decline rates.
Searching for a benchmark for Bakken decline rates
I spent a lot of time trying to find a benchmark for oil-well decline rates in the Bakken. It's hard because people tend to keep this information confidential especially if they are bidding for local acreage. Also the technology to fracture Bakken wells is relatively new and so to some extent decline rates (especially over the out years) are just unknown. Still, I have heard good stories about individual wells that are flowing well after three years. I have heard horror stories about wells that are flowing at less than a barrel per day after one year. The average almost certainly lies in this range.
I was getting hung up on a benchmark when one of my favorite bloggers Peter Sacha stepped in with a really useful post
. In it he picked apart the accounts of Petrobakken. Petrobakken is the major player in the Saskatchewan Bakken. What is more he had a decline curve.
Its only a decline curve for the first year - of production - and it is for a particularly big well (a seven stage frac). However, the initial flow rate was almost 250 barrels of oil per day and after 12 months this had declined to 75 barrels per day. It is a steep decline.
And it shows in Petrobakken’s accounts. Peter Sacha deadpans that the company likes to report “cash flow” which does not include the depletion allowance for wells. He then notes the company spent $812 million in capex most of which was to replace depleting wells. The cash flows are enormous - but the capital expenditure needed to maintain those cash flows are enormous - and that is because the depletion is rapid.
The key analytical assumption of this blog post is that we can compare decline rates for Petrobakken and Northern Oil. Both Petrobakken and Northern Oil are Bakken shale producers. Petrobakken is just much bigger and a little older and a little more experienced with decline rates. But both companies have a large spread of wells in similar areas so the average decline rate (relative to current production) should be similar.
Here are the accounts for the first nine months. (I don’t give the annual accounts because Northern Oil had an irregular depletion charge in the last quarter.)
For Petrobakken (and sorry you will need to click)
Note that over 9 months Petrobakken had 750 million in oil and gas sales before royalties and 654 million in net revenue. The depletion charge was 391 million. Depletion is 52 percent of gross revenue and almost 60 percent of net revenue. These wells deplete badly and the depletion charges are large.
Here is Northern Oil for the first nine months (and sorry again you will need to click).
Note that for Northern the gross revenue for the first nine months is 35.6 million (a very small fraction of Petrobakken). The depletion charge plus amortisation is 8.36 million.
Depletion and depreciation is only 23.5 percent of gross revenue.
Same field. Same geology. Less than half the depletion rate.
Same geology, same field suggests that the right approach to compare these companies is to assume the same depletion rate for both companies. Of course which depletion rate - Northern or Petrobakken?
We could assume that Northern is right and these fields could actually last a lot longer than Petrobakken thinks. In which case however you should buy Petrobakken as it will generate cash flow for years and will have already expensed the associated costs. If Northern is right don’t buy Northern - buy the company which is really earning far more than it says.
Altenatively Petrobakken is right and the right depletion rate is 52 percent of gross revenue. In that case for nine months the depletion plus depreciation at Northern should not equal 8.36 million it should equal that times 52/23.5 or 18.5 million. Instead of having income from operations of 14.4 million for the nine months as per stated you have income from operations of only 4.3 million. Income after tax is just over 3 million. That is for nine months - but hey - annualize it!
Oh, the market cap is over $1.7 billion.
If I do the depreciation as per Petrobakken (and I see no reason why I should not) then this stock is around 400 times earnings.
Obviously enough we are short.
It all swings on accounting for depletion - so who certifies the accounts?
All of this swings on accounting for depletion. If you use Northern’s accounting then this is a high growth high PE stock.
If you use Petrobakken’s accounting this is a lower growth stock with much higher PE and less prospects.
At that point I ask who is certifying the accounts?
Well the auditor is Mantyla McReynolds
. Have you never heard of them? Nor had I and I am a connoisseur of obscure audit firms.
First I checked whether they were a local firm. No such luck. They are based in Salt Lake City and Northern Oil is based in Wayzata Minnesota. That is 1254 miles away according to Google maps. They went out of their way to find this auditor.
So who does Mantyla McReynolds audit? I went through the SEC database
to see who else is audited by them and this is what I found:
You get the idea. We are critically dependent on an estimated accounting expense (depletion) that is low compared to the competition (the far more established Petrobakken) . Further, these estimates are certified by an auditor most associated with penny stocks. Moreover, that auditor was chosen despite being over a thousand miles away and having no obvious expertise in oil and gas.
And despite all this advantage the company took a small reserve adjustment in the fourth quarter. (Isn't a depletion charge the first warning? My guess - some well actually ran dry!)
You can imagine that all this made me uncomfortable with the stock.
Alas Melissa Davis (at the Street Sweeper) has found far more about the stock than me. And none of it raises my comfort level.
What the Street Sweeper found
Remember my issue here is the accounting for depletion. The whole valuation, indeed the whole story swings on the depletion numbers, and Petrobakken gives you a good reason to doubt the depletion numbers.
I recommend you read the Street Sweeper piece
. However, given that I am obsessed by the accounting, I thought you should focus on the CFO. Here is what the Street Sweeper says:
This guy looks like the least experienced guy ever to be appointed to all the finance positions at a nearly $2 billion oil and gas company.
I think I can stay worried about the depletion numbers. Short seems a reasonable bet to me.
PS. The Street Sweeper suggests there is a Part II. I look forward with anticipation as they found plenty of stuff I did not know about this stock.