BOOK COOKING MADE EASY, OR HOW TO SHOW A PROFIT BY HIDING COSTS.
While most folks would consider interest payments an expense, the folks at Seitel treat part of their interest payments as a capitalized asset.
From: "NOTE D-OIL AND GAS PROPERTIES" in the FY 2001 Q3 10Q
<<<For the nine months ended September 30, 2001 and 2000, interest costs of $1,630,000 and $2,202,000, respectively, have been capitalized to oil and gas properties.>>>
As you can see, this adds .08-.10 to the bottom line every year.
Also from NOTE D:
<<<For the nine months ended September 30, 2001 and 2000, exploration and development related overhead costs of $1,783,000 and $1,341,000, respectively, have been capitalized to oil and gas properties.>>>
What do these costs consist of? It's explained in the filing:
<<<salaries, benefits and other internal costs>>>
Now a lot of companies would expense these costs in the current quarter rather than treat wages, etc. as an asset, but not Seitel. Why not you may ask. Because it's hard to keep the price of the stock up if you don't make the numbers and an extra .07 per share makes a difference.
These two entries alone account for over .15 every year in reported earnings even though there is no possible way those costs will ever be recovered as "assets".
There's been a fair amount of attention paid to Seitel's data swaps by the media lately in both the Wall Street Journal and Forbes, so it's important to understand how "barter" works.
If Jim owns a 73 Pinto worth $300 and Bill owns a 73 Vega worth $300 they can trade cars and call it a barter transaction. For all practical purposes Bill and Jim are still relying on undependable forms of transportation, with no noticeable change in fortune, but that isn't the case with GAAP. For GAAP purposes, Jim can report the estimated value of Bill's Vega as a capital expense for the quarter making it an asset, and report the estimated value of his Pinto as $300 in income (revenue).
An additional problem with barter transactions is the actual value has not been market tested in a cash transaction. If Bill and Jim are unusually proud of their vehicles, they may "estimate" the value of the swap at $1000 rather than $300.
In Q3 Seitel swapped an "estimated" value of $30 million in data for an "estimated" $30 million in data. The $30 million was reported as revenue and depreciated at 40% ($12 million) for a profit of $18 million, or .72 per share. In the preceding 2 quarters data with an "estimated" value of $9.4 million was reported as revenue and depreciated at 46% ($4.3 million) for a profit of $5.1 million, or .20 per share.
This brings us to the question of depreciation. Why was data depreciated at 46% in the past and all of a sudden depreciated at 40% in Q3? Other than the fact it makes an .11 per share difference in reported EPS, there doesn't seem to be any explanation whatsoever. According to public statements made by officers of Seitel, the company expects to see a 200% return on investments in seismic data. In other words, for every dollar in reported revenue, fifty cents of that amount is the actual cash cost of the data, or 50%.
Seitel uses what is called "income forecast method" for depreciating the costs of library data. In other words, the actual cash cost to acquire the data is not expensed unless a sale is made. Even if the costs are not recoverable, the data remains on the books as an "asset".
The typical transaction would be where Seitel sets up a multi client deal where 60% of the cost to acquire the data is provided by the customers to license the data and Seitel puts up the other 40% in order to own the right to resell the data to other customers. In the typical multi client situation with Seitel, if the cost to acquire the survey was $10 million, it would work like this: Seitel collects and reports $6 million in revenue on the $10 million invested. The $10 million is capitalized as an asset and 40% of the $6 million ($2.4 million) is depreciated as an expense for that quarter. In other words, $7.6 million stays on the books as an asset for up to 10 years and the only time it will see another accounting entry as an expense is at 40% of the value of any future sales. If 3 years later there is a $2 million sale, another $.8 million would be written off.
The problem is if the data doesn't sell after the initial transaction, Seitel is out the $4 million that was their share of the costs to produce the data in the first place. Another problem is that the data becomes less valuable each time it is sold. The first customer to view the data is able to exploit the best prospects, while leasing mineral rights, before anyone else could gain any benefit from buying the data. Once the best prospects have been drilled, the data becomes worthless to anyone else and has to be replaced with fresh surveys.
Debt tells the story for Seitel. If Seitel's business model actually produced real profits, it would be unnecessary to borrow money to replace the surveys no longer able to generate revenue. Library resales would cover the 40% up front cost to engage in new multi client surveys. Unfortunately for those who own stock in Seitel, this has never been the case. The company has survived only through its ability to borrow more money. |
