The Oil Casino: SEC Heading for Monte Carlo, Part I 27 comments
an article to
-
Font Size:
-
Print
- TweetThis
This is a long article on the subject of oil and gas reserves and due diligence.
My purpose is to alert you to the revision of SEC Regulation S-K and Regulation S-X effective January 1, 2010. Concealed in a handful of benign new regs is a financial truck bomb that's going to blow away "proved reserves" as a meaningful metric of oil company assets.
Under the new SEC rules you don't have to drill a well and actually produce oil. An operator can establish levels of lowest known hydrocarbons and highest known oil through "reliable technology" other than well penetrations. It doesn't have to be 90% reliable or widely accepted by industry peers. It can be AVO bright spots, or a fuzzy patch of seismic that could conceivably be a mud volcano, or the ridiculous Russian hokum of "passive" hydrocarbon indicators. You don't even have to explain exactly what your technology does, if it's proprietary and trade secret.
We [the SEC] proposed to define the term ‘‘reliable technology,’’ expressed in probabilistic terms, as technology that has been proven empirically to lead to correct conclusions in 90% or more of its applications. Several commenters expressed concern that this proposed 90% threshold would be difficult to verify and support on an ongoing basis. We agree that a bright line test would be difficult to apply to a particular technology or mix of technologies to determine their reliability. Therefore, we are not adopting the 90% threshold as part of the definition... The proposal also would have required reliable technology to be ‘‘widely accepted.’’ However, some commenters were concerned that this requirement would exclude proprietary technologies that companies develop internally that have proven to be reliable. We concur with these commenters and have removed the ‘‘widely accepted’’ requirement from the final rule.” [Federal Register Vol. 74, No. 9, p 2166]
Who were the commenters in favor of playing dueces wild? Basically everybody. Oil companies, professional groups like SPE and AAPG, consultants, academics, Wall Street speculators and Bush Administration lawyers.
Why? -- because the Shell (RDS.A) reserves fraud made them duck and cover.
Two Plus Two Equals Seven
Under the old SEC rules, "proved reserves" were quantities that geological and engineering information indicated with reasonable certainty could be commercially recovered from known reservoirs. In 1998, the proved reserves outlook was scary. World reserves of conventional crude declined eight percent, from 1.26 trillion in 1997 to 1.07 trillion barrels in 1998.
It was a powerful incentive to start fibbing.
Less than three years after taking over the chairmanship of the Royal Dutch/Shell Group, the world's third-largest oil and gas company, Sir Philip Watts was swept out of office in early 2004 by revelations that the company had overstated its proved reserves by nearly 25 percent. Ousted with Watts was Walter van de Vijver, head of Shell's exploration and production operations. A third top Shell executive, chief financial officer Judy Boynton, was forced from office shortly after the departure of Watts and van de Vijver. It was alleged that Watts, van de Vijver, and Boynton had been aware of the reserves shortfall since early 2002 and had conspired to keep the problem a secret from investors. [Answers.com]
Matt Simmons tells an interesting tale about discovery of Ormen Lange gas field in Norway and how partners booked vastly different proved reserves based on data from two wildcat wells:
In 2003, the five owners of Ormen Lange all decided that they had enough knowledge to go ahead and begin booking that as proved reserves... BP, who had a 10.3% share, booked 83.7% of their reserves. Norsk Hydro, who had an 18% share, booked 79.6% of their reserves. Shell booked 64.3%. Exxon Mobil booked 32.9% and Statoil booked 25%.
Q: No eyebrows were raised?No, they couldn’t have been raised because no one knew this data. This was all comingled within the total reserves companies report. Companies don’t break out field by field. Well, then after Shell shocked the world with their 20% reserve reclassification, the board gets nervous and asks Ryder Scott to come in and do an audit. And two months after the big reserve write-down, they do an embarrassing write-down #2. Write-down #2 I think involved a couple of other areas, but the big area was Ormen Lange. They now only report 22.6% versus 64.3%. And also, guess why Statoil booked 25% and their neighbor Norsk Hydro booked 80%? Statoil had DeGolyer McNaughton as their third party engineering report and they reminded them that to be technically correct, the maximum you could book was 25%. [Global Public Media]
First gas at Ormen Lange came in July 2007. Everything went swell until April 2009, when production of those "proved reserves" didn't happen as booked:
Ormen Lange gas field's reserves in the Norwegian Sea could be less than previously estimated. Disappointing drilling results in the north of the field made it possible that official reserve estimates of 382 billion cubic meters of recoverable gas could be 100 billion cubic meters too high. [Offshore247.com]
When an entire industry goes haywire, like the "toxic asset" meltdown that almost killed commercial banking and made rating agencies look like sell-side hookers, investors want to know: What the hell went so horribly wrong?
The answer is straightforward. Statistical rocket science wrongly predicted a AAA happy meal from diseased meat, lean tails, and worthless guarantees. Something similar is happening now (very quietly, almost stealthily) in the energy sector.
Math Wiz: If you throw 21 darts at Africa, you have a 95% chance of finding oil.Q: Regardless of basins or known prospectivity?Math Wiz: It has nothing to do with geology. It's a statistical fact. If you drill 21 wells anywhere in Africa, there's a 95% probability of making a discovery.Q: What if your first three wells are dry holes?Math Wiz: That increases the odds of finding oil in the next well. After a dozen or so, it becomes almost 99% probable that the next random location will be highly successful, enough to justify the entire drilling program.
When the Bush Administration SEC succumbed to regulatory capture by Big Oil and discarded the proof in "proved reserves," they opened the door to reporting P1, P2, and P3 capital assets. These are not geological terms. They are probabilities derived from an aptly-named statistical software engine called Monte Carlo.
To cook a Monte Carlo black box project, you input whatever seismic data and well logs you happen to have, ask a petrophysicist to guess a range of porosities and permeabilities, push a button in Petrel, and get a snappy false-color reservoir model suitable for PowerPoint presentation to investors.
The Petrel polygons and a range of possible hydrocarbons (low case, high case) are ported via Excel to Monte Carlo, which runs a couple thousand iterations of probability and spews out P1 (90%) P2 (50%) P3 (10%). Investment decisions are usually made on P2 "more probable than not" estimated reserves. The process is seldom if ever led by rigorous science. Monte Carlo can't process fault picks or coarsening-and-fining facies transitions. Well data is crunched by computer, and assumptions are plugged in concerning production. Reservoir engineers ponder flow assurance and water flooding to maintain pressure. More assumptions (low case, high case). Make a guesstimate of WTI five years from now -- and presto!
None of this has any relation to the job of oil and gas production. It's a smooth curve of rocket science black box probability. Most of the input values are type averages.
A research report from analysts at Credit Suisse in New York stated that U.S. E&P companies are booking “new era,” lower-quality proved undeveloped [PUD] reserves for year-end 2006... “PUD locations are being booked simply in conjunction with ‘on-trend’ acreage acquisitions, often with no wells having yet been drilled and without specific capital budget allocations,” the report stated. “This acreage-driven booking process poses obvious questions regarding reserve quality.” The analysts noted that rising PUD ratios have been a familiar trend over the past eight years and are poised to increase an average of 35 percent, up from 30 percent the prior year and 23 percent in 1995. [Ryder Scott Reservoir Solutions, Vol 10 No 1, Mar-May 2007]
"Proved undeveloped" acquisitions, huh? Let's look at a notorious reserves player, who booked proved undeveloped reserves of 300 Bcfe plus 9,800 P3 undeveloped acres, 1400 square miles of 3-D seismic, and 12 producing fields in 1998.
If we are unable to obtain further concessions from our lenders and creditors, we would continue to be in default ... and would be subject to the exercise of remedies by our lenders and creditors on account of such defaults. The exercise of such remedies could result in the Company seeking protection under federal bankruptcy laws. [TMR 8-K, 9/10/09]
In 2006, TMR founder CEO and Chairman Joe Reeves was paid $1.855 million in salary and bonus -- earning more than the CEOs at Halliburton, FMC, El Paso and Smith International. He controls two million shares of TMR, still sits on the board of directors, and Reuters seems to think his fiscal year compensation was $6.69 million recently. TMR proved reserves have been cut to 80Bcfe.
Our shop is one of over 100 independent "competent person" outfits that certifies reserves. We don't use PRMS or Monte Carlo. Instead, we make subsurface maps and pick drilling locations. We sign off on reserves volumes if there's enough well data, core analysis, petrophysics, geochemistry, paleontology, and production to justify it. Period. We don't like sell-side. We work for exploration managers.
Typical bedtime reading:
The deposition model starts in the deepest water zone with dark coloured, crinoidal beds and abundant carbonate mud and argillaceous material (mudstone in the deepest zone facies F1 and wackestone F2). The biostromal zone is mainly built by stromatopoids. [Sedimentary Geology 214]
On a recent assignment, our client griped about his in-house team, saying "There's been too much button-pushing and not enough critical thinking." The new SEC rules are a joke and professional explorationists know it. Very few will speak out about it, because there was a wave of computer automation and black box software to cover up and gloss over the shortage of experienced oil & gas geoscientists.
The industry has been shedding geologists and engineers for the past two decades. “As an aging generation of workers retires, industry experts say the resulting shortfall in skilled labor could lead to an increase in delays and problems on mega oil and gas projects… Over the next decade, a wave of retirements will strip the industry of its most skilled project managers, just as some of the most complex operations ever attempted are supposed to come on stream.” [Rigzone]
For three decades, oil downsized its people and rarely hired. Focus on staying “lean and mean” trumped any concern that we missed three decades of new employees. [Matt Simmons]
The latest quarterly report from Hays indicates exploration professionals, in particular geoscientists and senior reservoir engineers are amongst the most sought after in Australia. [E&P]
I've seen it too many times. Promoters shop around and there are plenty of sell-side whores who exaggerate blue sky hand-waving on skimpy data.
Related Articles
|
-
The Seeking Alpha 100
See allThe top 100 stock
market authors
selected for publication in the last week
Seeking Alpha 100 » -
Fastest Climbers
Those with the largestSee all
increase in followers
in the last week
Fastest Climbers » -
Top Commenters
Top commenters,
as ranked by their peers-
1
User 353732
-
2
Moon Kil Woong
-
3
doubleguns
-
4
optionsgirl
-
5
Mayascribe
Top Commenters » -
1
-
Top Instabloggers
The 100 most activeSee all
Instabloggers.
Top Instabloggers » -
Top StockTalkers
TopSee all
StockTalkers.
Top StockTalkers »
There has been a revolution in Natural gas extraction technology and this has made the old conversion ratio artificially extremely favourable for companies with proven natural gas reserves. This artificial advantage is compounded greatly by the capacity that your article well describes to count shale extensive gas deposits as “proven” on the basis of what are preliminary tests.
Have I got this wrong?
> ... is the general industry going to follow the SEC definitions
> for 'proved reserves' or are we going to have two standards floating
> around?
The answer is probably three or four standards. Big integrated majors will stick to "proved" in the traditional sense of deterministic proof and commercial opportunity. Unconventional niche producers (tar sands, shale) will be tempted to restate larger reserves, egged on by investment bankers. Australia has extremely loose rules and IPOs or secondary offerings should be scrutinized closely.
What has happened occurred slowly, over the course of a decade, and irreversibly colored the exploration business. Investors should rate oil & gas firms by cashflow, not "forward-looking" models.
at the time no one would have questioned that?
believe the new hokum & i have a proven reserve of bridges i would like to sell you.
> jack
The way I understand the efficacy of the Monte Carlo method is that it makes no predictions about the future other than it will unfold in a random way.
While there is usually a model and model parameters, the nice thing about Monte Carlo is that the model itself does not have to do any predicting.
But, it is not quite random, and here is the rub: You DO have to make assumptions about the volatility, and, in some cases, the "drift".
After all, you are trying to replicate what you think the real world looks like, and natural occurring things usually follow some type of distribution.
If assume a parameter that is wrong, or you pick the wrong distribution, you get bad results.
But, you can deal with a lot of this by including what Graham calls "a margin of error" e.g. you might assume a lot higher volatility.
Since the method of analysis can be cloaked as "proprietary", there's no reason to debate. A prudent investor should just assume it's false unless proven otherwise.
On Nov 21 09:33 AM CaptainJJack wrote:
> I use Monte Carlo methods in testing my portfolio allocations, and
> I see nothing wrong in using them here, per se.
>
> The way I understand the efficacy of the Monte Carlo method is that
> it makes no predictions about the future other than it will unfold
> in a random way.
>
> While there is usually a model and model parameters, the nice thing
> about Monte Carlo is that the model itself does not have to do any
> predicting.
>
> But, it is not quite random, and here is the rub: You DO have to
> make assumptions about the volatility, and, in some cases, the "drift".
>
>
> After all, you are trying to replicate what you think the real world
> looks like, and natural occurring things usually follow some type
> of distribution.
>
> If assume a parameter that is wrong, or you pick the wrong distribution,
> you get bad results.
>
> But, you can deal with a lot of this by including what Graham calls
> "a margin of error" e.g. you might assume a lot higher volatility.
"Statistical rocket science wrongly predicted a AAA happy meal from diseased meat, lean tails, and worthless guarantees."
While no one wrote an analysis like Alan's warning investors in advance about potential or perceived problems with data rules for asset-backed securities, the good news is that this is now a topic for open discussion in the oil and gas sector. Without commenting on the merits of Alan's analysis, I'd note that the full text of the rule and its justification are available here: sec.gov/rules/final/20...
In the case of ABS, the rules allowed disclosure to be satisfied with ASCII data that was largely impossible for most investors to analyze (except for the short sellers who did). Even ratings agencies had problems using unstructured data from "tapes," meaning the pay-to-rate incentives overwhelmed incentives to get the analysis right.
A careful reading of the oil and gas rule and critical analysis should help investors arrive at their own balance among cash flow, various types of claimed reserves, and other information available to them to use to evaluate oil and gas companies. I would merely note Section 8 of the rule, which suggests a path toward a global standard to provide data in an industry standard format:
~~~
"VIII. Application of Interactive Data Format to Oil and Gas Disclosures
"In the Proposing Release, we sought comment on the desirability of rules that would permit, or require, oil and gas companies to present the tabular disclosures in Subpart 1200 in interactive data format in addition to the currently required format. Most commenters addressing the topic supported the use of XBRL for oil and gas disclosures.314 They believed using interactive data would be very helpful to investors and analysts.315
"However, they also recommended that the Commission wait until a well developed taxonomy exists.316 Some recommended that the Commission implement it in stages, initially with a voluntary program.317 One commenter recommended that the SEC work with other groups like SPE, IASB, and the United Nations to ensure tags ultimately become the industry standard.318
"We agree that much of the disclosures regarding oil and gas companies would be conducive to interactive data. We intend to continue to work on developing a taxonomy for such disclosure. Once a well-developed taxonomy is created, we will address this issue further. We are not, however, adopting interactive data requirements in this release. We will continue to consider whether to require interactive oil and gas disclosure filings in the future and, if so, when such filings should be required based on the development status of an oil and gas disclosure taxonomy."
~~~
The ABS industry wasn't too excited about creating a taxonomy for ABS, and only some participants now realize that an investor-friendly taxonomy would help restore the ABS market to its production possibility frontier (and keep future bubbles from expanding quite as far past the ABS PPF). Moreover, a taxonomy alone is not enough. To prevent GIGO, any taxonomy must be created in a way that allows for objective evaluation of the quality of its application, whether by auditors, a self-regulatory organization, effective deterrents to the use of bad data, sufficient openness to allow third party testing, other methods, or some combination thereof. Fortunately, it's easier, and in the long run less expensive, to audit machines and business processes than it is to audit human judgment. One of XBRL's key features is its potential integration with validation rules. That's not to say character isn't critical -- tone from the top can be a vital investment criteria -- but it is to say that the part of President Reagan's favorite Russian proverb after the conjunction is pretty important too: Trust but verify.
(There's a comment letter on the topic of XBRL for oil and gas reserves at www.sec.gov/comments/s... -- worthwhile reading.)
On Nov 20 07:03 PM Alan von Altendorf wrote:
> nerfer asked:
Mexicans has been cheating about reserves for long trying to create the idea than "reserves are bigger than reported feeling" but it seems now it was quite the opposite.
Rgds
On Nov 20 06:27 PM William M. Wright wrote:
> Petroleos Mexicanos, the state-owned oil company, said proven reserves
> of crude-oil equivalent may drop for an 11th straight year as existing
> fields dry up and discoveries are smaller than expected. Looks like
> Mexico needs to hire a GS Wall Street advisor to build them a new
> Black-Box Model.