In this post I present what I found from applying R/P (Reserves divided by [annual] Production) ratios for Light Tight Oil (LTO) for 3 big companies in Bakken/Three Forks/Sanish.
The companies are; Continental Resources, Oasis Petroleum and Whiting Petroleum, which operated 28% of total LTO extraction in the Bakken(ND) in December 2014.
- Undertaking oil and gas reserves assessments are just as much an art as a science.
From previous work with LTO from Bakken I kept track of the R/P ratio for wells/portfolios and generally found it was in the range of 3 – 4 after their first year of flow. This suggested that 25 – 35% of the wells’ Estimated Ultimate Recovery (EUR) was extracted in their first year of flow.
This made sense as extraction (production) from LTO wells are heavily front end loaded and have steep initial declines.
Examining some big Bakken companies SEC 10-K (SEC; Securities and Exchange Commission) filings for 2014 I noticed that these had R/P ratios for Proven Developed Reserves (PDP) that ranged from 7 – 9.
(Refer to the end of this post for more detailed explanations/definitions of PDP and PUD)
That did not make sense and R/P ratios give away powerful and very valuable information about likely future extraction trajectories.
About 50% of the companies’ total LTO extraction (flow) in Dec 2014 in Bakken (ND) were from wells started in 2014. In other words, the flow was dominated by “young” wells which decline rapidly. Therefore, whatever flow data (monthly, quarterly) that was annualized it should be expected a R/P ratio for total extraction around 4 for 2014.
What I present is how PDP, extraction data and R/P data derived from the 3 companies SEC 10-K statements compares to what was derived from actual data. Further, what actual data now is projecting for EUR for the average well for these companies.
LTO in Bakken will now generally work profitably with an oil price (WTI) above $80/b.
The willingness of several companies to sell more debt (obtain more credit), assets and equity to continue to manufacture LTO wells which estimates showed were not commercially viable have had many analysts puzzled.
Something was likely overlooked, and chances are that this is related to EUR driven incentives to expand assets/equity on the companies’ balance sheets (or “book to model”).
As companies drill wells and puts these in operation (production), it allows them to book reserves on the balance sheets. And reserves are the biggest portion of the LTO companies’ balance sheets.
The rush to use credit/debt to drill what likely would become unprofitable wells (applying project economics) with a lasting, low oil price appears driven by some perverse incentive to grow booked reserves to grow assets and thus equity on the companies’ balance sheets, overriding outlooks for poor profitability. High equity on the balance sheets allows for more debt.
Looking at actual, hard well data (from NDIC; North Dakota Industrial Commission) this strategy will at some point have to face up to the realities of physics and Nature. And physics and Nature do NOT negotiate.
- Using actual data for LTO wells strongly suggests that the PDP (and thus PUD) estimates in companies’ SEC 10-K filings for 2014 are grossly inflated. If so, this has inflated the assets/equity numbers on the companies’ balance sheets.
- The findings from this study suggest that the massive drilling activity funded by growing debt, was likely motivated by balance sheets expansions of assets, and thus the equity from inflated EUR numbers (“book to model”) which made room to take on more debt.
- An inflated balance sheet that allows for a debt load above the carrying capacities of the real underlying collateral, will at some point in time turn against their creators and call for revisions of future plans and expectations.
- It will be interesting to see how the LTO companies’ balance sheets and their profitability respond as it become Mother Nature’s turn with the bat.
NOTE: Actual well data used for this analysis are all from North Dakota Industrial Commission (NDIC). For wells on confidential list, data on runs were used as proxies for extraction (production).
Production data for Bakken, North Dakota: Monthly Production Report Index
Formation data from: Bakken Horizontal Wells By Producing Zone
Data on wells kindly made available by Enno Peters’ excellent and tireless work.
Continental Resources; Investor relations, SEC filings
Oasis Petroleum; Investor relations, SEC filings
Whiting Petroleum; Investor relations, SEC filings
The companies selected for this study were based on their number of producing wells (high portion of wells with 30 months or less of flow as per December 2014, their portion of total Bakken LTO extraction and these represent some spread of better and poorer than the average Bakken well.
Some companies’ 10-Ks PDP shows reserves by area/field split on oil, NGL and natural gas and are also totaled as BOE (BOE; Barrels of Oil Equivalents).
The R/P ratio
The R/P ratio for (oil, gas, coal) is a useful and powerful metric that gives away tons of information.
- The R/P ratio is a snapshot about how long the annual production level for any year versus the (estimated) remaining proven reserves at the end of that same year could be sustained.
The R/P ratio is a number which describes a theoretical rectangular production profile.
In the real world things do not work this way. The R/P number changes from one year to another and it also needs to be seen together with the production level described by it for the year in question.
The Bakken LTO companies looked at
Analyzing the time series for around 8,000 LTO wells in Bakken for 2008 – 2015 makes for a solid foundation to develop predictable trajectories towards their EUR as the time series grow. It is a Nature thing.
This study/analysis presents some data on and derived from R/P for 3 companies that are big in LTO extraction in Bakken(ND) and what to expect from actual data versus those reported on their SEC 10-K 2014 filings.
EUR trajectories for the average wells by vintage
Exclusive of the 2008 – 2010 vintage wells, the average for the younger ones are within a small trajectory band.
Wells of 2014 vintage pulls the average down. Wells started in 2015 have so far performed better, but are not included in this study. Exclusive of the vintages 2010 and 2011, the average for the younger ones are within a small trajectory band.
Whiting’s operated wells of 2014 vintage started out better than average and have in recent months moved to a trajectory converging with the average. Wells so far in 2015 are closely tracking the average for all.
Exclusive of the 2008 – 2010 vintages, the average for the younger wells are within a small trajectory band.
LTO Wells and Extraction
Table 1 shows that the wells in this study that had flowed 30 months or less as per December 2014 constituted a major portion of the total LTO flow (production) and total number of wells.
Long time series with actual data provides more reliable descriptions about what to expect than relying on extrapolations of Initial Production (IP) numbers and/or shorter time series of production followed by tweaking some exponential/hyperbolic factors in the equations for well models.
The R/P Analysis
How the R/P numbers based on actual data were derived?
By totaling the projected EUR of LTO for the average well that started to flow from Jan 08 to Dec 14, adjusting this with what had been extracted (this is measured and reported by NDIC!) for the same period, results in an estimate of PDP (the R for the R/P equation) at end 2014.
The P is total LTO extracted/produced in 2014.
NOTE: The estimated overstatement of PDP at end 2014 does not equate to a similar estimated divergence for the EUR of the average well, refer also table 1.
The estimated magnitude of PDP overstatements was tested and confirmed by alternative approaches, like using BOE as a basis, Q4 2014 numbers for production in the R/P ratio and more.
The Balance Sheets
Now let us move over to these companies’ balance sheets as these were filed with the SEC in their 10-Ks for 2014.
A balance sheet contains a lot of useful financial information about a company.
Here it will be kept simple focusing on the relation as described by the equation below:
- Equity = Assets – Liabilities
If, it is, as I have shown in this post/analysis that the presented companies’ PDP reserves for LTO (in Bakken) are grossly overstated (hard numbers and Nature do not lie!), then what follows from logic is that it should be expected that the numbers on Proven UnDeveloped (PUD) LTO reserves are subject to about the same inflation as the PDP numbers.
This leads to some interesting prospects.
LTO companies’ assets on their balance sheets are primarily described by their PDP and PUD numbers (the reserves).
If, this analysis by direction and magnitude reflects reality, then this with time will show up in financial performance (not only from oil price changes) through metrics for profitability. There are several good profitability metrics that with time will reveal imbalances from an inflated balance sheet with poor/none profitability. Profitability metrics will be the proverbial canary that will give away the true strength of the balance sheet.
If, PDP and PUD reserves are overstated by a very high percentage, ceteris paribus (all things equal, i.e. prices as per 2014), then the balance sheet assets at some point in time will have their day of reckoning and deflate to reflect reality.
Now add the effects from a much lower oil price at end 2015 relative to 2014.
(and…..equity is gone!)
As reserves are depleted (extracted), this becomes recognized on the balance sheets as depletion. From what has been shown in this post the unit depletion numbers are probably based on inflated EUR numbers which understates depletion adjustments which thus overstates assets and equity.
The reality of this arrives as the “non-existent” volumes do not show up.
The focus from Wall Street analysts on Initial Production (IP; production for a well during a defined period of time as the well starts to flow, normally during a 24 hour period early in the well’s life) is misguided from the belief that there are good correlations between IP and EUR in the shales. Statistical analysis for correlations between IP and expected EUR in shales from actual data has shown these to be poor.
In LTO extraction there has been little empirical data available for benchmarking of the models. This will with time change as longer time series of actual data become available to calibrate the models with.
Overly balance sheets focused (as in myopic) investors/creditors will continue to be confident from the numbers of IP and balance sheets and will in the near future spend sleepless nights wondering why such good IPs and strong balance sheets produces poor or no profits and/or why they do not fully receive the money lent.
Their worries will gradually morph from being focused on return on investment to return of investment.
The mysteries created by Nature’s lack of cooperation with the balance sheets will surpass any other existential questions; also the one about what there was before the “Big Bang”.
As observations of/feedbacks from Nature, hard data and logic demonstrate poor compliance with some human created EUR and/or financial models, the models become ripe for revisions to reflect these feedbacks.
Definitions for some acronyms for reserves used in this post.
“proved developed reserves” Reserves expected to be recovered through existing wells with existing equipment and operating methods.
“proved undeveloped reserves” or “PUD” Proved reserves expected to be recovered from new wells on undrilled acreage or from existing wells where a relatively major expenditure is required for recompletion.
12 thoughts on “Are the Light Tight Oil (LTO) Companies trying to outsmart Mother Nature with their Financial Balance Sheets?”
Rune, Thank you for all of the information.
So, in addition to PV10 plummeting due to the price crash, it should also plummet further due to the EUR in the reserve reports being exaggerated on the high side? I had not considered that the barrels reported in the SEC reserves reports could be so far off.
Hi Shallow, and thanks!
I found it odd that R/P ratios derived from the SEC 10-Ks were so high for portfolios (companies) dominated by high decline LTO wells where total production was in a build up phase. I tested several angles to it and ended up at the same place.
I suspect reserves are booked to models which overstates the EUR. Actual data does not support such high EURs derived from SEC 10-Ks.
Look at the actual developments of the totals by vintage for the companies (and I have lots of more that shows the same).
Since I started looking at LTO wells in Bakken, there was one number for the average well EUR that repeatedly came out: 300 kb (+/-) 10%.
I looked at big majors PDPs for 2014 which portfolios are dominated by conventional reservoirs and found R/P at 6.
Looking at Norwegian Continental Shelf (NCS), which is in general decline (now showing a slight reversal [due to a period of high oil price!]) and that has conventional reservoirs which in recent years have had a steady supply of new discoveries brought to flow, the R/P for PDP (only producing) has been stuck at 7 – 8 for years.
Something did not sound right with R/P in the same order for portfolios with high decline LTO wells.
We will soon learn more about the true nature of those PDP and PUD reserves in LTO and their effect on the balance sheets.
Very nice work! Just curious, have you accounted for the gas-to-oil (GOR) behavior? It takes a few years, but it can have a big negative impact on oil production as a well’s drainage volume drops below the bubble point downhole. This impact has started to manifest itself across the state as a whole, where GOR has increased from 1000:1 to 1350:1 cubic feet of gas per barrel oil since 2011. With the downturn, this increase is exacerbating quickly. Measuring the reserves in BOEs might close the gap a bit (albeit with a less valuable product). Note that although North Dakota oil production has been about flat this year at ~1.2 mmbopd, gas production has actually increased by ~ 200 mmcfd due to this effect.
On the other hand, the overstatement may be even more dramatic, as SEC 10-K reserves are also net of the leasehold royalty burdens, usually ~ 20%; not just net of the companies’ working interest as implied Table 2.
Hello Doug, and thanks!
The R/P used in this post are all based on oil (LTO).
The R/P expressed with BOE [includes all hydrocarbons; oil, NGL (Natural Gas Liquids) and Natural Gas] based on companies SEC 10-K filings are higher.
This should be expected as the GOR normally increases during pressure depletion.
Leasehold burdens I will have a closer look at.
Rune, very informative and well researched. Thank you. What i try distill from all the data is : what is the critical oil price ( WTI , spot month) for these LTO plays to get shut down? With WTI currently well below $50, are we going to see wells shut in and abandoned? Are new wells still being drilled at this oil price?
Wells (with low flow, like 10b/d) not covering all their operating expenses would normally be shut down. However, it may be that wells with negative cash flows are kept flowing if there are some other obligations attached, like holding acreage by production (HBP).
The effect on total extraction from shutting down cash flow negative wells is expected to be small.
Presently there are, according to NDIC, 74 rigs operating in Bakken(ND) and net cash flows from operations at present prices ($50/b WTI) may unabridged pay for 80 – 90 wells/month (from spud to flow).
What remains to be seen is how many of the added wells will make a profit with a sustained low oil price, like $50/b.
What oil price could shut down drilling (beyond “drill it or lose it”) is difficult to point to as the companies still will be cash flow positive from operations at lower oil prices, further access to more capital/money (from credit/assets/equity sales). This is also influenced by debt retirement and/or creditors willingness to defer such.
50% of production comes from same year wells. On average those wells would be six months old so the R/P ratio would be 8, not 4.
The other 50% comes from older wells and I am not sure what that ratio should be. On one hand the high first year flow is done, but on the other hand the ratio between R and P might be the same. Lets say I have a 300 b/d first year flow and a 1200 ultimate recovery, ratio 4. By year two if the flow was 150 and the recovery now 900 the ratio would be six.
If you build it up like this you could get an answer greater than 4.
BTW I have no expertise on the oil business, just doing some math here. I do agree with the conclusion that the equity in those LTO companies is threatened.
Read the post again and the portion about how the R/P ratio is calculated. The P is annual extracted (produced) the R is estimated remaining reserves at the end of the same year as the P.
You need the actual data to conduct further calculations. (That takes some time to collect, but they are there in the public domain.)
Added 05.08.15 at 00:55 am CET
The 50% flow number refers to 2014. The older wells (vintages) need to be weighted according to their relative contributions. That is their actual well performances and number of wells by vintage.
Gerd, applying your method would result in a R/P just above 2 for flow of 2014 vintage.
Rune, thank you for your good work, as always.
I have two comments, please, regarding EUR “modeling” and overstating LTO reserves:
Early in both of the two largest shale plays in the US the public was privy to lots decline curve analysis and EUR’s published in the media and in corporate reports, etc. I personally do not see a lot of decline curve analyis for shale wells anymore and I believe there is now some form of EUR “standardization” occuring for wells within given sweet spots, units, field designations and/or counties, etc. I have reason to believe that this standardization occurs from EUR “modeling” quite early in the life of a shale well, perhaps after only 6 months of production is realized, precisely as you have suggested in your article. Production data is normalized for some wells in a given area and the EUR model is built. Once built, the model applies to other wells in the area; the Likvern #8H will perform like the Likvern #4H, similar to the Norway #5H down the road 3 miles, according to the EUR model, and the ultimate recovery from all wells will be the same. Or close enough for SEC purposes. After all, who is auditing these EURs?
As an operator myself I can assure you that initial potential (IP) “management” is the easiest trick in the oilfield; I can, within reason, make any given well have any IP I wish it to have and it will be the absolute truth. High IP’s in the Eagle Ford play tend to correlate to steeper declines rates, so my research indicates. I believe the same holds true for the Bakken, as other have stated. It is not known whether high IP’s translate to higher UR in a given well nor is known whether steeper decline rates translate to lower UR’s. We can make some good guesses but only time will sort that out.
On the other hand, with regards to EUR modeling, higher IP’s and subsequently higher 6 month production rates will necessitate the use of higher hyperbolic exponents in the decline curve analysis and therefore higher IP’s can, in fact, imply higher EUR’s. If modeled EURs are the basis for PDP reserve reporting then the higher the well’s IP’s, the more reserves one can overstate. This EUR modeling would most certainly also apply to PUD reserve estimates. So higher IP’s and 6-12 month production averages mean more PDP reserves to book and that clearly is the mantra for shale companies in this current price environment, as you point out…book the reserves, even at the expense of long term reservoir management that might result in better UR. There is a very well know shale producer in S. Texas known for gutting it’s wells the first 6-12 months of production, then being applauded for prudent “choke management” practices thereafter.
Secondly, I think not enough attention is being paid to associated gas in EUR discussions nor in evaluating the overall health, or lack thereof, of a shale company. Better shale wells in sweeter spots, in both the Eagle Ford and Bakken, tend to be in areas of higher gas to oil ratios (GOR). Initially, within the 6 month time frame of the EUR model scheme, gas production is higher as a percentage of the total production stream, liquids lower, then liquids increase and gas decreases, then as the well depletes GOR increases and gas volumes increase again. LTO EURs are always reported in barrels of oil equivalent. Seldom, however, are we told what the conversion factor is. I think the 6:1 BTU conversion is fairly API (standard). At 6:1, higher initial gas production during the first 6 months of a well serves to distort the EUR model significantly and also lends itself to over exaggerated PDP reserve reporting, in my opinion. Reserve equity in an oil and gas company is based on dollars, not BTUs, therefore gas to oil conversions based on price is the far better metric. In the first 5 years of these LTO shale plays in the US the real conversion factor for gas to oil was almost 36:1; currently at todays prices, it is more like 20:1. BTU based conversions of gas to oil are a distortion of reported PDP reserves. For a large shale company the distortion is significant.
Finally, as kind of the ultimate overstatement of PDP reserves, please imagine a situation where fully 30% (associated gas) of the reported BOE PDP reserves for a shale well were allowed to literally go up in smoke, thru a flare stack, the first 4-5 years of production. That is precisely what happened in the vast majority of Bakken wells.
I would imagine the relationship between lender and lendee in the LTO business is pretty tenuous right now, and likely to become an all-out fist fight pretty soon.
Thanks a lot for yours far reaching comment. This is important.
I have very strong reasons to believe that if a selection of modeled wells were paired with their actual production data that would provide valuable documentation supporting overstatements of EUR for shale wells based on model.
If done as described above, one would find that the R/P number grew strongly as the wells aged.
In other words as the wells ages the growing R/P number would describe a long fat tail. That runs contrary to what to expect from these kinds of wells (there will always be outliers) and what actual data shows.
This would be the giveaway.
I agree; the R/P tool will be very telling and valuable to anyone needing to gauge the health of an LTO company. My apologies, by the way, for the length of my post. Being critical of the shale oil industry business model is not often a popular place to be in given my line of work.
Thank you again.
Worth the read and the article touch many of those factors affecting oil prices and why it may take a while for the market to rebalance.
”As shale has dramatically reduced time-to-build (the time between when producers commit capital and when they get production) from several years to several months, oil prices now need to remain lower for longer to keep capital sidelined and allow the rebalancing process to occur uninterrupted. This spring’s rally in prices did prove to be self-defeating. Not only did all the capital markets reopen as oil prices rose, but producers began to redeploy rigs and remained under hedged, which is a reflection that the industry simply had not faced enough pain to create real financial stress that would create change.”
Crude Carnage Continues As Goldman Warns “Storage Is Running Out”
Comments are closed.