Permian Basin Update, February 24, 2022

Much of the information for this post comes from data at shaleprofile , and assessments by the USGS. In addition a paper published in Jan 2022 by Wardana Saputra et al was an excellent resource.

The basic method used in the is analysis is covered in an earlier post, essentially the convolution of average well profiles with the monthly completion rate over time is used to model future output. I focus on the period starting in Jan 2010 and consider only horizontal tight oil wells in the analysis. Future well profiles are estimated and several future scenarios for completion rate are used, clearly the future is unknown so future completion rates and estimated ultimate recovery (EUR) for wells completed in the future can only be guessed at.

In order to make such a guess I start with the USGS assessments for the Permian basin where the mean estimate for prospective net acres as of mid 2017 was about 50 million acres. I use an estimate for average acres per well of 300 acres (about 9500 feet lateral length with spacing of 1320 feet between laterals) which gives an estimate of about 167 thousand wells. There were about 14 thousand wells already completed in the Permian basin by June 2017 so total completions would be about 181 thousand wells, if oil prices were high enough to make every potential well location profitable. Using the mean UTRR estimate (70 Gb) and number of potential drilling locations (about 160 thousand as of Dec 21, 2021 based on the data at shale profile where about 21 thousand wells were completed from July 2017 to Dec 2021), I find and estimate for the future decrease in EUR per well that will result in a UTRR of 70 Gb if all potential wells were completed.

After that step a discounted cash flow analysis using guesses of future costs and prices is used to determine whether a well will be profitable to complete to arrive at an ERR for a given scenario, typically ERR is less than TRR, but in rare high oil price scenarios they could be nearly equal.

Average well profiles have been developed by fitting an Arps hyperbolic function to the data from shaleprofile.com for the average 2010 to 2012 well and then for each individual year from 2013 to 2020. In my scenarios I assume EUR starts to decrease after Dec 2020 and assume no further increase in lateral length or change in average well spacing.

Since 2010 average new well EUR has been increasing, but note that when we normalize for increasing lateral length, the productivity growth stopped in 2018 and may be decreasing slightly, unfortunately I do not have access to average lateral length data so I rely on occasional updates at shaleprofile.com. Data for these well profiles can be found here.

My central scenario assumes the Permian basin horizontal tight oil well completion rate increases from 400 new wells per month (the rate for past 6 months) to 800 new wells per month by July 2025 with the rate increasing by 10 wells per month from July 2022 with a slower increase of 5 wells per month from Feb 2022 to June 2022, the completion rate remains at 800 new wells per month from July 2025 to January 2037 in my high oil price scenario case and then decreases to zero by April 2039. The EUR for the average new well for the high oil price (maximum price of $100/bo in 2020$) and mean USGS TRR estimate (75 Gb) from Jan 2022 to April 2039 is shown below. No wells are completed after this date for this scenario. Note that for other TRR assumptions (F95=45 Gb and F5=116 Gb) the decrease in EUR is different (it decreases less in the F95 case and more in the F5 case). This scenario has 182 thousand competed horizontal tight oil wells from Jan 2010 to April 2039, about 34,200 wells have been completed through December 2021 based on the shaleprofile.com supply estimate for the Permian basin.

The table below summarizes USGS estimates for the F95, mean and F5 cases from the Permian basin assessments of the Midland basin Wolfcamp(2016), Spraberry (2017), and Delaware Wolfcamp and Bonespring formations (Avalon formation is also included in Delaware basin assessment). UTRR is undiscovered technically recoverable resource, net acres are total acres multiplied by the success ratio for individual benches (1 million acres with a success ratio of 0.9 would be 900 thousand net acres) and wells are estimated by dividing net acres by 300 acres per well.

Note that I use June 2017 as the mid point for these assessments as I do not have the detailed data on which formations wells were completed as of the dates of the assessments, so that is an approximation. As of June 2017 there were 13,710 completed horizontal wells in the Permian basin, based on the most recent Permian basin update at shaleprofile.com, so for the mean USGS case total wells completed would be about 182 thousand wells (F95=99k, F5=305k). The total output from wells completed through June 2017 may be about 5 Gb, this would be added to the UTRR in the table above so TRR would be 45, 75, and 115 Gb respectively for F95, mean and F5 USGS estimates.

The details of the economic assumptions are as follows (all in 2020 US$):

average well cost=$10.8 millionOPEX=$11/bo+$16000/month (monthly cost)NGL price=35% of wellhead crude pricenatural gas price=$3.50/MCFtransport cost to refinery=$5/boroyalty and taxes=28.5% of wellhead revenuenominal annual discount rate=10%nominal annual interest rate on debt=7%dividend payout=25% of net revenue

I use a discounted cash flow analysis where the scenario oil price and the assumptions above are used to estimate discounted net cash flow (DCF) over the life of the well, wells are completed when the DCF is greater than or equal to the well cost for oil price and economic assumptions above. The average well is assumed to have a 9500 foot lateral spaced at 1320 feet (roughly 300 acres per well). The three oil price scenarios (very low, low and high) are given in the chart below. Note that the very low oil price scenario is only used for one case with a 45 Gb TRR with a 400 well per month completion rate, for all other scenarios either the low or high case scenario is used.

The various well completion scenarios are shown below, they are mostly similar over the 2022 to 2030 period and then the tails change depending on both the assumed TRR for the scenario (due to fewer wells in the lower TRR cases) and the oil price scenario (fewer profitable wells in the low price case). Note that only one case uses the very high completion scenario (1600 new wells per month maximum) to see if the an ERR close to the high TRR for the F5 scenario could be approached, this scenario is not likely to be realistic at any price ( and definitely not at $100/bo). Most of the scenarios have a maximum completion rate of 400, 600, or 800 per month with most scenarios either 400 or 800 completions per month. In the charts in this post I use the following notations:

e=ERR in Gbt=TRR in Gb,c=maximimum well completion rate in new wells per monthw=total wells completed from Jan 2010 to end of scenario in thousands

Note that where there are two scenarios with the same TRR and completion rate maximum, the different ERR and total well completions is due to different oil price assumptions (very low to high oil prices). Data can be downloaded here.

The high oil price scenarios are in the chart below. Note that the scenario with ERR=TRR=75 Gb depends on the high oil prices assumed, should there be a rapid transition to electric transport in response to high oil prices (which may rise to $150/bo in 2020$ by 2028 when oil output is likely to have peaked) in a world where the ramp up in battery production overcomes the many obstacles that exist we might see oil prices start to drop by 2035 and perhaps earlier if OPEC chooses to develop their resources more aggressively to sell their output before World demand starts to wane. We will come back to this later.

Low oil price scenarios in chart below

In chart below we look at the range of scenarios including the very low oil price scenario for the low completion and low TRR case and the very high completion rate scenario for the high TRR and high price case, in between we have the central 600 completion rate mean TRR scenarios averaged for low and high oil price and the all scenario average.

Below we consider 4 scenarios with completion rates of 600 and 800 completions per month at the high and low oil price scenarios, all are based on the mean USGS TRR estimate of 75 Gb and the ERR ranges from 62 to 75 Gb, the average of the 4 scenarios is also shown, this would be my best guess for future Permian output, the ERR of the 4 scenario average is about 70 Gb. Data for the various scenarios can be downloaded in a spreadsheet.

When we look carefully at the USGS assessments we can consider the various benches and which are the most productive volumes of rock. Of the 50.4 million net acres in the USGS mean estimate, roughly 31.4 million net acres have more prospective (higher EUR per acre) volume. These 31.4 million net acres have a UTRR of 52 Gb, when we add the 5 Gb that is likely to be produced from wells completed through June 2017, the TRR becomes 57 Gb, that leaves another 18 Gb of TRR to potentially be produced from the remaining 19 million acres, if half of this can be produced profitably that would bring the total ERR to about 66 Gb. The forecast by Saputra et al (2022) has an ERR estimate of 55 to 62 Gb somewhat lower than my estimate. Note that Saputra assumes a completion rate scenario of 400 wells per month, for my scenarios assuming the mean USGS TRR and 400 well per month completion rates at both high and low oil price scenarios, the average ERR is about 51 Gb. My guess is the completion rate will increase in the future to at least 600 wells per month where the average ERR of the low and high oil price scenarios is about 67 Gb or possibly to 800 wells per month where the ERR of the low and high oil price scenarios is 73 Gb. The low completion rate scenarios will leave a lot of oil in the ground if oil prices start to fall around 2036 to 2042 as in my oil price scenarios.

What happens if OPEC is able to increase their capacity by 2028 and/or the transition to electric transport occurs more rapidly than mainstream agencies such as the IEA currently forecast? We briefly consider this by looking at the TRR=75 Gb and 800 well per month maximum completion rate scenario under a modified "high oil price" scenario that sees prices drop rapidly ($1/month rate of decrease) starting in Jan 2031.

Below we have the resulting scenario with completion rate shown on right vertical axis. The ERR drops from 75 Gb to 53 Gb with this change in oil price scenario.

Many oil pros believe fewer completions would be a better approach to tight oil field development, let us consider the same oil price scenario and TRR assumption but reduce the completion rate to 400 new wells per month.

The lower completion rate scenario leaves about 15 Gb of oil in the ground that is unlikely to ever be produced, if oil prices follow the "new" high oil price scenario.

Many in the oil industry doubt demand for oil will fall faster than the supply of oil earlier than 2040 to 2050, that was the basis for my initial high oil price and low oil price scenarios, combined with my skepticism that OPEC will choose to increase their capacity substantially.

Switching back to my original high oil price scenario and considering the 75 Gb TRR and 600 well completion rate scenario and using the economic assumptions given earlier in the post, I can show cumulative net revenue basin wide from Jan 2010 to Dec 2035. Well cost is assumed to increase from $7.5 million in Jan 2010 (2020$) to $10.5 million in August 2017 and then remain at that level until September 2021, then real well cost(2020$) is assumed to increase at 1% annually, and increase of 3.5% at an annual inflation rate of 2.5%. Debt can be fully paid back by early 2025 under these assumptions and cumulative net revenue builds to about $750 billion (in 2020$) by 2036, this does not include money that could be earned on this pile cash if it were invested. Note that by some estimates drilling and completion costs per lateral foot have been falling in most tight oil basins. I assume no change in lateral length or well design after Dec 2020, so I would any cost increases long term would be marginal, close to the rate of inflation (so no change in real cost in constant dollars). The estimate below is conservative.

A final question was posed by Ovi about US tight oil and whether it might be able to meet world demand growth. If we assume the long term (40 year) trend of about 800 kb/d increase in demand for crude plus condensate continues into the future, this would be out target. I created the scenario below for the Permian basin, based on a TRR of 75 Gb, the standard high oil price scenario (first one presented in this post) and a maximum completion rate of 800 wells per month. The rate of increase was modified (reduced) from the initial scenario in the post as shown on right axis below.

The rate of increase from june 2022 to june 2028 will be shown later, first we combine this scenario with scenarios created for all other US tight oil basins to get a US tight oil scenario.

This scenario would only occur if oil prices remain high through 2040, this is not likely in my view, if oil prices start to fall in 2032 we would see closer to 75 Gb for US tight oil URR, this scenario is optimistic/unrealistic.

In any case form mid 2022 to mid 2028, the rate of increase in Permian and US tight oil is about 700 kb/d annually.