Is U.S. Oil Production Already In Decline?

by: Michael Maltby


The EIA oil production estimation methodology appears flawed as it focuses on drilling and not completions.

Completions in the Bakken collapsed in November 2014, implying that oil production is already in decline.

If replicated across other basins, U.S. production will decline year on year.

This article explores the question of whether oil production forecasts for the Bakken for 2015 have been overestimated by the US Energy Information Administration ('EIA').

The EIA Drilling Productivity Report is a key investor report providing short term guidance on production trends in the main oil shale basins. The most recent report was published on 12th January 2015.

Estimates are made for the current and the next month, in particular in respect of oil produced from new wells. This is estimated using data on the number of rigs drilling (the 'Rig count') and a productivity factor for each rig that estimates the average oil produced per rig ('Production per rig').

The EIA data for the Bakken published on 12th January 2015 is shown in the table below.

On closer review, the data appears to have two key weaknesses.

1. Is the EIA using the correct rig count?

The first weakness is in the implied rig count for January 2015 and February 2015. By mathematically working backwards through the data it appears the rig counts are the Baker Hughes rig counts of November 2014 and December 2014 respectively. However, the Baker Hughes total on 16 January is 164, less than the 189 assumed in the January estimate. Assuming the rate of reduction continues at current rates, this suggests a January average of 164 and a February average of 140. The table below adjusts the January and February forecasts using these revised rig count numbers.

The adjustment implies the EIA is overestimating short term production. As can be seen, a reduction in average rigs to 140 in February 2015 results in production stabilizing at approximately 1.27m boe/d, up from the average for the Bakken of 1.11m boe/d for 2014 (source: EIA productivity drilling report), but below the 1.30m boe/d assumed by the EIA. This is consistent with recent presentations from the North Dakota Oil and Gas Division about the number of rigs required to stabilise production.

2. Should the EIA be focusing on completions or the rig count?

A more fundamental issue is whether using the rig count as a proxy for incremental oil production is valid in the current climate and if not the likely implications. Implicit in the EIA data is the assumption that there is a direct correlation between the rig count and wells being completed. However if we review the data from the North Dakota Oil and Gas Division this is no longer the case.

The key report is the monthly Director's Cut, providing actual data on production in the state. In the most recent copy, the report notes that only 39 (23%) of the wells drilled during November 2014 were actually completed, with companies preferring to leave completion until a recovery in oil prices takes place. This is effectively a new factor that needs to be taken into account when determining the likely impact of drilling activities on production, and does not appear to be reflected in the EIA calculations.

For illustrative purposes, the possible impact of this is set out in the table below. Assuming a one month lag, and no changes to the November completion ratio of 23%, the EIA data for the period to the end of February has been recalculated by adjusting the rig count for the completion ratio, while leaving the other assumptions unchanged. I have also added a line for March assuming a further decline in rigs to help establish trend.


The key conclusion is that oil production in the Bakken peaked in Q4 2014 and is now moving into decline, with production for Q1 already trending below average production for 2014.

From a business perspective the conclusions should not be surprising. Even if companies are drilling wells, they appear to be deferring the decision to complete and produce. It might appear that for most of the drilled wells in November, there was no economic case for completing. If this is the case, then we should expect that drilling activity will continue to fall, potentially to very low levels, in the near term.

Further work is required to assess whether this is localized to the Bakken or part of a wider trend. However, assuming that this behavior is replicated in other basins (which may not be the case to such an extent given the price differential for oil in the Bakken to WTI), it appears that the supply side response to the fall in oil prices has been much faster than has been predicted and that oil production in the USA may already have started to decline.

Conversely of course, if there is an upswing in prices, the US oil industry will be well positioned to rapidly respond and complete a substantial inventory of already drilled wells with low incremental capital costs and short lead times.

What does this mean for the investor? Firstly, a more rapid supply side response than previously expected should help establish a floor under the price of oil. Secondly, investor focus at this point must shift to completions as a key indicator of the economics of producing oil in the Bakken and in other basins. Finally, the economics of the Bakken at least appear very unattractive at this point.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it. The author has no business relationship with any company whose stock is mentioned in this article.