What Goes Up? When Did U.S. Crude Production Start to Decline?

A question we get asked all the time these days is whether or not U.S. crude output has begun to decline yet and if so by how much? We don’t actually think the answer makes a lot of difference to the market - especially when you consider changing imports and inventory. But ever since the OPEC meeting last November (2014) failed to take action to reduce  output to support oil prices - market watchers have placed a lot of emphasis on when U.S. shale producers would respond by cutting production. So regardless of the merits of the question we are all living in a marketplace where knowing the “real” state of U.S. production – and whether it is up or down – has become a big deal. To that end today we look at crude production data from the Energy Information Administration (EIA).

The EIA makes a number of estimates of crude oil production for its various weekly, monthly, quarterly and annual publications. We will take a look at three of them today. The first of these is likely the most reliable – namely the EIA monthly crude balance production data. That monthly data is reliable in part because it is published at least three months in arrears (the latest report is for June 2015). And the monthly data just got a little more reliable this summer when EIA changed their collection approach to gather more data direct from producers rather than relying on data published by State agencies and to include more individual states into their data breakdown.  The better quality of data in the monthly numbers will certainly help EIA improve their reporting of crude oil production.   The big disadvantage of the monthly EIA data for the market is that 3-month lag – which no one used to care about when production just went up every month - but is now a big deal with the eyes of the market looking for turning points. For the record EIA monthly data shows total U.S. crude domestic production topping out in April 2015 at 9.6 MMb/d and then declining by 200 Mb/d in May and a further 100 Mb/d in June 2015.

A second source of production data and one that is potentially more valuable for market watchers looking for signs of lower crude output from shale producers – is the monthly EIA Drilling Productivity Report (DPR – see Every Rig You Take). The latest DPR for September 2015 indicates that overall crude production from 7 major domestic shale basins has been declining since reaching a peak of about 5.6 MMb/d in April 2015 (see blue bars in Figure #1 against the left axis). The DPR says that overall crude production in the 7 basins (Bakken, Permian, Eagle Ford, Niobrara, Utica, Marcellus and Haynesville) fell by 190 Mb/d between April and August 2015 and predicts a further 160 Mb/d decline through October 2015 (pink bars in Figure #1). As we discussed in “Every Rig You Take” the DPR uses drilling activity (number of wells drilled) and historic well decline rates to predict future shale production. Since we know that productivity in these basins is generally improving all the time – this approach (if not adjusted accurately) is likely to be inaccurate because it assumes new wells will only be as good as historic wells. The reality is that “actual declines may be smaller than they appear in the rearview mirror” – in other words actual production is likely to be higher than the DPR predicts. However, the backward looking bias is only going to impact the predictive data in the DPR – the further back you go, the EIA “actual” data (as used in the monthly reports we just discussed) replaces the predictive. Andin this case the DPR does accurately reflect the April 2015 turning point in crude production reported in the EIA monthly report.