North American crude oil has been in the news on several fronts this week, including some rapid price moves and an unexpected intervention by President Obama. Despite the publication of a new report projecting a much more rapid rate of tight oil supply growth than is generally expected and the entire Buffet-Railroad-Traffic-Pipeline meme relying on increasingly exponential dreams of the Bakken et al. saving us from our excess-energy-consuming selves, Barclays questions just how realistic these forecasts are, noting "it is perhaps wise to exercise a degree of caution over longer-term shale oil forecasts... partly because of the steepness of decline rates for shale oil wells, a lot of the very big productivity gains have already been made, and finally, skepticism around some of the more ambitious projections of US shale output due to the existence of numerous logistical barriers."
Authored by Barclays' Kevin Norrish,
There has been a distinct theme of North American oil running through much of the commodity news flow this week. Weather disruptions to Canadian crude and the return of several US mid-west refineries have helped push WTI prices up relative to Brent, pressuring the front-month spread between the two to its lowest for 18 months. President Obama’s surprise intervention in the Keystone XL pipeline debate has raised fresh concerns about its viability by placing a heavy emphasis on environmental criteria in its approval process. Related to this we also take a look this week at how Canadian oil producers have been rapidly adapting to pipeline capacity shortages by making intensive use of other options such as rail. Finally, the Kennedy School of Government at Harvard has released what appears to be the most bullish shale oil report yet, projecting that US output of shale oil could reach 5mbpd by 2017 making the US the world’s largest oil producer.
But how realistic are forecasts such as these? It is certainly true that the sharp increase in US shale oil production has taken many in the oil market by surprise over the past couple of years and that includes even those oil analysts and forecasters that are watching the sector most closely. Current rates of growth are exceeding expectations and bring in to question some of the more conservative assumptions made by the major oil forecasting groups.
As little as 18 months ago the US Department of Energy was forecasting a flat profile for domestic oil production and that US oil liquids output would grow this year by a very modest net 100k b/d.
Taking the latest figures from the DOE’s Short-Term Energy Outlook, that figure has now risen roughly nine-fold as it has for most other major forecasters (including ourselves). Although big upward revisions to the outlook for US crude oil supply were implemented by the DOE through 2012, the data so far this year suggest that reality is running a long way ahead and that even its latest forecasts are looking far too modest.
Growth in US crude oil production is running at almost 1.2m bpd above year-ago levels, according to weekly oil supply data, exceeding both the DOE and IEA’s latest full-year 2013 growth forecasts by almost 50% and OPEC’s by almost 100%. This rate of growth is almost solely down to shale and has been achieved despite the poor performance of producers in the Gulf of Mexico. Unplanned maintenance and a shut-in currently underway at Shell’s Mars B project where a new platform is being installed have forced some downward revisions to project GOM crude output, which is now expected to decline this year.
There is also an offset to the strong growth in crude oil output from the other components of US liquid fuels output. Natural gas liquids growth in the year to March at just 50k bpd is relatively sluggish at just half the level expected this year by the DOE and ethanol production has actually fallen by 100k bpd. However, these figures are modest alongside the pace of growth in US lower 48 crude oil output, the biggest driver of which is rapid growth in tight oil output.
Although the rate of drilling has slowed in shale plays such as the Bakken and Eagle Ford this year, output is still going up due to ongoing and rapid productivity improvements. Moreover in the Permian, where development started relatively late, drilling is still rising rapidly and increases here are likely to be the big driver of US oil production growth over the next few years. With domestic US oil prices (as represented by WTI) trading at well above $90/barrel and well-head discounts having contracted (eg, the discount that Bakken oil producers have to live with has shrunk to just -$3/barrel from -$28 in early 2012), prices are at a level that will continue to incentivise investments in new output.
The short-term implications of these trends are clear. It is likely that big upward revisions to forecasts of 2013 growth in US output by the major forecasters will need to be implemented soon. For now we are raising our US oil production forecast to 1.1m bpd from the previous level of 710k bpd and expect others to soon follow suit.
But what are the implications of this rapid growth for the longer term? The current EIA forecast from its Short-Term Energy Outlook report is that by December 2014 the US will be producing over 12m bpd of liquid fuels. That figure suggests there may be something of a mis-match opening up between medium- and long-term projections and perhaps not just for the DOE. The DOE’s own long-term reference or base-case forecast as published in its Annual Energy Outlook (May 2013), projects domestic production of liquids topping out at 11.8m bpd in 2019 and declining thereafter, implying growth of around 3m bpd in total domestic liquid fuels output over the 2011-2020 period (Figure 3) with tight oil contributing around half that growth. Both the IEA and ourselves see growth of a similar order of magnitude (apart from the widely circulated forecast made in March 2012 by another investment bank, and which was a “best possible case” scenario).
However, it is perhaps wise to exercise a degree of caution over longer-term shale oil forecasts and for now we are sticking with our 2020 projections which imply a sharp slowdown in tight oil production compared with current levels. This is partly because of the steepness of decline rates for shale oil wells. For example, after two years a typical well in the Bakken-Three Forks drilled in 2012 is likely to be producing at less than 30% of its initial production rate. It also reflects the likelihood that many of the best plays for crude oil extraction are being developed early (recent disappointing performance relative to expectations at the newer Utica shale play in the north-east US and Canada backs this up) and that a lot of the very big productivity gains (shortened well drilling times, increased initial production rates and better rig efficiency) have already been made.
A final very good reason for scepticism around some of the more ambitious projections of US shale output is the existence of numerous logistical barriers. These include a lack of pipeline capacity, which despite recent expansions, is still a major constraint forcing regional producers to accept discounts that can often be wide and volatile in order to get their oil to market. A sustained acceleration in shale output growth would likely stretch the US distribution system for crude oil to breaking point.
The associated uncertainty over price levels and take-away options would create a very difficult climate to invest in the intensive and costly drilling programs necessary to keep shale output growing.