(John Kemp is a Reuters market analyst. The views expressed are his own)
By John Kemp
LONDON, Jan 27 (Reuters) - Rig counts are a highly imperfect guide to future oil production but they are one of the few readily available statistics on oilfield activity so it is unwise to dismiss their importance entirely.
The sharp drop in crude oil prices since June 2014 and associated fall in rig counts published by state regulators and service companies such as Baker Hughes has sparked a lively debate about the short-term outlook for U.S. oil production.
Some analysts have forecast the drop in rig counts will cause production to peak in the first six months of 2015 then begin to fall in the second half of the year. I count myself in this camp.
Other analysts predict production will remain steady or continue to grow, albeit much more slowly than the 1 million barrel per day (bpd) increases recorded in 2013 and 2014.
In practice, the differences are smaller than the two sides imply. Most forecasters put the change in daily production between December 2014 and December 2015 at less than 250,000 bpd, whether up or down.
For example, the U.S. Energy Information Administration (EIA) predicts output in December 2015 will be just 90,000 bpd higher than in December 2014 (“Short-Term Energy Outlook” Jan 2015).
That would be a marked slowdown from the 1.28 million bpd increase between December 2013 and December 2014 or the 790,000 bpd increase between December 2012 and December 2013.
At a rough approximation, most forecasters expect U.S. production to be flat in 2015 - after two years of 1 million bpd growth, during which time North American shale producers were the marginal suppliers to the world oil market.
And every forecast for U.S. oil production contains an implicit forecast for what will happen to oil prices over the course of the year.
EIA thinks the number of active drilling rigs in the United States will fall by 24 percent between January and October 2015 before starting to recover in November.
That forecast is based on its assumption WTI prices will increase to an average of $53 per barrel by June and $67 in December.
WTI prices have so far averaged just under $48 so far in 2015, so EIA’s forecasts for rig counts and production are conditioned on a $20 price increase by the end of the year.
U.S. oil production “reflects more than just the rig count,” as EIA emphasised in a research note published on Monday analysing the combined effect of all the factors known to affect output. The note is essential reading for anyone trying to understand the likely production trend in 2015.
In fact, oil production reflects a constellation of factors, of which the most important are:
1. The number of rigs employed (raw rig count).
2. The speed with which rigs are able to drill wells on average (affected by the time lost moving location and setting up, incidents causing stoppages and type of rock drilled through).
3. Efficiency and capability of the rigs (maximum operating depth and available horsepower).
4. Average vertical depth of wells drilled and length of horizontal sections.
5. Speed with which wells, once drilled, are pressure pumped and completed, so they can start producing.
6. Quality of the rock in the neighbourhood of newly drilled and completed wells, affecting production rates.
7. Average number of stages fractured in each well (which can range from 10 to 30 or more).
8. Average production from the wells during the first 30 or 60 days and decline rates thereafter as natural energy in the reservoir falls.
9. Decline rates on production from old wells (average age and decline rates on the stock of existing wells drilled in both shale plays and conventional oil fields).
10. Wellhead oil prices relative to the full life-cycle breakeven costs of drilling new wells.
Rig counts are just one of many factors which determine production. Experts are right to remind readers that production is about much more than “just the rig count”. In a sense every well and every drilling team is different and the impact on production is complicated. Some simplifications must, however, be made for the sake of analysis.
In the face of a steep decline in oil prices, production companies have a number of strategic options to cut costs and improve recoveries.
The least-efficient rigs and crews can be idled first. The remaining rigs can be pulled from exploration work in frontier areas (where recoveries are uncertain) to focus on development work and infill drilling in existing plays (where likely production is more certain).
Within existing plays, rigs can be pulled back from the periphery to concentrate the most high-yielding “sweet spots”.
Production companies can negotiate and likely obtain big reductions in hire rates for rigs and pressure pumping equipment as well as the price of all their other inputs, from water and sand to diesel and trucking.
For all these reasons, the number of new wells and the output from them is likely to fall more slowly than the rig count.
At the same time, breakeven prices for new wells will come down substantially to reflect cheaper drilling and pressure pumping costs.
In some instances, however, producers may postpone the completion of already drilled wells to defer the costs and hope for a recovery in prices.
Completion accounts for two-thirds of the total cost of some of the very long horizontal wells now being developed. And one third of the well’s total production may occur in the first 12 months. So there are sharp financial incentives to slow completion where possible.
There is a substantial backlog of oil wells that were drilled in 2014 waiting for the arrival of pressure pumping crews and other completion services. U.S. production could continue rising for several more months as these wells are finished and put into production.
But there are also anecdotal reports from oilfields that some completions are being postponed to save costs and wait for a more favourable price environment. If widespread, slower completions could cause production to peak earlier than expected.
While production forecasts are about much more than just rig counts, rig counts are one of the few pieces of data readily available in near real-time. There is no comprehensive and timely information on drilling speed, well depth and horizontal length, initial production rates, and decline rates on the stock of existing wells.
There is some information on the number of new wells started (“spudded”) each month but that is only marginally more helpful than a rig count because production depends on where the wells are located and how many horizontal stages are fractured, among other factors.
Focusing on the rig count introduces a data availability bias into production forecasts. But since it is almost the only comprehensive and timely information on drilling activity, forecasters do not have much choice. Rig count data must be interpreted carefully but it would be unwise to ignore it completely.
On balance, the safest conclusion is that U.S. shale producers will be able to offset some of the fall in oil prices by drilling more efficiently, and that the amount of new oil produced per rig employed can be increased.
But the efficiency improvements are unlikely to fully offset the decline in prices and rig counts, at least in the short term.
For that reason, it seems reasonable to assume U.S. shale production will be fairly flat in 2015, give or take a couple of hundred thousand barrels per day, after several years of exceptional growth. (Editing by David Evans)