Friday afternoon, Baker Hughes (BHI) reported yet another significant weekly decline in the U.S. Lower 48 gas-directed rig count. As of October 12, there were 422 rigs drilling for natural gas, 15 less than a week earlier. The reduction in the horizontal rig category was particularly significant, with a 23-unit decline to 290 rigs from 313 rigs.
I have written recently that the gas-directed rig count possibly had already bottomed, given that the gas price recovery in September had reduced the economic incentive to operators to lay down additional gas rigs. I have also noted that the rig count reflects operators' decisions, and the price environment, with a several-month lag. This week's report proves that my suggestion regarding the inflection point in the gas rig count being achieved was a bit premature. It is important however to look inside of today's headline number for its correct interpretation.
A closer analysis reveals that the aggregate rig count decline masks the stable drilling activity in the "dry" gas plays:
- The combined horizontal rig count in the Haynesville (30 rigs as of 10/12/12), Fayetteville (13 rigs) and Arkoma Woodford (6 rigs) was little changed (-1 rig) from the prior week.
- The infrastructure constrained Marcellus (which has rigs drilling in both dry and wet areas) saw its horizontal rig count decline by 2 rigs to 74. If vertical and directional rigs are included, the rig count in the Marcellus was unchanged at 90 rigs.
- The Barnett lost one horizontal rig to 25 from 26.
The biggest loss was in the liquids-rich plays where NGL processing and off-take capacity has often lagged the drilling activity:
- The Eagle Ford made the largest single contribution to the rig count decline, with the gas-directed horizontal rig count losing 8 rigs to 66 from 74.
- Several other liquids-rich plays saw another 7 horizontal rigs leave the active status on a combined basis (the Permian, Cana Woodford, Ardmore Woodford and Granite Wash each lost one rig; the Williston Basin lost two rigs; and the Mississippi Lime and Utica had their rig counts unchanged). Total gas-directed rig count for these plays stood at 49 rigs as of this date.
All other areas lost a combined 9 horizontal gas-directed rigs during the week, from 80 to 71 (that includes 2 horizontal rigs in Wyoming).
The important conclusion from this closer review of the current and prior weeks' data is that the "dry" gas rig count has seen only minor contraction during the past six weeks. The ad-hoc "dry" shale rig count index has declined by 2% since the end of August, as shown in the chart below. Dry gas plays typically have much greater gas yields per rig than the liquids-rich plays (the ratio can be as high as three or four times). Therefore, dry gas rig count is an important driver and a leading indicator of future marketed natural gas production.
At the same time, the analysis indicates that the liquids-rich plays have continued to see steep reductions in the gas-directed rig counts. As illustrated in the graph below, the ad-hoc "liquids-rich" shale rig count has declined by 20% since the end of August. This can be explained by three factors:
- gradual slowdown of the "land rush" in the liquids-rich plays;
- unavailable fractionation and off-take infrastructure, which translates in extensive well backlogs;
- the deterioration in NGL economics, particularly in the regional markets without Gulf Coast access.
If the direction of the rig count change is any indicator of economic attractiveness of a play, the drilling activity dynamics would suggest deteriorating returns in the liquids-rich areas. This trend may come to the fore during the forthcoming earnings season in the E&P sector.
As I wrote in a note Friday morning, the reduction in the rig count does not get fully reflected in the production data for two to three months. Furthermore, once the inflection point in the rig count is finally reached, the incremental production from the rig count additions will not be seen for another several months. Therefore, from a fundamental perspective, today's Baker Hughes data is very constructive for the natural gas price. Capital discipline continues to be in place, and the rig count so far has not embarked on the growth trajectory. Moreover, the material additional contraction in gas drilling activity should help rein in the supply. Even if the drilling activity were to resume growth next week, the low point in the natural gas production will not be seen until January.
One week's rig report does not make a trend and does not change the likely "Goldilocks" scenario for natural gas in 2013 that I discussed earlier. However, everything else being equal, today's Baker Hughes data has positive implications for natural gas (UNG) and the natural gas producer stocks. My natural gas producer index includes: