Musings: Is the U.S. Rig Count about to Enter Freefall Territory?
Barely a month ago, Wall Street analysts became concerned about a drop in the U.S. rig count due to weakening commodity prices, and in particular natural gas prices, and the impact of the credit crisis on producer access to capital. The talk initially was about whether the rig count would fall by 100-200 rigs or experience a more severe correction of 300-400 rigs. As soon as investors began to focus on the impact of the larger drop, market conditions worsened with crude oil falling to the $50 a barrel level in response to weakening global oil demand. Falling oil prices and a further tightening of credit markets caused analysts to begin to think that the rig count drop might be even greater than they had been thinking.
Recently, the view began to target the potential for a 600-rig drop during the seasonally weak early months of 2009. Last week the prospect of a 700-800 rig drop emerged as a possibility and as that view began to gain some traction, the oilfield service stocks began to dive. The Philadelphia Oil Service Index (OSX) fell Thursday by 11.06 points to 104.14, a decline of 9.6%. Intraday the OSX dropped 13.47 points to 101.73, a fall of 11.7%. It dropped further early in Friday’s trading (to the 98 level) as crude oil plunged into the $41 a barrel range.
The Baker Hughes domestic rig count peaked in September and then slumped some, although some of that fall off may have been related to the impact of hurricanes Gustav and Ike on oilfield activity. The rig count then rallied back to within 2% of the peak count before starting its current slide. In the prior two weeks (excluding the week ending December 5), the rig count has started to fall like a rock, declining by 51 and then 45 rigs. These drops were 2.6% and 2.4%, respectively, of the active rig count the prior week.
When we have written about the stock market performance of oilfield service stocks, we have focused on the similarity of their track record since the start of 2000 to that of the 1973-1983 period. We thought we would look to see how the Baker Hughes rig count fared in the same comparison. The result is presented in Exhibit 4.
It shows a very similar pattern -- the peak in the rig count has come almost exactly to the week of that historical period peak. The rig count did not rise as much as it did in the 1970s. We also indexed the rig counts to see how they compared. That chart is in Exhibit 5 and shows a similar pattern.
What we thought would be interesting was to see how this rig count correction would look if we applied the pattern of the 1982-1983 decline period. That forecast is displayed in red for each of the two exhibits -- the raw rig count and the indexed rig count (Exhibits 6 and 7). Based on the forecast, the rig count, which peaked in September at slightly over 2,000 working rigs would fall to a low of about 1,000 rigs, or approximately a 50% correction. That pattern suggests that the estimates of the magnitude of the rig count drop might prove to be conservative. On the other hand, the 1982-1983 rig count drop initially fell by over 2,000 rigs but then rallied by about 300 rigs. It subsequently fell by about 700-800 rigs before hitting bottom in 1983 and starting to head back up. The rig count eventually sank a lot lower in 1986 when Saudi Arabia waged its war against fellow OPEC members who were cheating on their production and crude oil prices fell to the $10 a barrel level, but that’s another story.
Based on our methodology, the rig count should decline by slightly over 1,000 rigs from the peak to the trough. Since the rig count has already fallen by 165 rigs, the 800-rig drop estimate recently made by Nabors Industries (NBR-NYSE) could prove pretty accurate, albeit not pretty. On the other hand there is a mitigating factor that
could help prevent the rig count from falling as far -- steeper natural gas production decline rates. Unfortunately, natural gas production continues to grow at a significantly stronger rate and forecasts call for more LNG coming to the U.S. as international gas producers will shift cargoes to here to support the gas pricing strength in Europe and Asia. More LNG along with growing domestic production does not auger well for natural gas prices in the future, which is important to support a higher drilling rig count.
While we are not sure where the rig count may bottom out, we still are taken aback by the idea that the timing of the two rig-cycle peaks appears so close, especially when one remembers that we are talking about a 25-year gap between them. We will watch the rig count in ensuing weeks to see whether our forecast proves accurate, but it appears for the near term the direction of the rig count is straight down.