The NYMEX sold off hard in the wake of last week’s EIA report. Here at The Schork Reportwe switched our bias to bearish on the previous session, thus we are pleased. But as we analyze in today’s issue, we are not 100% sure this report is as bearish as Thursday’s price path seemed to indicate.
Key takeaway:
Unlike the weekly DOE/ API oil report, which the market consensus never gets right, the weekly EIA gas report usually falls within a reasonable tolerance of the market’s guess. It is therefore a concern when the consensus flubs this report; as it now has in three of the last four reports.
For example, per estimates off of Bloomberg, the consensus survey has forecasted a combined delivery of 540 Bcf for the last four EIA reports. The actual delivery was 561 Bcf. Therefore, upon first glance it would appear the consensus undershot the actual by 21 Bcf or just 3.7%. However, first looks can often be deceiving.
In reality, the consensus has missed the mark by 75 Bcf in absolute terms; through the first two reports the consensus underestimated the delivery by a combined 48 Bcf and then overestimated the last two deliveries by 27 Bcf. That works out to an average miss of 19 Bcf per week with a 9.5 Bcf standard deviation.
For the corresponding four weeks from a year ago, the consensus missed by a combined 44 Bcf or an average miss of 11 Bcf per week with a standard deviation of only 5.6 Bcf. Furthermore, you can make an argument that gas deliveries last year were harder to gauge given that the global economy was in an absolute freefall, yet the accuracy of the consensus surveys then was greater.
Therefore, current storage models have yet to identify the fundamental shift currently underway in the gas market, i.e. the advent of nonconventional onshore production impacting the supply side.
While weather-related demand is relatively clear, the supply side is not. The market knows we have a lot of molecules stuffed into the ground, but the market also witnessed an unprecedented drain of molecules last month. We have only seen four deliveries thus far this season, as opposed to eight at the start of a typical season. However, those four deliveries have been massive, 20.04 Bcf/d as opposed to last year’s 17.2 Bcf/d average for the same four weeks and the 16.5 Bcf/d average for the five seasons prior to that.
On the other hand, over the first four weeks of this season we saw four straight injections totaling 49 Bcf or 1.8 Bcf/d. That compares to last year’s 1.9 Bcf/d average delivery and the 3.8 Bcf/d average delivery for the 2004-08 timestep.
To confuse matters more, marketed production, which dropped by a seasonally high 3.17 Bcf/d in September, bounced back by 3.03 Bcf/d in October. The EIA noted that Louisiana reported strong numbers for October on extant drilling in the Haynesville shale and Wyoming production surged (+7.9%) to levels not seen since June “… as wells shut-in due to unfavorable market conditions were brought back on line…” On the other hand, the decline in production last September was noteworthy given the limited number of hurricane related shut-ins to production. For instance, the 2009 Atlantic hurricane season was the first since 2006 when hurricane force winds failed to reach the United States. Yet, the decline in production in 2009, 3.17 Bcf/d was nearly 2½× greater than in 2006, 1.3 Bcf/d. To be fair, the industry was still scrambling in the fall of 2006 to recover production lost to Hurricane Ivan in 2004 and the 2005 season, the most active Atlantic hurricane season on record.
Bottom line, the sharp production swing from last August to September to October has put the U.S. gas market on notice, i.e. producers might have attained their dy/dx moment. That is to say, it took them all of 2009 to figure it out, but they now have a better grasp on the correct proportion between the growing of production and the growing of output. That is important. As discussed an analyzed in many past issue of The Schork Report, last year was brutal for a lot of producers. They are obviously not in the mood to see a repeat this year.
_________________________
Stephen Schork is the Editor of, "The Schork Report"and has more than 17 years experience in physical commodity and derivatives trading, risk systems modeling and structured commodity finance.
Data also provided by 
