Shale Plays - Good Economics or Irrational Exuberance? June 7th 2010 J. Cameron Bailey, CFA Making Sense of Shale Plays After witnessing natural gas prices of over $9.00 per mcf and more than 1,600 natural gas directed rigs operating in June 2008, to a low $2.00 per mcf natural gas and natural gas directed drilling falling to 680 active rigs in the United States, you would logically expect a significant change in natural gas supply and a corresponding price response. Although a correction did occur in late 2009, it was modest and short lived, leaving operators in North America scrambling to reconstruct investment portfolios to more economic projects – oil and resource plays. The main culprit to the lack of expected recovery has been the significant supply of gas coming from horizontal drilling of US shale plays. There has been significant commentary and opinions about the potential economics of shale plays to justify operators’ claims of robust type curves, but little in the way of hard facts to demystify the underlying fundamentals driving operators’ behaviors. These questions are extremely important to Canadian producers when we are pushing molecules of methane into a pipeline from remote areas of northern Alberta and BC, while US producers are producing the exact same molecules of methane into a pipeline in closer proximity to 300 million consumers. The perversity really comes to light when an mcf of gas located in tight shale formation in the US commands many multiples of value compared to an mcf of gas located in sand reservoirs 1000 times more permeable in northern Canada. Canada has always had the perception of being the lowest cost source of supply, but this ‘conventional wisdom’ may have changed in North America, at least temporarily. The behavior of producers has everything to do with “Ebbs and Flow of Capital” and much less to do with the good sound economics. The following analysis evaluates some hard facts that help in understanding the question – Do Gas Shale Plays Make Sense? Shale Plays Dissected How Much Do You Pay for Growth? Barnett Shale Experience. From 2004 to 2009 Chesapeake was able to go from a stand-still to nearly 1 bcf/d of gross production (660mcf/d net) from robust drilling activity in the Barnett Shales. An absolutely remarkable achievement considering this came from a very modest land position of 200,000 net acres. This production growth represented 34% of Chesapeake’s growth over this short period. It is very difficult to suggest another play type anywhere in the world that could provide this kind of production growth. During this time, Chesapeake drilled approximately 1,100 net wells at a cost of approximately $3.4mm per well totaling $3.7 billion of investment (not including land cost). Did they make money? That depends on your definition… The on-stream costs were $33,000 per boe/d, great by anyone’s definition of making money, however almost 50% of the production comes from wells drilled in the past year with decline rates estimated to be 65% in the first year and two thirds from wells drilled in the past two years. This production profile is derived from 1,100 separate investment decisions related to drilling each and every well. The average production profile of Chesapeake wells drilled over this time is as follows: Wells drilled between 2004 and 2008 had an average peak monthly rate of 1.8 mmcf/d. Wells drilled in 2009 and 2010 have reached a peak monthly rate of 2.2mmcf/d. The rest of the industry’s average wells look as follows: Of the 9,500 wells drilled by the industry between 2004 and 2008 that were placed on stream, the average well peak monthly rate was 1.5 mmcf/d. Wells drilled in 2009 and 2010 have reached a slightly higher peak monthly rate of 1.7mmcf/d, lower than the average of Chesapeake’s, clearly indicating where to drill is important. Based on these observations we can determine the average decline rates to be as follows: Expected High Year One 61% 61% Year Two 34% 34% Year Three 24% 24% Year Four 16% 17% Year Five 11% 17% Thereafter 7% 17% The decline rates indicated in red are calculated from the observed rates and the later years are forecasted. Since very few wells actually have production histories past three years, I have run a sensitivity using higher assumed decline rates in the later years. The following is the production profile and typical well in the Barnett and the cumulative production. As can be seen, based on a $5.50 per mcf gas price it takes approximately 93 months for payout to occur resulting average production of 200 mcf/d per well on payout, then declining at an assumed rate of 7% per annum to generate a positive return. Of the 13,000 Barnett wells identified by the Rail Road Commission of Texas, only 601 or 4.6% have reached estimated payout producing greater than 1.1 bcf of cumulative production. Of the estimated $30 billion of investment in the Barnett shales, it is shocking that only 4.6% of the wells have reached payout, although 40% have achieved a peak production rate of 1.7mmcf/d or better. The bottom line is that there is a high probability of achieving very robust per well production rates in every well drilled targeting the Barnett. The concern revolves around the back end of the production years which drives estimated ultimate reserves, finding and development costs and rates of return. If it turns out the decline rates in the later years are higher than estimated, then the average well will only recover 1.4 bcf versus 1.9 bcf and the well does not achieve payout. It appears that there is not enough production history to confidently predict the ultimate recoveries. However Chesapeake and Devon, the two largest operators in the Barnett, have each added a 1.0 bcf/d of production in five years and spent billions in the process. Today Chesapeake has 24 rigs operating in the Barnett shales, not necessarily because the investment decisions make money, but because they must, otherwise production would begin falling quickly. I have assumed that Chesapeake drills 23 wells per month with 24 rigs operating and each well comes on-stream at 2.0 mmcf/d and the annual decline is 65% per month. The following is Chesapeake’s forecast production profile in the Barnett over the next twelve months: Chesapeake must keep 24 rigs active in the Barnett area or its production would fall dramatically and the stock price right along with it. Haynesville Shales The Haynesville shale is deeper and with higher associated pressures than the shallower Barnett shale and IP rate experienced is much higher. The average well cost is approximately $7.5 to $9.0 million for drilling and completion. The Haynesville drilling has been concentrated in an area 50 miles by 25 miles or approximately 1.0 mm acres. To date there are approximately 1,150 wells drilled into the Haynesville shale of which 758 or 65% have reached commercial qualities of gas. Of the 758 wells now on stream, the average peak rate of production is 6.5 mmcf/d but increasing well performance has been achieved with greater number of frac stages being applied over the past few years. Chesapeake, the most active operator in the Haynesville, has 22 rigs operating and has 169 wells currently producing. The average production from the wells is as follows: The average peak production for wells drilled over the past three years is 7.2 mmcf/d declining 82% in the first year of production and 66% in the second year. These results like the Barnett are better than industry average, so again, where to drill is important. For the purpose of generating economics and a well production profile I have used the following assumptions and plotted the actual wells performance that have been drilled and placed on stream. The decline rates used are as follows: Normal High Year One 81% 81% Year Two 47% 47% Year Three 22% 24% Year Four 12% 24% Year Five 7% 24% Thereafter 7% 24% The decline rates indicated in red are calculated from the observed rates and the later years are forecasted. Since very few wells actually have production histories past one year, I have run a sensitivity using higher assumed decline rates in the later years. I have used $9.0mm per well and $5.50 well head price and an average one month IP rate of 8.5mmcf/d. It takes approximately 82 months to payout a Haynesville well and it generates a modest 8.9% rate of return. Chesapeake has 22 rigs operating in the Haynesville shale and considering it is early days, we will expect to see a very significant growth in production leveling out in 15 months unless more rigs are deployed to drill. Over 12 months Chesapeake will spend approximately $1.6 billion and will have 800 mmcf/d of production, albeit declining by more than 80% per annum. The cost of a boe per day of production – a mere $12,000 per boe/d. Spectacular investment metrics by anyone’s standard, but hard to conclude from this one measurement alone that the investment makes money with a very modest 8.9% rate of return. In addition, none of the economics calculated take into account failed or abandoned wells; only wells that have made a commercial quantity of gas have been used in the statistical sample. Conclusion Do Shale Plays Make Money? Investors pay for production growth and production growth means increasing cash flow. The disconnect with shale plays comes when we examine the cost of growing production and the unsecure footing that production growth has built. When we examine the individual decision of drilling a shale well, I am hard pressed to see many wells making payout and generating a positive rate of return unless natural gas prices increase significantly. Well production histories seem to be suggesting lower ultimate recoveries and therefore lower rates of return and higher finding costs than the claims made by operators. As long as the growth rates are maintained which means rig activity must also increase, the music will play on. Irrational Exuberance = Natural Gas Bubble.
Pages to are hidden for
"Shale Plays - Good Economics or Irrational Exuberance"Please download to view full document