Impact of Low Prices on Shale Gas Production Strategies.

• Author: Ikonnikova, Svetlana
• Position: Report

1. INTRODUCTION

   In 2014, shale gas accounted for approximately one-third of U.S. natural gas production, up from a negligible contribution just a decade earlier. A base case estimate, presented by the Energy Information Administration in 2014, suggests that U.S. shale gas production could further increase and exceed 50% of domestic gas production by 2040 (EIA, 2014). Other studies present similar expectations (WoodMackenzie, 2014). At the same time, there are concerns regarding the ability of shale production to grow in an environment of persistently low natural gas and, now, oil prices and declining rig counts.

   Total U.S. dry shale gas production has been growing despite the decline in the number of rigs drilling for natural gas from close to 1,600 in September 2008 to about 880 in 2011, about 330 in late 2014 and about 220 by June 2015 in response to falling prices from almost $14/MMBtu in 2008 to about $4 in 2011 and $2.5 in April 2015 (Fig. 1). This trend has been widespread, though not universal, as some plays have experienced proportionally larger declines in drilling (e.g., Haynesville, with the highest cost of drilling) than others (e.g., Marcellus). The collapse of oil prices since October 2014 (roughly 50%) led to a dramatic decline in the number of rigs drilling for oil from roughly 1,600 in October 2014 to about 630 in June 2015. However, oil production has started showing some marginal decline of 2-3% per month since May 2015. (1)

   This apparent contradiction is often explained by operators focusing on better locations (known as "high-grading"), improved drilling techniques, and increased well productivity. Also, operators shifted rigs to more productive or liquid-rich areas within and across plays. Thus, production from higher-productivity wells, although drilled in smaller numbers, and associated gas from liquids and oil locations, have sustained shale gas production growth. While it is true that producers have advanced the technology and managed to drill cheaper wells more quickly, the argument about productivity may not necessarily hold true. Some operators, in response to lower prices, focused on lowering costs by using less inputs and drilling infill wells, taking advantage of existing infrastructure. Although, infill wells tend to produce less than the original wells in the same location, the reduction in costs suggests commercial viability. Looking at the individual well production across the Barnett, Fayetteville, Haynesville, and Marcellus plays, we find that productivity in the first three has been declining even in the best areas. Hence, a better understanding of drilling practices and well economics is required to explain the current trends and to predict future developments in the shale industry.

   While the engineering literature offers a broad range of studies on optimal completion practices, (2) including infill drilling, there has been almost no formal analysis that explores whether there is an economic basis for such practices based on the input costs and resulting production. The purpose of this paper is to explore how production practices could change during low-price periods, possibly allowing producers to maintain their production levels. In particular, we focus on infill drilling and how it can change play development dynamics and affect long-term resource recovery. We extend the previous studies on shale gas well profitability (Lake et al., 2013; Kaiser, 2012;

   Gulen et al., 2013; Ikonnikova et al., 2015a) and show the roles of resource scarcity and product pricing on drilling decisions and therewith on final resource recovery.

   As compared to previous analyses on shale production, we benefit from a longer production history and more comprehensive data. Namely, we add data on water used in hydraulic fracturing, provided by FracFocus. (3) Water usage provides a proxy for hydraulic fracturing costs in the absence of detailed data on proppant, chemicals and the number of fracturing stages. In part, the reduction in water usage per well can be explained by the reduction in the distance between wells, referred to as "spacing." While investment costs are reduced through reduced water usage and closer spacing in existing drilling locations, inventory is expanded through infill drilling. Hence, potentially, final recovery of the entire play could change.

   First, we highlight the most important trends in shale gas production, paying special attention to changes in per-well production and per-well costs and water usage. Then, we present a theoretical model to frame the discussion and address questions posed in section 1 such as optimal spacing and use of water for fracturing. Finally, we use real-world observations in an attempt to evaluate the economic rationale of the reviewed developments. We conclude with a discussion of the implications of our analysis.

2. EXISTING MODELS AND NEW DEVELOPMENTS

   Commercial viability, or profitability, of shale gas production, similar to profitability of any other natural resource production, depends on resource availability (geology), productivity (rock quality and technology), extraction costs (technology and completion practices), and netback (or wellhead) prices (markets for methane as well as natural gas liquids, in general, and availability and capacity of midstream and downstream infrastructure such as pipelines and processing facilities, in particular).

   It is important to understand how these four factors of supply come together in modeling production scenarios. First, the boundaries of a play combined with local (county or square-mile average) estimates of original resource in place (ORIP) yield the upper limit of extractable resource. Second, well productivity reflects resource extraction and technical efficiency. The ORIP estimate is essential for preventing the overestimation of potential production by mistakenly factoring in overly-aggressive technological improvement; per-well productivity must be limited by ORIP. Third, extraction costs normalized to well completion design and capital costs allow estimation of the return on investment, or when combined with wellhead prices, profitability. Finally, profitability serves as a proxy for producers' willingness to drill. The drilling pace is subject to market conditions and price signals, drilling location and infrastructure availability, producers' budget constraints, and competition with other investment opportunities. Hence, a change in either of the four factors would inevitably lead to change in the entire production outlook.

   Owing primarily to a lack of data on completion specifics by company or location and relatively short history of drilling in most plays, the majority of the literature analyzing shale gas production from major shale plays had to assume standardized completion practices (Browning et al., 2013; EIA, 2014; Gulen et al., 2014, 2015; WoodMackenzie, 2014; Ikonnikova et al., 2015b). In such analyses, two choices are made: (1) where to drill, given that locations are differentiated by standardized well productivity, and (2) how many wells of given productivity to drill. Recent developments indicate that these assumptions may not hold true across all plays and locations. A review of emerging drilling practices suggests that a completion choice, particularly water use as a proxy to fracturing approach and spacing, can play a significant role in a play's development outlook if it can be sustained and implemented more universally.

   Until recently, producers in many U.S. States were not obliged to report all the details regarding well completion, including quantity and type of water used in hydraulic fracturing. Well length was the only readily available piece of information, and it has been used to normalize well production and to approximate producer costs. Per unit cost appeared to be declining as production grew along with well length. But, for the most part, a smaller number of longer wells were being drilled in high OGIP areas (Fig. 2). (4) Without knowing more about completion practices and investigating incremental costs and revenues, it is difficult to conclude whether technological advances were improving production efficiency beyond geology.

   In times of price fluctuations, however, producers are driven to adjust their production. Classic microeconomic theory suggests that not only will producers change the quantity produced, but also the combination of inputs owing to potential changes in capital budgets and input costs. Hence, we should expect changes in the number of wells drilled; the combination of water, proppant, and other ingredients used in completion; and completion design (e.g., sequencing and the number of fracture stages). While there are reports on the decline in the number and productivity of wells in some shale-gas--producing regions, a detailed discussion of the causes and justifications of this decline is missing from the literature. We discuss some reasons for the decline in productivity and the implications of that practice in terms of shale gas development and production forecasting.

   We start by providing an example of infill drilling pattern from a high-productivity area in the Barnett play (Fig. 3). The example illustrates how a well pad, originally perceived to be exhausted in terms of drilling locations, has been reused to expand the inventory of wells and to increase resource recovery from the area. The original wells were drilled in 2009 with an average estimated ultimate recovery (EUR) of over 4 Bcf each (Fig. 3a, lighter lines). In 2013, an operator returned to the site and drilled five more wells among the original wells, cutting the spacing between them from about 900 ft to 400 ft or less (Fig. 3a, darker lines). Completing new wells (with hydraulic fracturing), the operator used significantly less water per well than the original well completions (Fig. 3b). The result was a significant decrease in the production of new wells...