Breakeven points are defined and partitioned. Their dynamics are explained.
Sources of tight oil market inelasticity are explained.
Tight oil production has little effect on short-term market stability.
Tight oil production can stabilize markets in the medium term.
R.L. Kleinberga, S.Paltsevb,C.K.E.Ebingerc,D.A.Hobbsd, T. Boersmae
aSchlumberger-Doll Research, One Hampshire Street, Cambridge, MA 02139, United States. bJoint Program on Science and Policy of Global Change, Massachusetts Institute of Technology, Cambridge, MA 02139, United States. cAtlantic Council, 1030 15th Street NW, Washington, DC 20005, United States. dKing Abdullah Petroleum Studies and Research Center, Airport Road, Riyadh 11672, Saudi Arabia. eCenter on Global Energy Policy, School of International and Public Affairs, Columbia University, 420 West 118th Street, New York, NY 10027, United States
Received 20 December 2016 Accepted 25 November 2017
When comparing oil and gas projects – their relative attractiveness, robustness, and contribution to markets – various dollar per barrel benchmarks are quoted in the literature and in public debates. Among these benchmarks are a variety of breakeven points (also called breakeven costs or breakeven prices), widely used to predict producer responses to market conditions. These analyses have not proved reliable because (1) there has been no broadly accepted agreement on the definitions of breakeven points, (2) there are various breakeven points (and other benchmarks) each of which is applicable only at a certain stage of the development of a resource, and (3) each breakeven point is considerably more dynamic than many observers anticipated, changing over time in response to internal and external drivers.
In this paper we propose standardized definitions of each breakeven point, showing which elements of field and well development are included in each. We clarify the purpose of each breakeven point and specify at which stage of the development cycle the use of each becomes appropriate. We discuss in general terms the geological, geographical, product quality, and exchange rate factors that affect breakeven points. We describe other factors that contribute to tight oil market dynamics, including factors that accelerate the growth and retard the decline of production; technological and legal influences on the behavior of market participants; and infrastructure, labor, and financial inelasticities.
The role of tight oil in short-term and medium-term oil market stability is discussed. Finally, we explore the implications of a broader, more rigorous, and more consistent application of the breakeven point concept, taking into account the inelasticities that accompany it.
From 2011 to mid-2014, Brent crude oil generally traded above $100 per barrel (1 bbl = 0.159 m3). During that period, U.S. crude oil production increased from about 5.5 million barrels per day (bbl/d) to about 8.9 million bbl/d. Most of the increase was due to the growth in production of tight oil, which is often erroneously termed “shale oil” (as explained in Kleinberg, forthcoming) but is correctly defined by the U.S. Energy Information Administration as oil that is produced from rock formations that have low permeability to fluid flow (EIA, 2016i).
Tensions among oil producers, which originated in the oil price collapse of the mid-1980s, have weakened the ability and willingness of the Organization of Petroleum Exporting Countries (OPEC) to act as an oil market stabilizer (McNally, 2015). By the third quarter of 2014 it had become apparent that the rate of increase of supply of U.S. tight oil had significantly outstripped the rate of increase of worldwide demand, leading to persistent increases in the amount of oil sent to storage, see Fig. 1. This was an unsustainable situation. In light of the tight oil boom, numerous publications declared America to be the world’s marginal producer (e.g., The Economist, 2014), and when oil production had to decrease, it seemed that burden would fall on the U.S. tight oil industry, whose per barrel costs were far above those of Middle East, and most other, producers.
Fig. 1. The growth of United States tight oil production (upper curve) (EIA, 2016i) upset the global balance between supply and demand, leading to persistent additions of stored oil after early 2014 (lower curve) (EIA, 2016g).
Many analysts suggested that the oil price needed to maintain the economic viability of the preponderance of U.S. tight oil projects – the breakeven point – was in the range of $60/bbl to $90/bbl (e.g., EY, 2014, Wood Mackenzie, 2014c, Bloomberg, 2014). It was further widely believed that once the oil price fell below $60/bbl, many investments in tight oil projects would end and “since shale-oil [sic] wells are short-lived (output can fall by 60–70% in the first year), any slowdown in investment will quickly translate into falling production” (The Economist, 2014). Thus the $60–$90 range for the U.S. tight oil breakeven point was thought to act as a shock absorber, with tight oil projects quickly coming onto production as prices increased, and dropping out of production as prices decreased through this range.
With tight oil accounting for roughly 4% of global production, and seemingly able to respond to price signals considerably faster than conventional projects, analysts predicted that this new resource could bring welcome stability and price support to oil markets (see e.g. IHS, 2013a, Krane and Agerton, 2015, Ezrati, 2015, The Economist, 2015). There is no documented evidence that the Organization of Petroleum Exporting Countries acted on these assessments, but we can speculate that these considerations might have influenced their decision late in 2014 to preserve their share of the international oil market by increasing oil production.
If the conventional wisdom were to hold true, moderate increases of OPEC oil production, accompanied by a moderate oil price decline, would result in prompt declines of tight oil production, thereby preserving both OPEC market share and profits.
In reality, markets did not respond to a modest increase of supply as smoothly as had been predicted. The West Texas Intermediate benchmark oil price fell from $108/bbl in mid-2014 to $32/bbl in early 2016, well below tight oil minimum breakeven points calculated by energy economists. Moreover, tight oil production did not start to decline until mid-2015, when it started falling at a moderate rate in the Bakken region, see Fig. 2a, and more rapidly in the Eagle Ford region (EIA, 2016k). Remarkably, oil production from the Permian Basin continued to increase through 2016, see Fig. 2b. As OPEC reported in October 2016, “… the resilience of supply in the lower oil price environment caught the industry by surprise, particularly tight oil in North America.” (OPEC, 2016).
Fig. 2. a. A sharp decline in Williston Basin oil-directed rig count, which is dominated by Bakken field activity (dotted curve) (Baker Hughes, 2016), followed a drop in WTI crude oil price (lower solid curve) (EIA, 2016m) with a lag of less than three months. Bakken oil production (upper solid curve) (EIA, 2016k) started falling in mid-2015. b. As in the Williston Basin, the Permian Basin oil-directed rig count (dotted curve) (Baker Hughes, 2016) swiftly followed the decline of WTI crude oil price (lower solid curve) (EIA, 2016m). However, oil production (upper solid curve) continued to increase slowly (EIA, 2016k), defying expectations.
The industry was “caught by surprise” in part because the dynamics of breakeven points were not broadly understood. The effects of other market drivers were also incompletely understood, including factors that accelerated the growth and retarded the decline of tight oil production. Technological, legal, infrastructure, labor and financial influences must also be considered. The goal of this paper is to provide a consistent methodological approach to understanding the costs of oil production, and to show, in a systematic way, how those costs change with time and circumstances.
We analyze the various breakeven points and other benchmarks, show how they are calculated, and point out how they can sometimes provide misleading signals to analysts and markets. We also explore the difference between the decline rates of a single well and a field, and remark on other inelasticities inherent in the production of crude oil in general and tight oil in particular. Finally we remark on how tight oil influences short-term and medium term market stability.
When evaluating the economic viability of a resource or project, one of the most commonly used economic concepts is benchmarking. We discuss how various benchmarks are appropriately used. When comparing projects, companies may wish to prioritize short term cash flow per dollar of investment, reserve additions per dollar, or the robustness of project economics to price declines. In the latter case, the most commonly used measure is the “breakeven point”, also called breakeven cost or breakeven price.
The breakeven point is the combination of project costs and market prices for which the net present value of a project is zero (Brealey et al., 2009). In this paper the breakeven concept is analyzed as follows. We start with the definitions of breakeven points; in many publications they are presented without adequate disclosure of what exactly is meant by breakeven. While we realize we cannot promulgate rigorous definitions by fiat, in this paper we offer definitions we believe to be in the mainstream of analyst and corporate practice; the proposed scheme can and should be modified according to individual circumstances. We discuss how breakeven points are partitioned, and when the various breakeven points are appropriately used. We show how breakeven points change with time, due to internal and external drivers.
We discuss other inelasticities that accompany expansions and contractions of output. To address a misconception of fast decline of tight oil production, we provide a simulation that contrasts individual oil well declines with the collective declines of conventional and tight oil fields. Finally, we assess how a misreading of breakeven points, and lack of insight into the ways in which companies use benchmarks to prioritize investment, may have contributed to the sudden, unexpectedly large change of oil prices in 2014–2016. Although this paper is couched in terms of oil markets, the same principles apply to natural gas resources, and to some extent to other commodities.
Oil market dynamics
Investments in fossil fuel production constitute a multitrillion dollar part of the global economy (IEA, 2014). The largest single segment is occupied by crude oil, which in 2015 provided about one-third of global primary energy use (BP, 2016). Not only is oil consumed at a high rate – roughly a thousand barrels per second – but the demand for it is relatively inelastic (Labandeira et al., 2016). This means demand is relatively insensitive to price.
Conversely, a small but persistent imbalance between demand and supply – sometimes as little as 1% of total production – can result in dramatic price changes. Moreover, long lag times inherent in large, risky, capital-intensive exploration and development projects cause substantial, long-lived supply overshoots. Thus the oil price collapse of 2014–2016, when West Texas Intermediate benchmark crude oil prices fell by 70%, was accompanied by substantial increases in production from long‑lead-time projects in the U.S. Gulf of Mexico (EIA, 2016b) and elsewhere. These were not unprecedented events.
Also contributing to market instability is the complication that a barrel of oil with a relatively high cost of production can enter the market before another barrel that can be produced more cheaply. It is true that the lower the cost of the resource, the more likely it is to be exploited by a producer who holds a range of resources, and lower-cost resources present less risk of loss in the event of a decline of market price. However, dispersal of resources among a wide variety of independent actors, as in the United States, implies that oil and gas resources are not developed in seriatim order of cost.
If oil sells for $100/bbl, the small producer with costs of $90/bbl will sell as much as possible, regardless of lower-cost resources owned by others. Thus, given a range of producers acting independently of each other, any resource with a marginal cost of production below the prevailing market price can be produced. It was this reality that enabled the creation of the shale gas and tight oil industry in the United States (Wang and Krupnick, 2013). The extensive experimentation that led to the commercialization of Barnett shale gas would never have occurred if left to commercial entities each of which had a wide variety of resources to exploit.
It is in this context that the advent of abundant North American tight oil resources, brought to market by horizontal well construction and massive hydraulic fracturing, was believed to be a market stabilizer (Maugeri, 2013). Unlike deepwater and Arctic projects, for which lead times are typically a decade or more, a tight oil well can be planned, drilled, and completed in months. Furthermore, unlike wells in conventional reservoirs, which decline at around 6% per year (IEA, 2013) and continue producing for decades, tight oil wells typically decline by about 60% in the first year and 25% in the second year of production (IHS, 2013b), see Fig. 3.
As a result, nearly half of Lower 48 U.S. oil production in 2015 had originated from wells drilled since the start of 2014 (EIA, 2016d); much of this new production came from tight oil plays. To maintain tight oil production at a constant level, wells must be drilled and completed at a rate beyond that required in conventional fields, a phenomenon colorfully called “The Red Queen Race” (Likvern, 2012). Thus it had been thought that tight oil production would follow the price of oil with a short time lag.
Fig. 3. Type curves for oil production from individual conventional wells with an annual rate of decline of − 6%, and from individual tight oil (Bakken) wells (IHS, 2013b).
The oil market developments of 2014–2016 in some respects confirmed these views, and in other respects contradicted them. In response to the rapid decline of oil prices after June 2014, U.S. rig counts in tight oil plays declined rapidly, following falling oil prices with a lag of two to three months, as expected for this very nimble industry. Tight oil production peaked in the Eagle Ford play in March 2015 (EIA, 2016k), a lag of nine months, and it peaked in the Bakken play (Fig. 2a) in June 2015, a lag of twelve months. In the Permian Basin, tight oil production continued to increase, as shown in Fig. 2b. Production from these regions was sustained by the relatively slow decline of a substantial number of legacy tight oil wells, by improvements in rig productivity (EIA, 2016k), by reduced costs of oil production (EIA, 2016c), and by a dynamic redefinition of breakeven point. We discuss each of these factors below.
Cost per unit productive capacity
When companies compare projects to choose those in which they intend to invest, the benchmarks they use depend on their corporate priorities. One is the cost per unit of productive capacity. The cost of productive capacity is of particular interest to oil market forecasters trying to relate changes in capital expenditures to likely levels of future supply. The crude oil market does not care whether the barrels supplied made profits for their producers, only that they are available. Capacity is added both to accommodate increasing demand for petroleum and to compensate for the natural decline of mature fields. Recently an average of 5 million bbl/d of new capacity has been added each year, at a cost of more than $500 billion: $100,000 per barrel per day. Therefore it might be expected that a cut back of $100 billion in capital expenditures would reduce production capacity in the future by 1 million bbl/d. However, these forecasts are complicated by the fact that the impact could be spread over multiple years, e.g. as reductions of 200,000, 300,000, and 500,000 bbl/d over a three year period.
Depending on companies’ view of future prices, they might favor one investment over another, even at the expense of damaging the ultimate value of a resource, because they need to meet debt covenants or other factors that are influenced by net operating cash flow. In the market example above, it is quite possible that the projects that are cut are the ones with above average costs of capacity and thus the expected aggregate cutback would be less than 1 million bbl/d.