Uncertainty and investment: Evidence from domestic oil rigs

• Published date: 01 February 2024
• Author: Asad Dossani,John Elder
• Date: 01 February 2024
• DOI: http://doi.org/10.1002/fut.22474

Received: 5 September 2023

|

Accepted: 25 October 2023

DOI: 10.1002/fut.22474

RESEARCH ARTICLE

Uncertainty and investment: Evidence from domestic

oil rigs

Asad Dossani|John Elder

Department of Finance and Real Estate,

Colorado State University, Fort Collins,

Colorado, USA

Correspondence

John Elder, Department of Finance and

Real Estate, Colorado State University,

Fort Collins, CO 80528, USA.

Email: john.elder@colostate.edu

Abstract

We provide new evidence on the response of investment to uncertainty, using

granular and high‐frequency (weekly) data on domestic oil drilling and oil

prices since 2012, corresponding to the period of widespread horizontal

drilling and hydraulic fracturing in the United States. Weekly data permits

much weaker identifying restrictions than is required with monthly data that

is common in the literature. We measure domestic drilling activity by the

number of rigs drilling for oil, and we measure oil uncertainty by implied

volatility from options on oil futures and the return on delta‐neutral straddles

from options on oil futures. We show that the number of oil drilling rigs are

tightly linked to both oil prices and oil uncertainty, and we find that oil

uncertainty significantly decreases the number of drilling rigs, with a one

standard deviation increase in uncertainty reducing the number of drilling rigs

by up to 5%.

KEYWORDS

drilling rigs, implied oil volatility, uncertainty

JEL CLASSIFICATION

C32, Q43, Q40

1|INTRODUCTION

The theory of real options developed in Majd and Pindyck (1987) views a firm's decision to invest in the presence of

irreversibility as an option, with a choice between investing immediately, or delaying investment until more

uncertainty about the evolution of the investment's payoff is resolved. The optimal decision rule suggests that firms

should delay investment until there is a significant difference between the investment's expected benefits and the

expected costs, at least equal to the option value of delaying. As uncertainty increases, the incentive to delay also

increases. The assumption that the investment has some element of irreversibility is necessary but not restrictive, since

it is common for investment decisions to have some irreversible component, as discussed in Dixit and Pindyck (1994).

At the macroeconomic level, a large body of research finds a countercyclical relation between various measures of

uncertainty and economic activity. For example, Bloom (2009) finds that equity market uncertainty leads to decreases

in industrial production; Elder and Serletis (2010) find that oil price uncertainty reduces GDP and investment; Jurado

J Futures Markets. 2024;44:323–340. wileyonlinelibrary.com/journal/fut

|

323

This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any

medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made.

© 2023 The Authors. The Journal of Futures Markets published by Wiley Periodicals LLC.

et al. (2015) show that macroeconomic uncertainty has large and persistent recessionary effects. Bansal et al. (2014)

show that asset prices tend to fall in response in uncertainty.

1

Empirical measures of uncertainty include various

measures, such as newspapers word counts (Baker et al., 2016) and econometric measures such as an average of the

volatilities of the residuals from a set of factor augmented regressions (Jurado et al., 2015) and the semivariances of

industrial production (Segal et al., 2015).

The countercyclical pattern of uncertainty is often attributed to a real options effect, although there is also evidence

that uncertainty may raise risk premia, as in Ioannidis and Ka (2021), and increase risk aversion resulting in lower

aggregate demand, as in Fernández‐Villaverde et al. (2015). Uncertainty related to oil prices has been particular

prominent, given the role of oil prices in macroeconomic activity. Bernanke (1983) shows uncertainty about oil prices

can generate macroeconomic fluctuations through a real options type effect. Gao et al. (2022) develop a two‐sector

model in which an increase in oil supply uncertainty induces firms to increase oil inventories, which curbs oil used for

production, leading to a fall in aggregate equity prices and general investment.

Empirical evidence on the effects of uncertainty at the microeconomic level is more sparse. One example is Moel

and Tufano (2002), who examine the behavior of Brazilian mine closings. They find mixed evidence of the effects of

uncertainty on investment, using a measure of uncertainty based on realized historical variance. Schaal (2017)

examines the effect of establishment‐level uncertainty on employment, while Elder (2019) estimates the effect of oil

price uncertainty on components of industrial production. In a related article, Kellogg (2014) finds oil price volatility

causes oil production to decline in a sample of selected wells in Texas before 2003.

In this paper, we analyze the relation between the decision to invest and the level of uncertainty, using granular

data on the weekly number of active onshore rotary drilling oil rigs in the United States, and contributing to both the

microeconomic and macroeconomic literature on uncertainty. Our drilling rig data is compiled by industry participants

and is cited often in the press. The decision to drill for oil represents an observable variable that is early in the

production process, providing a relatively tight link between the decision to invest and our observation of an active

drilling rig. This link has been especially tight since recent advancements in on‐shore oil exploration have greatly

compressed the production time frame, especially with firms maintaining an inventory of drillable sites.

Our primary measure of uncertaintyabout oil prices is implied volatility from options on oilfutures. Implied volatility

has several advantages over other measures of uncertainty. For example, implied volatility is market‐based,incorporating

information in prices that may otherwise be unknown to an econometrician. Implied volatility may also permit a better

match between the horizon of uncertainty withthe horizon of the economic decision, although Barrero et al. (2017)show

that firms react to both short and long‐run uncertainty with oil volatility contributing to uncertainty at both horizons.

Innovations in implied volatilitycan also be modeled as separate and distinct from an innovation in the level of oilprices,

unlike in some time‐series models such as generalized autoregressive conditional heteroskedasticity (GARCH). Implied

volatility is not without disadvantages, however. Measures of implied volatility are available for a relatively small number

of assets and commodities, and with more limited samples. Implied volatility also includes a risk premium, causing

implied volatility to differ from expected future realized volatility, as in Doran and Ronn(2008). As alternative measures

of uncertainty, we use GARCH and the return to delta‐neutral straddles from options on oil futures.

We make several contributions to the literature. First, we provide direct evidence on the effects of uncertainty on

investment decisions. As previously discussed, there is relatively little empirical evidence on this topic, with one

notable exception being Kellogg (2014). Kellogg (2014) analyzes the role of oil volatility on a select sample of wells in

Texas, terminating in 2003, by developing and estimating the parameters of a structural optimizing model derived from

economic principles. We build on Kellogg (2014), but impose less econometric structure than is associated with

estimating parameters of an optimizing model. Rather, we investigate the effects of uncertainty on oil exploration in

richly parameterized empirical models, which permit informative characterizations of the dynamics between

uncertainty and investment and controls for contemporaneous innovations in oil prices and oil demand. We focus our

sample on the period during which US onshore energy exploration has been dominated by the technological revolution

in the exploration for shale oil, typically via widespread horizontal drilling and hydraulic fracturing since about 2012.

Second, we show that our results are robust to various measures of uncertainty, including implied volatility from

options on oil futures, the return on delta‐neutral straddles from options on oil futures, and time‐series measures of

uncertainty such as GARCH. Results from delta‐neutral straddles suggest that investment in drilling rigs responds

more strongly to future implied volatility than realized volatility.

1

Related studies include Bachmann et al. (2013), Basu and Bundick (2017), and Carriero et al. (2018).

324

|

DOSSANI and ELDER