WEAI/AERE 2012 - Individual Paper Abstract
Title: Races to Extract in Oil and Natural Gas Production
Author(s): Andrew BALTHROP, Department of Economics, Andrew Young School of Policy Studies, Georgia State University, P.O. Box 3992, Atlanta, GA 30302-3992, abalthrop1 at gsu dot edu [Photo credit: NASA; Yates Oilfield, West Texas]
Abstract:
This paper employs recent spatial econometric techniques to quantify the spillovers in production and injection in oil and natural gas recovery in Slaughter field of West Texas. Oil and natural gas deposits are concentrated spatially in reservoirs, yet it is often the case in Texas that no single producer has rights to the entire reservoir. Because property rights to the oil and natural gas are administered according to the 'rule of capture,' ownership is not fully secured until the resource is extracted. While in the ground, it is an example of a common good: rival but nonexclusive. Under these circumstances the resources can be the subject of fierce competition as neighboring producers race to extract. Economic rents are dissipated in the drilling and operation of more wells than are needed to efficiently drain the reservoir. Additionally, owing to the dynamics of recovery, overly rapid extraction can result in damage to the reservoir and lowered cumulative recovery.
The spatial interdependence of oil and natural gas production brings the potential for economic inefficiency because competing producers discount the value of leaving resource in situ for future periods resulting in a 'race to extract.' The goal of this paper is to demonstrate how a race to extract can be prevented through unitary ownership. I use a spatial econometric model developed by Kelejian and Prucha (1998) and Kelejian and Prucha (1999) to explicitly characterize spillovers in production (the spatial autoregressive lag) while controlling for unobserved spatial autocorrelation (spatial autocorrelated errors), by using spatial weighting matrices to estimate a single parameter for each. Under some specifications the spillover parameter can be interpreted as a geological transmissivity coefficient , similar to models that have been used recently to estimate spillovers in production of fossil groundwater aquifers. The coefficient estimates allow for policy simulation of quotas, well spacing and ownership concentration.
The model is estimated using extensive and novel data provided by HPDI Corporation. Previous literature has found Slaughter field to be a field of interest because it is one of the largest in the US in terms of proved reserves, and to have a history of common pool problems. The main result is as expected: when neighboring wells are under unitary ownership, the spillovers in production are insignificant and extraction proceeds at a comparatively slower pace than when wells have competing operators. These results are consistent across a variety of spatial weighting specifications after controlling for spatial fixed effects and injection.