Oil “Rents” and Political Development: What Do We Really Know About the Curse of Natural Resources?

• Author: Sarah M. Brooks,Marcus J. Kurtz
• Published date: 01 September 2022
• Date: 01 September 2022
• DOI: http://doi.org/10.1177/00104140211060281
• Subject Matter: Articles

Article

Comparative Political Studies

2022, Vol. 55(10) 1698–1731

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00104140211060281

journals.sagepub.com/home/cps

Oil “Rents”and Political

Development: What Do

We Really Know About

the Curse of Natural

Resources?

Sarah M. Brooks

1

and Marcus J. Kurtz

1



Abstract

Research on the political implications of oil wealth is now over two decades

old, and yet still unresolved. Despite many advances in this literature, a core

concept—resource rents—remains under-theorized, and even less well

measured. This mis-measurement, we contend, has direct and signif‌icant

consequences for theoretical f‌indings on the resource curse. Although rents

refer to excessive prof‌its, analyses to date have not directly or indirectly

focused on the prof‌itability of oil extraction; instead they rely more often on

estimates of gross receipts or exports of natural resources. The largely

untested assumption that oil sales consistently produce enormous prof‌its has

been the bridge used to treat these as measures of rents, even as the empirical

foundations for this claim are shaky. In contrast, we measure the diff‌iculty of

oil production using data from over 3800 major oil f‌ields around the globe

between 1980 and 2012 to more accurately estimate the income generated by

oil production and to calibrate its rent-generating potential using important

correlates of its cost of production. The results show that where oil’s cost of

production is low, and its prof‌itability high, pernicious regime outcomes are

observed in keeping with resource curse claims. However, as production

shifts to more challenging sources, oil production loses this association and

1

Ohio State University, Columbus, OH, USA

Corresponding Author:

Marcus J. Kurtz, Department of Political Science, Ohio State University, 154 N. Oval Mall, 2140

Derby Hall, Columbus, OH 43210-1373, USA.

Email: kurtz.61@osu.edu

may well at the extremes be associated with modestly more democratic

outcomes. In other words, it is only “easy oil”that may be a curse for

democratic development.

Keywords

resource curse, democracy, political development, oil

Introduction

More than a quarter century after Sachs and Warner’s seminal research

launched a discussion about the relationship between natural resource en-

dowments and slow economic growth (Sachs and Warner, 1995,1999) a rich

literature in political economy has emerged examining the “resource curse”in

democratic development. Since that time, scholars have linked substantial

natural resource wealth to myriad domestic political ills, from corruption and

administrative bloat, to civil conf‌lict, pervasive clientelism, and the persis-

tence of authoritarianism (e.g., Collier & Hoeff‌ler, 1998;Dixit, 2007;Ross,

1999,2001,2013;Sokoloff & Engerman, 2000;Torvik, 2001,2002).

1

This

has been especially the case for one particular (and particularly important)

natural resource: oil. And it takes only a brief glance at the political char-

acteristics of the petro-states of the Middle East and Central Asia for the prima

facie linkage between oil and authoritarianism to become apparent.

At the same time, cracks have begun to emerge in the once-regnant

consensus linking oil to pernicious and authoritarian politics. Indeed, a re-

cent wave of resource curse scholarship has suggested that the negative effects

of oil wealth are instead conditional—on public versus private ownership

(Luong & Weinthal, 2010), pre-existing levels of administrative capacity

(Karl, 1997), human capital (J. P. C. Stijns, 2005,2006), institutional quality

(Bulte et al., 2005;Mehlum et al., 2006;Robinson et al., 2006;Snyder, 2006;

Torvik, 2009), local context (Brollo et al., 2013;Diaz-Rioseco, 2016;

Dunning, 2008;Gervasoni, 2010;Goldberg et al., 2008;Orihuela, 2018), and

international context (Hendrix, 2018). Others have suggested that while

natural resource wealth does not directly induce authoritarian political de-

velopment, natural resources instead stabilize whatever the incumbent regime

type might be (Morrison, 2009;Smith, 2007,2017;Ulfelder, 2007)—which

tends to be authoritarian in much of the developing world. In a study of the

long-term linkages between oil wealth and political regime, moreover, Haber

and Menaldo (2011) do not detect a clear association (either positive or

negative) between oil wealth and democracy when the statistical analysis is

properly specif‌ied to emphasize within-country dynamics.

2

Brooks and Kurtz 1699

After many years of serious scholarship, therefore, research on the “re-

source curse”is perhaps further from a consensus on the political effects of

natural resource wealth than it was a decade or more ago. This dissensus

should not surprise us, however, if we consider that few existing treatments of

the political resource curse have converged on a set of concepts and measures

that would enable scholars to put the theory and its proposed mechanisms to a

clear empirical test. Scholars disagree, for example, about whether the “curse

of natural resources”is def‌ined by their abundance, or by the country’s

“dependence”(on resource exports), or by the amount of income accruing

from the sale of those resources. And, the measures have not always matched

the concept at issue very directly. Critical here is the concept of “rents.”Most

strains of research have argued that certain types of natural resource

production—particularly that of oil—produce economic “windfalls”that are

easily captured by governments (Rosser, 2006). And it is primarily from such

allegedly easy money that malign political outcomes derive, including,

conf‌lict, corruption, clientelism, malgovernance, or authoritarian persistence

(Boschini et al., 2007;Greene, 2010;Paler, 2013;Ross, 2013;Snyder, 2006;

Snyder & Bhavnani, 2005).

There are, however, both theoretical and empirical questions about the way

in which this basic claim has been treated in the literature. For instance, most

scholarship takes as an assumption that oil extraction produces rents for

producers, and that such rents are easily appropriated by the state in large

quantities (either through direct ownership or taxation of private oil-producing

f‌irms). This income is generally labeled as “non-tax”revenue, to distinguish it

from the “harder-to-get”resources extracted through direct and indirect

taxation.

3

A critical challenge, however, is that this empirical work has not generally

measured the oil “rents”that are at the core of the resource curse theory and

analysis. This is problematic insofar as rents—returns to a productive factor in

excess of its next best utilization—are the critical piece of most resource curse

causal mechanisms. Without large volumes of “easy money”accruing to the

state, few if any of the presumed political consequences of oil endowment

should actually be expected.

4

Instead of measuring the actual prof‌its produced by oil (or other natural

resources), scholars have instead relied on a variety of output, export, and

receipts indicators. For instance, much of the foundational work on resource-

induced problems focused on the gross value of resource exports (typically as

a share of GDP). But this measurement strategy raises two challenges: f‌irst, it

counts only exported oil as revenue producing, and second, it assumes that oil

is costless to produce (or, at least equally costly to produce everywhere if one

takes the measure to be a correlate of the rents produced). More recent work

has improved on these assumptions, however, relying on the sale value of oil

production—normed to GDP or population, but still requiring the assumption

1700 Comparative Political Studies 55(10)