ArticlePDF Available

Curses, Diseases and Other Resource Confusions


Abstract and Figures

Natural resource rents, development assistance and unrequited foreign exchange inflows such as remittances relax the balance of payments constraint on economic growth. The failure of some governments to translate these resources into successful development has been attributed to an affliction called ‘Dutch disease’, or, more ominously, to a ‘curse’ associated with the availability of natural resources. This paper examines the disease/curse analysis and rejects it in favour of a political economy explanation of the problems associated with resource use. We argue that conventional analysis of resource-rich countries is misleading because its various manifestations are based on inappropriate assumptions and flawed logic. In practice the ‘curse’ and the ‘disease’ are outcomes of policy decisions, rather than manifestations of deep structural weaknesses, and they are more likely to be suffered in countries whose governments pursue neoliberal economic policies.
Content may be subject to copyright.
This article was downloaded by: [School of Oriental and African Studies]
On: 01 February 2013, At: 03:43
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Third World Quarterly
Publication details, including instructions for authors and
subscription information:
Curses, Diseases and Other Resource
Alfredo Saad-Filho a & John Weeks a
a are, respectively, in the Department of Development Studies
and the Department of Economics, School of Oriental and African
Studies (soas), University of London, Thornhaugh Street, London
Version of record first published: 01 Feb 2013.
To cite this article: Alfredo Saad-Filho & John Weeks (2013): Curses, Diseases and Other Resource
Confusions, Third World Quarterly, 34:1, 1-21
To link to this article:
Full terms and conditions of use:
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
Curses, Diseases and Other Resource
ABSTRACT Natural resource rents, development assistance and unrequited
foreign exchange inows such as remittances relax the balance of payments
constraint on economic growth. The failure of some governments to translate
these resources into successful development has been attributed to an afiction
called Dutch disease, or, more ominously, to a curseassociated with the avail-
ability of natural resources. This paper examines the disease/curse analysis and
rejects it in favour of a political economy explanation of the problems associated
with resource use. We argue that conventional analysis of resource-rich countries
is misleading because its various manifestations are based on inappropriate
assumptions and awed logic. In practice the curseand the diseaseare
outcomes of policy decisions, rather than manifestations of deep structural weak-
nesses, and they are more likely to be suffered in countries whose governments
pursue neoliberal economic policies.
The economic and social consequences of a sudden increase in foreign
exchange inows have been a subject of considerable discussion over the past
four decades, producing their own pejorative terminology, including Dutch dis-
easeand resource curse. The essence of these hypotheses, accepted in most
of the literature, is that large foreign exchange ows almost inevitably create
problems that most developing country governments fail to resolve. As a result,
what supercially appears as a great benetlarge inows of foreign
exchangealmost always results in varying degrees of development disaster.
We address these hypotheses and reject them as vaguely specied, unproven
and potentially misleading. We review the disease and curse writings in order to
extract a clear specication of the problem allegedly caused by large foreign
exchange inows. Notwithstanding the common use of the terms Dutch dis-
easeand resource curse, specication proves difcult.
The essential aw in these hypotheses is that the diseaseand the curseare
outcomes rather than causes. Once this is recognised, foreign exchange ows
can be placed in their appropriate context, as economic circumstances requiring
Alfredo Saad-Filho and John Weeks are, respectively, in the Department of Development Studies and the
Department of Economics, School of Oriental and African Studies (SOAS), University of London, Thornhaugh
Street, London, WC1H 0XG, UK. Email:;
Third World Quarterly, Vol. 34, No.1, 2013, pp 121
ISSN 0143-6597 print/ISSN 1360-2241 online/13/000001-21
Ó2013 Southseries Inc., 1
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
a purposeful policy response. It follows that what writers call diseases and
curses are the result of failures to implement effective macro-management poli-
cies. Specically, neoliberal policies, including oating exchange rates and
unregulated capital ows, allow the (avoidable) negative effects of external
resource ows to overwhelm their potential benets.
Curses and diseases
The hypotheses
Until the mid-1980s most economists and policy makers believed that natural
resource abundance, or a large increase in resource prices, provided substantial
economic advantages. They would relieve the foreign exchange constraint on
growth, and the capture of so-called rents could be used to increase investment,
foster infrastructure modernisation and expand domestic markets. The neoclassi-
cal theory of international trade was consistent with this view, arguing that com-
parative advantage offers the most efcient route to development.
Raúl Prebisch and Hans Singer, the founders of Latin American structuralism,
famously questioned the benets from primary product exports,
and suggested
that commodity exporters suffered from a secular deterioration of the barter
terms of trade. While it did not emphasise price booms, the PrebischSinger
hypothesis implied that volatility could result in pro-cyclical scal policies and
patterns of output, employment and consumption, hampering economic
Also from the structuralist school, Hirschman suggested that resource-abun-
dant economies tend to develop insufcient backward and forward linkages
among sectors, and that mining, in particular, was often an enclave.
resource products, notably petroleum, may generate little employment, and with
few linkages the spreadeffects from its extraction may be minor. The combi-
nation of large foreign revenues and low employment tends to foster high
imports of inputs and consumption goods. From a structuralist perspective,
growth driven by natural resources appeared contrary to a more socially desir-
able, broad-based and sustained development. While the structuralists did not
consider natural resource endowments as inherently negative for growth, they
argued that active policies were necessary to stabilise export revenues and foster
In the early and mid-1980s criticism came from the mainstream, with claims
that natural resource abundance led to a range of ills, including low growth,
low savings, limited export diversication, high unemployment, ination and
external debt, authoritarianism, corruption and vulnerability to internal conict.
These claims gained acceptance rapidly, drawing support from the tensions and
conicts in resource-exporting areas of the former USSR, and in countries as
diverse as Cambodia, Chad, Mali and Nigeria.
To investigate whether natural resource wealth is a boon or a curse,
mainstream work focused on the correlations among natural resource wealth
(however dened) and economic growth, income inequality, rent-seekingand
Early studies found strongly negative correlations, with sub-Saharan
Africa allegedly being extremely vulnerable to such effects.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
These dysfunctional outcomes have been variously attributed to Dutch dis-
the quality of institutions,
and/or badpolicy making.
Auty explained
badpolicy by the excessive optimism of governments during price booms.
Other writers used the alleged short-term focus to explain the inability of
governments to manage adverse shocks.
However, in order to be more than
simply one unknown rationalising another, the bad policy argument requires an
explanation of why policy makers (and, at a further remove, voters) do not learn
from their mistakes.
As the list of negative side-effects of resource abundance lengthened, the
neoclassical focus narrowed to the association between resource abundance and
Because neoclassical economic agents are utility maximisers, their
objective functions would lead them to pursue rent-seeking activities whenever
possible. These rents raise the return to holding power, and lead to the
misallocation of public resources.
A group called urban elitesis alleged to
be a major beneciary. As with the bad policystory, the rational-rip-off
hypothesis is incomplete. If public ofcials are crass personal maximisers, the
analyst must explain why the problem is worse in resource-rich countries. The
suggestion that this is because there is more to steal, or it is stolen more easily,
is ad hoc and invites similarly simplistic objections; for example, that there is
more to steal in developed countries than in the most resource-rich developing
economy. This would not be conducive to a constructive debate.
A similar argument is made for the private sector, in which companies in
rentier statesseek to pressure governments to pursue policies serving their nar-
row interests.
In these states wealth accumulates and men decay, as Oliver
Goldsmith wrote in The Deserted Village(1770): resource rents discourage
taxation, foster a bloated civil service, and promote public sector income redis-
tribution through welfare programmes.
To give the argument some nuance,
these outcomes are viewed as especially likely when primary product booms are
concurrent with state formation, weak institutions and unaccountable govern-
ments. As these countries are to hastening ills a prey, Bannon and Collier
argued the logical extension: that natural resource abundance is correlated with
an increased risk of conict.
These arguments for a resource curseare supplemented by the large amount
of mainstream work on policy distortions in resource-rich countries. By deni-
tion a distorted outcome must imply a non-distorted alternative. The neoclassical
denition of the latter is, invariably, full employment general equilibrium,
although that is rarely stated since it might raise questions of relevance. Leaving
that aside, one of the putative distortions of resource booms is ination; a sec-
ond arises when the resource wealth supports foreign borrowing. This allegedly
explains the perverse situation of many oil-exporting countries, in which large
foreign exchange inows are accompanied by unsustainable external debts.
At the heart of the resource curse hypothesis, then, are the uncertainties
caused by commodity booms for balance of payments sustainability. When
prices rise, this allegedly creates incentives for excessive borrowing; when
prices drop, the balance of payments deteriorates. Support for these generalisa-
tions is mixed, and one would expect to nd that resource abundance has differ-
ent implications for growth, depending on a countryshuman capital,
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
institutions and policies.
Indeed, Weeks found that, after one accounted for
the effects of conict, oil exporting African countries grew 1.5 percentage
points faster than the regional average during 19902008.
In a study covering
the same time period for the UN-dened Least Developed Countries, he found
a statistically signicant difference between groups of countries divided by type
of resource export.
An obvious explanation for these mixed results is imprecisely specied
hypotheses, so that any association between a negative outcome and some de-
nition of resource richis interpreted as evidence. With this analytical critique
in mind, in the following section we focus on terminology, denition and speci-
cation of causality to nd the content in the allegations of curses and diseases.
Weaknesses of the resource cursehypothesis
In the mainstream literature specialisation in the export of primary products is
normally identied with resource abundance, deriving from an exogenously
given endowment of natural resources. The associated export revenues are
tainted by the use of the pejorative terms rentsand windfalls.
These sug-
gest that such revenues are unearned, temporary and undeserved. This approach
also predicts retribution-like outcomes, including low growth, institutional dis-
tortions and civil unrest. This analysis fails on ve levels.
First, a resource-abundant country is difcult to dene, and vague denitions
can lend spurious support to the cursehypothesis by their undiscriminating
inclusiveness. For example, if abundanceis dened by the resource output/GDP
or the resource export/GDP ratios, low income countries will be more likely to
qualify. Developing countries with and without resources are also more prone to
a range of maladies, including civil unrest, than developed ones. This is an
instance of coincidence masquerading as causality, providing spurious support
for a curse.
Second, clear specication of resource abundance creates its own problem:
contradictory empirical results. For example, if abundance is measured through
the ratio of natural resource exports to total exports or GDP, the resource curse
is often conrmed (see above). If it is measured by the output of primary prod-
ucts, total reserves, reserves per capita or exports per worker, the hypothesis
tends to be rejected. Interestingly, while the hypothesis claims that adverse
outcomes are the result of the rents generated by primary exports, rent-based
measures of resource abundance have failed to support the curse.
inconsistencies are magnied by arguments that the curseis associated with
specic types of resources, especially minerals (copper, coal, diamonds or oil),
rather than agricultural products (wheat, rice, cocoa or coffee).
Third, most empirical studies of the curseseek to demonstrate that resource
abundance, however measured, is correlated with poor growth. Even if this cor-
relation could be demonstrated, causation could plausibly run the other way.
For example, poverty, low skills and weak infrastructure can result in reliance
on primary product exports, in which case the latter becomes a symptom of
underdevelopment rather than its cause. It is also possible that poor performance
could be caused by a third (omitted) variable. These ambiguities suggest that a
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
clear theoretical framework is required to avoid confusing resource abundance
and comparative advantage.
Fourth, it is often assumed that public revenue in resource-rich countries is
more reliant on rents than on taxation, giving the government greater autonomy
from pressures for political and economic reform. There is no reason why the
absence of direct taxes should reduce the social pressures upon the government,
especially with regard to the allocation of revenue from royalties and export
taxes on natural resources. It also seems contradictory to argue that resource
rents reduce the accountability of governments, and that such governments are
vulnerable to populist pressure for redistribution through excessive social
Fifth, mainstream economic theory assumes that societies are composed of
rational maximising individuals who organise to pursue their individual and col-
lective interests. The inference is drawn that the pursuit of these interests results
in predatory behaviour, specically in corruption triggered by competition over
rents. While supercially appealing, this inference is devoid of content if it is
not placed in the specic conditions of each country. When it is so placed, the
role of natural resources may become secondary to other sources of conict
based on class, gender, ethnicity, religion or region.
In summary, most discussions of the resource curse are based on ambiguous
evidence, and allegations of causality tend to be static and deterministic. The
curse hypothesis is reduced to the proposition that the resource base determines
social, political and institutional outcomes. However, as Rosser rightly put it:
Scholars have been asking the wrong question: rather than asking why natu-
ral resource wealth has fostered various political pathologies and pro-
moted poor development performance, they should have been asking what
political and social factors enable some resource abundant countries to utilise
their natural resources to promote development and prevent other coun-
tries from doing the same.
Analysis of resource rich economies should focus on the country-specic media-
tions among institutions, classes and economic performance. In the absence of
this political economy approach, discussions of resource-abundance economic
growth tend to be reductionist, generating policy recommendations of limited
practical substance:
The resource-curse hypothesis seems anomalous since it has no clear
policy implication but stands as a wistful prophecy: Countries aficted with
the original sinof resource endowments have poor growth prospects. The
danger of such ruminations is that they may inuence sectoral poli-
cies. Minerals themselves are not to blame for problems of rent-seeking and
corruption If minerals are conceived as xed stocks, and mineral abun-
dance as a windfallunconnected to past investment, then the problem
becomes one of divvying up the bounty rather than creating more bounty.
Minerals are not a curse at all in the sense of inevitability; the curse, where
it exists, is self-fullling.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
Given its obvious limitations, the cursehypothesis might have been an inter-
esting curiosity. Instead, its frailty was interpreted as suggesting generality,
prompting its defenders to extend it to any commodity-exporting economy or,
indeed, any country passing through a period of large, unanticipated foreign
exchange boom.
A hypothesis that cannot be sustained in its clearest specication is unlikely
to gain strength through generalisation. This exercise creates major problems of
discrimination between those countries to which the hypothesis should apply,
and those to which it does not. If the hypothesis applies to any country under-
going a resource boom, it is necessary to dene how long is sufcient for the
curseto work and how substantial the boom must be to qualify. Equally
important, as the number of countries dened as aficted by the curse
increases, alternative explanations also multiply. The search for a theory gener-
ating the curseleads to a specic form of the hypothesis—‘Dutch disease’—
which has itself fallen victim to near-random generalisation (see below).
What needs explaining?
This section offers a narrow and reasonably obvious specication of the curse
hypothesis to see if there is evidence to inspire more rigorous analyses. If there
were a resource curse, it should show itself in countries whose exports are dom-
inated by hydrocarbons (petroleum and natural gas). The potentially negative
consequences of resource booms, absence of backward linkages, low employ-
ment, high inequality and public sector corruption should be signicant for
petroleum exporters if they are so for any country.
The countries with data relevant for inspecting the hypothesis between 1980
and 2010 include those dened by the United Nations Conference on Trade and
Development (UNCTAD) as major petroleum and gas exporters.
To these we add
Azerbaijan, Bahrain, Brunei, Cameroon, Congo Brazzaville, Ecuador, Equatorial
Guinea, Gabon, Indonesia, Kazakhstan, Mexico, Russia, Sudan (including South
Sudan), Syria, Trinidad and Tobago, Turkmenistan and Yemen, whose petro-
leum share in merchandise export value either exceeded the UNCTAD threshold
for extended periods since the 1980s, or who were large global producers dur-
ing this period (see Table 1). This sample excludes countries with recent discov-
ery of petroleum reserves (eg Chad), developed countries (eg Canada), and
major producers which are not substantial exporters (eg China).
For at least 14 of these countries the resource curse would be at most a
secondary explanation of their social and economic development. Seven are
monarchies characterised by varying degrees of authoritarian rule. This would
seem a more important explanation of corruption and inequality than the com-
position of exports (it may be that oil revenues helps to perpetuate these
regimes, but that would be a specic version of the curse hypothesis). Six of
the seven also have small populations, whose correlation with lack of economic
diversication is well documented and unrelated to petroleum.
Three countries with substantial populations, Angola, Iraq and Sudan, have
long histories of conict only secondarily related to petroleum wealth. The con-
ict in Angola, like the contemporary one in Mozambique (where there are no
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
known resource riches), was a product of the Cold War, of ethnic tensions, and
of the policies of the South African apartheid regime. The two major conicts
involving Iraq, rst against Iran and then against the USA and its allies, were
TABLE 1. Hydrocarbon exporters, overview
Population in 2010
PCY growth
(millions) 198099 200010
Saudi Arabia 27 95 2.6 0.2 na
UAE 8 81
2.5 3.9 na
Oman 3 93 3.1 2.7 na
Kuwait 3 95 5.1
na na
Qatar 2 90 na 1.2 na
Bahrain 1 74 0.4 0.4 na
Brunei 0.4 100 3.1 0.7 na
Conict and transition
Russia 142 58 5.0
5.5 45.1
Sudan 44 75 1.0 4.0 na
Iraq 32 98
na 3.9 41.5
Angola 19 87
Azerbaijan 9 88 7.9
13.3 50.8
Kazakhstan 16 69
7.5 41.4
Turkmenistan 5 70
12.8 na
Latin America & Caribbean
Mexico 113 55 0.8 0.9 51.1
Venezuela 29 93 1.4 1.4 47.6
Ecuador 14 63 0.3 2.8 53.4
Trinidad and Tobago 1 81 0.3 5.2 40.3
Sub-Saharan Africa
Nigeria 158 98 0.6 3.8 43.7
Cameroon 20 54 0.7 1.1 44.6
Gabon 2 86 1.0 0.2 44.1
Equatorial Guinea 1 na 14.7
13.3 na
Congo Brazzaville 4 92
0.7 2.3 na
North Africa and Western Asia
Iran 74 86
0.8 3.4 38.4
Algeria 35 98 0.4 2.1 35.4
Syria 20 74 1.0 2.1 n.a.
Libya 6 98 n.a. 2.2 n.a.
Yemen 24 95 1.4
1.3 37.7
Southeast Asia
Indonesia 240 64 3.6 4.0 39.4
Notes: na= data not available;
Average of the 10 years with the highest share of fuel in total exports between 1980 and 2010. Starred
data include only the top ve years;
Average annual growth rate of per capita income. Starred data include only 199099;
Latest data point on wider Income Distribution Database (since 1990).
Sources: World Development Indicators, April 2012; and WIDER Income Distribution Database.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
wars between states, which is not what the curse hypothesis seeks to explain.
Furthermore, the petroleum-fosters-conictargument would be absurd for
Sudan, where rebellion leading to the secession of South Sudan dates from the
civil war of 195572, long before the oil discoveries. Russia is another case in
which resource wealth would be at most complementary to a broader analysis
of the transition from central planning to a market society, while Azerbaijan,
Kazakhstan and Turkmenistan have been subjected both to authoritarian regimes
and to complex economic, social and political transitions in which oil plays a
signicant but not necessarily determining role.
Of the remaining 15 candidates for the curse hypothesis, fourCongo Braz-
zaville, Equatorial Guinea, Gabon and Trinidad and Tobagoare quite small,
which may be the principal explanation for their curse-like problems. For the
remaining 11, outcomes are mixed. Of the three Latin American countries,
Ecuador had a per capita income growth rate above the average for the non-
petroleum exporters, Mexico was at that average, and Venezuela below it.
Among the sub-Saharan countries, Nigeria was above the regional growth aver-
age in the recent period, and Cameroon and Gabon below it, but all three had
contracted during the previous two decades. In North Africa and Western Asia,
Iran grew faster than the non-petroleum average, and Algeria, Libya and Syria
close to it. Finally, the only Southeast Asian country, Indonesia, had a signi-
cant growth rate both during and after its oil exporting phase, with signicant
economic diversication; but it also suffered heavily through the East Asian
crisis in the late 1990s.
The absence of distributional data for some of these countries, and the
large variation in the available data, make it difcult to assess the link
between petroleum exports and inequality, but it is possible to look in detail
at a few countries (see Table 2). Statistics from Indonesia, Mexico,
Venezuela and Nigeria do not obviously support the hypothesis that oil
dependency is associated with greater inequality. In the rst three cases
inequality changed little over the period for which there are statistics. For
Nigeria substantial changes occurred between 1975 and 1980, which could
be linked to increased petroleum dependence. As for growth, the evidence
suggests varied outcomes.
Figures 1 and 2 inspect the evidence for the hypothesis that petroleum
dependence results in failure to diversify the production structure. Figure 1
provides the crude oil exports for the four countries, showing that each
remained a substantial exporter into the 2000s. Figure 2 shows a dramatic
difference in outcomes. The external trade of Indonesia and Mexico diversi-
ed spectacularly; in contrast, Nigeria and Venezuela were unable to sustain
any degree of diversication that cannot be explained by uctuations in
petroleum prices.
The lesson to take from this review of petroleum exporters is that, if a
resource curse exists, it does not afict most of the countries that should fall
victim to it. Further, those countries that display the symptoms of the curse may
do so for reasons specic to themselves, associated with their social institutions,
class structures and government policies. It is to these that we now turn.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
Analytics of Dutch disease
The disease as such
After reviewing the less-than-rigorous resource cursehypothesis, we consider
a more precise specication of a macroeconomic problem of resource-rich
countries: Dutch disease. This hypothesis considers the situation in which a
country passes through a resource boom, resulting either from new exports or
from a rise in the price of its existing exports. Although the term Dutch
diseaseis most frequently used to refer to natural resource booms, it has also
been applied to substantial increases in the price of any exported good or
service, or large inows of foreign investment, assistance or remittances.
Without loss of generality, this article focuses on export price booms.
The term Dutch diseasewas suggested by The Economist, drawing upon
the experience of the Netherlands in the 1960s.
When gas was discovered in
the North Sea, the Netherlands went through a resource boom that substantially
appreciated its currency and eroded the competitiveness of other exports.
Similar outcomes can be attributed to capital ows. For example, nancial
investors can borrow in low-yield currencies and invest in high-yield ones,
leading to the appreciation of the latter regardless of the countrys fundamentals
TABLE 2. Gini coefcient and share of hydrocarbon exports, Indonesia, Mexico,
Venezuela and Nigeria
Indonesia Mexico Venezuela Nigeria
Gini Oil/X Gini Oil/X Gini Oil/X Gini Oil/X
1964 33.3 na 1950 52.3 na 1976 53.2 93.8 1959 48.6 0.0
1967 32.7 33.2 1963 53.0 na 1977 52.5 92.6 1975 43.0 93.3
1970 30.7 32.8 1968 54.2 3.1 1978 49.6 94.5 1980 51.2 96.0
1976 31.8 70.3 1975 55.9 15.5 1979 47.0 92.8 1985 47.9 96.7
1978 34.8 68.6 1984 50.6 61.8 1980 47.5 94.0 1992 54.2 96.6
1980 31.8 71.9 1989 53.1 33.9 1981 43.7 92.8 1996 52.2 95.6
1981 30.9 79.8 1994 55.7 11.9 1982 44.5 95.0
1984 30.8 71.7 1996 53.7 12.0 1983 45.9 95.4
1990 32.0 44.0 1998 55.4 6.0 1984 51.2 93.1
1993 33.0 28.4 2000 55.6 9.7 1985 44.9 80.0
1996 36.0 25.8 2005 55.7 14.9 1986 45.2 84.1
1999 32.0 23.0 1987 43.6 87.1
2002 33.9 24.4 1988 45.0 81.1
2005 39.4 27.5 1990 44.0 80.1
1991 44.2 80.9
1992 42.6 80.0
1995 46.6 76.7
1998 47.2 71.7
2000 44.1 86.1
2001 46.4 83.2
2002 47.5 80.3
2003 46.2 82.4
2004 45.4 83.7
Sources: World Development Indicators, April 2012; and wider Income Distribution Database.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
or the sustainability of the current account.
These exchange rate movements
create no wealth, can be destabilising and complicate the task of policy makers,
who know, as well as the market operators, that these ows are unsustainable.
FIGURE 2. Petroleum as a Percentage of Total Merchandise Export Value, Indonesia,
Mexico, Nigeria and Venezuela, 19672010.
Source: World Development Indicators April 2012.
FIGURE 1. Crude Petroleum Exports, Indonesia, Mexico, Nigeria and Venezuela, 1986-
2009 (thousand barrels/day).
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
The argument proceeds as follows. A resource boom increases exogenously
the inow of foreign exchange and raises the income received by producers and
the government, allowing for a higher level of imports and greater expenditure
on domestic goods. The increase in domestic activity provokes a real
appreciation of the currency, which can erode the competitiveness of non-boom-
ing (lagging) exports. Since the non-booming tradable sector is typically
manufacturing, natural resource booms are alleged to cause deindustrialisation.
The analysis concludes that resource booms can have an adverse impact on the
non-booming sectors and create a structural inexibility in the balance of
payments, increasing the countrys dependence on the booming sector. This will
reduce the economys ability to cope with adverse shifts in resource ows,
including export price declines. While it has long-run consequences, Dutch
disease is usually seen as a short-run problem which may or may not be related
to the (long-run) natural resource curse.
Dutch disease has been treated formally. Let the real exchange rate (RER) be:
where E is the nominal exchange rate (domestic currency per unit of foreign
currency), and P
and P
are the foreign and domestic price indices. Hence, the
real exchange rate measures the relative price of goods produced in two
countries. For example, a nominal devaluation of the local currency vis-à-vis
the dollar improves the economys competitiveness relative to the USA (the RER
rises). By the same token, if domestic ination is higher than US ination the
RER falls, suggesting that the country has lost competitiveness.
We can rene the concept of RER. First, we group the goods produced in both
economies into tradables (including exportables, importables and domestic
import-competing goods) and non-tradables. The former can be further
disaggregated into boomingand non-boomingtradables, with prices P
. Second, we distinguish between the prices of non-tradables (such as
construction, real estate and haircuts) in the foreign and the domestic economies
and P
). Third, we assume that the economy is small and open, and that
foreign and domestic tradable goods are perfect substitutes, so P
and P
xed with their world prices. In contrast, the non-tradable sector includes
services and domestically produced goods for which world supply and demand
are effectively zero. Thus, the price of non-tradables is determined solely by
domestic supply and demand:
R¼EðafPTB þbfPTN þcfPNf Þ
ðadPTB þbdPTN þcdPNdÞ
where α,βand γare the weights of booming tradables, non-booming tradables
and non-tradables in the foreign and domestic price indices, with:
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
If we assume that the price of non-booming tradables is xed and that the
booming resource has the same weight in the foreign and domestic price indices
), the impact of a resource boom on the RER will depend on the changes
in P
and P
The Dutch disease model presumes that the resource boom triggers excess
demand in the domestic economy. This excess demand cannot raise the domes-
tic price of tradables, but it raises the price of non-tradables (ΔP
Consequently, the RER falls (overvaluation), and the non-booming tradables sec-
tor loses competitiveness. This is illustrated by the lower-right quadrant of the
SalterSwan diagram in Figure 3. This quadrant displays the quantities of trad-
able and non-tradable goods produced and consumed. We assume a starting
equilibrium in which the economy is at point where the indifference curve (ID)
is tangential to the Production Possibilities Frontier (PPF) of non-booming trad-
ables and non-tradables (point A), suggesting that, before the boom, there is full
employment of every resource.
In the upper left quadrant, representing the
market for lagging tradables, domestic supply (ST) equals domestic demand
(DT) at point QT, so the trade balance is initially zero. In the upper right quad-
rant, depicting the market for non-tradables, the economy is initially at point Z,
where the supply of non-tradables (SNT) equals demand (DNT).
FIGURE 3. The SalterSwan tradable and non-tradable sectors analysis.
Source: Adapted from M Nkusu, Aid and the Dutch Disease in Low-Income Countries,
Working Paper 04/49, Washington, DC: IMF, 2004.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
The resource boom affects the economy through three channels.
The rst of
these is the income (spending) effect. The additional income resulting from the
export boom will raise national income from Y to Yin the lower-right
quadrant. As income rises, the demand for non-tradables increases, represented
by the shift from DNT to DNTin the upper right quadrant (the demand for
tradables also increases, but this is met by an increase in imports nanced by
the additional revenues in the booming sector). The rise in non-tradables prices
will trigger ination because of their rigidity of supply in the short-run; there-
fore, the RER appreciates. At the same time the balance of payments deteriorates
because of the import growth and the squeezing of the lagging tradables sector.
However, this deterioration is by denition smaller than the export gains, since
national income has increased. On a net basis national income and the domestic
price level rise, and the balance of payments improves.
The second channel is the resource transfer effect. The resource boom and
the increased demand for non-tradables will attract factors of production to the
booming sector, shrinking the non-tradables and lagging export sectors. Note
that this is based on the assumptions that the economy is operating at full
capacity and that labour is the only mobile factor between sectors.
the resource transfer effect is towards deindustrialisation. This is summarised by
a shift from A to P on the domestic PPF in the lower-right quadrant which, in
the two upper quadrants, is associated with a contraction in the supply of trad-
ables from ST to ST, and by an expansion in the supply of non-tradables from
The third channel is the expenditure switching effect. As national income
rises from Y to Y, consumption increases to C on indifference curve IDand
the demand for tradables rises from QT to QDTin the upper left quadrant. The
increased consumption of tradables, combined with a decrease in domestic sup-
ply to QST, thanks to the spending and resource transfer effects, leads to a
deterioration of the trade balance, from zero (before the boom) to QDT-QST
(after the boom). Overall, the economy will have expanded its possibilities of
consumption because of the higher rer and the shift of resources towards non-
The closure of traditional industries brings bankruptcies and job losses, and
eliminates skills.
This may not be a loss to the economy, especially if the
resource boom is permanent, but it will be costly to many and, inevitably, dif-
cult to manage. It is more serious if the boom is temporary. Should the price of
the booming exportable good return to its initial level, or if the booming
resource becomes exhausted, the economy will nd itself with reduced diversity,
lost skills and foreign markets, and with a more precarious balance of payments
position than was the case before the boom. It is even worse if the squeezed
sectors include those which can generate positive externalities and learning by
doing, which, again, tend to include manufacturing.
The processes leading to Dutch disease depend on the exchange rate regime,
although the mechanics need not concern us here.
Substantively, the local
currency must appreciate because this is how the additional inows of foreign
currency are absorbed. The appreciation cheapens foreign goods and, together
with the higher national income, leads to additional imports of non-booming
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
goods and services. A real appreciation does not, itself, demonstrate the Dutch
disease. However, there would be cause for concern if the appreciation led to
declining exports and a widening trade gap. For example, in a sample of six
oil-exporting countries, Gelb found that their average real effective exchange
rate had risen by nearly 50% between 1970 and 1984.
As this was not offset
by productivity growth, their manufacturing sectors were severely hampered.
Dutch disease can be difcult to neutralise and costly to correct, and it may
be compounded by the volatility of prices and quantities of natural resource
exports. If the economy is heavily geared towards the production of a narrow
range of goods, the interaction between the overvaluation of the RER, the
(exogenous) volatility of resource prices, and the scant linkages between the
export sector and the rest of the economy can lead to uncertainty, low
investment, inadequate skills and weak diversication.
The long-term consequences of Dutch disease can be severe; at the very least,
it can create rich countries with poor people.
In contrast, a diversied
economic base generally brings a more rened division of labour and higher
levels of employment, and it can lead to increasing returns and stronger exter-
nalities, which can more easily support distributive development strategies.
Reviewing the model
The Dutch disease model is excessively simplistic and may not offer useful
policy guidance for countries dealing with resource booms. Five reasons are
especially important.
First, the division of the economy into tradable and non-tradable sectors is
conceptually untenable. For example, it is impossible to classify construction,
which includes homes and ofces, but also multiple-use hotels, tourist facilities,
ports and roads.
Second, if domestic rms use imported components to assemble goods for
export, the RER appreciation can be benecial. For example, if US income rises,
local demand for computers will also go up. This would increase the shipment
of US components to the country where the assembly plants are located, say
China, which is counter-intuitive according to the model. Likewise, since the
RER appreciation implies that the USA can buy foreign components more
cheaply, the export price of US-assembled goods could fall as the dollar rises.
Third, economic growth in developing countries is normally associated with a
more than proportionate increase in the demand for money. As the economy
expands, more areas of activity are brought into the market, nancial depth grows,
and the monetary and nancial needs of households and business expand. Since
the income elasticity of demand for money is likely to exceed unity, the economy
should be able to absorbto a certain extenta rising money supply without
undue inationary pressures or Dutch disease phenomena.
Fourth, the model assumes that the economy is perfectly competitive and
operating at full-employment with external balance. This is unrealistic, because
developing countries always have a pool of unemployed, underemployed and
informal sector workers willing to work at the going wage (which is generally
too low to lift them out of poverty), and unable to nd full-time positions.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
These varieties of unemployment often coincide with spare capacity in several
sectors. Since output is normally well within the Production Possibilities Fron-
tier, domestic output and non-booming tradables prices may rise, as long as
imports are imperfect substitutes for domestic goods.
Fifth, windfalls can help the developing countries address three constraints on
growth: the availability of foreign exchange, the supply of savings and scal
sustainability. For example, a resource boom can enable the country to address
localised supply problems through additional imports until the expansion of pro-
ductive capacity (which may also be nanced by the windfall) takes hold.
Hence, the macroeconomic impact of a resource boom is likely to depend upon
circumstances of time and place, and on government policies. It is impossible to
say ex ante whether a boom is likely to lead to sustained economic growth, ina-
tion or Dutch disease. The outcomes are likely to depend on the policy mix, the
governments priorities, and its capacity to implement consistent policies.
Economic policy beyond diseases and curses
Several policies can be deployed to avoid Dutch disease and the resource
curse, and to address the macroeconomic side-effects of resource booms. All
the major policy instruments come into play, requiring purposeful coordination.
Diseases and curses result from the failure to use these instruments, which is
often the result of an ideological opposition to policy intervention.
Balance of payments management
Countries beneting from temporary resource booms should save a large pro-
portion of any windfall, especially if they are marginal producers. If part of the
windfall is not internalised, it will not put pressure on the exchange rate, the
domestic money supply or the non-monetary liabilities of the central bank,
avoiding both the resource curseand Dutch disease. Both outcomes can be
achieved, for example, through the build-up of the countrys international
reserves, nancial investment abroad (including sovereign wealth funds, see
below), anticipated debt repayment (which will be benecial when export prices
decline) or the taxation of export revenues and/or of imports of non-competing
goods and services. Controls on capital ows can also help to stabilise the bal-
ance of payments, because these ows tend to be both pro-cyclical and volatile.
These controls should include the requirement that all foreign resource ows
must pass through the central bank, at least to ensure adequate accounting of
the countrysnancial relations with the rest of the world. They should also
ensure that banks and rms do not build up unsustainable positions for specula-
tive reasons. In these cases, the windfall may not bring any adverse macroeco-
nomic consequences.
Exchange rate policy
The co-ordinated deployment of economic policies to neutralise the resource
curseand Dutch disease is problematic because of the so-called impossible
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
trinity: countries cannot, simultaneously, have an open capital account, xed
exchange rates and independent monetary policies. This translates into the need
to lose of one of three potential goals: exchange rate stability, monetary policy
autonomy or global nancial integration.
Simplistic conclusions about this trinityare insufcient for two reasons.
First, the ability of developing country central banks to oat the exchange rate
is limited because of the original sin,
currency substitution, poor credibility
and other factors. However, a xed exchange rate may not be the best
alternative, because the currency ows create tensions for monetary and
nancial policy.
Second, with an open capital account independent monetary
policy can become impossible regardless of the exchange rate regime, because
developing countries often cannot bear the ensuing balance of payments
pressures under a xed exchange rate regime, or the costs of exchange rate
volatility under oating exchange rates. The managed uctuation of the
currency, or an adjustable peg vis-à-vis a basket of currencies reecting the
countrys trading and nancial relations, allows the government to limit
the adverse consequences of the foreign resource inows, expand its policy
space, and diffuse the impact of the resource boom across multiple channels.
Fiscal policy
Fiscal policy can play a key role in transferring rents, productivity improve-
ments and gains from trade to the non-export economy and to strategic
economic sectors. This includes linkages across existing areas of activity, and
support for the emergence of sectors bringing new competitive advantages and
offsetting the dependence on traditional exports. In resource-rich developing
countries scal policy should be the primary tool for short-term economic
stabilisation, which includes smoothing the impact of the windfall on public
expenditures, and promoting stabilising and equalising outcomes through health,
education, welfare and employment policy. Fiscal policy can also support large-
scale public infrastructure projects which the private sector may be unable or
unwilling to complete.
In order to achieve these outcomes, it may be useful to establish a domestic
stabilisation fund in addition to an external sovereign wealth fund. The former
should accumulate temporary inows of rents, royalties and taxes in the form of
private and public sector securities (including the repurchase of the outstanding
domestic public debt), mass housing and other modalities of anti-cyclical saving
in local currency. These will curtail the scope for Dutch disease, and they can
provide a long-term revenue stream to compensate, in part, the exhaustion of
primary export revenues.
In turn, external resource (sovereign) funds can help to cushion the balance
of payments against market uctuations and prevent Dutch disease.
funds operate under permanent income assumptions: a baseline price for the
exportable product is estimated, and surpluses are saved in the fund. In contrast,
when resource prices are below the baseline, withdrawals from the fund can
help to smooth consumption and investment. In the long term the fund should
provide alternative sources of hard currency when the countrys reserves are
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
exhausted, or if its competitive advantages are eroded by changes in the world
Monetary and nancial policy
The sterilisation of foreign exchange ows blocks Dutch disease, because
changes in the non-monetary liabilities of the central bank sever the link
between the balance of payments and the domestic money supply. Sterilisation
requires some degree of nancial market development,
and frequent
adjustments in the domestic interest rates, which might have adverse
macroeconomic implications. However, as long as sterilisation is reasonably
well managed, these difculties are likely to be minor compared with the
potentially severe destabilising impact of resource booms and busts.
The regulated development of the nancial sector simultaneously with rising
currency reserves can support the expansion of the domestic demand for money
and nancial assets, and improve the efciency of the nancial system. These
outcomes can be readily linked to the governments industrial policy. Regulation
can also help to absorb windfalls, especially through changes in the money
multiplier. This can be achieved through adjustments in bank reserve require-
ments, interest and discount rates, oron a one-off basisthrough the transfer
of central government deposits from the commercial banks (where they count as
part of the money supply) to the central bank (where they do not).
These policy initiatives are especially signicant because monetary and
nancial policies in primary export-dependent countries are often dysfunctional.
For example, in most Gulf countries monetary policy plays little role in promot-
ing growth.
In contrast, developing country governments are often obsessed
with ination stabilisation, even though the literature shows that permanently
contractionary policies compromise growth and equity, and that ination rates
as high as 40% annually have no adverse implications for growth or equity.
Developing country nancial systems also tend to lock resources into
speculative circuits which do not benet investment, production or social
welfare. For example, instead of lending for productive purposes many banks
concentrate their assets in liquid paper, consumer loans and nancial
speculation. These difculties are unlikely to be resolved through nancial
liberalisation, especially in the current circumstances of global instability.
Industrial and employment policy
Higher levels of investment, employment and capacity utilisation, sustained pro-
ductivity growth and export diversication will require additional nance, which
will absorb a large part of the resource boom. If industrial policy is judicious it
can also remove bottlenecks that might trigger ination and shift relative prices
towards non-tradables.
These policies should be supplemented by the regulation of the uses of funds
in order to avoid potentially destabilising capital ight, overconsumption (espe-
cially of imported luxury goods), the explosive growth of consumer credit, and
speculation with nancial assets and real estate resulting from the wealth effect
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
of the windfalls and the revaluation of the currency. It is especially important to
avoid using windfalls as an excuse to reduce taxes on the private sector.
Temporary windfalls should be either saved (see above) or invested in one-off
short-term infrastructural projects that can relieve bottlenecks, support the
balance of payments and expand productive capacity: new roads, railways,
bridges, ports, storage depots, power plants, oil reneries or other key
infrastructural or industrial facilities. In contrast, subsidies to consumption
(including tax cuts) and open-ended non-revenue-generating commitments
cannot be adequately nanced through windfalls, and attempts to do so are
likely to trigger macroeconomic disequilibria. Governments should always make
sure that any additional commitments can be nanced, even if the resource
boom is reversed.
The impact of windfalls can be modulated through scal, monetary and
exchange rate policy tools, and through capital controls and balance of
payments management techniques. Industrial policy tools can also direct these
ows in order to remove bottlenecks and promote strategic sectors, turning
windfalls into sustained growth, or at least avoiding the resource curseand
Dutch disease. Since these adverse outcomes can be ltered out by
coordinated scal, monetary, nancial and exchange rate policies, it is within
the power of policy makers to decide whether or not, and to what extent, there
will be a curseor a disease. In this sense Dutch disease and the resource
curseare not only avoidable: they are policy outcomes, or consequences of
misguided policy choices and perverse private sector behaviour.
Economic and social policies are not merely technical tools. They can also
play an instrumental role in the reproduction of poverty and inequality,
especially in the context of unduly contractionary mainstream policy strategies.
Conversely, economic policies can also make an important contribution to the
implementation of distributive development strategies, especially through their
support for growth-accommodating policies, and their inuence on the
allocation of resources.
This article has reviewed the literature on the (presumably long-term)
resource curse, and the (short-term) Dutch disease, and found it wanting. It
follows that adverse macroeconomic outcomes cannot be blamed entirely on
exogenous or unavoidable factors. They are, instead, primarily the result of
domestic policy decisions. The other side of the coin is that there is
considerable scope for selecting and implementing non-neoliberal economic
policies, which may avoid the curseand the disease, and facilitate the
achievement of distributive outcomes.
The article has also shown that the choice between the mainstream and pro-
poor approaches to economic policy involves not only their internal consistency,
but also the political constraints inuencing the selection of economic policy in
each country. In particular, the most important constraint on the introduction of
distributive strategies in developing countries is not resource scarcity, resource
curse,Dutch disease, international constraints or the lack of government
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
capacity. It is, rather, the lack of political will to build distributive economic
policy alternatives in each country.
1 Rents are the returns to a factor of production above the amount necessary to bring that factor into use.
Depending on the context, rents are returns higher than those accruing from the next-best alternative use
of that factor (eg the production of one crop rather than another in a piece of land). Alternatively, they are
returns exceeding the competitive cost of production (for example, selling oil above the domestic extrac-
tion cost). Rentsare always the result of market distortions in relation to the perfectly competitive ideal.
2 R Prebisch, The Economic Development of Latin America and its Principal Problems, New York: ECLA,
1950; and H Singer, The distribution of gains between investing and borrowing countries,American Eco-
nomic Review, 40(2), 1950, pp 473485. For earlier hints of a curse, see M Boianovsky, Humboldt and
the Classical Economists on Natural Resources, Institutions and Underdevelopment, 2010, at http://collo-
3 A Hirschman, The Strategy of Economic Development, New Haven, CT: Yale University Press, 1958.
4 See JD Sachs & AM Warner, Natural Resource Abundance and Economic Growth,NBER Working Paper
No 5398, Cambridge, MA: NBER, 1995; and Sachs & Warner, The big push, natural resource booms and
growth,Journal of Development Economics, 59, 1999, pp 4376. See also RM Auty, Resource Abun-
dance and Economic Development, Oxford: Oxford University Press, 2001; G Nankani, Development
Problems of Mineral Exporting Countries, World Bank Staff Working Paper 354, Washington, DC: 1979;
and D Wheeler, Sources of stagnation in sub-Saharan Africa,World Development, 12(1), 1984, pp 123.
5 A Rossser, The Political Economy of the Resource Curse,IDS Working Paper 268, Brighton: Institute of
Development Studies, 2006.
6 Sachs & Warner, The big push, natural resource booms and growth.
7 H Mehlum, K Moene & R Torvik, Institutions and the resource curse,Economic Journal, 116, 2006, pp
8 P Collier & B Goderis, Commodity Prices, Growth, and the Natural Resource Curse, OxCarre Research
Paper 14, University of Oxford, 2008.
9 RM Auty, Patterns of Development, London: Edward Arnold, 1995.
10 M Perälä, Persistence of Underdevelopment: Does the Type of Natural Resource Endowment Matter?
WIDER Discussion Paper 2003/37, Helsinki: WIDER.
11 S Bhattacharyya & R Hodler, Natural Resources, Democracy and Corruption, OxCarre Research Paper
20, University of Oxford, 2009; and C Leite & J Weidmann, Does Mother Nature Corrupt? Natural
Resources, Corruption, and Economic Growth,IMF Working Paper WP/99/85, Washington, DC: IMF, 1999.
12 JA Robinson, R Torvik & T Verdier, Political foundations of the resource curse,Journal of Development
Economics, 79, 2006, pp 447468.
13 TL Karl, The Paradox of Plenty: Oil Booms and Petro-States, Berkeley, CA: California University Press,
1997; and Perälä, Persistence of Underdevelopment.
14 See A Abderrezak, Colonisations long-lasting inuence on economic growth: evidence from the mena
region,Journal of North African Studies, 9(3), 2004, pp 103112; M Humphreys, JD Sachs & JE Stiglitz,
Escaping the Resource Curse, New York: Columbia University Press, 2007; and ML Ross, The political
economy of the resource curse,World Politics, 51(2), 1999, pp 297322.
15 I Bannon & P Collier, Natural Resources and Violent Conict, 2006, at
16 V Polterovitch, V Popov & A Tonis, Resource Abundance: A Curse or Blessing? DESA Working Paper 93,
New York: DESA 2010. See also M Alexeev & R Conrad, The elusive curse of oil,Review of Economics
and Statistics, 91(3), 2009, pp 586598; CN Brunnschweiler, Cursing the Blessings? Natural Resource
Abundance, Institutions, and Economic Growth, Working Paper 06/51, Zurich: Swiss Federal Institute of
Technology, 2006; and D Lederman & WF Maloney (eds), Natural Resources: Neither Curse nor Destiny,
Washington, DC: World Bank, 2006.
17 J Weeks, Policy Options for Reducing Unemployment and Mitigating the Social Impact of the Global
Financial Crisis, Addis Ababa: UNECA, 2010.
18 J Weeks, Enabling Recovery and Macro Stability in LDCS: A Study for the LDCR 2010, Geneva: UNCTAD.
19 A windfall is an economic gain realised without sacrice, or without the expenditure of resources. This
suggests that it results from luck rather than effort. The term is also used to imply that the gain is
20 Alexeev & Conrad, The elusive curse of oil, p 589.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
21 P Collier & A Hoefer, Resource rents, governance, and conict,Journal of Conict Resolution, 49(4),
2005, pp 625633; and M Herb, No Representation Without Taxation? Rents, Development and Democ-
racy, 2003, at
22 Rosser, The Political Economy of the Resource Curse,p3.
23 G Wright & J Czelusta, Why economies slow: the myth of the resource curse, Challenge, 47(2), 2004, pp 638.
24 This group includes those countries whose share of petroleum and gas exports (SITC code 33 and 34) was
above 50% of their total exports, and whose share of world exports in 200406 was at least 1%. UNCTAD,
Handbook of Statistics, Geneva: UNCTAD, 2011, p xi. It includes Algeria, Angola, Iran, Iraq, Kuwait, Libya,
Nigeria, Oman, Qatar, Saudi Arabia, UAE and Venezuela.
25 The macroeconomic impact of remittances is examined by Y Abdih, R Chami, J Dagher & P Montiel,
Remittances and institutions: are remittances a curse?World Development, 40(4), 2012, pp 657666; IMF,
World Economic Report 2005,IMF website ; and UNCTAD,Least Developed Countries Report 2012, Geneva:
UNCTAD, 2012.
26 The Dutch disease,The Economist, 26 November 1977, pp 8283.
27 See UNCTAD,Trade and Development Report 2011, New York: UNCTAD,ch6.
28 Indifference curves are used to summarise aggregate demand. The model ignores the fact that changes in
income distribution will also cause these curves to shift.
29 These channels were initially described by WM Corden & JP Neary, Booming sector and de-industrialisa-
tion in a small open economy,Economic Journal, 92, 1982, pp 829831. For an overview in the case of
oil booms, see C Ebrahim-zadeh, Dutch disease: too much wealth managed unwisely,Finance and
Development, 40(1), 2003, at
30 The slope of the income line Y is less steep than that of Y. Their slope represents the weight of tradable
and non-tradable sectors in the economy, and the RER appreciates because tradables have become cheaper
relative to non-tradables.
31 For alternative congurations of capital and labour mobility and marginal productivity between sectors, see
Corden & Neary, Booming sector and de-industrialisation in a small open economy.
32 Barder, A PolicymakersGuide to Dutch Disease, Center for Global Development Working Paper 91,
Washington DC: Center for Global Development, 2006, at
33 See N Shaxson, New approaches to volatility: dealing with the resource cursein sub-Saharan Africa,
International Affairs, 81(2), 2005, pp 311324. See also Å Cappelen & L Mjøset, Can Norway be a Role
Model for Natural Resource Abundant Countries? UNUWider Research Paper no 2009/23, Helsinki, 2009;
and T Gylfason, The International Economics of Natural Resources and Growth, CESifo Working Paper
no 1994, Munich: CESifo, University of Munich. 2007.
34 A Chowdhury & T McKinley, Gearing Macroeconomic Policies to Manage Large Inows of ODA: The
Implications for HIV/AIDS Programmes,UNDP International Poverty Centre Working Paper No 17, New York:
undp, 2006; and JP Neary, Deindustrialization and the Dutch Disease, at
35 A Gelb, Windfall Gains: Blessing or Curse? Oxford: Oxford University Press, 1988.
36 J Stiglitz, We can now cure Dutch disease,Guardian, 18 August 2004.
37 A Saad-Filho, There is life beyond the Washington Consensus: an introduction to pro-poor macroeco-
nomic policies,Review of Political Economy, 19(4), 2007, pp 513537.
38 The positive implications for global convergence of the commodity boom starting in 2003 support the
argument that windfalls are not necessarily bad for long-term growth. See Report of the Secretary-General
of UNCTAD to UNCTAD XIII, 2012, at
39 Barder, A PolicymakersGuide to Dutch Disease; and AJ Venables, Resource Rents: When to Spend and
How to Save, OxCarre Research Papers, University of Oxford, 2010.
40 J Frankel, No Single Currency Regime is Right for All Countries or at All Times,NBER Working Paper
7338, Cambridge, MA: NBER, 1999.
41 Original sinis dened by B Eichengreen & R Hausmann, Exchange Rates and Financial Fragility, nber
Working Paper 7418, Cambridge, MA: NBER, 1999, as the inability to borrow abroad in the domestic cur-
42 E Levy-Yeyati & F Sturzenegger, Exchange Rate Regimes and Economic Performance,IMF Staff Papers,
47, Washington, DC: IMF, 2001.
43 KP Gallagher, S Grifth-Jones & JA Ocampo, Capital Account Regulations for Stability and Development,
Boston, MA: Boston University, Frederick Pardee Center for the Study of the Longer-Range Future, Issues
in Brief 22, 2011; and S Grifth-Jones, SouthSouth Financial Cooperation, 2012, at http://cgt.columbia.
44 These difculties should not be exaggerated, as sterilisation is used even by very poor countries. See
Chowdhury & McKinley, Gearing Macroeconomic Policies to Manage Large Inows of ODA, p 13.
45 UNESCWA,Survey of Economic and Social Developments in the escwa Region 20092010, New York:
United Nations, 2012.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
46 R Pollin & A Zhu, Ination and Economic Growth: A Cross-Country Non-Linear Analysis,PERI Working
Paper 109, Department of Economics, University of Massachusetts Amherst, 2005.
47 Barder, A PolicymakersGuide to Dutch Disease, p 11, claims that rather modest productivity [growth]
would be sufcient to offset the possible dynamic costs of export contraction resulting from Dutch Dis-
ease. See also Shaxson, New approaches to volatility, p 319.
Notes on Contributors
Alfredo Saad-Filho is Professor of Political Economy at the School of Oriental
and African Studies (SOAS), University of London. He has research interests in
the political economy of development, industrial policy, pro-poor economic
policies and the economic and social development of Latin America. He has
published widely on these and closely related topics. He is also the editor of
Anti-Capitalism: A Marxist Introduction, 2003; The Elgar Companion to
Marxist Economics (with B Fine), 2012; Economic Transitions to Neoliberalism
in Middle-Income Countries (with G Yalman), 2009; and Neoliberalism: A
Critical Reader (with D Johnston), 2005.
John Weeks is Professor Emeritus of the University of London, School of
Oriental and African Studies. His recent books include Capital, Exploitation
and Economic Crises (2011) and The Irreconcilable Inconsistencies of
Neoclassical Macroeconomics (2012). His recent policy work includes acting as
adviser to the Minister of Finance and Economic Cooperation of the Republic
of Sierra Leone and consultancy with the Central Bank of Kenya. He is author
of over 80 scholarly articles, and has authored or co-authored over 10 books.
Downloaded by [School of Oriental and African Studies] at 03:43 01 February 2013
... Some scholars have investigated the association among natural resources and corruption (Robinson et al. 2006). Saad-Filho and Weeks (2013) argued that natural resources become a curse only in countries that choose not to diversify their streams of revenue. This literature review will discuss several articles showing pragmatic evidence that having great quantities of natural resources could affect the economy, democracy levels, and quality of public institutions. ...
Full-text available
The Natural Resource Curse literature started with a clear consensus that dependence on natural resources have clear direct negative effects on economic growth and levels of democracy. However, the literature today reflects that the debate is still going where several papers reflects evidence that is against the consensus of the Natural Resource Curse hypothesis, which provides many open avenues for further research. This paper surveys the literature of the natural resource curse and identifies the main arguments and findings of both streams (curse stream and blessing stream). The main point that this paper highlights is that the literature is still not clear 100% whether the natural resource is a curse or a blessing. The econometric technique and the way how you define natural resources could lead to different or opposite results.
... The centrality of ground rent cycles and its appropriation forms is related to important debates in the international literature such as rentier states (Peters, 2017), resource curse/blessing (Ross, 1999) and Dutch disease (Saad-Filho and Weeks, 2013). This literature has mainly focused on the relation among oil-exporting countries, political institutions, and rent-seeking strategies (Cooley, 2001;Gilberthorpe and Papyrakis, 2015). ...
Full-text available
This paper measures with a novel methodology the amount of agrarian ground rent intersectoral transfers in Uruguay and its relative weight in the national amount of surplus value during the period 1955-2019. In order to this, the paper identifies the mechanisms through which ground rent is transferred from landowners to industrial and commercial capitals. The main results show that agrarian ground rent transfers, in particular through exchange rate overvaluation, was a recurring mechanism that has been central for capital accumulation. These transfers complement the surplus value appropriated by individual capitals, however, at the same time, consolidate the rentier nature of Uruguay due to the chronic overvaluation of exchange rate. KEYWORDS: Dutch disease; Latin America; land rent; capital accumulation; intersectoral transfers
... Similarly, resource curse studies have a number of theoretical problems. They make estimations based on selective and/or a sometimes narrow reading of history (Saad-Filho & Weeks, 2013). They fail to adequately explain how agricultural crops, energy and mineral resources promoted industrialization in some countries due to their ability to extract from colonies, exploit racial slavery, or violently settle in majority white and settler-colonial states like United States and Great Britain (Du Bois, 2014;Williams, 2014). ...
Full-text available
This paper offers an alternative explanation of the performance of the steel industry in Trinidad and Tobago (T&T). The steel industry emerged due to social mobilizations demanding changes to the colonial economy by investing natural resources in state‐led industrialization and welfare improvement. This investigation contends that the “resource curse” account that describes so‐called intrinsic social pathologies whereby postcolonial actors interfere with market freedom is misguided. A “resource curse” fundamentally offers a racialized conception of state corruption and incompetence within peripheralized economies. This study challenges this explanation showing how postcolonial arrangements reflect persistent colonial entanglements and contradictions, as resource dependence emerges from a long history of colonial exploitation and racialized subjugation within a global market predicated on a “global color line”. The T&T government's industrial effort was circumvented by both domestic and international economic and political pressures revealing a racial hierarchy and power asymmetries. Anti‐competitive sanctions from the (colonial) global economic powers United States and Europe, structural adjustment that maintained this system and improper dealing by foreign players were among these discriminatory practices. Similarly, elite behavior and political decisions reflecting (neo)‐colonial tendencies also undercut the sector's long‐term development. Together, these configurations reveal continuities, changes and contradictions within the island's postcolonial economic structure.
... Other approaches, such as the resource curse or those of the institutionalist IPE, have looked in more depth at the institutional conditions and material interests that perpetuate commodity dependence and its negative impacts on development (see, for example, Johnson, 1999;Rosser, 2006). However, although IPE has produced rich studies of resource dependency and theorized its development limitations (Nem Singh, 2010Saad-Filho and Weeks, 2013), the persistence and stickiness of commodity production demands a deeper explanation than these studies have provided. We know historical pathways create institutional patterns that 'lock in' societies (Liebowitz and Margolis, 1995;Mahoney, 2000); we also know that private firms and investors in the resource sectors can play a key role in maintaining political economies of extraction (Veltmeyer and Petras, 2014). ...
Full-text available
Many developing countries continue to rely on export‐oriented growth strategies based on primary commodities, despite the many limitations of such policies. The persistence of this model is inherently related to the dominance of ‘commodity imaginaries’. This article focuses on Argentina, an emblematic case of commodity dependency, where the soybean imaginary has dominated for the past 30 years. This imaginary has framed mainstream understandings of Argentina's path to growth and progress, shaped political contestation and ensured that a particular understanding of science and technology sits at the centre of the meaning of national development. In the process, it has transformed the country's geography in ways that normalize soy's dominance and invisibilize people and places located at the margins of the imaginary. The soybean imaginary renders a deeply political project of economic growth as ‘common sense’. This article concludes that closer attention to the way national development projects are shaped, consciously and unconsciously, by commodity imaginaries could help explain the puzzle of how national governments can become locked into development choices that are environmentally unsustainable and that reproduce inequalities.
... More generally, multi-disciplinary research is required to explore how the rapid articulation/disarticulation of illicit economies in rural peripheries of the global South compares with better-known dynamics underlying uneven development, such as well-studied historical patterns of boom-and-bust, or so-called "Dutch Disease," whereby national economies are destabilised by the influx of foreign exchange from natural resource rents (Saad Filho & Weeks, 2013). ...
Full-text available
Among geographers, recent focus on the illicit and illegal has tended to fall into two camps. Economic geographers focus on regimes of illicitness and corresponding production of specific forms of economic space; political ecologists and land change scholars, on the other hand, have scrutinized how illicit commodity flows shape land and resource use, especially in the global South. This paper offers an initial integration of these two relatively separate subfields, specifically in terms of their complementary attention to uneven development. Specifically, it uses the concept of ‘global commodity chains’ to explore the ways in which the regulation of agricultural commodities shapes how they are trafficked and embed in space, with particular attention to sites of international transshipment. When a commodity is illegal, spaces of transit take on significant analytical importance, particularly with respect to their socio‐ecological embedding. As a heuristic, we present a comparative mapping of two agricultural commodity chains linking Colombia and the U.S.: coffee and cocaine. Their comparison highlights how ‘illicitness’ fundamentally transforms cocaine’s spatiality, requiring risk evasion that results in characteristically enlarged transit spaces and a huge differential between producer and consumer price. We show how rents circulate in those transit spaces, socially and geographically embedding the commodities in diffuse, fluid networks with severe consequences for people and environments. We conclude with implications for work on illicit commodities and the collateral social and environmental harms they produce.
Full-text available
The goal of this study is to analyse the role of financial inclusion, governance, andremittances on growth in Sub-Saharan African countries (SSA), as well as themoderating influence of the role of financial inclusion on the remittances-growth nexus,using panel data spanning 1996–2020. Data were collected through secondary sources,including World Bank and IMF reports for the period 1996–2020. By using a principalcomponent analysis method, we constructed composite FI indexes to measure the degreeof FI. Cross-sectional dependence, slope homogeneity, and pooled mean grouping (PMG)are employed to evaluate the stated objectives. The study findings showed thatremittances have a significant positive relationship with economic growth, and thatpositive financial inclusion moderates the remittance-growth nexus in SSA countries.Given the findings that the level of governance has an adverse influence on growth, hencestakeholders should improve financial infrastructure, which provides the underlyinginstrument for financial inclusion, and protect customers by instituting controls andprocedures for reporting, fairness, and resorting to SSA countries. Governments shouldalso promote the larger global remittances agenda, which includes leveragingremittances for better consumer and business financing, and exposure to global financialmarkets through refinancing and the issue of diaspora bonds.JEL Classifications: C1; F6; G2; O4
Cette thèse a pour objectif principal d’apporter une meilleure compréhension de la relation entre les ressources naturelles (le pétrole) et le risque de défaut souverain en retenant comme cadre d’analyse la période 2010-2017 qui renferme la chute des prix du pétrole de 2014. Pour concrétiser cet apport ,elle est découpée en trois parties. La première partie correspondant au chapitre 1 expose la notion de défaut ou encore de risque de crédit et ses méthodes de mesures et les mécanismes d’ajustement aux chocs des prix du pétrole à travers une large revue de la littérature tout en dressant le fonctionnement du marché pétrolier. Le chapitre 2 qui est la première partie empirique de notre thèse mobilise des régressions simples ainsi que des études de cointégration et des analyses causales dans un cadre bivarié pour étudier l’ampleur et le sens de la relation existant entre la volatilité des CDS (proxy de la qualité de crédit des pays exportateurs) et les variables du marché pétrolier dans un premier temps. Dans un second temps, nous utilisons une modélisation multivariée avec des restrictions appropriées reflétant le fonctionnement du marché pétrolier pour étudier la réaction de la volatilité des marchés des CDS (Proxy du marché des CDS) aux chocs des prix. Nous montrons que la réaction de la volatilité des CDS est différenciée selon la nature du choc qui sous-tend le choc du prix du pétrole et que la chute de 2014 à contribuer à renforcer cette liaison pour certains. Le chapitre 3 complète l’analyse précédente en étudiant le comouvement entre le marché des CDS (credit default swaps) des pays exportateurs de pétrole et les différents compartiments du marché pétrolier (marché physique et marché à terme) en combinant la théorie des copules aux modèles de corrélation dynamique. Nous établissons dans un premier temps que la dépendance entre les rendements des CDS et les rendements du marché du pétrole sont caractérisés par des dépendances différentes s’éloignant d’une copule normale exhibant une dépendance asymétrique. Enfin nous établissons que cette dépendance est variable dans le temps et instable en fonction du pays considéré dans l’échantillon avec des phénomènes de contagion marqué par la dernière chute de pétrole de 2014. Un résultat complémentaire que nous avons pu établir sur le plan économétrique, est la supériorité des modèles de corrélation dynamique augmentée par des distributions de type copule et la nécessité de retenir des mesures afférentes comme le tau de Kendall pour apprécier la dépendance entre les marchés surtout en période de stress.
Full-text available
By focusing on a half-century of recent Latin American economic history, this book presents a multidisciplinary approach to examining the relentless pursuit of development in the Global South and aims to revitalise the academic debate on whether having abundant natural resources is a blessing or a curse. Its pioneering diachronic comparative approach of analysing two Ecuadorian oil booms, those in 1972–1980 and 2003–2014, reveals processes of continuity and change in the capacity of this peripheral state to intervene in its national development process and the consequences of this on its social formation, framed by the contemporary trends of global capitalism and the irruption of environmental thinking into development policymaking.
Full-text available
By focusing on a half-century of recent Latin American economic history, this book presents a multidisciplinary approach to examining the relentless pursuit of development in the Global South and aims to revitalise the academic debate on whether having abundant natural resources is a blessing or a curse. Its pioneering diachronic comparative approach of analysing two Ecuadorian oil booms, those in 1972–1980 and 2003–2014, reveals processes of continuity and change in the capacity of this peripheral state to intervene in its national development process and the consequences of this on its social formation, framed by the contemporary trends of global capitalism and the irruption of environmental thinking into development policymaking.
Violent conflict can spell catastrophe for developing countries and their neighbors, stunting and even reversing the course of economic growth. Recent World Bank research on the causes of conflict and civil war finds that the countries most likely to be blighted by conflict are those whose economies depend heavily on natural resources. Natural Resources and Violent Conflict: Options and Actions first explains the links between resource dependence conflict and then considers what can be done to help reduce the risk of civil war in these nations. In this collection of previously unpublished essays by experts in the field, contributors consider the risks of corruption, secessionist movements, and rebel financing. They also consider the roles played by government, the development community, and the country’s population and propose an agenda for global action. Focusing on what we can do collectively to diminish the likelihood of civil war, contributors to this volume suggest practical approaches and policies that could be adopted by the international community—from financial and resource reporting procedures to commodity tracking systems and enforcement techniques, including sanctions, certification requirements, and aid conditionality. A fascinating look at the results of important new World Bank research, this book represents an important addition to the dialogue on development.
Case studies as well as cross-country studies suggest that countries with an abundance of natural resources are more prone to violent conflict. This collection of articles analyzes the link between natural resources and civil war in a number of different ways. So far the literature falls broadly into two camps. First, in the economics literature the well-documented “resource curse” leads to low-income growth rates and low levels of income. These in turn constitute low opportunity costs for rebellion and make civil war more likely. On the other hand, political science literature concentrates on the link between natural resources and weak institutions. States with natural resources often rely on a system of patronage and do not develop a democratic system based on electoral competition, scrutiny and civil rights. Based on further empirical evidence in this volume we conclude with a brief overview of current policy initiatives.
Countries in the Middle East and North Africa (MENA), known for their natural resource riches, have often shown classical symptoms attributed to oil rentier states: heavily distorted economies, chronic political unrest and undemocratic governments. To explain the inverse relation of these countries' economic performance to their natural resource endowment, analysts have offered the Dutch disease hypothesis; this study, however, explores a historically oriented approach. The alternative thesis is that institutions in these countries were imprinted at their ‘rebirth’ or decolonisation by conditions with long-lasting influence on their economic performance. Using a cross-country growth model, the study offers statistical support indicating that contemporary economic growth variations have been impacted by MENA's ‘rebirth’ conditions through variations in government, investment, population growth and trade strategies. The evidence also depicts a weakening of the above relation over the more recent years, an outcome that may be attributed to economic reform initiatives that most MENA members have embarked on since the late 1980s.
This paper addresses the complex and overlooked relationship between the receipt of workers' remittances and institutional quality in the recipient country. Using a simple model, we show how an increase in remittance inflows can lead to deterioration of institutional quality - specifically, to an increase in the share of funds diverted by the government for its own purposes. Empirical testing of this proposition is complicated by the likelihood of reverse causality. In a cross section of 111 countries we document a negative impact of the ratio of remittance inflows to GDP on domestic institutional quality, even after controlling for potential reverse causality. We find that a higher ratio of remittances to GDP is associated with lower indices of control of corruption, government effectiveness, and rule of law.
Much attention has been given in recent years to the paradoxical fact that huge flows of money from petroleum appear not to have brought prosperity to the African countries that produce it, but may instead have helped cause poverty, economic decline and conflict. Issues such as human rights abuses near oil installations and environmental damage have often captured the headlines, but these, while important, are peripheral to the main problems: the Dutch Disease, whereby an influx of oil money causes real exchange rates to appreciate, making local industry and agriculture uncompetitive; the damage that petroleum money causes to institutions, incentives and overall governance; and the volatility of oil prices and revenues. This article will look at the volatility problem, and how oil contracts tend to make matters even worse. They are like this for long-established technical, political and historical reasons, and there is consequently a widespread belief in the industry that change is not possible. This defeatist attitude needs to be vigorously challenged.
Our goal is to show that contrary to the claims made in several recent papers, the effect of a large endowment of oil and other mineral resources on long-term economic growth of countries has been on balance positive. Moreover, the claims of a negative effect of oil and mineral wealth on the countries' institutions are called into question. Copyright by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
This paper investigates the sources of economic stagnation in sub-Saharan Africa during the 1970s. Several policy variables are considered, as well as a set of ‘environmental’ variables which include climate, violence, export prices and other exogenous factors. The paper uses statistical analysis in an attempt to identify the relative importance of the policy variables and the ‘environmental’ variables, taken as groups. It also seeks to identify particular policy variables which seem to have been most significant in determining growth outcomes.