Pergamon PIh S0301-4207(97)00036-6
Resources Poli¢y. Vol. 23, No. 4, pp. 191-199. 1997
© 1998 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
0301-4207/98 $19.00 + 0.00
Explaining the resource curse, with
special reference to mineralexporting
countries
Raymond F Mikesell
Department of Economics, College of Arts and Sciences, 1285 University of Oregon, Eugene, OR 97403-
1285, USA
Recent studies have found an inverse correlation between economic growth and natural
resource abundance among developing countries. There appears to be no obvious explanation
for this finding, such as an important growth variable that is common in resource-poor countries
and deficient in resource-abundant countries. The resource curse hypothesis is closely
related to the problem of sustainability for resource-exporting countries since periods of high
growth are frequently followed by long periods of stagnation. Models are examined showing
how increases in export income can affect relative prices in the trade and nontrade sectors
and how the changes in relative prices affect investment and growth. It is the thesis of this
article that there is nothing inherent in resource-abundance that condemns countries to either
low growth or nonsustainability. © 1998 Elsevier Science Ltd. All rights reserved
Keywords: mineralresources,economicgrowth,per capitagrowth,resourcescarcity,resourceabundance
The purpose of this paper is to explore the reasons
for the relatively poor growth performance of mineral
exporting countries. The mineral exporting countries
are the most visible victims of the resource curse
hypothesis, which holds that resource-abundant countries
tend to perform more poorly in terms of sustainable
growth than resource-poor countries. Most
mineral exporting countries had relatively high rates
of growth during the 1960-1980 period when mineral
prices were rising. However, their growth rates
declined sharply during the 1980-1993 period, and a
substantial proportion of them experienced negative
per capita growth rates during the latter period. For
these countries, periods of high export earnings have
not given rise to sustained growth.
Traditionally, abundant natural resources have been
regarded as contributing to successful development
and have provided an explanation for the growth of
Western countries, such as Australia, the U.S. and
Canada (North, 1963). However, during the past three
decades, the star performers among developing counJThis
paper was presentedat the annual Western Social Science
Association Conference,23-27 April 1997,in Albuquerque,New
Mexico.
tries, such as Korea and Taiwan, are resource poor,
while the majority of the resource-rich countries, such
as Argentina, Mexico, Peru, Saudi Arabia and Venezuela,
have had lower than average annual rates of
gross domestic product (GDP) growth in the 1980-
1993 period, and negative per capita gross national
product (GNP) growth. Several econometric studies
covering nearly all developing countries have provided
impressive evidence of the relatively slower
rate of growth of the resource-rich countries over the
past several decades (Nankani, 1979; Gelb et al.,
1988; Auty, 1986, 1993; Sachs and Warner, 1995).
For example, Sachs and Warner (1995) found a negative
relationship between per capita growth rates and
the ratio of natural resource exports to GDP for
eighteen developing countries over the 1971-1989
period. The authors found this result to remain significant
in cross-country growth regressions, after taking
account of a large number of additional variables,
including initial GDP, trade policy, investment rates,
and trade volatility. While relatively poor per capita
growth performance characterized resource-rich
developing countries generally, poor per capita
growth was especially manifest in mineral exporting
countries (Table 1). Thus, over the period 1980-1992
191
Explaining the resource curse: R F Mikesell
Table 1 Exports of fuels, minerals and metals as a percentage of total exports, per capita GNP (1992 dollars), and average annual GNP
growth rates, 1980-1992
Share of exports GNP Average annual growth
Country (%) (1992 $) (%)
Sierra Leone 34 160 - 1.4
Niger 86 280 4.3
Nigeria 96 320 0.4
Togo 45 390 - 1.8
Mauritania 84 530 -0.8
Indonesia 38 670 4.0
Bolivia 66 680 1.5
Papua New Guinea 52 950 0.0
Peru 49 950 2.8
Congo 92 1030 -0.8
Ecuador 45 1070 -0.3
Jordan 34 1120 -5.4
Colombia 29 1330 1.4
Algeria 97 1840 0.5
Iran 90 2200 1.4
Chile 47 2730 3.7
Venezuela 86 2910 - 0.8
Mexico 34 3470 0.2
Trinidad/Tobago 64 3940 2.6
Gabon 89 4450 3.7
Oman 94 6480 4.1
Saudi Arabia 99 7510 3.3
Botswana na 2790 6.1
All low- and middle-income -- -- 0.9
countries:
Source: World Bank (1994).
Note: To the above list we may add Zaire and Zambia, which have a highproportion of mineral exports, but negativeper capita GNP growth rates.
the average annual per capita GNP growth rate of
twenty-three mineral exporters was a negative 0.5%,
and only five of these countries (Indonesia, Colombia,
Chile, Oman and Botswana) had positive rates of per
capita GNP growth (Table 1). For all low- and
middle-income countries, the average rate of growth
in per capita GNP for this period was 0.9%. In a
World Bank Staff Working Paper, Nankani (1979)
found that mineral economies have been less successful
in economic performance: lower rates of growth,
lower levels of social welfare, and more highly
skewed income distributions than nonmineral LDCs.
Gelb et al. (1988) found the same thing specifically
for oil exporting countries, and that for these countries
the prudent use of boom windfalls was the exception
rather than the rule.
It is not my purpose to discuss the statistical findings
supporting the resource curse. Rather, I want to
examine the reasons that have been advanced for the
growth bias, specifically for mineral-exporting countries.
Any single explanation for the relatively slow
growth of resource-abundant countries must be shown
to be uniquely characteristic of these countries and
have a negative correlation with growth. Simply being
a resource-abundant country is not a satisfactory
answer. There are important exceptions to this slow
growth bias among resource-abundant countries, and
the explanations advanced for the resource curse
cover a range of economic, structural, micro and
macro policy, and political factors: the same combination
of which rarely applies to more than one country.
Therefore, it is difficult to identify growthretarding
factors common to a large number of
resource-abundant countries.
The resource curse is paradoxical because production
of natural resources (1) has been the initial
source of nearly all development, (2) provides an
almost immediate source of foreign exchange, (3)
attracts foreign capital and skills, and (4) provides
both raw materials for processing and a market for
manufactured inputs. The traditional position that
natural-resource exports are a primary engine of
growth began to be questioned in the development
literature following World War II. One of the first
challenges to the efficacy of primary commoditydriven
export growth was made by Raul Prebisch
(1964) who argued that over the long term, prices of
primary commodities decline relative to the prices of
manufactured goods, thereby causing primary
exporting economies to stagnate. However, evidence
supporting Prebisch's terms-of-trade hypothesis is
very weak. Firstly, there is no significant correlation
between long-term declining real export prices and
the rates of economic growth of exporting countries
(Maizels, 1968). Secondly, Kindleberger (1956) and
others have not found statistical support for the existence
of long-term declines in the terms-of-trade of
developing countries. In addition, the significance of
long-term changes in terms-of-trade is doubtful
because of constant improvements in the quality of
192
Explaining the resource curse: R F Mikesell
manufactured goods. Also, there have been substantial
reductions in production costs for many primary
commodities as a result of increased productivity, so
that the economic rents produced by the natural
resources have not declined with the decreases in primary
commodity prices.
Another challenge to the contribution of natural
resource industries to growth is provided by Hirschman's
(Hirschman, 1958) analysis of forward and
backward linkages, and his finding that resource
industries have lower linkage effects than manufacturing
and are, therefore, less effective in transmitting
growth to the rest of the economy. Yet, primary commodity
exports have provided the initial stimulus to
manufacturing and there are many examples of parallel
and symbiotic development of resource industries
and manufacturing, such as that between mining and
industrial development in the American West. Without
the foreign exchange income from primary commodity
exports, most countries would have been
unable to buy the capital goods and intermediate
inputs to develop the manufacturing sector.
Substantial volatility in export revenue is sometimes
advanced as the reason for slow growth in primary
commodity exports, and it is certainly true that
terms-of-trade volatility is higher for primary commodity
exporters than for industrial countries. Over
the period 1960-1980, the standard deviation of
annual percentage price changes for forty-nine primary
commodities was 26.4%, while the deviation in
the World Bank Overall Commodity Price Index was
half that percentage. The terms-of-trade volatility of
the regions with the highest primary export shares
(Latin America, sub-Saharan Africa, the Middle East,
and North Africa) were two to three times that of
industrial countries over the 1970-1992 period.
[Terms-of-trade volatility is defined as the standard
deviation of the percentage growth rate of the terms of
trade (Westley, 1995)]. However, Sachs and Warner
(1995) did not find a strong relationship between
terms-of-trade volatility and per capita growth.
Some mineral-exporting countries do not save and
invest enough to compensate for the depletion of their
reserves. However, most mineral-exporting countries
have been able to expand their mineral output over
the past three decades, and in only a few cases have
countries been forced to curtail output because of
reserve depletion. The latter will undoubtedly occur
in a number of countries over the next few decades,
but it cannot be regarded as a significant cause of
the relatively slow growth in most mineral-exporting
countries over the past three decades. Gross investment
as a percentage of GDP declined for most mineral
exporting countries between 1970-1980 and
1980-1993. This suggests they are not investing a sufficient
amount in productive industries to replace the
reduction in export earnings following a decline in
export prices. For example, in Saudi Arabia, much of
the oil revenues have gone for consumption by the
middle- and upper-income classes, for defense, and
for public works that yield a low rate of return. The
country has failed to direct enough oil revenue for
investment in manufacturing, agriculture, and education.
Although I fail to find any single explanation
as to why countries heavily dependent on mineral
exports should have relatively low growth, these
economies do require special policies in dealing with
adjustments arising from fluctuations in mineral
exports.
The economics of export booms
Important to the explanation of the resource curse is
the distortion created by export booms in the economies
of the primary-commodity-exporting countries.
These distortions are not simply transitory and disappear
when the boom subsides, they affect the structure
of production and investment required for future
growth in a manner that may impair sustainability
(Sachs and Warner, 1995, Appendix A).
If revenues from mineral exports were stable or
expanding slowly, there would be no special economic
impacts generating distortions that inhibit sustainable
growth. Mineral-exporting countries are subject
to temporary surges in export income, mainly
from increases in world mineral-prices, but in some
cases from new discoveries, such as a large petroleum-reservoir.
Whatever the cause of the rise in
export income, it levels off or declines so that it is
no longer a source of increased income. Export booms
not only increase domestic income, but affect savings
and investment, government expenditures, and relative
prices in different sectors of the economy. These
economic impacts differ significantly among countries,
and depend on the structure of the economy and
the policies adopted by the government. However, in
most cases, distortions occur that tend to depress output
and investment in key sectors. The following
paragraphs outline a typical set of responses to an
export boom by a developing country heavily dependent
on primary-commodity exports.
(1) A surge in foreign exchange sold in the foreignexchange
market by both government and private
firms causes the exchange rate to appreciate and
domestic income to rise. The combination of a
rise in the nominal exchange rate and domestic
price inflation results in a rise in the real
exchange-rate. (I define a rise in the real rate of
exchange as a decline in the ratio of the price
index of tradeable goods to the price index of
nontradeable goods.2)
(2) The real exchange-rate appreciation tends to
reduce relative prices for tradeable products
2This definition of the real exchange rate is used by Edwards
(1989). The reason for my preference for this definition of real
exchange rates, as contrasted with the purchasing power parity
method, is that it measures the effects on sectoral prices which, in
turn, have a major impact on the economy.
193
Explaining the resource
(manufactured goods and agricultural products)
relative to prices of nontradeables (construction
and services), so that labor and capital are withdrawn
from the tradeable sector and flow into the
nontradeable sector.
(3) Exports of nonresource tradeables decline and
imports rise. Governments often respond to an
increase in imports and fall in exports by
imposing import restrictions and subsidizing
exports. This brings about further distortions by
attracting investment to high-cost importing-substituting
manufacturing. Higher prices for manufactured
goods depress agriculture and further
reduce the competitiveness of all tradeables in the
export markets, including mineral exports.
(4) Foreign capital may be attracted by investment
opportunities in the export boom sector, which
may cause further appreciation of the real
exchange-rate. There may also be an increase in
foreign borrowing made possible by an improved
credit standing of the government and private
business. The proceeds of foreign borrowing by
the government are often channeled into the nontradeable
sector in the form of low-yield public
works, defense outlays, and social projects that
expand consumption. The increased capital
imports may lower interest rates, which induces
domestic capital to go abroad in response to
higher earnings.
(5) The movement of resources between sectors may
reduce capital accumulation. If the nontradeable
sector is relatively labor intensive, while the tradeable
sector is capital intensive, the movement
in favor of the nontradeable sector will tend to
raise wages and lower returns to capital, thereby
reducing capital accumulation.
(6) When the export boom subsides and the current
account turns from surplus to deficit, the central
bank may prevent a decline in the nominal
exchange rate by intervening in the market so that
the real rate of exchange remains high. Even if
the real rate of exchange declines, it will require
several years for the tradeable sector to recover
from the price distortions occurring during an
export boom. It may also require several years for
the government's fiscal expenditures to decline in
line with the decline in income and export taxes.
The typical effects of an export boom outlined
above have been called the 'Dutch Disease', a term
derived from the experience following a sharp rise in
Holland's natural gas exports. When the export boom
subsides and primary-commodity exports return to
their pre-boom level or lower, growth may be reduced
below the rate existing before the boom for the following
reasons: (a) the economy is distorted,
especially by the reduced output of tradeables; (b) tradeables
become less competitive in world markets
(including the domestic market) owing to the appreciation
of the real exchange rate; (c) and saving and
curse: R F Mikesell
investment decline. There may also occur an increase
in external indebtedness, which increases payments
abroad and may lead to debt crises and capital flight.
Over time, adjustments are likely to be made to eliminate
the distortions, perhaps with the assistance of
the International Monetary Fund or the World Bank.
However, meanwhile, the country has lost ground as
a consequence of reduced investment, and may never
recover the rate of growth experienced before the
export boom.
Policies for avoiding or moderating Dutch
Disease symptoms
The symptoms of Dutch Disease noted above can be
moderated or largely avoided by adopting appropriate
governmental policies. The increased foreign
exchange income can be prevented from having a
sudden large impact on domestic income and on the
real exchange rate. Most of the export windfall can
be channeled into a stabilization fund, or otherwise
prevented from suddenly increasing domestic
demand. If the government is the net recipient of the
windfall, as in the case of state mining enterprises,
the government could accumulate a large portion of
the increased receipts, or the receipts could be sterilized
by the central bank. A second policy action
would be to prevent a significant appreciation of the
real exchange rate. The central bank could require the
export revenue windfall to be sold directly to the central
bank, or the central bank could purchase foreign
exchange on the exchange market to prevent an
increase in the nominal exchange value of the domestic
currency. Even if the central bank prevents the
nominal exchange value of the domestic currency
from rising, the increase in domestic demand may
have the effect of increasing prices of nontradeables
relative to tradeables so that there is real exchangerate
appreciation, as I have defined it. This may call
for the nominal depreciation of the exchange rate by
the central bank.
A third policy measure is for the government to
allocate its increased income to productive investment.
If the government uses the increase in its
receipts arising from the increase in exports to
increase social programs that increase consumer consumption,
or channels the increased revenues into
low-yield infrastructure projects or into defense, the
increased export income will not be invested in ways
that will sustain income. Temporary export windfalls
should be invested in income-earning projects or in
foreign exchange reserves, which can be used to
maintain investment in periods of low export earnings.
These policies should be integrated with appropriate
fiscal and monetary policies to prevent inflation.
A legitimate question is whether the shocks caused
by primary commodity export booms are more serious
than the shocks experienced by resource poor countries.
Perhaps they are, but they can be handled by
adopting appropriate policies. Several mineral-
194
Explaining the resource curse: R F Mikesell
exporting countries have experienced Dutch Disease
symptoms, but some have been able to avoid impairment
of long-term growth. All countries experience
shocks from a loss of major export markets, from
sharp increases in prices of essential imports, or from
shifts in capital movements. How the shocks are
handled determines whether they will prevent sustainable
development.
Growth experience of mineral-exporting
countries since 1960
The growth records of mineral exporting countries
over the past three decades reflect, in part, the effects
of fluctuations in export revenue, but they also reflect
symptoms of the Dutch Disease and various structural
changes in the individual economies. The 1970-1980
period was one in which substantial export booms
occurred in nearly all mineral-exporting countries.
During this period, the average price of metals
increased three-fold and the price of crude petroleum
rose fifteen-fold. The price of copper approximately
doubled and prices of most other metals, including
aluminum, lead, tin, nickel and zinc rose two- or
three-told. However, between 1980 and 1985, metal
prices declined sharply. Prices rose moderately in the
late 1980s and 1990s, but never regained the earlier
peak levels. Crude-petroleum prices declined drastically
in the 1980s and remained quite low into the
1990s. During the 1960-1970 decade, both metals and
petroleum were fairly stable and were, on the average,
about the same as they were in 1970-1971 before the
export booms took place.
A comparison of the average annual GDP growth
rates of major petroleum and nonfuel mineral
exporters for the periods 1960-1970, 1970-1980, and
1980-1993 (Table 2) shows that for nearly all the
countries listed, average annual GDP growth rates
declined during the 1980-1993 period from those of
the boom period, 1970-1980. 3 Only in the case of
Chile, Jamaica, Papua New Guinea (PNG), and Oman
were the growth rates higher during the 1980-1993
than in the 1970-1980 period. For nearly all these
countries, average annual GDP growth rates during
the pre-boom period, 1960-1970, were higher than
the growth rates for the post-boom period of 1980-
1993. The only exceptions were Chile and Indonesia,
whose average annual GDP growth rates in the 1980-
1993 period were significantly higher than in the pre-
3If GDP for the mineral exporting countries was adjusted for the
depletion of mineral reserves, decreases in GDP over the 1980-
1993 period would have been even greater. For example, Repetto
et al. (1989) adjusts the Indonesiannationalaccountsover the period
1971-1984 for the depletion of the three principal naturalassets:
hydrocarbons,forests, and soils. Inclusion of natural-asset
depletion reduced the Indonesianeconomicgrowth rate from7.1%
per annum, using the standard national accounting system, to a
sustainablegrowthofonly4% per annum.About70% ofthe downward
adjustmentis due to hydrocarbondepletion,which accounted
for a reduction of around 2.1% in the annual GDP growth rate.
boom period. This record is in line with the resourcecurse
thesis that growth rates following resource
booms do not return to pre-boom levels. However,
there are obviously other factors involved since the
growth pattern for all low- and middle-income
developing countries for the three periods was similar
to that for the major mineral exporting countries (see
Table 2). Average annual per capita GNP growth was
negative for the 1980-1993 period for more than half
the mineral exporting countries listed, as contrasted
with respectable per capita GNP growth rates during
the 1960-1970 period. There was also a sharp decline
in per capita GNP growth rates for all low- and
middle-income countries between the two periods, but
the decline was not as large as for most of the mineral-exporting
countries (Table 2).
The effects of export booms on individual
mineral-exporting countries
An analysis of the economic conditions and policy
responses in the individual countries listed in Table
2 reveals major differences in the factors responsible
for growth performance. While most experienced
some symptoms of the Dutch Disease, other factors
(such as changes in mineral production caused by the
opening or closing of major mines, changes in mineral
export policy, or shifts in governmental objectives
such as promoting social programs at the expense of
investment), were as much or more responsible for
their growth patterns than the factors described by the
export boom model.
PNG and Botswana
PNG's GDP growth rate has fluctuated with mine production
declining sharply with the closure of its Bougainville
copper mine in 1988, but rose to an annual
average of 9.3% in the 1990-1995 period with the
expansion of the Ok Tedi gold/copper mine and the
recovery of copper prices. The Dutch Disease played
no discernible role. Nor did the Dutch Disease appear
to have played a significant role in Botswana. The
initial double-digit rate of growth in the 1960-1980
period was generated by the opening of diamond
mines. There was a decline in the rate of growth after
the large initial investments, but growth was sustained
at nearly 10% over the 1980-1993 period.
Jamaica
Jamaica's average annual GDP growth rate declined
from 4.4% in 1960-1970 to a negative rate of growth
in 1970-1980. This decline occurred despite the fact
that the country's bauxite and alumina exports peaked
in 1973. Exports declined sharply in the second half
of the 1970s, largely as result of the government's
imposition of a levy on bauxite in 1974, which made
the product less competitive on world markets. The
decline in Jamaica's GDP growth rate from 6.7% in
1970-1973 to -2.2% in 1974-1979 was also due to
195
Explaining the resource curse: R F Mikesell
Table 2 GDP growth rates for major mineral-exporting countries, 1970-1980 and 1980-1993 (%)
GDP Average per capita GNP
growth rate
1960-70 1970-80 1980-93 Increase/decrease Change: 1970- 1960-77 1980-93
80 to 1980-93
Major oil exporting LDCs: a
Algeria 4.3 4.6 2.1 -2.5 - 54 2.1 0.8
Ecuador 2.2c 9.5 2.4 -7.1 - 75 3.1 0.0
Indonesia 3.9 7.2 5.8 - 1.4 - 19 3.3 4.2
Mexico 7.6 6.3 1.6 -4.7 -75 2.8 -0.5
Nigeria 3.1 4.6 2.7 - 1.9 - 41 3.6 - 0.1
Oman 21.9c 6.2 7.6 1.4 23 -- 3.4
Saudi Arabia 11.2~ 9.0 0.4 -8.6 -96 6.7 -3.6
Trinidad & 4.0~ 5.9 -3.6 -9.5 - 161 1.6 -2.8
Tobago
Venezuela 6.0 3.5 2.1 - 1.4 -40 2.7 0.7
Total average 7.6 6.3 2.3 -4.0 -63 3.2 -0.9
annuala
Major nonfuel mineral exporting LDCs:b
Bolivia 5.2 4.5 1.1 - 3.4 - 76 2.3 0.7
Botswana 14.8c 14.5 9.6 -4.9 -34 -- 6.2
Chile 4.4 1.8 5.1 3.3 183 1.0 3.6
Jamaica 4.4 - 1.3 2.3 3.6 -- 2.1 -0.3
PNG 6.7 2.2 3. I 0.9 41 3.4 0.6
Peru 4.9 3.5 -0.5 -4.0 - 114 2.3 2.7
Zambia 5.0 1.4 0.9 - 1.3 - 21 1.5 -3.1
Total average 6.5 3.8 3.1 -0.7 18 2.1 0.0
annuald
All low- and 5.6 5.2 2.9 -2.3 -44 3.0 0.9
middle-income
countries: d
a Excludes Iran, Iraq, Bahrain, Libya and Qatar for lack of data. Excludes Israel, Hong Kong, Kuwait, Singapore and the United Arab Emirates
which are regarded as high-income countries.
h Countries for which nonfuel minerals constitute 30% or more of export income. Excludes some countries for lack of data.
c 1965-1973.
o Unweighted.
Sources: World Bank (1979, 1983, 1995).
the policies of the social democratic government,
which neglected investment in favor of social programs
and adopted monetary and fiscal policies which
generated a 27% rate of inflation. The Jamaican economy
recovered under a new right-of-center government,
which came into power in 1980, but growth was
impaired by an appreciation of the real rate of
exchange of almost one-third between 1980 and 1983
(Findlay and Wellisz, 1993).
Peru
Peru, whose GDP growth rate declined from 3.5% in
1970-1980 to a negative 0.5% in 1980-1993, is closer
to the Dutch Disease model than Botswana or PNG,
since real appreciation of the exchange rate during the
1980s had a depressing effect on Peru's agriculture
and manufacturing sectors. Despite sharp increases in
copper and other metal prices, Peru's GDP growth
rate in 1970-1980 was less than that in 1960-1970
when metal prices were lower. Extensive governmental
intervention, high import protection, fiscal
deficits, and high inflation also played a role in
depressing productive investment.
In addition, Peru had debt crises from time to time
because of borrowing for current expenditures not
directed to productive investment.
Venezuela
Increases in oil prices in 1973-1974 and 1979-1980
provided Venezuela with two export booms with the
windfalls going mainly to the government. Paradoxically,
Venezuela's average annual GDP growth rate
declined to 3.5 percent in 1970-1980, from a growth
rate of 6 percent in 1960-1970 when oil prices and
export revenues were lower. There was a further
decline in the country's GDP growth rate to 2.1 percent
for the 1980-1993 period, and average per capita
GNP growth declined from 3.2 percent over the period
1960-1977, to a negative 0.9 percent in 1980-
1993. Venezuela might be regarded as a classic
resource curse case in the sense that a surge in
resource-based exports was accompanied by
reductions in the rate of growth and in per capita
income. However, the poor performance of the nonoil
sectors of the economy, including agriculture and
manufacturing, was due more to faulty governmental
196
Explaining the resource curse: R F Mikesell
management than to distortions stemming directly
from the export booms. Much of the windfall from
the export booms went for increased consumption,
and the windfall used for investment went largely for
government-sponsored industrialization (e.g. steel and
aluminum) which yielded low returns (Auty, 1986).
There was real exchange-rate appreciation in 1979-
1981 which, together with increased real wages,
brought about a contraction in nonoil output and discouraged
investment. Controls over nominal interest
rates in the face of inflation led to a reduction in real
interest-rates which, along with an overvalued currency,
induced substantial private-capital outflow in
the early 1980s. An analysis of the Venezuelan economy
published by the World Bank concludes that
"nothing appears to have been gained from the windfalls
in terms of non-oil GDP during 1973-1982. Consumption
has been the only winner, and even that gain
is probably temporary" (Gelb et al., 1988, p. 322).
Have the oil windfalls been a curse to Venezuela
and would the country have been better off if petroleum
prices had stayed at the 1960 levels when
Venezuela's GDP growth rate was much higher?
Would the absence of the export boom have promoted
good government? I doubt if Venezuela's performance
would have been much better even without the
direct effects of the export booms.
Chile
Chile's favorable record was due to the policies
adopted by the government to correct foreign
exchange-rate appreciation and moderate the other
distortions generated by export booms (IMF, 1996,
pp. 404~05). Chile's first export boom (1973-1974)
occurred during a period of severe economic and
political instability, which led to the overthrow of the
Allende government in 1973. Despite an increase in
copper prices, real GDP declined during the Allende
regime, there was substantial inflation, and production
was disrupted by the nationalization of the banks and
industries. The military government under Pinochet
instituted a severe anti-inflation program and devaluated
the real exchange rate, which had appreciated
significantly as a consequence of the 100% inflation
rate during the Allende regime. Nominal exchangerate
depreciation tended to lag behind inflation so
there was periodic real exchange-rate appreciation
during the 1970s. Nevertheless, trade liberalization
and reduced domestic demand boosted noncopper
exports in 1976-1978. Chile's second export boom,
which occurred in 1979-1980 with a sharp rise in
copper prices, was accompanied by substantial
growth in real GDP. The decline in copper prices in
1981 led to a rise in the current account deficit
financed by foreign debt. However, GDP grew at
about 5% annually during the remainder of the 1980s
and there was only a modest appreciation of the real
exchange rate. Since the mid-1980s, the Chilean
government has managed the economy well in the
face of export booms and subsequent low copper
prices. The government retained export windfalls in
a stabilization fund and prevented the non-mining tradeable
sector from being suppressed from the effects
of the Dutch Disease.
Indonesia
Indonesia experienced an appreciation of its real
exchange-rate in two oil booms (1974-1978 and
1979-1981). However, in 1983 the government sterilized
a substantial portion of the windfalls, substantially
devalued the currency, and began trade liberalization.
GDP growth recovered in the second half of
the 1980s, despite the sharp fall in petroleum prices
in 1982 (Gelb et al., 1988, Ch. 12). Following the
decline in petroleum prices, the Indonesian government
cut public spending and reduced tariff barriers,
both of which promoted competitiveness. Overall,
Indonesia managed the fluctuations in its petroleum
revenue with considerable skill as contrasted, say,
with Venezuela or Mexico, both of which are also oil
exporters with diversified economies.
Oman
Among the major oil exporters, only Indonesia (4.2%)
and Oman (3.4%) had positive rates of growth in per
capita GNP during the 1980-1993 period (Table 2).
Yet all the oil exporters had positive per capita GNP
growth rates in the 1960-1977 period, the highest
being Saudi Arabia with a 6.7% rate. While the
experience of most oil exporters is certainly in
accordance with the resource curse, a comparison of
the economic structures of good performers with
those of poor performers provides little hint of the
sources of the resource curse. For example, Oman has
very little manufacturing or agriculture and is almost
wholly dependent upon petroleum for export income.
Oman has had good financial management and has
accumulated significant foreign exchange reserves
over the 1980-1993 period. Its GDP growth rate in
1980-1993 exceeded its GDP growth rate in 1970-
1980 and the per capita GNP growth rate was 3.4%
in 1980-1993. By contrast, Saudi Arabia failed to
invest its large export windfall in either domestic production
or foreign exchange reserves.
Conclusions from country reviews
My conclusions from this brief analysis of the experience
of a group of mineral exporting countries are
as follows:
(1) The Dutch Disease was not the major factor in
explaining the growth pattern for more than half
of the countries reviewed.
(2) Differences in the growth performance of the
countries are not closely related to differences in
economic structure, such as the degree of diversi-
fication.
(3) Major determinants of performance include how
197
Explaining the resource curse: R F Mikesell
the govemment disposes of the windfalls and the
policies it adopts to avoid inflation and maintain
incentives for investment and production in the
tradeable sectors.
These conclusions suggest that the Dutch Disease
model provides only a partial explanation of the
recent growth performance of mineral exporting
countries and that we should examine other factors
inherent in mineral abundance to explain why the
mineral exporting countries have had less favorable
growth performance than resource-poor countries.
Other explanations of the resource curse
Most of the studies seeking to explain the resource
curse (Rainis, 1991; Krause, 1995; Sachs and Warner,
1995) place major emphasis on the failure of
resource-rich countries to promote a highly productive
manufacturing sector, which is regarded as
the principal source of technological progress. The
knowledge and skills generated in this sector spread
throughout the economy and increase productivity in
all sectors. Ample export income reduces the incentive
of resource-abundant countries to develop manufacturing
industries that are competitive in international
markets. They tend to be protectionist in
order to support import-substituting industries whose
prices are noncompetitive in world markets.
A second explanation is that natural resource
exploitation creates large economic rents, which
accrue either to the government or to a few private
owners of the resources. In the former case, the rents
are often distributed to those in control of the government
and their relatives, rather than being used for
broadly based national development. (Kuwait and
Saudi Arabia are good examples of this.) Concentration
of rents in the hands of a few private owners
directs revenue away from human resources and
infrastructure, traditional agriculture, and small
enterprise, in favor of consumption by the wealthy.
Since the rulers of oligarchies are likely to favor maximizing
resource rents for the benefit of the few rather
than promoting broad development, it is suggested
there is a positive relationship between growth performance
and the political system, and that mineralexporting
countries are more likely to be oligarchies
than democracies (Lal, 1995, Ch. 14). None of these
relationships appear to be valid. Of the four good performers
listed in Table 2, two are oligarchies
(Indonesia and Oman) and two are democracies
(Botswana and Chile). Among the poorest performers,
Peru, Trinidad/Tobago, Mexico, Ecuador and Zambia
are democracies, while Nigeria and Saudi Arabia are
oligarchies. Some mineral-exporting countries have
shifted from one political orientation to another over
the past three decades. This has been particularly the
case with Chile and Peru. Democracy does not insure
good government, nor are all oligarchies poorly governed.
Thus, the form of government does not provide
a satisfactory explanation for the resource curse.
A third explanation is that governments of countries
with widely fluctuating export revenues have
special problems in adopting and carrying out the proper
policies to promote economic growth. This suggests
that governments of resource-poor countries are
better able to adopt policies to promote sustainability.
However, there is no convincing evidence for this
generalization. Nearly all developing countries
receive advice and assistance from the World Bank
and the IMF. The issue is one of will rather than of
knowledge or ability.
One of the factors often overlooked in explaining
the resource-curse paradox is the development history
of the individual countries. Except for city/states such
as Hong Kong and Singapore, virtually all development
begins with exploitation of the resource base,
which initially may provide little more than subsistence
agriculture. The sudden discovery of petroleum
or other minerals in a primitive country results in substantial
export income and initially a high rate of
growth. If this growth rate is to be maintained, exports
must continue to grow at the initial rate or other economic
sectors must expand. Mineral exports do not
expand indefinitely and development of the nonmineral
sector requires considerable time, even if the
export windfall is wisely invested. Meanwhile, the
growth rate declines as mineral exports level off or
fall. Can we assume that development will take off in
a subsistence economy in the absence of an external
stimulus? While some mineral exporting countries
have not maintained the initial rate of growth promoted
by suddenly becoming a large exporter, these
countries might not have got much beyond the subsistence
level in the absence of mineral windfalls.
Saudi Arabia's GDP rate of growth fell from an average
of about 10% in 1960-1980 to nearly zero in
1980-1993, but what would its per capita GNP have
been in the absence of oil? Can we say that Saudi
Arabia would have been better off without oil? Even
though the country has not used its mineral resources
for sustained growth, it is certainly better off for having
the resource.
Conclusions
There is no single or even dominant explanation for
the resource curse as applied to mineral-exporting
countries. None of the explanations suggested apply
to all mineral-producing countries and for most of the
mineral-rich countries that experienced poor performance,
the principal explanation was incorrect government
policies or exogenous conditions that have little
or nothing to do with mineral exports and could just
as well have occurred in resource-poor countries. It
may be that the bias against the growth performance
of mineral-exporting countries arises from a number
of factors which, while not applying to all these countries,
applies as a group to more of them than would
apply to resource-poor countries. For example, the
Dutch Disease is more prevalent among mineral-
198
Explaining the resource curse: R F Mikesell
exporting countries, but it does not provide a single
explanation of the resource curse. What may best
explain differences in performances among
developing countries is the advantage of endogenous
growth over growth stimulated by external forces
such as export booms.4 Without the policy conditions
for endogenous growth, including free trade, incentives
for investment, technological advance, and
stable prices, sustained growth will not take place.
However, the causal factors underlying endogenous
growth are too complex to be satisfactorily explained
by a country's resource endowment.
Acknowledgements
I greatly benefitted from comments by Richard Auty
of Lancaster University, Lancaster, England in the
preparation of this paper.
References
Auty, R M (1986) Resource-based industrialization and country
size: Venezuela and Trinidad and Tobago. Geoforum 17,
325-338.
Auty, R M (1993) Sustaining Development in Mineral Economies,
Routledge, London.
Edwards, S (1989) Real Exchange Rate, Devaluation and Adjustment.
MIT Press Cambridge, MA.
Findlay, R and Wellisz, S (eds.), (1993) Five Small Open Economies,
World Bank, Washington, DC.
Gelb, A and Associates (1988) Oil Windfalls: Blessing or Curse?
World Bank, Washington, DC.
Hirschman, A O. (1958) The Strategy of Economic Development,
Yale University Press, New Haven, CT.
International Monetary Fund (1996) Estimating potential output
growth in Chile, IMF Survey, 16 December, pp. 404-405.
Kindleberger, C (1956) Terms of Trade: A European Case Study,
Wiley, New York.
Krause, L B (1995) Social capability and tong term economic
growth. In Social CapabiliO, and Long Term Economic Growth,
eds. B H Koo and D H Perkins, Macmillan, New York, pp.
310-327.
Lal, D. (1995) Why growth rates differ. In Social Capability and
Long Term Economic Growth, eds. B H Koo and D H Perkins,
Macmillan, New York, pp. 288-309.
Lucas, R E (1988) On the mechanics of economic development.
Journal q]"Monetary Economics 22, 3-42.
Maizels, A (1968) Experts and Eeonomic Growth q['Developing
Countries. Cambridge University Press, Cambridge, MA.
Nankani, G T (1979) Development problems of mineral-exporting
countries, World Bank Staff Working Paper No. 354, World
Bank, Washington, DC.
North, D C (1963) Aspects of economic growth in the United
States, 1815-1860. In Experiences tq'Economic Growth, ed. B
E Supple, Random House, New York.
Prebisch, R (1964) Toward a new trade policy for development. In
Proceedings of the United Con~erenee on Trade and Development,
New York, Vols I-VIII, United Nations, New York.
Rainis, G (1991)Towards a model of development. In Liberalization
in the Process of Eeonomic Development, eds L B Krause
and K Kim, University of California Press, Berkeley, pp. 59-
101.
Repetto, R et al. (1989) Wasting Assets: Natural Resources in the
National Income Account, World Resources Institute, Washington,
DC.
Roemer, P (1990) Endogenous technological change. Journal of
Political Economy 98, 71-102.
Sachs, J and Warner, A (1995) Economic Convergence and Economic
PoliQ,, Working Paper No. 5039, National Bureau of
Economic Research, Cambridge, MA.
Westley, G (1995) Economic volatility from natural resource
endowments. In Development Policy, Inter-American Development
Bank, September, Washington, DC.
World Bank (1979) World Development Report, World Bank,
Washington, DC.
World Bank (1983) World Development Report, World Bank,
Washington, DC.
World Bank (1994) World Development Report, World Bank,
Washington, DC.
World Bank (1995) World Development Report, World Bank,
Washington, DC.
4For a discussion of endogenous versus exogenous long-term
growth, see Robert E. Lucas (1988) and Paul Roemer (1990).
199