Resource abundance in the Caribbean:
Curse or blessing?
Sabine Klok
Abstract
This thesis investigates whether resource abundant countries in the Caribbean Community grow significantly slower than resource poor countries in this region as a consequence of their relative richness in natural wealth. I investigate the resource curse
channels for the individual resource abundant countries and I compare the group with resource abundant countries with the resource poor group by estimating two fixed panel regression for the period 1980 – 2006. This investigation makes me conclude that there is
no evidence that resource abundance in the Caribbean Community has a negative effect on economic growth. However, there is some evidence of the presence of resource curse
channels like Dutch disease effects, low levels of genuine saving, high levels of government spending and lower trade levels. All of these are a result of low institutional
quality, so this suggests that by improving institutions, the effects of these channels can be mitigated.
Resource abundance in the Caribbean
Curse or blessing?
Master thesis Economics
University of Groningen
August 2008
Sabine Klok
First supervisor:
Studentnumber 1322761
dr. H.J. de Jong
klok.sabine@gmail.com
Laan van Meerdervoort 267
Second supervisor:
Preface
This thesis is intended to investigate whether natural resource abundance has a negative influence on economic growth in the Caribbean region. The motivation for this subject originates from my internship at the Dutch Embassy in Paramaribo, Surinam.
When I did this internship I got the opportunity to visit several companies in the natural resource sector, as the Canadian gold company IAMGold and the Surinam State Oil company. During these visits I experienced the contrast between these successful companies and the relative poor state of the economy. After these visits I joined a discussion evening about the impact of the bauxite and alumina industry on the financial and socio-economic conditions of the Surinam society. A local central bank member suggested that Surinam would have experienced higher economic growth rates without its abundance in bauxite. This statement made me want to know more about the negative influence of natural resources on economic development. For this reason, I decided to write my thesis about the existence of a so-called resource curse in the Caribbean region. When I was back in Holland to start this investigation it was though to go every day to the university library. I missed the friendly Surinam people, my hammock and the beautiful Surinam interior. Luckily I had many people who supported me during this process. I want to thank my supervisor Dr. H.J. de Jong for all his assistance. He was enthusiastic about the subject and helped me to find the right direction for this research. Furthermore I want thank my colleagues of the Embassy, especially my supervisor Zevoera Khodabaks and the head of my department Peter Derek Hof for motivating me for this subject. Finally, I want to thank my friends and family, with special attention to my brother and his girlfriend for all the hours spending with me in the university library.
Contents
1 Introduction 5.
1.1 Background 5.
1.2 Hypothesis and scope of research 6.
1.3 Outline 7.
2 Review of the literature 8.
2.1 Type of resource 9.
2.2 Theoretical explanations 9.
2.2.1. Economic channels 10.
2.2.2 The role of institutions 15.
2.3 Empirical findings 17.
2.3.1 Dutch disease effects 17.
2.3.2 The role of institutions 18.
2.3.3 Neglect of diversification 21.
2.3.4 Inappropriate use of resource rents 22.
2.3.5 Protectionist behavior 26.
2.4 Policy measures 27.
3 A resource curse in the CARICOM? 28.
3.1 CARICOM economy 28.
3.1.1 CARICOM economy 28.
3.1.2. Economic growth in the CARICOM 31.
3.1.3. Surinam, a case study 33.
3.2 The resource curse channels within the CARICOM 34.
3.2.1 Dutch disease effects 34.
3.2.2 Institutions of the resource abundant member states 39.
3.2.3 Policy failing 43.
3.2.3.1 Neglect of diversification of exports 43. 3.2.3.2 Neglect of investment in human capital 45. 3.2.3.3 High Government consumption 46.
3.2.3.4 Neglect of saving 47. 3.2.3.5 Neglect of Investment 48. 3.2.3.6 Protectionism 49. 3.3 Conclusions 50. 4 Empirical investigation 53. 4.1 Data 53.
4.2 Model and Hypothesis 56.
4.3 Results 57.
4.4 Discussion of the results 60.
5 Conclusions 63.
6. Bibliography 66.
1. Introduction
1.1
Background
Currently, prices of food, minerals and oil are booming. There are some counties who take of advantage of this. According to the International Monetary Fund (IMF) Surinam is among these countries1. The World Bank has ranked Surinam as the 17th richest country in the world in terms of natural resources, because of its possession of bauxite, gold, oil, timber and other unexplored resources2. This might suggest that Surinam is a very wealthy country. However, the contrary has been true; for a long period of time economic growth was low, much economic activities took place in the informal part of the economy and the country experienced a period of hyperinflation.
A member of the Central Bank of Surinam (CBoS) even suggests that it is due to the richness in bauxite that economic development has stagnated. He found evidence that the revenues of this sector negative influenced GDP growth in Surinam3.
This feature seems not to be uncommon in the world. Especially in African countries like Nigeria, Congo and Sierra Leone natural resource abundance appears to be an obstacle for economic development. There is a general tendency that resource-rich countries grow slower than resource-poor ones. This paradox is referred to as the resource curse, first investigated by Richard Auty (1993)4. He described how countries rich in natural resources were unable to use their wealth to boost their economies and how, counter-intuitively, had lower economic growth than countries which were resource poor. Sachs and Warner (1995) were the first who find empirical evidence for the negative
relationship between natural resource abundance and economic growth. This relationship was even true after controlling for variables found to be important for economic growth, like initial per capita income, trade policy, government efficiency, investment rates, geography and climate.
There are several explanations for this negative relationship. The best known is the so-called Dutch disease theory. The Dutch disease explains the negative relation between
1 Helbing et al. (2008). 2
http://www.surinameembassy.org/111502.shtml
3 Eckhorst (2004).
natural resource abundance and economic growth, due to an appreciation of the real exchange rate (RER) as a result of a boom in the natural resource sector. This
appreciation makes the non-booming sectors less competitive at world level, which hurts economic growth. However, there are several economists who argue that economic channels alone can not cause a resource curse. Additionally, there must be some mismanagement of resource rents or other kind of wrong policies arising from weak institution quality. Examples of wrong policies are: the neglect of diversification, inappropriate use of resource rents like the neglect of investment in human capital, low savings rates and high government consumption and protectionist behaviour. This suggests that the resource curse can be avoided by improving institutional quality.
Mehlum et al.(2006a) found evidence that institutions are indeed decisive for the resource curse to happen or not.
1.2
Hypothesis and scope of research
In most literature on the natural resource curse, the focus lies on African countries. In this thesis I am interested in the question whether resource abundant countries in the
Caribbean are cursed by their richness in natural resources as well. My aim is to study whether they indeed performed worse than other counties in the region between 1980 – 2006. However, when there is a difference in per capita growth rates, this does not necessarily have to be a consequence of their resource abundance. For this reason I will investigate several channels of the traditional resource theory. I will not include all Caribbean countries, but only the member states of the Caribbean Community (CARICOM), because of the similarities between the member states. All of the
economies within the CARICOM are highly specialized (in natural resources or tourism), were former colonies, have population sizes of less than one million people5 and are highly sensitive for exogenous shocks. This brings me to the following research question:
Do resource abundant countries within the CARICOM grow significantly slower than resource poor countries? Can this be explained by the relative richness in natural resources?
5 Exceptions are Haiti, which has a population of 8.6 million people and Jamaica, which had a population
1.3 Outline
2. Review of the literature
In the traditional growth literature natural resources are seen as supportive to economic growth. These are inputs that can be used in the production process to generate high levels of output. However, empirical evidence for this is mixed. There are some countries that indeed turned their natural resources into wealth and economic growth. Middle-East countries like Saudi-Arabia and the United Arabic Emirates profited substantially from their oil abundance. Furthermore, Norway and Australia succeeded in turning their natural resources into wealth by using their rents for diversified investment projects. Earlier, the US, Britain and Germany experienced technological development because of their natural resources.
On the other hand, the lack of these natural resources has not proven to be a barrier for economic success6. Japan and the Asian Tigers7 have shown high growth rates without large natural resource reserves while many African countries abundant in oil, diamonds and metals have experienced economic and social decline. This experience of these African countries suggests that there might exist a negative relationship between natural resource abundance and economic growth. The idea that natural resources are like a curse rather than a blessing first emerged in the 1980s. Auty (1993) introduced the so-called paradox of the plenty (or the resource curse)8. He described how countries rich in natural resources have been unable to use that wealth to boost their economies and how, counter-intuitively, have lower economic growth than countries that are poor in natural resources. Sachs and Warner (1995) were the first who found an significant negative relationship between resource abundance9 and economic growth. They showed with a cross-section regression model that countries abundant in natural resources (agricultural products, minerals and fuels) tended to grow slower than countries without these resources. Economies with a high ratio of natural resource export revenues to GDP in 1970
experienced low economic growth in the period 1970 till 1990. This negative relationship is even true when controlling for many variables found to be important for economic
6 Humphreys et al. (2007). 7
Taiwan, Hong Kong, Singapore and South Korea.
8 Cited in Sachs and Warner (1995).
growth, like openness and institutional quality. Sachs and Warner (2001) also showed that this relationship was not influenced by climate or geographical variables.
In this chapter I focus on several possible explanations for the negative relationship between natural resources and economic growth. I start with a theoretical framework, in which Dutch disease effects are explained and the complex relationship between natural resources and institutions will be discussed. After that, I concentrate on the main
empirical findings. Before doing so, I shortly describe the differences in point-source and diffused resources.
2.1 Type of resource
There are different types of natural resources. In this thesis I will only focus on the abundance in oil and mineral resources and neglect the abundance in agricultural products. Is important to distinguish between resources which have concentrated
production and revenue patterns; so-called point-source resources (oil and minerals) and resources which are more likely to be produced throughout the whole economy; so-called diffused resources (agricultural products). Murshed (2004) shows that especially
countries that are abundant in point-source resources have shown growth failure, while he could not find empirical evidence for countries abundant in diffused resources. A
possible reason can be that point-source resources can retard democratic and institutional development, which impede economic development. Also, point-source resources like diamonds and gold are attractive to anyone interested in short-term illegitimate gains, which are not supportive to economic growth10.
2.2 Theoretical explanations
In this section I will present an economic model that shows the consequences of a boom in the natural resource sector11. I will loosen some assumptions to make this model more realistic. In addition I will show some policy measures in the model. After that I will give theoretical explanations for the complex relationship between resource abundance and institutions.
10 Boschini, Petterson and Roine (2003).
2.2.1. Economic channels
The negative effect of resource abundance can be a direct consequence of economic channels, like increased volatility or declining terms-of-trade (Prebisch-Singer theorem12). The most famous economic channel related to the natural curse theory is the Dutch
disease channel. This channel can be explained best by the core model of Corden (1984).
Basic model, the Dutch disease
The model of Corden (1984) has a booming sector (B), a lagging sector (L) which both produce tradables (prices are fixed at world level) and a non-tradable sector (N) (prices are determined at the local market). Output is produced with a factor specific production factor and labour. Labour is mobile between the sectors and will move between them until there is a fixed wage. A boom in B often has a direct effect on the exchange rate. The higher export value of the natural resource sector will generate additional demand for the domestic currency which will lead to an appreciation13. Beside this effect there will be two other effects: a spending effect and a resource movement effect. The spending effect arises if some of the extra income of the boom in B is spent on the domestic market, which increases demand. When the income elasticity of demand for N is positive, the demand curve shifts outward and the price of N must rise relative to the price of B and L. This causes an appreciation of the RER14. This appreciation will make domestic sectors less competitive at world level. Especially L will be hurt by this. The resource-movement effect arises when there is a movement of labour out of N into B. This shifts the supply curve of N inwards which leads again to a rise of Pn, reinforcing the appreciation due to the spending effect. This may cause de-industrialization15.
Elwood (2001) shows that shocks in an economy of a resource abundant country are hard to predict with the traditional model of aggregate demand and supply (AD/AS model),
12 This theorem states that terms of trade between primary goods and manufactured goods tends to
deteriorate over time. The explanation for this phenomenon is the observation that income elasticity for manufactured goods is greater than the for primary goods (especially food). So, when income rises, the demand for manufactured goods will rise faster than that for primary goods (Toye and Toye 2003).
13 Under the assumption that demand for natural resources is inelastic and that the income effect of this
appreciation is bigger than the negative effect of declining demand.
14 The RER is defined as: e (P*/P), where e is the nominal exchange rate, P* foreign price level and P
domestic price level. When domestic price level rises compared to the foreign price level, the real exchange rate goes down, which means a real appreciation.
because this model assumes that a resource boom will increase production costs16, while this is not necessarily the case for resource abundant countries. He rejects the assumption of the traditional AD/AS model that there are constant relative prices. According to him this might be the case when there is a closed economy. However, when an economy trades with other countries, it absorbs another basket off goods than it produces. If in this case the price of one of the domestically produced goods will increase, the relative price increases.
Figure 1 shows the difference in consequences between a natural resource (oil) abundant country and a resource poor (importing) country when the prices of this particular resource will increase. When for example the price of the main commodity of a resource abundant country rises, this can greatly increase the value of its national production (income effect). The costs of production will also rise but for a resource abundant country the income effect will exceed the production cost effect, so the terms-of-trade of that country will improve and the supply curve shifts outward. Due to the higher export value of demand this curve will shift outward as well and the economy will end up at point B; higher prices, but also higher output. On the contrary, in a resource poor country there is besides a negative production cost effect also a negative income effect due to higher export prices. This economy will end up with lower output and higher prices.
The consequences of a resource boom in an oil exporting country as predicted by Elwood (2001) are not consistent with what is expected by the Dutch disease theory. I will use his model and incorporate Duct disease effects in this. The results are shown in figure 2. For the resource abundant sector it is true that the income effect is larger than the production cost effect. However, this is not the case for the resource poor sector, which experiences a loss in competitiveness due to higher domestic prices and will produce less; as a result AS shifts back and the economy ends up at point C, higher prices but the same level of output as in the initial situation, since I expect demand to stay at high levels due to the additional revenues from the resource sector in the economy. When there is in addition a
16 Providing only this cost-push interpretation of oil-price shocks within the AD/AS framework is almost
resource movement effect, there are less production factors in the resource poor sectors, so production of these sectors declines even more and as a result domestic prices will rise, amplifying the appreciation already caused by the spending effect, which causes the economy to end up in point D, much higher prices and lower output.
Figure 1, Consequences of an oil shock in a resource exporting and – importing country
Source Elwood (2001) Source: Elwood (2001)
Variations of the basic model
There are several variations possible when loosening some of the basic assumptions or introducing policy in the model. I will shortly discuss some of them below. What happens in the economy of the resource abundant country is shown in figure 2.
Enclave industries and diversification. Natural resource industries tend to operate as enclaves with few forward and backward linkages with the rest of the economy. Most of these industries have immobile labour, because of the need for specific skills. In this case the resource movement effect is less likely and only the spending effect will arise. When there are no linkages with the rest of the
economy, this extra revenue will not be spend into the economy. No spending effects occurs and the demand shift will be less likely, so in this case I predict
AS1 AS2
Prodcost effect
Income effect AD2
nothing to happen and the economy to stay point A. Furthermore, when the economy is diversified, the natural resource sector does not have such a strong impact on the real exchange rate. I expect the spending effect to be present, so in this case the economy ends up in point C.
Neglect of investment in human capital. There is an additional problem with the resource-movement effect apart from the reinforcement of the exchange rate appreciation. Sachs and Warner (1995) suggest that skilled labour and high-quality capital are less common in R. When this is the case, there are less
learning-by-doing effects and technological advantage. Also, in an overlapping generations model when there is a large primary sector, there is less private incentive to invest in education. Besides the private incentive to invest in education, there is, according to Gylfason (2000) also lower public incentive, because governments easily receive enough revenue from the resource sector, so they become overconfident. In the long-run this will lead to a lower steady state level of output. This shifts the AS curve inward once more and the economy ends up in point E.
Learning-by-doing effects in the booming sector. Torvik (1999) shows that the results are quite different when the assumption, that the mining sector is not characterized by learning-by-doing, is rejected. When there are learning effects in the booming sector, productivity will rise in this sector. Part of this productivity improvement will spill over to the non-traded sector. This results in lower prices on the domestic market and a depreciation of the RER. Due to this depreciation the non-resource sector becomes more competitive and as a consequence the AS curve will shift out once more. The economy will end up in point F. Due to this depreciation imports become more expensive, so I expect that AD shifts
somewhat inwards. The economy will end up in point G.
discovered resources. In the short-run the same effects of the model of Corden (1984) will arise; demand will increase, which leads to a real appreciation and possibly deindustrialisation, so the economy will end up in point C or even point D. However, over time, when the discovered resources are consumed, the net foreign asset position will deteriorate. To restore this, high debts are accumulated on which interest has to be paid. Due to these interest expenditures, expenditure will be lower than before the discovery and so will be demand, so the AD curve shifts to the left, even more than the initial level. This leads to a real depreciation and probably to less production in this sector, which releases factors for the manufacturing sector and this increases production in the manufacturing sector; pro-industrialisation. In a reaction to this the AS curve will shift outward and the economy will then end up in point H; higher level of output as in the initial situation but lower prices.
Protection of the lagging sector. In many cases the government tries to protect the lagging sector to avoid real income loss or to help infant industries. According to Corden (1984) the best way to do this is to subsidize output of the sector directly. He criticizes the most often used policies of protection, which are
exchange rate protection or protection with tariffs or quota’s. The objection to the first policy is that it does not only protect the lagging sector, but also the booming sector. The objections to the second is that due to these restrictions the lagging sector export products will be hit twice, because this sector is then missing out on trade and by this the AS will shift inward once more and the economy will end up in point E. However, it is good to mention that when a natural resource sector is booming, it often receives higher levels of foreign direct investment (FDI)17, which is stimulating for economic growth, do the trade argument might not be so evident anymore.
17
Figure2, Effects of a resource boom
2.2.2 The role of institutions
Economists like Wijnbergen (1984) find the term ‘disease’ misleading. A disease
suggests that it is something that happens without possibilities to avoid it. However, there are many economists who believe that a resource curse can be avoided if there are strong institutions.
There are three views on the role of institutions in the resource curse theory18:
1. Institutions as an intermediate causal link; the main reason for the resource curse is a decay of institutional quality in resource rich countries.
2. Institutions have a neutral role; they do not influence the curse.
3. Resource abundance interacts with the quality of institutions; natural resources richness does not necessarily cause institutional decay, but it might put the institutional agreements to a test.
citizens, so there is less need for taxation. It is often believed that taxation makes a state strong, because the state then has to clarify where the tax money has been spend on. Another reason to support the first statement is that it might raise conflict in a country, which is explained by Collier and Hoeffler (2005). Especially when there are secessions in a country, the presence of large amounts of natural resources can lead to conflict. Secessions often claim to be the owner of the natural resources instead of the government and they also do claim that the government is misusing this money. A third argument is that it hinders democracy because, according to Acemoglu et al. (2004), higher resource rents make it easier for dictators to buy off political challenges. The final argument is that natural resource abundance will lead to rent-seeking behaviour19, which will result in corruption. Gylfason (2004) indicates that corruption is endogenously determined by natural resources via rent-seeking behaviour.
Most of the above arguments do not seem to be an option when there is high institutional quality in the first place; Acemoglu et al. (2004) shows that perverse political incentives are mitigated in countries with adequate institutions . Furthermore, institutions within a country are determined in the past and are hard to change. Mo (2001) suggests that, since institutions tend to evolve slowly, corruption can not be seen as endogenous determined by natural resource abundance. Acemoglu et al. (2001) argue that former colonies have often inherited their institutions from their settlers and they often still have these. They show that Europeans were likely to introduce institutions facilitating the extraction of resources instead of establishing institutions which contributed to economic growth and wealth for the citizens of the colony. In the 19th century, there were new investment opportunities in these countries, because the traditional plantation economy became less important. However, poor economic institutions persisted, because the political power of elite was linked in the existing system and a change in this system would make them political losers.
19 The term rent-seeking occurs when an organization or a firm seeks to make money by manipulating the
2.3 Empirical findings
In this section I will present some empirical evidence for the relationship between natural resource abundance and economic growth via the various channels that were mentioned in the previous section. From the findings of other authors I will formulate my own hypothesis.
2.3.1 Dutch disease effects
Stijns (2005) shows that in many countries oil and gas reserves indeed draw away resources from the manufacturing sector as is predicted in the Dutch disease theory. He finds evidence for Venezuela, Iran, Congo and Trinidad and Tobago. However, he also finds that Dutch disease effects are not present when there are institutions of good quality. In this case oil and gas revenues will allow countries to afford better education and higher savings and investment rates. Examples are Norway and Australia. The evidence for the presence of Dutch disease symptoms is not that strong for mineral economies. There are some countries where high mineral reserves are correlated with a slower rate of real growth in the non-natural resource sector and a smaller change in the share of
manufacturing exports in total exports; examples are Australia, Canada, South Africa and the United States. However, there are also countries where mineral reserves have a slightly positive correlation with growth in the non-natural resource sector and the share of exports in total exports. These are Chile, Bolivia, Peru, and Spain. Davis (1995) did not find evidence that twenty-three mineral economies20 in 1970 or 1990 underperformed compared to fifty-three resource poor countries. He concludes that mineral economies in a group are not cursed and have not performed poorly in the long run. He showed also that twenty-two of these countries have not been decimated by traditional Dutch disease symptoms. He shows that these countries have experienced some short-term Dutch disease effects as an appreciation of the RER, but these countries do not pop in and out the production and exports of minerals, as the traditional Dutch disease literature suggests. The only country who did indeed showed a traditional Dutch disease cycle, was Tunisia. However, there is no evidence for this country that traditional export sectors are
20
permanently damaged since this country could successfully diversify its exports more towards manufactured goods instead of minerals.
I will investigate whether there are Dutch disease effects visible in the resource abundant CARICOM countries after a boom in the resource sector and whether this has a negative impact on economic growth.
Hypothesis 1 : When there occurs a boom in the natural resource sector, the real exchange rate (RER) of the resource abundant country appreciates, which hurts other sectors in the economy and this is negative for economic growth.
2.3.2 The role of institutions
A prominent example of economic growth due to natural resource abundance of the latest period is Botswana. Chile, Brazil and Australia also have shown that the mineral sector can indeed contribute to growth. Additionally, Malaysia and Peru seem to have escaped the curse. Mehlum et al. (2006b) suggest that the explanation can be found in institutional differences. They show that good performers as mentioned above do also have much higher scores on institutional and political indicators than bad performers as Ecuador, Nigeria and Zambia. Botswana has the best African score on the Corruption Perception Index (CPI) of Transparency International21. Also, the World bank shows that it is not a coincidence that resource rich countries as Angola, Nigeria, Sudan, Sierra Leone and Congo performed poorly, but that countries like Norway, the US, Canada, Iceland and Australia performed rather well; five of the top eight countries according to natural resource wealth were also among the top fifteen of institutional quality22. In addition, Robinson et al. (2002) indeed show that political incentives as a result of resource abundance are the key to understanding whether or not there is a curse. They show that politicians in resource abundant countries tend to over-extract natural resources because they discount the future too much. Further, politicians in these countries can use these
21 Since 1995, Transparency International has published annual the corruption perceptions index (CPI)
which orders countries on a scale from 1 to 10 (whereby 10 is the less corruptive) according to the degree to which corruption is perceived to exist among public officials and politicians.
resource rents to influence election outcomes, which leads to resource misallocation. The extent to which this happens depends on the quality of institutions; countries with
institutions that promote accountability and state competence will tend to benefit from resource boom. They find empirical evidence for this by investigating the institutional quality index23. They show that Botswana, Chile, Malaysia and Thailand have higher institutional quality than Mexico, Nigeria, Trinidad and Tobago, Venezuela and Zambia and that the first group did escape the curse, while the second did not.
For this reason I do not believe that, when investigating the relationship between natural resource abundance and economic growth, the role of institutions can be neglected. However, there is also not much empirical evidence that natural resource abundance has a negative impact on institutional quality. Brunschweiler (2006) indeed shows that natural resources, and in particular mineral resources, have a positive association with real GDP growth over the period 1970-2000 if there are sound institutions. However, she could not find evidence that resource abundance can influence institutional quality. She tested this relationship by using two different institutional quality indicators, the rule of law and government effectiveness. When treating resource dependence as an endogenous variable, Brunschweiler and Bulte (2007) again show that with good institutions, resource
abundance has a positive influence on economic growth. According to them the problem off the proxy used by Sachs and Warner (1995)24 was that the denominator explicitly measures the magnitude of other activities in the economy, so that it was not independent of economic polices and institutions. Countries with large resource stocks may derive high income from extraction and as a consequence they specialize in these resources. By this these countries become dependent on its resources. They do find a negative
relationship between institutions and resource dependence. Countries with poor institutions are unlikely to develop non-primary production sectors to reduce their dependence. So, causality is then from dependence to institutions, not the other way around. Gylfason (2004) also shows that dependence on natural resources is directly
23 The institutional quality index is an unweighted average of five indexes based on data from Political Risk
Services; a rule of law index, a bureaucratic quality index, a corruption in government index, a risk of expropriation index, and a government repudiation of contracts index, available at
http:/www.cid.harvard.edu.
associated with corruption, inequality and political oppression that all have a negative impact on economic development.
I will give some examples of negative effects of low institutional quality and the endowment of natural resources. According to Mehlum et al (2006b) the difference in good institutions and bad institutions depends on whether there exist grabber friendly institutions or producer friendly institutions. When there are producer friendly institutions, rent-seeking and production are complementary activities, while they are competing when there are grabber friendly institutions. When there exists grabber friendly
institutions, there are gains from specialization in unproductive influence activities, due to for example corruption or an inadequate court system. A paradoxal relationship between institutions and natural resource abundance is found by Collier (2007), who suggests that natural resources (oil rents) make democracy malfunction. He showed that surpluses of natural resources are unsuited for the pressures generated by electoral competition. In this case autocracies outperform democracies. He finds that if an autocratic country with resource rents worth 20% of national income switches to
democracy, this will cost almost 3% of GDP. Reasons for this are that democracy induces an incredibly large public sector and that a democratic government is likely to under invest.
Norway is often used as an example of a country who did escape the curse due to its sound institutions. Larsen (2003) indeed shows that the exploration of oil is the reason for this growth acceleration and that Norway escaped the curse because of the well-suited political and economic institutions. All the oil resources were publicly owned and the right policies were conducted to make a blessing out of the oil discovery. The oil wealth was not converted into domestic purchasing power, but was instead kept into foreign denominations and used to build up the Petroleum Fund in assets abroad. This fund was used to finance further investments into know-how and technology. However, Larsen (2003) states that he can not fully conclude that Norway totally escaped the curse, because Norwegian growth is nowadays back in line with Sweden and Denmark, so one can speak of a relative curse.
This literature shows that the relationship between institutions, natural resource
resource abundance has a negative influence on institutions. However there is some evidence that when a country becomes highly dependent on its resources, this can have a negative influence on economic growth. The relationship the other way around seems to be true for many countries; weaker institutions cause that resource abundant countries may suffer more from a curse than countries with sound institutions.
Hypothesis 2 : Resource abundance make institutions weak. Due to these weaker institutions, resource rich countries are not able to turn their resource abundance into economic wealth.
2.3.3 Neglect of diversification
Gylfason (2004) shows that natural resource abundant countries are by nature more specialized than resource poor countries. As shown earlier, in undiversified economies Dutch disease effects will be more present. Not only a boom will hurt the economy more, also, a slump in the natural resource sector (like lower demand or a decline in prices) will hurt an economy which is very specialized in one resource more. The empirical research focuses especially on export concentration.
Lederman and Malony (2003) find a negative relationship between export concentration and economic growth in a panel regression with 65 countries for the period 1980-1999. They showed this by using the Herfindahl index25 and the amount of natural resource exports in total exports. They did not find a direct negative effect of natural resource abundance on economic growth26. In addition, Isham et al.(2003) find evidence that countries with highly specialized exports in point-source natural resources and plantation crops are cursed double. These countries are more vulnerable, but they are also exposed to heightened social divisions and weakened institutional capacity, which in turn impedes their ability to respond effectively to shocks, which is negative for economic growth. They show that export concentration in general is not a problem; export concentration in
25 The Herfindahl index is a measure for concentration (market share) in a branch. It is defined as the sum
of squares of the market shares of each individual firm. It ranges from 0 till 1, the lower, the more competition.
26
other goods, like manufactured goods, contributes to economic growth.
I will investigate whether the resource abundant countries in the CARICOM indeed have concentrated exports, if this makes their economies more vulnerable and whether this has a negative impact on economic growth.
Hypothesis 3 : Resource abundant countries in the CARICOM are more specialized than resource poor member states, which makes their economies volatile and this harms economic growth.
2.3.4 Inappropriate use of resource rents
Due to the easiness of which these rents are obtained the resource rents will not be used optimally for long-term investment projects, but are used for government consumption (high public wage bill) instead. Several economists found evidence for this.
Neglect of investment in human capital
Gylfason (2000) shows that resource rich countries can live well of their resource rents even with poor economic policies and when there is strong commitment to education. He investigates the relationship between resource abundance, education and economic growth. Three measures of schooling have been used; government expenditure on education, the number of females in education and secondary-school enrolment rates. He shows that an increase in natural capital is associated with a decrease in public
expenditure on education, a lower number of females in education and a decrease in secondary-school enrolment. All these effects are associated with lower GDP per capita growth.
Scandinavian countries, while the Latin-American countries showed low levels of investment in human capital. The higher the level of human capital, the smaller the crowding-out effect on the labour force of the industrial sector. A well-educated labour force facilitates the movement of workers across different economic activities and because of this, can anticipate faster on exogenous shocks. A good example of this is the change of Denmark’s export of grains into the export of dairy products and meat, when the price of grain decreased in 1864. Stijns (2005) shows that it also works the other way around; countries with high levels of human capital can more easily explore and exploit their natural resources and in turn create more wealth.
Hypothesis 4 : Resource abundant countries in the CARICOM have less human capital than resource poor countries in the region, which is an obstacle for economic development.
Neglect of saving and investment
Sachs and Warner (2001) showed that resource abundant governments have often higher debts, because they borrow more with future rents. Manzano and Rigobon (2001) show that this becomes a problem when prices of natural resources decline and countries are not able anymore to repay their loans. They found evidence for this by using fixed effects in the model of Sachs and Warner (1995). In this model the effect of resource abundance disappeared. However, the negative growth effect of resource abundance did survive the cross-sectional regression. When distinguishing between two different periods of time (1970-1980 and 1980-1990), they show that in the first period resource abundance had no effect on growth, while it did in the second period. In the beginning of the 1970s
commodity prices were high and in that time countries used their resources as collateral to borrow money. Later on, prices declined. Countries like Mexico and Venezuela are good examples of high borrowers because of their resource rents. Buffe and Krause (1989)27 show that Mexico found large reserves of oil in 1977. Due to this discovery the Mexican government increased expenditures; in 1977 it was only 29.5% of GDP, while it was 46.4% in 1982. Most of the additional expenditure had gone in current expenditure,
which rose from 22% of GDP to 36% in this period. Public expenditure on investment hardly increased. When oil revenues decreased the country had built up high debts. In the theory it is suggested that resource abundant countries have insufficient levels of savings and that this is the reason that they borrow with future rents. However, there is not much empirical evidence that supports the idea that resource rich countries do have lower levels of gross savings than resource poor ones. On the contrary, Atkinson and Hamilton (2003) show that resource abundant countries have on average high gross savings rates. However, for economic development gross savings are not the best indicator to look at. Another savings rate, the genuine savings rate, is a better indicator for the development of resource rich countries. This savings rate is developed by the World Bank. Box 1 shows the composition of the genuine savings rate. Atkinson and Hamilton (2003) do show that resource abundant countries have relative low levels of genuine savings. A positive genuine savings rate is needed for sustainable development28. Dietz et al. (2007) show that despite high gross savings rates, resource abundant countries are the poorest genuine savers in the world. This leads to unsustainable development. The savings response depends on the quality of institutions. Resource-abundant countries with weak institutions often use these rents for high government consumption (public wage bill) and will have lower rates of genuine saving. Dietz et al. (2007) show that the rate of genuine savings rate for resource abundant countries can be higher, when institutional quality is improved.
28 Weak sustainable development is built on the assumption that different forms of capital are substitutable
Box 1: composition of genuine saving
Gross national saving
↓ - consumption of fixed capital
Net national saving
↓ + current operation expenditures on education
↓ - value of natural resource depletion ( energy,
minerals, metals, net forest depletion)
↓ - value of damages form pollutants, + pollutants
carbon dioxide
Genuine saving
Source: Hamilton et al. (2006)
Feldstein and Horioka (1980) studied the relationship between saving and investment rates and found a high correlation between the two29. They found that low savings in an economy will result in low investment. However, this is only the case when there exist imperfect capital markets. When there is full capital mobility, this is not the case.
Papyrakis and Gerlach (2004) show indeed the negative link between resource abundance and investment. They find empirical evidence that increasing resource rents lead to a decrease in investment. This decline in investment is not offset by the generated resource revenues. As a consequence of these low levels of investment, capital accumulation will be low and the resource abundant country will end up with a lower steady-state growth rate.
Hypothesis 5 : Resource abundant countries in the CARICOM have higher government consumption, lower genuine savings rates and lower investment rates, than resource poor countries in the region. All of these are negative for economic growth.
2.3.5 Protectionist behavior
The Dutch disease theory explains that, when there is a windfall in the natural resource sector, this will often lead to higher prices in other domestic sectors. Auty (1993)30 showed that as a result it was quite common for mineral rents to be used for the
protection of these sectors by lowering subsidies or protectionist actions. However, the inadequate performance of the weakened non-booming sectors during the post-boom downswings required unsustainable levels of subsidy from the mining tradable sector. Sarraf and Jiwanji (2001) show that this was the case in South America (Bolivia, Guyana, Chile) and Africa (Zambia, Zaire and Nigeria). Arezki and van der Ploeg (2007) showed with a linear regression model that growth in resource abundant countries positively depends on the openness of the economy. Resource abundant countries that have monopolies of major exports, black market exchange rate premiums, and average tariff rates higher than 40% or quotas covering more than 40% of imports, suffer more from a resource curse. Gylfason (2004) shows that due to protectionist actions an economy misses out on trade and this will hurt economic growth, because it misses out the imports of capital, technology and know-how. He shows that by changing trade policies, the negative relationship between resource abundance and economic growth might disappear. I will not only focus on the missing out on trade, but also on FDI, because these countries try to protect infant mining companies and make it hard for foreign companies to invest directly in the country. Navaretti and Venables (2004) show that for developing countries, FDI becomes a more important source of financing their investments.
I investigate whether resource abundant countries in this region indeed are less open and miss out on trade and FDI.
Hypothesis 6 : Resource abundant countries in the CARICOM have
protectionistic policies and as a result low levels of trade and FDI, which harms economic growth.
2.4 Policy measures
Davis (1995) points out some outstanding policy implications that follow from what Auty (1993)31 calls the counter-intuitive outcome of resource abundance on economic growth. The most extreme one is to make mineral exploration an illegal activity in resource abundant countries and any accidental findings should have been confiscated by the government and secured against further development. More realistic measures proposed by Davis (1995) are: to count mineral discoveries as a debit in these countries green accounts, to promote multinational enclaves, to stimulate the World Bank and IMF to penalize these economies for developing their mineral deposits and to stimulate governments of these countries to develop optimal depletion paths.
The most important policy measure to avoid Dutch disease effects is, according to Miksell (1997), preventing the rise in domestic demand arising from the windfall in the commodity sector. One solution for this is sterilization of the government revenues by the central bank. This can be done by forcing the mining companies to sell the export
revenue windfall 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. However, then there still will be the rise due to the increase of domestic prices. In this case the central bank should depreciate the exchange rate. Since many resource abundant countries have pegged their currency to the US dollar this is not always an option. The last option is channeling the export windfall in a stabilization fund, which is used for productive investment in income-earning projects or foreign exchange reserves to maintain investment in periods of low earnings. Velculescu and Rizavi (2005) note that a resource fund will be not very helpful when there are inadequate rules and there is insufficient commitment. However, if there is a willingness to adopt a policy that promotes national saving, a fund could be modified to help manage resource wealth in accordance with the principles of fiscal sustainability and intergenerational equity.
3. A resource curse in the CARICOM?
In the previous section I have explained the reasons why natural resource abundance can be an obstacle for economic growth. In this thesis I want to find out whether resource abundant countries in the CARICOM region are indeed cursed by their resources. I will do this by discussing in what way the resource channels apply to these countries
compared to the resource poor countries. In the next chapter I will use the same channels in a simple regression model. I will start with a short outline of the economies of the CARICOM member states and a case study of Surinam.
3.1 CARICOM economy
To promote trade and economic relationships within the Caribbean region, the
CARICOM was established in 1973. Table 2 gives an overview of the member states. All of them were former colonies (mainly British), but are nowadays independent.
Montserrat is an exception; this country is still a colony of the UK. For this reason I will exclude Montserrat from my analyses. Table 1 also gives the population size of all the member states. It shows that almost all countries are small states according to their population size.
3.1.1 Composition of the economies of the member states
Table 1 in appendix 1 shows the sectoral composition of the economies of the member states in 1990, 2000 and 2002. Most of the economies of the member states are highly dependent on services of which tourism is the most important. Offshore banking is an upcoming activity. Agriculture is declining, due to increased competition at world level and changed preferences. Only Belize, Dominica and Haiti still have a sizable
agricultural sector. Other countries where tourism is not the main foreign exchange earner are Guyana, Jamaica, Surinam and Trinidad and Tobago. These countries are abundant in natural resources and most of their economies depend strongly on mining and quarrying.
production expanded. In the beginning of the 1990s bauxite determined about 35% of total exports, gold 10%. In 2004 bauxite determined only 10% of total exports and gold 25%32. Besides this natural resource sector Guyana has a large public sector.
The economy of Jamaica is somewhat more diversified. However, also this country is highly concentrated on services (tourism) and mining (bauxite and alumina production). In 1998 Jamaica ranked third in the production of alumina after Australia and the United States33. Bauxite and alumina production determined in 2000 more than 55% of total exports34. Mining and tourism in Jamaica are based on large scale investments, which are isolated from the rest of the economy.
For Surinam bauxite mining and refining is the backbone of the economy and is an important source of foreign exchange and government revenues. Due to the opening of a new goldmine in 2004, gold is the second most important source of foreign exchange and government revenues. There is also some oil production in Surinam. In 2006, these primary products determined about 85% of the total export value, which consisted of 46.2% of bauxite, 31,7% of gold and 6.9% of oil35. As with Guyana, Surinam also has a large public sector.
In Trinidad and Tobago oil exploration and exploitation is the most important economic activity. Since 1917 the energy sector has played a central role in the economy. In 2003 it accounted for 40% of GDP, 83% of exports and 41% of government revenues36. With its oil revenues Trinidad and Tobago diversified its economy and became a major financial center in the region.
Despite the impact of natural resources on the economies of Guyana, Jamaica, Surinam and Trinidad and Tobago, the mining sector contributed only little to employment in these countries. In Trinidad and Tobago the contribution to employment was in 2005 only 5%37, in this year the mining sector of Surinam contributed only to 3.4% of total
employment38. Guyana and Jamaica have similar employment numbers.
32 Atoyan, R., J. Gold and C. Staritz (2007).
33 http:/minerals.usgs.gov/minerals/pubs/country/1998/9515098.pdf 34 Cueva et al. (2001)
35 Stichting Planbureau Suriname (2007). 36
Carvalho Filho et al. (2005).
37 Carvalho Filho et al. (2005).
Table 1 in appendix 1 shows only the formal activities. However, in these countries many economic activities are undertaken in the informal part of the economy. Vuletin (2008) estimated the informal economy of 32 Latin-American and Caribbean countries. Table 2 in appendix 1 shows the outcome for the CARICOM member states. The Bahamas has with an informal economy of about 15% of GDP the lowest score. Belize and St Vincent and Grenadines have sizes of about 50%. Also, Jamaica and Guyana have large informal sectors; about 35% of GDP. He did not include Surinam in his analyses. The general bureau of statistics of Surinam estimated that value of the informal economy was about 13.85 % of total GDP in 200539. As a consequence of these large informal sectors I should be careful with the interpretation of the data. Findings might turn out differently in reality. Especially Dutch disease effects will be hard to define, because it is difficult to estimate what the effect will be of a boom in the natural resource sector on the informal sector of the economy. When there is a large informal sector, diversification might be underestimated.
Table 1 : CARICOM Member states Country Number
of citizens
Government type Country Number of citizens Government type Antigua and Barbuda 83.000 constitutional monarchy with parliamentary system Jamaica 2.682.000 constitutional parliamentary democracy The Bahamas 323.000 constitutional parliamentary democracy Montserrat 6000 parliamentary democracy Barbados 292.000 parliamentary democracy St Kitts and Nevis 43.000 parliamentary democracy Belize 280.000 parliamentary democracy St Lucia 460.000 parliamentary democracy Dominica 68.000 parliamentary democracy StVincent and Grenadines 120.000 parliamentary democracy Grenada 105.000 parliamentary democracy Surinam 452.000 constitutional democracy
Guyana 739.000 republic Trinidad and
Tobago
1.328.000 parliamentary
democracy
Haiti 9.296.000 republic
Source: population size IMF International Financial Statistics (IFS) (2007), government type CIA World fact book (2007).
3.1.2 Economic growth in the CARICOM
Table 3 in appendix 1 shows that average GDP per capita of the CARICOM (excluding Montserrat, Barbados and Haiti) was in 2000 with 5319.1 US$ much higher than the 392 US$ of low income countries. There is high divergence between the member states. GDP per capita of the Bahamas was in 2000 almost 35 times as high as that of Haiti. Other countries that perform quite well are Antigua and Barbuda, Saint Kitts and Nevis and Trinidad and Tobago. Guyana performs almost as poor as Haiti. On the basis of GDP per capita four clusters can be made as shown in table 2.
Table 2 : Income categorization CARICOM
high middle income medium middle income low middle income low income
Antigua and Barbuda Belize Dominica Haiti
The Bahamas Grenada Guyana
Saint Kitts and Nevis St Lucia Jamaica Trinidad and Tobago St Vincent&Grenadines Surinam
Source: Tacone and Nogueira (2002)
Table 3 below shows real GDP growth rates40. Despite the high average per capita GDP, economic growth of the CARICOM has lagged behind when comparing this with
economies with similar small economies as Singapore, Ireland and Cyprus. Singapore grew in the period 1980 -2000 three times faster than the average of the CARICOM member states. Almost all member states suffered declines in per capita incomes in the 1980s and for most countries this continued in the 1990s. However, this is not a specific characteristic of the CARICOM member states, the whole Latin American and Caribbean region did not show high growth rates during this period. The 1980s in this region is often referred to as a lost decade. The lost decade was a result of a financial crisis that occurred in the early 1980s (for some Latin American countries already starting in the 1970s), when countries is Latin America reach a point where their foreign debt exceeded their earning power and the were not able to repay it41.
40 In this section I will look at GDP growth rates, in my empirical model I will use per capita rates.
However, since population in the member states hardly changed there is hardly a difference between the two.
Table 3 : Real GDP growth rates, annual % Country Name 1975 - 1980 1980 - 1985 1985 – 1990 1990 – 1995 1995 - 2000 2001 2002 2003 2004 Comparables regions Low income 3.4 3.5 5.1 3.9 5.0 4.7 3.5 7.0 7.4 South Asia 3.7 5.2 5.7 5.0 5.4 4.6 3.7 7.6 8.0
Latin America & Caribbean 5.4 0.6 1.8 3.5 3.1 0.3 -0.5 2.1 6.2
World 3.9 2.6 3.6 2.4 3.3 1.5 1.9 2.7 4.1
CARICOM42 - 1.3 4.5 1.9 3.7 0.1 2.0 4.1 5.6
Member states
Antigua and Barbuda - 5.3 6.9 2.1 4.7 0.4 2.5 5.2 7.2
Bahamas, The 12.2 4.0 2.1 -0.4 4.3 -2.0 0.7 1.9 3.0 Barbados 4.6 -0.8 3.4 -0.3 3.4 -2.1 -2.1 1.9 4.5 Belize 7.4 0.4 9.7 6.1 5.9 4.9 5.1 9.3 4.7 Dominica - 5.1 5.6 1.5 2.1 -3.8 -4.0 2.2 25.5 Grenada - 4.3 6.8 0.9 6.6 -4.9 2.2 8.0 -6.9 Guyana -0.6 -4.2 -2.3 7.1 2.8 2.2 1.1 -1.0 3.3 Haiti 5.7 -1.0 0.2 -4.7 2.4 -1.0 -0.3 0.4 -3.5 Jamaica -3.2 0.4 5.0 4.0 0.0 1.4 1.6 2.7 1.1
St. Kitts and Nevis - 3.1 7.9 3.8 4.5 2.0 0.9 0.5 9.6
St. Lucia - - 12.3 3.4 2.9 -4.7 3.0 3.0 6.7
St.Vincent& Grenadines 6.1 5.5 6.7 1.3 5.2 0.6 4.1 3.2 6.2
Suriname - -2.0 0.3 -0.1 1.5 4.5 4.8 6.0 8.0
Trinidad and Tobago 7.9 -2.1 -2.2 1.4 5.0 4.0 8.0 14.4 8.8
Source: World Development Indicators of the World Bank (WDI) 2006
Table 3 shows that the CARICOM countries in this period did not perform worse than the Latin American average43. From this table it is also not evident that the resource rich countries in the CARICOM region do show slower growth rates than resource poor countries in this region. By only reviewing growth rates it seems at first sight that Trinidad and Tobago escaped the resource curse and turned its oil wealth into economic growth., since this country shows remarkable high growth rates from 1995 onwards. For the other resource abundant countries conclusions from table 3 are hard to make. It seems that Guyana performs poorly. However, during the period 1985 – 1990 it had the highest growth rate of the CARICOM. Jamaica showed high growth rates between 1985 and
42 Excluding Montserat for the whole period, St. Lucia in the period 1980 – 1985. 43
1995. However, after that growth declined. Surinam performed poorly from 1980 till 1995, but shows high growth rates from 2001 to 2004. Further research is needed to make strong conclusions about this. Before discussing the resource curse channels from chapter 2 I will first describe the impact of natural resource abundance in Surinam.
3.1.3 Surinam, a case study
Economic development of Surinam is strongly influenced by its abundance in natural resources. The main natural resource is bauxite, which has the largest impact. Besides bauxite, gold and oil production becomes more and more important. In 2006, the mining sector accounts for more than 90% of Surinams exports and about 30% of the
government budget was financed by revenues from this sector44.
Martin (2001) investigated the influence of the bauxite sector on the Surinam economy during the 1990s. In this period there were two alumina booms. Dutch disease effects were evident because of these booms. Due to the higher prices, export earnings increased and the real exchange rate appreciated. In addition the boom increased the amounts of corporate taxes the alumina companies had to pay the government. With this additional revenue, the government increased its spending on the salaries of civil servants, transfers and subsidies and goods and services. This combination of increased foreign-exchange availability and government revenues fed through to higher inflation, higher real wages (not only in the booming sector, but also in other sectors of the economy) and higher output. Rising wages increased consumer’s disposable income and generate consumption and import booms. These booms increased the demand for output of the sectors of non-tradables and causes rapid economic growth. However, this came at the cost of other sectors; the agriculture sector declined. Table 1 in appendix 2 shows that the RER was very volatile due to booms and busts in alumina prices. The IMF45 showed that volatility of the real effective exchange rate (REER) of Surinam was the highest of a group of 22 countries in the Western Hemisphere. Between 2000 and 2005, the REER appreciated by a cumulative 33% as a result of increasing bauxite prices. Due to this appreciation, the external competitiveness of the non-mining sector has been adversely affected. Especially
the agriculture sector was hurt by this loss of competitive advantage. From 2002 till now this sector (apart from the banana industry) is declining.
A member of the CBoS, Eckhorst (2004) showed with a linear regression model in log-log form46 that bauxite exports negatively influenced GDP in the period 1970 – 2001. He investigated the channels of the traditional resource curse theory and he showed that during the period 1980 -1986 there were some Dutch disease effects. The real exchange rate appreciated from US$ 2.3 to US$ 0.0014947. This worsened the competitive
advantage of the Surinam economy. He showed that government investment as well as consumption had a positive effect on economic growth. However, almost all government investment came out of resources from official development aid (mainly Dutch grants) instead of resources like bauxite rents. In Surinam, there was a perfect example of the problem described by Manzano and Rigobon (2001). The government built two large bridges, costing of approximately US$ 100 million. This was financed with future bauxite revenues of the government. The bauxite companies directly deposited their transfers to the government on an account of the bridge builder. Current expenditures of the
government suddenly lost a fundamental source of financing. Therefore budget deficits accelerated and growth was deteriorating.
The findings of Martin (2001) and Eckhorst (2004) suggest that there is some evidence for the existence of a resource curse in Surinam. In the remaining part of this chapter I will investigate whether I can find this evidence as well for Surinam and for the other resource abundant countries in the CARICOM.
3.2 The resource curse channels within the CARICOM
In this section I will investigate to what extent the resource curse channels mentioned in chapter 2 apply to Guyana, Jamaica, Surinam and Trinidad and Tobago.
3.2.1 Dutch disease effects
I will analyze whether there are Dutch disease effects visible in the four resource abundant CARICOM member states, by looking at booms in the resource sector.
Spending effect
A boom in commodity prices can, according to the Dutch disease theory, results in an appreciation of the RER. Figure 3 shows the movement of the resource prices, GDP per capita and the RER of the resource abundant countries48.
Table 4 summarizes the periods when a boom occurred in bauxite, gold or oil prices. Table 4 : Booms in commodity prices
Commodity First boom Second boom Third boom Fourth boom Bauxite 1985 - 1989 1993 - 1995 1999 - 200049 2002 - onwards Oil 1986 - 198750 1988 - 1990 1998 - 2000 2002 - onwards Gold 1985 - 1987 1992 - 1994 2001 - onwards
In will start with a description of the effects in countries abundant in bauxite.
During the first boom, the RER of Surinam appreciated strongly, but this began already in 1980. 190 till 1993 was a period of hyperinflation and low economic growth. The appreciation of the RER might be a result of the high bauxite revenues in the period 1985 – 1989. However, when prices declined, the exchange rate did not depreciate. A
remarkable period in the exchange rate is shown in 1993, when there was a huge real depreciation. This was on the one hand a result of the low bauxite prices in the period before and of measures taken by the government which led to economic stabilization. As a result of the second boom, from 1994 the RER appreciated again as a result of high inflation rates. Table 4 in appendix 1 shows that in 1995 inflation was over 200 %. However, despite this Dutch disease effect, GDP rose. From 1998 there was a
depreciation of the RER and GDP per capita declined. The latest boom brings high GDP per capita growth rates. From 2004 Surinam becomes more dependent on gold and Surinam experienced a boom in all the three commodities from 2002. These booms let to high growth rates. One of the reasons for the high economic growth was stable economic policies which resulted in a positive budget balance and a positive current account in 2006. Despite high growth rates, the traditional Dutch disease effect was present; the real
48 When analyzing these figures it is good to note that a decrease in the RER means an appreciation of the
currency.
exchange rate appreciated again. As a result agricultural sector competitiveness declined. So there was evidence for the de-agriculturalization effect as mentioned by Corden (1984).
The RER of Jamaica depreciated from 1983 till 1986. As a consequence of the first boom in bauxite prices the RER appreciated strongly till 1989. In this same period GDP per capita did not decline until 1995. There was a strong appreciation of the RER between 1992 and 1998, which might be a result of high bauxite prices. However, this is hard to conclude, because between 1994 and 1998 bauxite prices declined. The appreciation from 2003 onwards is more likely to be linked to the current boom. Not only the RER
appreciated strongly, also labour costs increased. The combination caused a decline in competitiveness. As an illustration of the impact of this appreciation (35% from 1990 to 2001), the World Bank reports a 50% decline in Jamaica’s market share of world merchandise exports from 1994 tot 2001. Unit labour costs rose twice as much as in the economies of its main trading partners51. The negative impact on GDP is not evident from figure 3.
Guyana is not dependent on one single commodity, but possesses besides bauxite also a remarkable amount of gold. The share of gold in total exports was about 10% in 1990 and 30% at present. Bauxite share in total exports was about 35% in 1990 and is about 10% nowadays. Guyana experienced a long period of depreciation from 1985 till 1992. GDP per capita declined strongly from 1981 till 1990, while commodity prices were not very low during this period. In this period, there is no obvious reaction of the RER on commodity booms visible. From 1992 till 1995 the exchange rate appreciated. This appreciation could be a result of the high bauxite and gold prices. However, this had no negative impact on GDP per capita, since this increased strongly from 1990 till 1997. In reaction to the latest commodity boom the exchange rate appreciated and as a
consequence the rice sector lost competitiveness. This sector declined strongly from 2000 onwards. Furthermore, GDP decreased. This whole combination suggest that there were some Dutch disease effects as a result of the last commodity boom..
Trinidad and Tobago is the only country abundant in oil. It experienced the first rise in oil prices in 1988. The RER of Trinidad and Tobago appreciated in this period as well until
1991. Trinidad and Tobago has, in contrast to the other resource abundant countries, pegged its currency against the US dollar, so shifts in the RER can be directly linked to changes in prices. GDP was stable in this period and did not decline as a result if the appreciation. Also, in contrast with the Dutch disease literature, unit labour costs have declined faster than those of its trading partners. This decline was especially evident in the non petrochemical sectors, allowing them to mitigate the adverse effects related to a RER increase. From 1998 the exchange rate appreciated strongly until now. Despite this appreciation economic growth increased from 1991 onwards. Trinidad and Tobago has introduced a heritage and stabilization fund, which must lead to a reduction of the negative Dutch disease effects52.
From figure 3 I can conclude that frequently after a boom there were some spending effects which resulted in an appreciation of the RER. Non-booming sectors, especially agricultural ones, were hit by this appreciation and lost competitiveness at world level. However, from figure 3 I can not conclude that these appreciations and its spending effects had a strong negative effect on GDP per capita rates, which is predicted in the Dutch disease literature.
