There are a large number of hypotheses that were presented by various studies which show a negative or positive relationship between resource abundance and economic growth.
One of the earliest assumptions was affirmed by the French philosopher Jean Bodin in 1576, who declared that:
“Men of a fat and fertile soil, are most commonly effeminate and cowards; whereas contrariwise a barren country make men temperate by necessity, and by consequence careful, vigilant, and industrious”.
The 1950's and 1960's studies, and the Dutch Disease models of the 1970s and 1980s are concentrated on negative externalities provided by the natural resource sector, compare to manufacturing. Hirschman [1958], Seers [1964], and Baldwin [1966] encourage the idea that positive "forward and backward linkages" from primary exports to the rest of the economy seem to be small. The idea behind is that manufacturing leads to a more complex division of labor than natural-resource production and reflects in a higher standard of living.
The global commodity price booms of the 1970's caused further research about the economics of natural resource abundance. A review of this literature is available in a summary created by Neary and Van Wijnberge in 1986. A part of this literature is focused on how macroeconomic policy reacted to commodity price booms. It tries to analyze the long-term growth effects of natural resource production and natural resource booms. The main subject was, if de-industrialization is encouraged by natural resource production. According to Sachs and Warner [1997] these studies did not perform a cross country analysis in order to examine the relationship between economic growth and natural resource abundance.
This section is divided in four parts. The first provides an overview of the past literature which was related to the Dutch Disease model. Second part describes studies concerning environment and governance importance. Third concentrates on the importance on how resources are produced and the final part gives a short literature overview which shows the various measurements of the resource abundance.
2.1 Dutch Disease models
Matsuyama [1992] develops model which is described as the “linkages approach” in order to analyze the impact of natural resources on economic growth. In the model, he examines the function of agriculture when manufacturing is described by learning-by-doing. The conclusion shows that the growth rate of the economy declines when it is pushed away from manufacturing and towards agricultural sector. The reason for the decline is the reduction in the learning process which is important for the growth of manufacturing. Furthermore, the learning effects turn out to be external to the firm and the market equilibrium becomes inefficient.
Feenstra [1990] and then Grossman and Helpman [1991] have elaborated models where a country that remained behind technologically is capable to be influenced by trade to specialize in traditional goods and for that reason it faces a reduction in its long run growth rate. These types of models, according to Rodriguez and Rodrik, [1999], are formalization of older arguments concerning infant industries and the necessity for provisory protection in order to increase the economic growth and reach the level of more advanced countries.
Sachs and Warner [1995, 1999] simplify Matsuyama's model by means of the “Dutch disease” model.
The “Dutch disease” relies on the dreads of deindustrialization that involved the Netherlands as a result of the appreciation of the Dutch guilder because of the discovery of natural gas deposits within the country's jurisdiction in the North Sea in the late 1950s and early 1960s. One the one hand, this discovery generated a gas export boom and currency appreciation, but on the other hand it reduced the profitability of manufacturing and service exports.
The model is characterized as a one which has a tradable natural resource sector, a tradable (non-resource) sector, and a non-traded sector. In case of high natural resource endowment, the demand for non-tradable goods increases. As consequence, there is a smaller allocation of resources like labor and capital in the manufacturing sector.
The two authors explain that in Matsuyama's model, the negative effects of agricultural production occur because the agricultural sector directly involves the factors of production that otherwise would be in manufacturing. In Addition, they suggest that such framework would be useful for studying labor intensive production of natural resources, such as in agriculture, but is less significant for a natural resource sector like oil production, which uses very modest amount labor, and for that reason does not directly subtract employment from manufacturing.
This “Dutch disease” can be a problem for a country's economy if there is something unique about the sources of growth in manufacturing, such as the learning by doing mentioned by Matsuyama.
Contrary to Matsuyama, Torvik.R [2001] obtains a model in which learning-by doing can be created in all sectors, and learning transfer is permitted to occur between sectors. He concludes that “the existing literature on learning-by-doing and the Dutch disease may yield overly pessimistic conclusions for some countries, but at the same time may be too optimistic for others.” Furthermore, he also discovers that “in contrast to standard models of the Dutch disease, production and productivity in both sectors may go up or down”.
2.2 Environment and governance importance:
There were a plenty of studies that analyzed the importance of institutional quality for the economic development of a nation.
Mauro [1995] gathered a set of data including indices of corruption, the amount of bureaucracy, political stability and the efficiency of the judicial system for a cross section of countries. The analysis shows that corruption lowers investment, consequently lowering economic growth. Other findings show that bureaucratic efficiency may be at least as important a determining fact for investment and growth, as political stability is.
Knack and Keefer [1997] sustain that social capital is very important in order to measure the economic performance of a country, by using indicators such as trust and civic norms. They establish that these indicators are stronger in nations with higher and fairly distributed incomes, with efficient educational system, ethnically homogeneous population and with institutions where the chief executives are controlled in order to not perform actions that could lead to rent-seeking.
As Davis and Tilton [2002] affirm that “the appropriate public policy question is not should we or should we not promote mining in the developing countries, but rather where should we encourage it and how can we ensure that it contributes as much as possible to economic development and poverty alleviation.”
In case of high transportation costs for a natural resource, its presence within the economy is crucial for the introduction of a new industry or a new technology.
On the one hand, De Long and Williamson [1994] explain that the prerequisite for the development of a local steel industry in the end of the 19th century were coal and iron deposits. Resource-rich economies such as Britain and Germany grew rapidly at the end of the 19th century.
On the other hand, Carlo Bardini [1997] demonstrates the motive for Italy's weak performance before the First World War. The reason was the scarcity of domestic coal reserves, which resulted in “a backward economic structure of production”.
Lane and Tornell, [1996], utilize as leading example of the negative effect that natural resources can have on economic growth and development, the economic failure of resource-abundant Sub-Saharan Africa and Latin America. A potential explanation is that an extra coming from a discovery of natural resource deposits or improvement in terms-of-trade can lead to a “feeding frenzy” in which rivals groups fight for natural resource rents and end up inefficiently exhausting the public good.
Earlier studies by Gelb [1988] and Auty [1990] support the possibility of these political channels of influence. Particularly, Gelb is convinced that governments received the major part of the rents from natural resource exploitation. A country with a well defined and functioning property system, probably will not allow that its valuable resources are wasted by the governments or people with high political influence. However, many societies have a weak social infrastructure and weak economic policy that allow particular groups of interest to receive huge revenues coming from price booms. These revenues probably would raise a substantial inequality among the population, support rent seeking and favor policies that are difficult to reach.
Sachs and Warner [1997] added to the above mentioned issues that rent seeking is a dead weight loss. For that reason, an increase in rent seeking lowers steady state income and the growth to a higher the steady state. Another point raised by the authors, is the idea associated with the commodity price forecast in 1970's and 1980's which was too optimistic. This forecast was used as an instrument in order to support large public investments in projects that actually turned out to be totally inefficient because the forecast turned out to be over estimated. The outcome was that natural-resource abundant countries found themselves with more unsuitable capital in their pockets than other economies.
Finally, if the resource revenues are consumed or poorly invested, the GDP growth in natural resource abundant countries should be inferior than it would be in the same countries with best possible policies. Nevertheless, this kind of economies must not necessarily grow slower than other resource-poor economies.
A further discussed topic is the degree of ethnic fractionalization which may play an important role in determining a country's social infrastructure, and thus, it may affect the impact of the resource abundance on economic growth.
Hodler [2004] concludes that “theory, empirical evidence and casual observations suggest that natural resources tend to be a blessing with respect to incomes in homogenous countries, but a curse in fractionalized countries.”
Deaton [1999] describes the problems coming from revenues from resource exports that turned out to be inefficiently invested and the lack of balancing factors, like education in Africa. The level of exporting process in Africa is generally low. The possession of natural resources is often concentrated and an increase in commodity prices leads to increase in income inequality.
However, he stresses the fact that there is a strong relationship between GDP growth and commodity prices growth. In other words, the commodity prices growth leads economic growth.
Interesting findings are presented by Bulte, Damania, and Deacon [2005]. They discover that natural resource abundance, and particularly mineral resources, have an ambiguous direct impact on various measures of human development, but some negative indirect effect by means of two measures of institutional quality. They conclude that there is a connection between natural resource abundance and high quality institutions measured by an aggregate indicator, which leads to a positive growth effect. However the direct negative impact of resource abundance on economic growth seems to be persistent.
Robinson, Torvik and Verdier [2006] built up a model concerning the political economy and show that the impact of the resource abundance is dependent on the quality of political institutions, particularly on the level of political clientelism. They conclude that countries with discrete institutions quality have a greater probability to suffer.
Ross [2001] performed a cross-country study and discovered that the negative resource effects of mineral abundance on institutions diminish when there is an increase in income levels and significant quantity of past mineral exports.
Ross [2001] finds that countries rich in mineral resources, particularly oil, do not to make the transition to democracy or at least their score on an index of democracy tends to be low. The reasons he identifies are the following. There is a lack of .modernisation. as economic wealth does not translate into social and cultural change. Second, there is a repression effect; mineral and oil rich states can engage in higher levels of military and internal security expenditure to suppress internal dissent. Finally, and most importantly, there is a rentier effect. Revenues from oil and mineral resources create rents that can be utilised to bribe the population into silence regarding authoritarianism. This may also engender contests over the right to enjoy resource rents as in the model in section 4. Also, public goods may be provided alongside low taxes because resource rents are the main source of revenue for the state. Taxation normally results in pressures to introduce democracy.
Atkinson and Hamilton (2003) consider that in resource rich countries with low levels of genuine saving and poor institutional quality there is a negative effect on economic growth, because of high government spending in consumption and not in investment.
Stijns [2005] argues that there are both positive and negative channels through which natural resource abundance affects economic growth. He discovers that land abundance has a tendency to affect negatively all decisive factors of growth, by also introducing various measures of institutional quality. On the other hand the effects of mineral abundance are less precise. As a final point, Stijns explains that learning process is the decisive factor which determines how countries exploit and develop their natural resources.
Acemoglu, Johnson and Robinson (2001) affirm that after performing across country analysis they did not find a significant effect of natural resource abundance on income. Their findings suggest that institutional quality alone gives enough evidence to explicate differences in economic development.
According to the literature, there are ambiguous effects of natural resource abundance on growth, when the institutional quality factor is incorporated in the analysis.
The “curse” is more evident when resource rich countries have a low institutional quality.
2.3 The importance of quality and not quantity:
Sachs and Warner [1999] assume that learning-by-doing is proportional to the relative size of the manufacturing sector. The authors are convinced that countries which managed in the past to industrialize, based on their natural resource abundance, should still be experiencing higher growth rates because of the hereditary size of their industrial sectors. In conclusion, they believe that learning-by-doing is limited to the manufacturing sector and is absent in other sectors, in example resource production or agriculture.
Ross [2001] believes that “the best course of action for poor states would be to avoid export-oriented extractive industries altogether, and instead work to sustainably develop their agricultural and manufacturing sectors.”
De Ferranti et al. [2002] conclude that “the recurrent lesson of the successful natural resource developers, and of contemporary theory, is the necessity of engendering a high level of human capital and developing a capacity for national learning and innovation.”
Another line of argument is the hypothesis of Raul Prebisch [1950] and Hans Singer [1950] where the authors argued that resource-based growth would be disturbed by secular decline in world prices of natural resources and the demand for primary for primary products would grow slower that demand for manufacturers.
The "Prebisch hypothesis" of declining relative prices influenced developing countries which started to promote industrialization in order to decrease dependency on natural resource exports. According to Sachs and Warner [1997], there was an inaccuracy from the side of United Nations Commission for Latin America which suggested industrialization not through export promotion, but through extended import-substitution behind tariff and quota barriers.
According to economic historian Gavin Wright [2001], “what matters most for resource-based development is not the inherent character of the resources, but the nature of the learning process through which the economic potential of these resources is achieved.”
2.4 Measuring natural resource abundance
Sachs and Warner [1995a] constructed a cross country model where the resource abundance variable was described as the ratio between primary exports over GDP (sxp.). The authors provided an ordinary least squares (OLS) regression by implementing the sxp indicator at the beginning of the period and the initial income. The result of the study shows that “by holding the initial income constant a higher share of the primary exports at the start period is associated with a lower growth in the long run.” Moreover they implemented additional controlling variables in order to see if the result would have changed. Finally, the overall negative relationship persisted.
Several other studies use the same resource intensity variable and confirm the negative relationship. For that reason, the sxp variable became one of the preferred indicators in the natural resource literature.
In another study, Gallup and Sachs [1998] regress levels of per capita income on non-conventional explanatory variables.
The result of the regression shows that levels of per capita income across countries in 1995 are positively related to deposits of some natural resources. To put it differently, natural resources are correlated with GDP per worker, which is a variable usually included to capture conditional convergence effects. This result has a tendency to reduce the beneficial role natural resources play for growth, even controlling for initial income per capita. As example, the authors choose the economic failure of resource abundant Sub-Saharan Africa where natural resources affected harmfully the development, thus the economic growth.
Sala-i-Martin and Subramanian [2003] by studying study the case of Nigeria, find an excessive amount of corruption caused by oil abundance. The authors provide evidence which shows that the resource curse is visible in case of mineral and especially oil abundance, but not agricultural products and food. All included variables are based on export data which is a share of total exports or GDP. However, they are convinced that the economic performance in this country is due to poor institutional quality caused by oil abundance.
They propose a solution in order to stop resource course, by directly distributing the oil revenues to the public. To conclude, the presented proposal can be appropriate also for other countries which are dependent on oil and minerals.
Other studies show that by a variation in measurement of the exports variable the resource course hypothesis changes.
Davis [1995] shows a positive relationship for economic development by using the share of mineral exports in total merchandise exports.
Ledermann and Maloney (2003) use the share of primary exports in total exports and primary exports over total labor force. They result is a positive effect for economic growth.
Neumayer (2004) take the genuine income as dependent variable (GDP minus depreciation of produced capital) and resource abundance variable Sachs and Warnes's sxp ratio. Their calculations provide evidence of a weak negative effect.
Various other studies present alternative measures for resource abundance which are not based on exports data.
Atkinson and Hamilton [2003] discover both positive and negative growth effects after having used resource rent data. They distinguish between various classes of resources and categorize them the by indices. In the end, the result demonstrates that countries dependent on point source extraction, in example minerals, differentiated by restricted concentrated local production, show a non satisfactory performance on economic and institutional level. Contrary to this, Isham et al. [2005] show that countries dependent on more diffused resources perform better that those dependent on point source extraction.
Davis [1995] discovers a positive relationship between economic growth and mineral production weighted by GDP. The same variable is used by Papyrakis and Gerlagh (2004) who find positive and negative effects.
Ding and Field [2005] try recalculated Sachs and Warner's model by using World Bank data and use a three-equation model in order to observe the effects of resources on human capital. They provide a result which shows a positive growth effect of natural resources per capita and a negative effect of natural resources as a share of total produced capital.
In conclusion, Brunnschweiler [2006] studies in her paper the effect of natural resource abundance by using new measures of endowment accompanied with the role of institutional quality. She finds a positive relation between natural resource abundance, measured by subsoil wealth and total natural capital.
Brunnschweiler, Christa N. [2006], “Cursing the blessings?”
Natural resource abundance,institutions, and economic growth.
Working Paper 06/51, May 2006, page 4.
Moreover, her results provide no indirect effect through the institutional channel and the resources do not influence negatively the institutional quality through ret-seeking behavior.
An interesting fact is that by an increase in institutional quality the growth seems to diminish, but the positive relationship seems to be overall persistent.
3. Cross country estimations
3.1 Data and descriptive statistics
This section describes the key variables used for calculations and presents the descriptive statistics. Furthermore, it presents and illustrates additional variables included in the calculations which according to the literature might influence the outcome of the computations. In the sample there are 26 oil producing countries. The reason for excluding other countries in the analysis is the lack of available data. In one case during the regression analysis there will be a further reduction of countries number being short in data availability.
The key variables used in the analysis can be observed in Table 2.
The first raw shows the average real GDP growth per capita between 1985 and 2003. The data is taken from the Penn World Tables 6.2 and calculated by the author according to the formula presented by Barron d Sala-i-Martin [1995] and then used by other authors. The formula is Gi = (1/(T-t))* LOG(YiT / Yit) where G is the average real GDP growth per capita, t is the start year and T is the end year respectively. Y is described as real GDP in a given year. This variable will be the dependent one for succeeding calculations.
The second variable is the Log real GDP per capita in 1985 which was taken from Penn World Table 6.2 and transformed by the author.
The third and the fourth raw show the resource abundance variables. The first is the average of produced oil tonnes per capita calculated for the same period as for real GDP. The second is the value of produced oil tonnes per capita which is calculated by the author by multiplying the amount produced with each year's oil price, which is adjusted to 2006 prices. The reason for these calculations is to see in the price influence the outcome and in which amount. The data is provided by the Statistical Review of World Energy June 2007 presented by British Petroleum (BP). The reason for choosing production data instead of exports or reserves is that production shows the effective presence of the resource in economy which according to governmental decisions may be exported, used as a factor of production in other industry sectors or deposited.
The last three rows illustrate the most important variables which determine the institutional quality. These variables are described by Kaufmann, Kraay and Mastruzzi in 2005 and then updated in 2007 as Worldwide Governance Indicators (WGI) research project.
The first variable is the governmental effectiveness (GE) which “measures the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment to such policies.” The second institutional variable is the Rule of Law (RL) that evaluates the “extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, the police, and the courts, as well as the likelihood of crime and violence.”
The last indicator is described as Control of Corruption (CC) that explain the amount to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as "capture" of the state by elites and private interests.
All three indicators are calculated as averages between 1996 and 2003, and have a score which lies between - 2,5 and 2,5. Higher values imply a better institutional quality. The importance of these World Bank Indicators lies in their objectivity provided by an extensive study samples and the outstanding country reporting.
There are several additional variables included in the computations which will be briefly presented.
According to Christa Brunnschweiler [2006], latitude seems to have a positive impact on economic growth. For that reason the latitude variable will be included in the calculation. This variable is described by La Porta at all.[1999] as theoretically justified measure of geography. It is calculated as the capital's distance from the equator and the transformed in order to reach values between 0 and 1.
The next additional variable is country's trade openness which is described by Sach and Warner [1995b]. A country is said to be integrated if it maintained reasonably low tariffs and quotas, and did not have an excessively high black market exchange rate premium, was not socialist, and avoided extreme state-control of its export sector A country that was open every year between 1965 and 1989 received a value SOPEN = 1. A country that was always closed during these years received a value SOPEN = 0.
An other openness variable is presented by Penn World Tables 6.2 ad is described as:
A further introduced variable in the model is the ethnical fractionalization describer and provided by Alesina et all. [2002]. who suggest a measure of ethnic fractionalization has become a “standard” control in regressions explaining cross-national differences in economic success Measure of ethnic fractionalisation ranging from 0 (least fractionalised) to 1 (extremely fractionalised) based on racial or linguistic characteristics, determined country by-country. Most data for mid-1990s.
An additional included variable is the schooling variable described and presented by Barro and Lee [2001]. The authors are convinced that human capital that mainly obtained through education is a decisive determinant of economic progress. A greater amount of educational attainment implies more skilled and productive workers, who in turn increase an economy's output of goods and services.
The last variable which was included in the calculation is the schooling variable described and presented by Barro and Lee [2001] which is calculated as the average yeas of Educational Attainment of the Total Population Aged 15 and Over between 1985 and 2000.
3.2 Defining the model
3.3 Regression calculation and analysis
