Thesis Supervisor: Dr. Sijeong Lim Second Reader: Dr. Seiki Tanaka
Political Conditions on the Resource Curse: What
Determines the Incentives to Invest in Economic
Diversification?
By Derk-Jan Verhaak, 10384340
Master’s Thesis Political Economy, Political Science June 2018
Abstract
This thesis analyses the effects of resource abundance on diversification and how institutional quality could play a key role in making a resource-rich country more diversified. So far, scholars have solely focused on either the effect of institutional quality on the prosperity given by resources or the effect of diversification on this prosperity. However, does institutional quality also contribute to a higher diversification of a specific country and therefore mitigate the damage caused by natural resources in two manners? This will be discovered with multiple datasets of 179 countries for a timeline ranging from 1980 to 2016. Overall, this thesis finds that an abundance of resources has a negative effect on the diversification of a country. Next to this, a higher institutional quality implies a higher diversification of a country. Institutional quality does also influence how an abundance of resources affects the diversification, as countries experience smaller negative effects of an abundance on the diversification if the institutional quality is high.
Table of Contents:
Abstract... 1
Table of Contents... 2
I. Introduction... 3
II. Literature... 5
A: Possible Solution to the Resource Curse B: Seeking Diversification C: Policies and Instrumental Powers D: How to Fill the Gap? III. Theoretical Framework... 13
A: The Gap Discovered B: Case Study: Venezuela C: Conclusion IV. Methodology... 21 A: Data B: Estimation Technique C: Descriptive Statistics V. Results... 26
A: Estimates on the Diversification Index B: Estimates on the Share of Fuels on Total Exports C: Zooming into the Results VI. Discussion... 36
VII. Conclusion... 37
References... 39
1) Introduction:
Venezuela is a country that is characterized by its vast amount of natural resources, theoretically worth billions of dollars (OPEC, 2018). It has benefited significantly from historically high oil prices until the end of 2014. This amount of resources enabled, until that time, an increased public spending and ,economically speaking, ambitious social programs. Economic growth and redistribution policies resulted in significant declines in poverty, from a poverty rate of 50 percent in 1998 towards 30 percent in 2013 (World Bank, 2017). However, this came to an end after the collapse in international oil prices. The economy became in a severe crisis and the country is nowadays characterized by hyperinflation, violent clashes, and shortages of medicines and basic household supplies (Garcia, 2017). It seems that Venezuela did ultimately not profit from its natural resources and according to multiple studies, Venezuela is not on its own (Mehrara, 2009). Historical records show that there is less economic growth in many resource-rich countries than in resource-scarce countries. This corresponds with a much debated paradox, which is the resource curse. This thesis is centered around the phenomenon that some countries with an abundance of resources handle their resources significantly better than other countries. This would, for example, result in vast differences in diversification among countries that have the somewhat same resource abundance. An example of this occurrence are Russia and Iran: In 2010, 66 percent of Russia’s total exports were fossil fuels. In the same year, Iran’s exports consisted for 71 percent of fossil fuels, which is around 8 percent more than Russia. However, according to an export diversification index, Iran is performing, with a percentile rank of 29, significantly worse than Russia, which has a percentile rank of 46 in diversifying its exports. How is it possible that, while both countries have an abundance of natural resources, one is succeeding relatively more in diversifying its exports, while the other fails in this? In this thesis, this is one of the questions that will be discussed.
Scholars have researched the effects of natural resources extensively. Many scholars indeed propose the existence of the resource curse (e.g. Sachs & Warner, 1995; Karl, 1997; Auty, 1994). Sachs & Warner (1995) find, for the period 1970-1990, that economies with a high ratio of natural resources to GDP experience slower economic growth. Even when controlling for variables such as capital accumulation and global commodity price shocks, the results remain significant. Other scholars also researched the resource curse and combined several control factors, such as institutional quality (Arezki & van der Ploeg, 2007; Brunnschweiler, 2008). Here, they find, when they control for institutional quality, a positive relationship between natural resource abundance and economic growth. Even with this many scholars investigating the paradox of resources, a clear conclusion has not been stated yet. Also, it seems that more recent studies disagree with studies from a longer time ago. An overall conclusion is hard to state and depends on the controlling factors of the studies. Besides the relationship between a resource abundance and economic prosperity, has the effect of resource abundance on diversification also been discovered. Though only a few researches about this relationship are done, the relationship is generally negative, indicating that a resource abundance decreases diversification (Gylfalson, 2006; Auty & Pontara,
2008; Van der Ploeg, 2011; Venables, 2016; Weinthal & Luong, 2006).
Next to these studies about abundance and growth, there have also been studies on both diversification and institutional quality and their effects on economic growth in general. The effects of diversification on economic growth and other relationships such as diversification and the index of democracy have been studied thoroughly. Generally, scholars find a positive relationship between these indicators (Hesse, 2009; Gylfalson & Nguessa Nganou, 2016). Also for institutional quality and economic growth, most of the studies find a positive linkage (Butkiewicz & Yanikkaya, 2006; Acaravci & Erdogan, 2017).
Overall, it seems that a lot of research has already been done about the general effects of a resource abundance, economic diversification, and institutional quality on economic prosperity. Next to that, it has also been tested how the effects of resource abundance changes when one takes institutional quality and diversification in account. For these reasons, this thesis will not empirically test these estimated effects again. Instead, this thesis seeks how diversification is affected by institutional quality as it argues that the effect of institutional quality will offset the negative effects of a resource abundance on diversification. Higher institutional quality would increase a country’s diversification, which would both increase a country’s economic growth and mitigate the negative effects of a natural resource abundance. In this way, institutional quality would transfer its effect into the diversification index. If this would be the case, then this would be the first step towards a different view on institutional quality and the resource curse. Next to this factor being discovered, this thesis will also highlight the effects of resource abundance on diversification, as only a small number of research has been done about this topic. This research could be valuable and interesting for the fact that natural resources could offer major opportunities for economic development of developing countries. If the aim of this thesis reflects the truth, these countries would have more information about functioning focus points concerning the prosperity of natural resources, namely institutional quality and diversification.
This thesis will now continue with section 2, which is the literature section. This surveys the literature and empirical studies, discussing papers done in the recent years. The other sections will describe the theoretical framework, method, results, discussion and lastly a possible conclusion.
2) Literature:
This section gives an overview of existing literature on multiple topics varying from solutions for the resource curse towards diversification and policies that make diversification more effective. In subsection A, “Possible solutions to the resource curse”, the resource curse that was introduced in the previous section will be discussed further. Possible solutions that are advocated by scholars are highlighted here. This is done in order to grasp the already existing proposed solutions for the resource curse and get an idea of the already existing research about this. After all, this thesis tries to fill the gap and help countries by deciding on policies, meaning existing solutions should be discussed. Then, after this, diversification is introduced as it is the main focus point of this thesis. Therefore, subsection B, “Seeking diversification” discusses the general advantages of diversification. It is important to have an overview of the nexus between diversification and its effects on, for example, economic growth; namely, as this thesis assumes in its analyses that diversification is generally good for economic growth, certain conclusions concerning economic growth will be drawn out of these analyses as well. As was discussed in the introduction, this thesis will not empirically test these findings again, as many scholars have already researched this (Gylfalson, 2006; Auty & Pontara, 2008; Van der Ploeg, 2011; Venables, 2016; Weinthal & Luong, 2006).
Next to diversification in general, the incentives but also the disincentives and obstructions of governments to seek diversification are considered. In this way, it becomes clear why governments seek diversification in the first place. After a framework of export diversification has been made, subsection C “Policies and instrumental powers” discusses what makes a policy more effective in addressing export diversification. One of the topics discussed here is what has already been found on the effects of institutional quality on diversification. Several instrumental factors are discussed, as powers and interests could affect the effectiveness of certain policies. After these topics, an overview will have been made of diversification, the incentives of governments to diversify and the effectiveness of a government’s policies. At last, subsection D, “How to fill the gap?” gives a small summary and elaborates what the exact remaining gaps are and how this thesis seeks to fill in these gaps.
A. Possible Solutions to the Resource Curse
In the introduction, the resource curse was brought up. There, an ambiguity became clear about natural resources and whether they deliver extra economic growth or not. The main reason to make this thesis comes, though partly, from the resource curse. For now, this section discusses potential solutions for the resource curse, brought by multiple scholars. This is done to have a more informative background and this thesis will then be able to eventually come up with a possible solution itself.
One of the factors that could turn the resource curse around is highlighted by Tsani (2013). She investigates resource funds and the role these funds could play in the addressing of the resource curse. The results show that countries that have established a recourse fund perform substantially better in terms of institutional quality. Even though this could be a circular relationship, it is still a promising finding. The circular relationship comes from the fact that countries with a higher institutional quality are more likely to
assign a resource fund in the first place. Next to a better performance, the presence of a resource fund provides protection against the resource curse as it may be associated with changes in the behavior of economic agents. In that way, a country could become less vulnerable for corruption and rent-seeking, whose effects are elaborated later in this literature section.
Another possible solution is mentioned by Sala-i-Martin & Subramanian (2003) who used Nigeria as a case study to address the Resource Curse. They propose that the government should stop managing the revenues from natural resources and directly turn them to the people. This would translate into every Nigerian having a constitutional right to an equal share of the proceeds. They argue that there will be certain difficulties such as corruption and inefficiency which will definitely disturb the actual implementation. However, even with those difficulties, they show that this proposal would at least be “vastly superior to the status quo”. They also state that the case of Nigeria could be extended to other countries if these countries have a deep dependence on natural resources and have suffered from declining institutional quality subsequently.
Weinthal & Luong (2006) do not agree with above-standing solutions as the success of these solutions have been limited when implemented. They argue this is the case because these solutions assume strong institutions, which are absent in most developing countries. They state that these scholars are neglecting a possible solution, namely domestic private ownership. In this way, resource rents are taken out of the state’s direct control. Next to that, it immediately creates an incentive for governments to build strong fiscal and regulatory institutions and it creates new societal actors with the power and potential to demand these institutions. By being able to demand such institutions, it is likely that institutional quality will eventually increase as well, making a country better off.
As was mentioned in the introduction, scholars generally don’t look at diversification as a possible solution for the resource curse. Therefore, this thesis will now discuss diversification and its general economic and political advantages. Also, it will discuss why governments seek diversification in the first place. What incentives and disincentives do governments have to diversify an economy and what obstructions do they face when they try to diversify.
B. Seeking Diversification
In their paper “Regional Economic Diversity and Diversification, Siegel et al. (1995) discuss the definition of economic diversity and diversification. As there is no single definition of diversification, it is important to get the definition of diversification in this thesis clear. First of all, diversification refers to economic diversification, which is defined as the process of structural transformation (Siegel et al.). This structural transformation takes the form of resources that are shifted from primary (natural resource-based) sectors into secondary and tertiary sectors, which are respectively the manufacturing and service sectors. In political economy, diversification normally refers to export diversification and more specifically, the spread of production over many sectors. Therefore, in this thesis, economic diversification is translated into the spread of sectors in the export or how many sectors are present in the export and what their
weights are in the export.
There are multiple scholars that investigate the effects of export diversification on an economy. One of the most studied subjects surrounding this is the relation between economic development such as economic growth and export diversification. From many studies, results show that export diversification promotes economic growth (Agosin, 2007; Hesse, 2009; Lederman & Maloney, 2007; Aditya et al., 2013). In almost all these cases, the results are robust and significant. Some of the mentioned scholars also research the number of trade partners, which is again positively correlated with economic growth. Not only growth is positively influenced by export diversification; it also mitigates economic risk in both the long and short-term (Samen, 2010). In the short term, a higher diversification decreases the volatility and instability of the foreign exchange rate which has opposing positive macroeconomic effects, such as a higher foreign direct investment and a higher employment (Nasir & Hassan, 2011; Samen, 2010). In the long term, trade becomes more predictable, which reinforces the short term effects even more. Also, the World Bank states that not only economic- but also political risks are alleviated: Both the chance of a worsened governance and the risk of civil wars decrease with a higher export diversification. Also, when looking at an abundance of natural resources, scholars find it to be an important factor whether a country with a natural resource abundance succeeds or not. Gylfason (2006) states that many resource-rich developing countries struggle with finding ways to reduce their dependence on resources and diversify their economic activity. There are multiple scholars who concur with this statement and state that economic diversification could offset the negative effects of natural resources. They argue that capital obtained with natural resources should be used to diversify the exports in order to mitigate the negative effects of a natural resource abundance (Auty & Pontara, 2008; Van der Ploeg, 2011; Venables, 2016; Weinthal & Luong, 2006).
It is clear from the scholars stated above that diversification has a positive effect on a country’s economy. However, resource-rich countries are not always implementing these diversifying recommendations. These incentives for governments to diversify become partly clear because of all the conclusions made above: Namely, export diversification prospers economic development and it decreases economic and political risk. Next to this, there is another political risk that diversification policies would partly cover, namely the demands of citizens. In order to keep the trust of citizens and to hold social order in a country, a government needs to react adequately on these demands. Therefore, this could incentivize governments to undertake certain activities and policies. An example of this would occur when a country has a highly educated workforce. This would incentivize a government to develop industries that require a high-skilled workforce such as the technology sector. This would then create a labor demand that matches the pool of workforce, which would meet the demands for high-skilled work by the population. This would then automatically result in a higher export diversification, as the country specializes itself in new secondary and tertiary sectors. This is in line with the self-discovery theory mentioned by Hausmann & Rodrik (2003) which implies the discovery of new export products by either firms or the government. According to the self-discovery theory, export diversification can be shaped by policies that are
implemented by the government. With this, governments could solve the issues that are related to a low diversification without a lot of help from other countries. However, a particular population with certain demands is needed for a government to feel these incentives. Without a population that is able to demand such things and with a low institutional quality, countries usually do not feel the need to immediately diversify.
This last note brings this thesis to the disincentives that governments have when looking at export diversification. In this, institutional and financial incentives play an important role. Andersson & Ostrom (2008) show that there are relatively many actors that have an active role in multi-level dynamics of natural resource governance regimes. The complexity of many natural resource cases requires sophisticated and strategical governance systems. Because of these many actors, actors that try to govern face a variety of incentives which often complicates and counteract the collective efforts governing actors make to reach a certain outcome. Besides this, they show that when a resource is more complex, with complexity measured as the number of goods and services it provides, more perverse incentives are likely to be present. As a whole, the incentives of other parties often counteract with the governing parties. Even if these parties have no practical control, they can influence the processes making the government less effective. Andersson & Ostrom emphasize the possible complexity of cases with natural resources involved. As a lot of value is entangled, many interested actors influence the decisions concerning extraction and distribution of a country’s resources.
Next to these disincentives caused by other actors, there are also other obstructions for export diversification. According to Wiig & Kolstad (2012), certain political and elitist powers may influence the incentives of a country’s regime to diversify its economy. They mention a significant influence of elitist powers when it comes to democratizing policies. These democratizing policies are usually accompanied by redistribution practices, which makes certain economic activities more equally spread, resulting in the elite having a relatively smaller share of these activities (Barro, 1996). If the rents from those economic activities are sufficiently important enough for the elite of a country, they would have an incentive to impede policies that might damage these rents. In this way, it proves difficult for regimes to implement policies to change a country’s diversification, as many policies towards higher diversification are combined with a higher democracy, and thus a higher redistribution. Especially when looking at developing resource-rich countries, this is the case; in these countries, resources are usually a relatively big rent for the elite, meaning they will not easily give this up. Besides, this effect seems to be bigger for immobile assets than for mobile assets, as the elite generally has less to lose from a further democratization in mobile assets (Wiig & Kolstad, 2012). Especially in immobile assets, that are controlled by the elite, policies may encounter opposition to democratization.
Besides these struggles for power, the incentives of the government themselves could harm the implementation of policies. Especially in natural resource-rich countries, this could be the case: Arezki & van der Ploeg (2007) state that natural resources invite rent-seeking and corruption. These happenings are, in general, costly for economic growth as is supported by multiple scholars. Murphy et al. (1993) argue
that high levels of rent-seeking may be self-sustaining and that especially public rent-seeking may torment innovation. Also, more recent papers agree with this statement and also found that rent-seeking forms an obstruction for economic growth (Carson, 2016; Iqbal & Daly, 2014). Besides this, corruption has been discussed in a much-quoted paper by Mauro (1995), who finds a negative significant association between corruption, investment and economic growth, which is again supported by more recent studies as well (d’Agostino et al., 2016; Dzhumashev, 2014). Thus, an abundance of resources invites corruption and rent-seeking, which again causes less economic growth. Next to this, when looking at natural resources and disincentives, Hvidt (2013) argues that there is a distance between the state and the national economy: Because of the relative high rents from for example oil or gas, the state is not forced to tax the local economy to finance its activities. Because of this, there is less need to create an efficient economy, with the corresponding consequences. Altogether, it would seem that, indeed, resources could lead to less economic growth in the form of rent-seeking behavior, corruption and less need for an efficient economy that is diversified. In these ways, natural resources could cause an economic downturn instead of the desirable economic prosperity. As was already stated, natural resources naturally invite corruption and rent-seeking. Collier & Hoeffler (2005) make an argument when these resources tend to be looted, stating there are four circumstances under which rents are looted earlier, having the corresponding aftermath for the country in question. These circumstances are when time horizons are short, which is obvious as elites then have little incentive to invest in the public good of growth. Second is when the elite is a relative small, fixed, and identifiable group. When an elitist group is smaller, the incentive to prioritize this group is higher. An example of such an identifiable factor is ethnicity. Thirdly, a circumstance under which rents tend to be looted earlier is when the assets are very valuable in relation to the average income in the society. This predicts more rent-seeking in resource-rich, low-income countries. The last circumstance is when democratic competition deteriorates into patronage politics. According to Collier & Hoeffler, at least one of these circumstances is always found in a developing country. That shows that developing resource-rich countries generally suffer a higher amount of rent-seeking than expected. According to them, this is partly caused by the lack of institutional quality, which will be discussed more elaborately in subtopic C.
Altogether, it is clear that there are multiple incentives and disincentives for governments to implement policies in order to diversify their economy. The fact that diversification is generally good for an economy does not automatically mean that countries will naturally diversify. There are many interests in an economy and therefore multiple incentives for governments to either diversify or not. In the next subtopic, this thesis will discuss what kinds of factors influence the effectiveness of policies.
C. Policies and Instrumental Powers
Overall, it is clear that the diversification of an economy and its export is important for an economy’s growth and soundness. However, what are the determinants of export diversification? And how can a country achieve the desired export diversification? Arezki & van der Ploeg (2007) find that trade policies
directed toward more openness could make the resource curse less severe or even turn it into a blessing. Agosin et al. (2012) concur with this statement and try to answer the same question by using a large dataset of countries, covering the period 1962-2000. They find evidence that trade openness and a country’s access to the open markets support a higher specialization and therefore a higher diversification. Also, human capital accumulation contributes positively to export diversification (Parteka & Tamberi, 2011; Cabral & Veiga, 2010). They state that human capital and the accessibility to other markets are the most relevant and robust determinants of export diversification. Whether policies would be able to directly influence the export diversification, is not completely clear as results are insignificant in these papers. Rodrik (2005) researches this in his paper “Policies for economic diversification”, and states that governments can, though with difficulty, influence the diversification of an economy. He argues that governments need to only provide incentives and subsidies for “new” activities. In this way, new sectors will arise and be able to prosper. However, he mentions that certain conditions are needed to make this strategy succeed. An example of this is a good institutional quality to make the policy-making work. Policies that are self-correcting and governments that recognize mistakes to react on them are needed for this, he argues. This last point, in particular, is what differentiates good policy regimes from bad ones. It seems that governments could influence diversification, but is there also a relationship between institutional quality and diversification? This exact question has not been studied thoroughly yet. Cortinovis et al. (2017) make a start with this as they look at the influence of institutions on regional diversification. They study 118 European regions in the period 2004-2012 and find that institutions do matter for regions to diversify their industries. Though they did not look at export diversification in particular, it is a basis for the effects of institutional quality on diversification. Méon & Sekkat (2008), on the other hand, do look how different types of exports are affected by different institutional quality proxies. These proxies are Control of Corruption, Rule of Law, Government Effectiveness, Political Stability, Regulatory Power, and Voice & Accountability. They find that manufactured goods are positively affected by Control of Corruption, Rule of Law, Government Effectiveness and Political Stability. Their results show that a lack of institutional quality hurts a country’s capacity to export manufactured goods. These results do not hold for non-manufactured goods and total exports. Again, they do not show the effects of institutional quality on the export diversification itself, but it gives an indication of how different export sectors are affected and therefore how capable a country would be in the first place to diversify its exports. It is clear that more research is needed about the relationship between institutional quality and diversification, as only a few scholars have done research that relates to this.
Institutional quality itself proves to be an important factor of success for countries. Returning to the cases of corruption and seeking, it can be observed that the rate at which resources attract rent-seekers and corruption, is naturally dependent on institutional policies and institutional quality. Bjørnskov (2011) looks into this phenomenon and explores under which conditions institutional quality leads to lower levels of corruption. He shows that a higher formal institutional quality is implied with a smaller likelihood of bureaucrats being offered and accepting bribes. This would mean that the chances of
corruption being present decreases when institutional quality is high. However, he also shows that the willingness among firms to pay a large amount increases when institutional quality is high. This creates an ambiguity when looking at the relationship of institutional quality and corruption. Torgler & Schneider (2007) also look into governance and institutional quality and find that a higher institutional quality helps to lessen the incentives towards the shadow economy, translated into black market transactions and undeclared work. Therefore, overall, one can state that higher institutional quality helps lessening corruption. Thus, when natural resources invite rent-seeking and corruption, this effect could be neutralized again by a good institutional quality.
It is clear that institutional quality could play a role in countering corruption and rent-seeking practices, but there are more important aspects surrounding institutional quality and the effects of it on countries’ practices. For example, institutional quality matters for foreign direct investment (FDI) as they are positively and significantly related (Buchanan et al., 2012; Masron & Abdullah, 2010). Buchanan et al. provide evidence for this by examining 164 countries from 1996 to 2006. Not only do they find a significant positive effect of institutional quality on FDI, they also find that institutional quality is negatively and significantly associated with FDI volatility. Therefore, when a country has a good institutional quality, its FDI flows are more consistent as well. In general, FDI has a significant positive effect on economic growth (Alfaro et al., 2004) while FDI volatility has a significant negative effect on economic growth (Lensink & Morrissey, 2006) meaning it is in a country’s own interest to maintain a high and consistent FDI. Asiedu (2006) also investigates the effects of FDI and does so in relation to natural resources in Africa. She finds that FDI is concentrated in the natural resources, which is usually not contributing to the positive spillovers, such as technological or human capital transfers, that are associated with FDI. Next, by having a concentrated FDI, there is a substantial risk of having a volatile FDI as well. As the FDI is predominantly in one sector, minor implications in this sector could cause the FDI to get significantly smaller. This could be a reason for the resource curse happening, as investments in the country are not diversified enough. Once they dry up because of an implication, a country’s whole finance will risk to almost dry up as well. Another point worth mentioning is that she finds that improving institutions and policies could attract FDI.
Thus, certain policies and investments could help a country’s policies to become more effective and thus have a higher effect on the diversification of an economy. However, here lies immediately a problem for certain countries: Namely, developing countries usually struggle with correctly implementing these policies. As they are characterized by weak public institutions and high levels of corruptions (Huang & Wei, 2006), implementing effective policies in order to incentivize differentiating investments may prove to be difficult. There are many examples which show that developing countries with poor institutions struggle to achieve a substantial impact with otherwise viable policy regimes. Huang & Wei provide such an example, as they research the impact of institutional quality on the effects of several popular monetary regimes. Their main results indicate that countries with poor institutions cannot always find the solution to their problems by viable policy regimes. As they lack confidence and credibility from
the public, it is difficult for them to implement policy regimes successfully. Wiig & Kolstad (2012) also mention the importance of a sound diversification strategy when a country wishes to diversify its economy. They argue that a regime should strengthen their own political power by focusing on industries that are concentrated in immobile assets controlled by the regime. In that way, a regime can retain its power step by step and increasingly influence the elitist powers in the country. By doing so, institutional quality will be able to gradually increase without too many powers working against it. Eventually, the institutional quality will be high enough to implement effective policies and a government will be able to diversify its economy and make the economy more efficient. This will then also make it less vulnerable for the negative effects of a natural resource abundance.
D. How to Fill the Gap?
This literature section has tried to form an overview regarding export diversification and the incentives of governments to diversify. It was shown how diversification could make an economy prosper and how these economic incentives lead governments towards diversification. However, it also showed why regimes sometimes do not diversify as there are also many disincentives to diversify. After that, policies, strategies and investments were discussed that could positively affect the diversification of a country’s export. Here, the obstructions for a proper implementation of these policies was discussed as well. Lastly, this literature section gave possible solutions to the resource curse and the difficulties related to diversification.
In the literature section, one element was still missing. This is also the gap this thesis will try to solve. Resources generally have an effect on a country’s economy. Whether this effect is negative or positive, is indistinct and dependent on multiple factors. Scholars state that diversification or a proper institutional quality could mitigate the negative effects caused by a resource abundance on a country’s prosperity. However, the relationship between this institutional quality and the diversification of a country has hardly been studied yet, especially when relating it to the resource curse. As institutional quality influences many aspects of a country, this thesis will explore whether it also influences the diversification of a country and how countries with different institutional qualities experience different effects of a natural resource abundance on the country’s diversification. If it would be shown that a higher institutional quality decreases the expected negative effects that a resource abundance has on a country’s diversification, this could be a helping hand for countries wanting to solve their resource curse, as the influences on this curse become more clear.
3) Theoretical framework:
In the theoretical framework, this thesis will highlight its main focus, show how it fits in the already existing literature, and formulate its expectations. First, the gap that scholars are currently missing is further elaborated and it is explained how this thesis will try to fill this gap. In this, the research question and hypotheses will be stated as well. Lastly, this theoretical framework will review a real-world example as a case study. In this, it will discuss the problems that Venezuela is currently facing and what the potential causes are of this crisis. After this case study, it will combine the knowledge from the case study and the expectations to form a conclusion.
A. The gap discovered
In the literature review, it could be seen that the resource curse is a much-debated topic. As was already stated in the introduction, merely mixed results about the consequences of having an abundance of natural resources are found. Therefore, a clear conclusion about the effects of natural resources cannot be stated. Other scholars have looked at certain effects that might determine the success of natural resources. A well-known example of this, also used by this thesis, is institutional quality. They often found that institutional quality could partly resolve the resource curse. But even here, some ambiguity remains as the consequences of this higher institutional quality were often not formulated. This is a strange thing to observe as many scholars find relatively many positive effects related to institutional quality when related to other indicators such as economic growth. As was shown in the literature review, a higher institutional quality, for example, reduced corruption levels and created higher and more stable FDI levels (Bjørnskov, 2011; Buchanan et al., 2012). However, little is mentioned about the effects of institutional quality on the export diversification. Even though it was shown that policies towards a higher trade openness and the accessibility to a market caused an economy to be more diversified (Agosin et al., 2012). If certain policies could affect an economy’s diversification, then why does the impact of institutions on diversification receive so little attention? This thesis expects this impact to play a significant role in export diversification and therefore in the economic prosperity of a specific country.
When looking at a natural resource abundance, corruption, rent-seeking and a country’s incentive to diversify, Mehlum et al. (2006) claim that institutions are indeed of significant importance here. They argue that natural resources push aggregate income down when the institutional quality is low or when the institutions are “grabber friendly”. This last notion means that institutions allow gains to be made of unproductive influences, caused by for example a weak rule of law or corruption. Good institutions that are “producer friendly” create circumstances that make the rent-seeking activities complementary instead of competing. With “producer friendly” institutions, resources attract entrepreneurs into productive activities instead of into unproductive activities. This causes a country to also experience growth because of resources. Thus, when a country has good institutions, it also attracts different types of entrepreneurs that create more economic growth. When a country already has a relatively low institutional quality when the natural resources are found, it also attracts relatively malicious entrepreneurs, creating less growth and
incentivizing regimes to not invest in institutional quality and loot instead. This causes a downward spiral, as institutional quality gets even lower. Eventually, the country will be contaminated with corrupt and inefficient activities, causing the mentioned resource curse. Therefore, institutional quality is of importance in this thesis: If a country already had developed institutions when they found the natural resources, it is less likely that the mentioned downward spiral occurs. This means that developing countries, in comparison to developed countries, more often make the decision to loot the resources and allow for corruption and rent-seeking rather than use it for growth such as investments in diversification, which was proved to have a significant effect on economic growth. Eventually, rent-seeking and corruption will have the upper hand and the country will receive less than expected from the resources. Because of this, the following research question will be examined:
𝑅𝑒𝑠𝑒𝑎𝑟𝑐ℎ 𝑄𝑢𝑒𝑠𝑡𝑖𝑜𝑛: 𝑊ℎ𝑎𝑡 𝑝𝑜𝑙𝑖𝑡𝑖𝑐𝑎𝑙 𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛𝑠 𝑑𝑒𝑡𝑒𝑟𝑚𝑖𝑛𝑒 𝑡ℎ𝑒 𝑖𝑛𝑐𝑒𝑛𝑡𝑖𝑣𝑒𝑠 𝑜𝑓 𝑟𝑒𝑠𝑜𝑢𝑟𝑐𝑒 − 𝑟𝑖𝑐ℎ 𝑐𝑜𝑢𝑛𝑡𝑟𝑖𝑒𝑠 𝑡𝑜 𝑖𝑛𝑣𝑒𝑠𝑡 𝑖𝑛 𝑒𝑐𝑜𝑛𝑜𝑚𝑖𝑐 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑎𝑛𝑑 𝑑𝑖𝑣𝑒𝑟𝑠𝑖𝑓𝑖𝑐𝑎𝑡𝑖𝑜𝑛?
To answer this research question, there are multiple factors that are important to regard. Therefore, this thesis states two hypotheses in order to fully answer the research question. These hypotheses are pieces of the whole research and the answers of these hypotheses will help answer the research question stated above. Firstly, it must be determined whether resources affect a country’s diversification. This must be known to be able to connect resources with the diversification of a country. In that way, it is examined whether countries with many resources also experience a low diversification. This piece of knowledge is needed to continue with the second hypothesis and the research question. The following hypothesis is stated:
𝐻𝑦𝑝𝑜𝑡ℎ𝑒𝑠𝑖𝑠 1: 𝐴𝑛 𝑎𝑏𝑢𝑛𝑑𝑎𝑛𝑐𝑒 𝑜𝑓 𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑟𝑒𝑠𝑜𝑢𝑟𝑐𝑒𝑠 ℎ𝑎𝑠 𝑎 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒 𝑒𝑓𝑓𝑒𝑐𝑡 𝑜𝑛 𝑎 𝑐𝑜𝑢𝑛𝑡𝑟𝑦′𝑠
𝑖𝑛𝑑𝑒𝑥 𝑜𝑓 exp 𝑜𝑟𝑡 𝑑𝑖𝑣𝑒𝑟𝑠𝑖𝑓𝑖𝑐𝑎𝑡𝑖𝑜𝑛.
Next to the research question above, the effect of institutional quality must be specified. As was claimed above in both the literature review and the theoretical framework, institutional quality is important to determine whether a specific country is a growth winner or growth loser because of its natural resource abundance. Multiple papers state that institutional quality is the main reason for the ambiguity of the resource curse (Yang & Lam, 2008; Arezki & van der Ploeg, 2007; Mehlum et al., 2006). New evidence that connects the institutional quality with another measure of economic policies, such as the diversification of an economy, could be helpful for the already extensive collection on the resource curse. The following hypothesis is stated:
𝐻𝑦𝑝𝑜𝑡ℎ𝑒𝑠𝑖𝑠 2: 𝐴𝑛 𝑎𝑏𝑢𝑛𝑑𝑎𝑛𝑐𝑒 𝑜𝑓 𝑛𝑎𝑡𝑢𝑟𝑎𝑙 𝑟𝑒𝑠𝑜𝑢𝑟𝑐𝑒𝑠 𝑑𝑜𝑒𝑠 𝑛𝑜𝑡 ℎ𝑎𝑣𝑒 𝑎 𝑛𝑒𝑔𝑎𝑡𝑖𝑣𝑒 𝑒𝑓𝑓𝑒𝑐𝑡 𝑜𝑛 𝑑𝑖𝑣𝑒𝑟𝑠𝑖𝑓𝑖𝑐𝑎𝑡𝑖𝑜𝑛 𝑤ℎ𝑒𝑛 𝑖𝑛𝑠𝑡𝑖𝑡𝑢𝑡𝑖𝑜𝑛𝑎𝑙 𝑞𝑢𝑎𝑙𝑖𝑡𝑦 𝑖𝑠 𝑟𝑒𝑙𝑎𝑡𝑖𝑣𝑒𝑙𝑦 ℎ𝑖𝑔ℎ
The hypotheses are both answered by measuring multiple indicators, which is elaborated in the methods section. By answering both the hypotheses, a conclusion about the research question can also be
stated. Now that the hypotheses are both stated, this thesis will give an overview of what it exactly is discovering. This is because it remains quite complex to understand the exact remaining gap in the literature. Therefore, two comprehensive figures must give clarity about the gap. Below, one can see the literature that has been discovered so far.
Figure 1: Existing research on the resource curse
As can be seen, both diversification and institutional quality and their general effects on
prosperity have been researched. Also, it has been discussed how these factors could offset the effects of a resource abundance on prosperity. Besides these points, the assumption that resource abundance affects diversification has been discovered as well. Instead of researching these points again, this thesis will look at how institutional quality affects diversification and, in that way, make the offsetting effects of
institutional quality more transparent. This translates into the underlying graph, whereas the purple line represents the new research this thesis aims to do.
Figure 2: Existing research plus the gap this thesis will research
Now that this thesis has shown the gap and formulated both the research question and hypotheses, it will finish this theoretical framework with a case study about the current situation of Venezuela. It will, among other things, be discussed how the country is affected by the indicators that are researched by this thesis. In that way, the expectations of this theoretical framework can be fully completed.
B. Case Study: Venezuela
In mid-2014, the drop in oil prices began whereas oil went from approximately 110 to a low of approximately 26 (Baumeister & Kilian, 2016). Venezuela’s oil-reliant economy was in an acute financial crisis which is now characterized by hyperinflation and a whole humanitarian crisis as the quality of public health is declining and crime is skyrocketing (Halff et al., 2017). In the meantime, the government of Nicolas Maduro has refused to recognize the National Assembly which was elected in 2015 and won by the opposition by a majority of two-thirds. Instead, Maduro called for a new election on July 30, 2017, which was characterized by a historically low turnout as the opposition boycotted the vote. In this case study, it will be shown how this crisis could happen and what could have been done to prevent it from happening. In this way, this thesis provides a viable example to its argument.
To understand the current situation in Venezuela, it is important to look at the country from 1999 and on. From that time, Hugo Chávez was elected president and introduced Chavism, a left-wing political movement in Venezuela, named after himself. This was characterized by two major transformations (Dachevsky & Kornblihtt, 2017): Firstly, there was a change in the forms of tax collection based on a revaluation of the royalty revenue. Secondly, there was an expansion of social expenditures and the financing of investment, both directly done by the PDVSA, the state-owned oil company of Venezuela. With these procedures, Chávez achieved quasi-full control of the oil sector and with that the oil rents in order to redistribute more effectively. However, the problem was that this did not guarantee that the owner had the capacity to run this industry effectively. This is supported by Salas (2017) who states that the governments of Chávez and Maduro achieved a more distributed income from oil than previous governments but failed in running the industry efficiently and in diversifying the economy accordingly. This quasi-full control of the oil sector is an example of the lack of institutional quality; in 2013, Chavez’s self-appointed successor, Nicolas Maduro became president. According to Halff et al. (2017), he does not have the same credibility, communication skills, or support as Chavez. This partly caused corruption and inefficiency to damage the oil industry. These factors caused the costs of the oil industry in Venezuela to increase from 6 billion dollars in 2006 to 25 billion dollars in 2015 (Halff et al.). The country lacks both in financial and political capacity to structurally fix these problems, causing the country to be in a severe crisis. Riera (2018) also researches the Venezuelan crisis and Chavez’s role in this; she argues that the political establishment, which was based on bipartisanship, was dismantled with the rise of Chávez. And instead of allowing a participatory democracy to replace it, Chavez built their political organization over the already existing institutions. This led to a relatively bureaucratic system, inevitably creating a parallel state with the expected consequences: Patronage, cronyism, and corruption (Riera, 2018). She continues that there are multiple examples of bad government from, for example, a worldwide criticized exchange rate policy to a relatively high inflation rate. There was a substantial failure by the government to control basic economic variables, resulting in an unavoidable crisis. In recent years, things have turned even worse caused by a concentration of power (Ellner, 2017). Government corruption is ubiquitous and law enforcement is unable to stop crime.
Riera (2018) also states that the government of Venezuela has become an authoritarian regime, which is generally characterized by central power and limited political freedoms. The creation of an authoritarian regime would have been impossible without a military force. She states that this statement can be traced back to Chavez’s decision making: There are several examples of Hugo Chavez giving the military more power in exchange for loyalty towards the regime. For instance, Chavez allowed high-ranked officials to financially benefit from corruption and public resources. In these cases, retired and active members of the military forces control oil- and mining industries and even the food distribution services. Not only does this show the lack of institutional quality, it also explains why there was a massive failure of the financial situation of the country and why the oil sector failed to operate efficiently. Another illustration that shows both the relatively high power of the military and the weakness of the institutional quality is the “Plan Bolivar 2000”, which was a social intervention program. This program was carried out by the military and preluded a period where the military took over the processes in key decision-making and policy implementations. This was done by the argument of providing relief of the crisis the country has been facing in the 90s. Because of this, elected officials increasingly lost authority and resources to govern. When combining the literature section with this case study, it can be seen that there are several factors that influenced the processes in Venezuela. Venezuela has the fifth largest hydrocarbon reserves in the world (Mann et al., 2006), causing more than 95 percent of the total exports to consist of primary resources, making it relatively vulnerable to primary resource shocks. Next to this fact, the abundance of resources causes there to be many interested parties seeking resource rents as well. Not only elitist parties in the country itself, but also international companies trying to pursue their share on these resources. This, as was shown in the literature review, enhances corruption and malicious practices of actors that attempt to govern. Also, in the literature section, it was shown that a good institutional quality could counteract this corruption and rent-seeking. But, as is shown in the table below, the World Bank states that the Governance Indicators of Venezuela are relatively low. When these indicators are compared to the surrounding countries, the results are striking. For every governance indicator, Venezuela scored significantly lower: When looking at the percentile rank, the scores are almost always at least twice as low. See the figure below.
Figure 3: Governance Indicators (Percentile Rank 2016)
Source: World Bank, 2018
Next to the relatively low institutional quality in 2016, a clear trend of institutional quality can be seen in the last 20 years, which is illustrated in figure 4. In figure 4, the percentile rank of the governance indicators in Venezuela for the period 1996-2016is shown. See the figure below.
Figure 4: Governance indicators for Venezuela (Percentile Rank 1996-2016)
Source: World Bank, 2018
In the figure, a downward trend is clearly visible: Whereas the average of all the governance indicators was around 40 in 1998. In 2016, the same governance indicators have an average of approximately 20. However, until mid-2014, the problems of Venezuela were not shown physically. The gradual decrease in institutional quality was the only factor that forecasted the eventual problems that Venezuela is facing at this moment. When combining the literature review with these observations, the trend shows that Venezuela has become more vulnerable to, for example, a weak rule of law and corruption. Also, when comparing it to surrounding countries, Venezuela is the country that generally had
the lowest institutional quality of the continent. This means it had the lowest chance to oppose corruption and rent-seeking. These types of institutions also lured malicious types of entrepreneurs to the country, who had less incentive to structurally support economic growth and more incentive to loot and seek rent, obstructing economic growth. This made the whole economy of Venezuela more vulnerable for economic shocks, which was exactly what eventually happened concerning the oil prices.
Next to this, one would expect that because of the low institutional quality, the FDI is relatively low and more volatile as well. The World Bank shows that Venezuela had a volatile FDI fluctuating from -1.1 billion to 6 billion, meaning there were years where there was more divestment than investment as well. When comparing this to two South American countries that have a somewhat same or smaller population, namely Peru and Chile, it can be seen that the FDI is low: Peru has an FDI of almost 1 billion to 12 billion and Chile has an FDI of 2.5 billion to 30.5 billion. Besides this point, the FDI in Venezuela has not shown the growth as it did in the compared countries. In 2003, Peru and Venezuela had approximately the same FDI, but Peru’s FDI has shown much higher growth than Venezuela. Also, the FDI of Venezuela is much more volatile than the compared countries. According to the literature, both the relative low FDI and the volatility of Venezuela’s FDI would have a significant negative effect on the economic growth of the country. See the figure below.
Figure 5: FDI Flows in dollars from 2000 – 2016 for the countries Chile, Peru, and Venezuela
Source: World Bank, 2018
As the mentioned reasons were the catalyst of some of the major problems for the Venezuelan oil industry, it can be argued that Venezuela’s current crisis is mainly caused by its poor and neglected institutional quality over the years. If Venezuela had paid attention to its institutional quality and improved its institutions with the capital obtained from resources, the corruption and rent-seeking could have been reduced. Then, the ones trying to govern could have done this much more efficiently, which would counteract the practices of malicious entrepreneurs and of those that do not support structural economic growth. Then it would also have been easier to diversify the economy and implement policies or investments that support the country’s economic productivity and diversification. With these measures, fuel exports as a percentage of all exports would decrease and the shock of a reduced oil price would have
been survivable as the economy would have been diversified enough. However, the opposite happened: Before 2000, the average fuel exports as a percentage of all the exports was around 83 percent. After 2000, it has increased to an average of approximately 90 percent. Their dependency on fuels merely increased instead of the wished decrease. Even though Venezuela’s problem might not be directly caused by institutional quality, the connection between the institutional quality and the performance of a country’s economy seems clear. This relationship will be further examined in the results section.
The case of Venezuela is not on its own. There are many other examples of countries that have an export defined by a high percentage of natural resource but do not experience the same growth as other countries with fewer resources (e.g. Kuwait, Libya, & Saudi Arabia) (Sachs & Warner, 2001). According to Mu & Hu (2018), this is usually because of a lack of institutional quality. They state that policy reforms are needed to address and improve the institutional quality. An example that they bring is the fixed exchange rate policy which directly exposes the country to the volatility of oil prices. They argue that countries with differing institutional quality obtain differing levels of prosperity from resources. Therefore, improving the overall institutional quality should be a focus points for such countries (Mu & Hu, 2018).
C. Conclusion
Overall, the stated conclusion for an economy like Venezuela could be generalized to every country with a relatively low institutional quality and a resource abundance. It is expected that economies with a low institutional quality generally have less incentive to diversify their economy and take action to counter the possible risks that come with a resource abundance. As their institutional quality is often already relatively low when the resources are found, it is difficult to implement policies that would prevent rent-seeking and the looting of resources. Also, when a regime of such countries decides to loot the natural resources, the chances of a further decrease of the overall institutional quality increases. This eventually could cause a downward spiral; as lower institutional quality generally increases the looting again. It seems that institutional quality could be a starting point for every government. When this quality is decent, it influences many factors of a country. Multiple points were already mentioned in the literature review such as the attraction of FDI or the counteraction of corruption and related malicious practices. This thesis will examine whether export diversification is part of these influences and therefore part of a possible solution of the resource curse. In the next section, the method for this examination will be elaborated.
4) Methodology:
A. Data
This thesis uses a quantitative method. Because the abundance of natural resources and the effect on diversification will be tested, certain variables and instruments are needed to measure this. These differing measurements will be clarified in this section. Preferably, multiple measurements are used for these measurements in order to get more reliable results. First of all, to measure resource abundance, this thesis will use two variables; these are the differences between fuel production and fuel consumption per capita and the total reserves per capita. Also for the measures of dependency, two indicators are used; these are the diversification index and the share of fuel exports on total exports. These measures will be elaborated later in this section. Firstly, the differences between fuel production and fuel consumption will be elaborated. For this variable, data is obtained from the U.S. Energy Information Administration (EIA) of which this thesis uses two datasets: First, this thesis uses a dataset that contains the total primary energy consumption per country for the period 1980-2015. Second, it uses a similar dataset for the primary energy production per country for the same period. Both datasets are measured in barrels of oil equivalent. These datasets contain data for a total of 227 countries, including countries that no longer exist. To make this data more useful, as many missing gaps as possible have to be filled in. This means that, for example, former East and West Germany will be merged into one dataset and merged with Germany after 1989. In this way, the data will become more continuous, more complete and easier to read. These two datasets will be compared with each other by subtracting the production of each country by the consumption of the same country, resulting in the gap between production and consumption per country. By doing so, it can quickly be observed which countries produce more than they consume and thus have a surplus of primary fuels. This means that they export at least a part of their produced primary energy. After this, the population of the countries for the same period will be extracted from a dataset of the World Bank. This dataset covers 263 countries for the timeline of 1960 – 2016. By using this dataset, this thesis is able to get a country’s natural resources per capita. In that way, the relative abundance of resources of a country can be observed, which is needed for a relative and objective comparison between countries. Because the data of every country’s population and the data of a country’s fuel production and consumption does not match, some countries have to be dropped from the datasets. These are countries that are either part of another country, such as Saint-Helena, or states that are not recognized as countries by some datasets, such as Monaco and Liechtenstein, or countries that do not exist anymore, such as former Czechoslovakia. Next to that, there is a lot of data missing for certain countries in specific years. Because of this, a total of 48 countries are deleted from the dataset. This means the dataset contains a total of 179 countries. To summarize, this dataset measures, when all adjustments are made, the difference between a country’s fuel production and consumption per capita. It represents the first measure of a natural resource abundance in countries.
measured in barrels of oil equivalent. This dataset is obtained from the Oil & Gas Journal which has data that goes from 1800 until 2008. Though the data runs from 1800 and onwards, not that many countries started measuring their reserves yet at that time. Most of the countries started showing consistent yearly data at around 1980. Therefore, data is dropped such that the dataset of reserves also covers the timeline from 1980 until 2008. Again, the population of every country is used, in order to get the relative resource reserves per country. A number of countries are dropped to make it match with the other abundance measure. This is done to make the abundance measures as equal as possible. By implementing these adjustments, both the abundance measures and the observed countries are comparable making a correct analysis possible.
Now that data on abundance is collected, it is time to look at dependency. Namely, this thesis aims to design a model which shows when a resource-abundant country actually becomes dependent on these resources. As was already mentioned, there are also two indicators of dependency. The first indicator is the share of primary fuel exports as a percentage of the total exports, obtained from the World Bank. This dataset covers 257 countries, including countries that do not exist anymore. It has a timeline from 1960 until 2010, but most countries show consistent data from 1975 and on. The dataset measures, per country, how big the share of fuels is as a share of the total exports. Because this does not fully cover all the possibilities – a country can have a big share of fuels as exports, but can still be diversified in the rest of its exports, causing the overall economy to be somewhat diversified – this thesis will add an overall export diversification index, which measures the overall diversification of the exports. This index is obtained from the IMF, which created this list covering 187 countries ranging from 1962 to 2010. This index varies between approximately 0.00 and 6.50, with 0.00 implying a full diversification and 6.50 implying that the exports are concentrated in one good solely. To make data on diversification better readable and understandable, the data will be inverted and a percentile rank will be made. This means that the new index goes from approximately 0 towards 100, with 0 equaling a concentration on one good solely and 100 equaling a perfectly diversified economy. Both the indicators, that are the share of primary fuel exports and the diversification index, are used in order to make inclusive arguments that regard multiple data sources.
To proceed and make conclusions about how institutional quality influences these matters, a proxy for institutional quality is needed. This is obtained via an index made by the World Bank. The index measures institutional quality along six different dimensions: Voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law, and finally control of corruption. They aggregate from around -2.5, which is a weak performance, to around 2.5 which is a strong performance. Again, this index is not clearly legible and interpretable. Therefore, another percentile rank will be created, now for institutional quality. 0 equals a weak institutional quality and 100 equals a strong performance. Then, a mean will be created of all the institutional indicators which will serve as the indicator for general institutional quality. This thesis will test this mean and several indicators of this index on the indicators for dependency. So, for example, whether the institutional quality would increase the
diversification of a country. In that way, it can be seen whether the institutional quality would influence these factors and thus also influences the phenomenon when a resource-rich country becomes dependent on their resources. Having summarized all the relevant datasets that will be used, one can find a table below with all the different indicators that are used. This table is made in order to give the reader a clear overview of the data being used. As can be seen, different timelines and numbers of countries are obtained. However, to keep the consistency in this research, one number of countries will be chosen, corresponding with one timeline. These variables are set on 179 countries and a timeline ranging from 1980 to 2015.
Table 1: Datasets that are used in this thesis Number of countries
Indicator observed Timeline Source
Fuel Production per 227* 1980 – 2015 IEA
country
Fuel Consumption 227* 1980 – 2015 IEA
per country
National fuel 167 1800 – 2008 Oil & Gas
reserves Journal
Population per 263* 1960 – 2016 World
country Bank
Fuels as share of 257* 1962 – 2010 World
export Bank
Diversification index 182 1962 – 2010 IMF
Institutional Quality 213* 1996 – 2016 World
indicators Bank
Overall dataset 179 1980 – 2015**
* These numbers include countries that do not exist anymore. (e.g. East-Germany)
** Not every dataset covers this full timeline
B. Estimation Technique
After these proceedings, the abundance, dependency and institutional quality will be measured in several forms. These tests will be done on a major number of countries in order to correct for certain econometric biases. Examples of these are selection biases and omitted variable biases. The dataset covers yearly data, meaning the research will be a variant of time-series research. For this, a panel data fixed effects model will be used to estimate the effects. There are multiple advantages to this method: First of all, by controlling for fixed effects, time trends that influence the analysis will be regarded for. Examples of these are interest rates and general growth, making countries more prosperous over time. These effects would normally make the estimates on diversification biased as the analysis would assign these effects to,
for example, institutional quality. Second, every observation that is used is regarded, even if some data is missing. This thesis will test multiple indicators on the dependent variable, which is either the diversification index or the share of fuels on total exports. For example, in the first regression, it will test whether the average resources per capita, which is the independent variable, has a significant estimated effect on the diversification index, the independent variable.
When one is looking at the diversification of a country, there are multiple variables that influence this dependent variable. For example, the presence of a high-skilled labor force influences the demand for other industries such as the technology sector. This would cause a country to diversify relatively more, causing a higher diversification. These kinds of factors will be translated into the error term of the regressions, as it is not controlled for. Because of this, the model suffers from omitted variable bias, which could cause the OLS estimator to be biased and inconsistent. However, there are hidden factors that could incentivize a country to diversify more and hidden factors that could incentivize a country to diversify less. Because of that, it could also be assumed that the sum of these factors is near zero, making the estimates relatively consistent. However, overall, one should consider the endogeneity problems that come with such a research. As there is not enough data available and as it is beyond the scope of this thesis, endogeneity may be present.
This thesis expects that the effect of resource abundance on resource dependency is different for different values of institutional quality. To measure this hypothesis, this thesis will utilize an interaction variable between institutional quality and resource abundance. In that way, this interaction variable could indicate whether the effect is different for different values of institutional quality.
Next to the general analyses, multiple variations will be done on these datasets. For example, the same regressions will be done per continent to look whether there are differences. Also, this thesis will look at countries with low, that is lower than 40 percent on the percentile rank, or high, which is translated into more than 60 percent on the percentile rank, levels of institutional quality. In that way, it is shown how a relatively low institutional quality would differ the effects of a resource abundance on a resource dependency.
C. Descriptive Statistics
In table 2, the mean, standard deviation, minimum and maximum value, and the total number of observations is shown per category. For every indicator that is used in this thesis, the data is shown. This is done to get a more transparent insight of the data that is being used. As was said earlier, the diversification index and the government indicators have been reshaped to make them better readable. As one can see, the diversification ranges from 1.37 to 82.49. This is because an economy, that is dependent on one good or perfectly diversified, does not exist. It can also be seen that the institutional ranks have some negative outliers. These numbers merely represent one or two countries in a few years. It is expected that this has no real effect on the results and is therefore maintained.
