Economic growth, economic freedom, and the resource curse.

Author:Campbell, Noel D.
  1. Introduction

    Many studies have demonstrated the negative relationship between natural resource abundance and economic growth. Countries with a high ratio of natural resource exports to GDP have experienced slow growth, even when controlling for other variables that may affect economic growth (Sachs and Warner, 1997; 1999). Countries with a large share of natural capital as part of their total capital (physical, human, and natural capital) have experienced a low economic growth rate (Gylfason and Zoega, 2006). Even at the state level in the United States, there seems to be a resource curse, as those states that had a high share of their output from the primary sector had a low economic growth rate (Papyrakis and Gerlagh, 2007). Countries that export energy, or rely less on imported energy than other countries, have experienced a low growth rate (shown later in Table 1A). This study tests whether this "resource curse" occurs from natural resources' negative effect on economic freedom.

    A resource curse may appear for several reasons. As mentioned in Sachs (1997), those that are endowed with natural resources enjoy easy riches, which may lead to laziness and sloth, whereas those who lack natural resources may be productive by necessity. Natural resource abundance may also lead to what is referred to as the "Dutch Disease," in which a discovery of natural resources (or natural resource boom) negatively affects manufacturing exports via the real exchange rate and loss of labor in the manufacturing sector (Corden and Neary, 1982). If the manufacturing sector is characterized by learning-by-doing, then a country that has a comparative advantage in the natural resource or agricultural sector may not grow as fast as the industrialized nations (Krugman, 1987; Matsuyama, 1992). If the manufacturing sector is characterized by increasing returns to scale with low returns at low levels of investment, a poor country or state may choose to produce primary products because they may not have enough capital or access to enough demand to make investment in the manufacturing sector worthwhile.

    Alternatively, the resource curse may work through a political channel, which is the focus of this paper. Tornell and Lane (1999) describe how political groups may create distortionary redistributive activity after a positive price shock. Baland and Francois (2000) describe how a resource boom may increase rent-seeking activities and lower entrepreneurship. Similarly, Torvik (2002) describes how entrepreneurs may engage in rent seeking instead of running productive firms. Political leaders can also be short sighted and over-extract the resources (Robinson, Torvik, and Verdier, 2006). Butkiewicz and Yankkaya (2010) show that developing countries suffer from a mineral resource curse, with evidence that the curse occurs because of rent seeking or weak institutions. Gylfason (2004) shows, among other things, that natural resource "intensity" is positively related to political corruption. Isham et al. (2005) show that those countries exporting "point-source" natural resources have poor institutions.

    Collier (2010) describes how natural resources may affect the political structure of a country, and vice versa. For example, natural resource abundance may lead to a less accountable government than a natural resource-scarce government that depends on tax revenue, which can lead to corruption and other adverse consequences. However, poor governance and property rights can lead to violence and rapid depletion of a resource. Haber and Menaldo (2010) point out that rulers who inherited weak institutions typically have pressing fiscal needs and short time horizons, which encourages them to extract resources at high rates today instead of saving them for tomorrow.

    Sometimes the effect of natural resources on government policy can be extreme. As reported by Human Rights Watch (2009), in Zimbabwe

    police coerced local miners to join syndicates that would provide the police with revenue from the sales of diamonds that the miners found. In seeking to end illegal mining and maintain control of the fields, police engaged in killings, torture, beatings, and harassment of local miners in Marange, particularly when police "reaction teams" carried out raids to drive local miners from the diamond fields. In 1980s Sudan, to clear out farms for oil production, the government sent armed militias to "loot cattle and burn, and to kill, injure, and capture Nuer and Dinka, whose men resisted on foot, mostly with spears" (Human Rights Watch, 2003, p. 51). Other explicit, but less extreme, examples of natural resource abundance involve nationalization, taxation, corruption, regulations, and trade protectionism. Once a country becomes dependent on a natural resource, such as Venezuela's dependence on its oil revenues, the dependence can lead to nationalization of the resource and then subsequently lead to additional government intervention in the economy to control and manage the resource.

    The question of whether natural resources affect institutions is similar to the question of whether foreign aid affects institutions. After all, a discovery of a natural resource is akin to receiving a financial gift. In a study by Djankov, Montalvo, and Reynal-Querol (2008), foreign aid led to a decrease in democracy. One distinction, however, is that foreign aid may be given to influence the political structure of a country, whereas a discovery of a natural resource is more of an exogenous event. Therefore, the question in our paper is similar to but distinctly different than the question of whether foreign aid affects institutions.

    Because natural resource abundance may cause government to intervene in its economy in many ways, an appropriate measure to test for a political channel for the resource curse must be a comprehensive measure of government activity. Examples of such broad measures of government intervention in an economy are the economic freedom indices of the world. As a measure of economic freedom, we use the Economic Freedom of the World (EFW) index by the Fraser Institute (Gwartney and Lawson, 2009). (1) The political and institutional difficulties that resource abundance spawns (as discussed above) will be reflected in a country's EFW value. The EFW consists of 42 third-party indicator variables grouped into five broad areas: size of government (expenditures, taxes, and enterprises); legal structure and security of property rights; access to sound money; freedom to trade internationally; and regulation of credit, labor, and business. The index ranges from 0 to 10, with 10 indicating that a country most closely approximates the free market ideal of various institutional measures. For example, whenever a government engages in violence, nationalization, protectionism, or other types of intervention in the economy, it is restricting economic freedom, and that country's EFW value will decline. This paper primarily uses the EFW summary index, a comprehensive index that averages all of those institutional components.

    Economic freedom has been shown to be positively related to economic growth. Numerous studies have used the EFW, and they have generally found a positive relationship between economic freedom and economic growth. De Haan, Lundstrom, and Sturm (2006) surveyed the literature that used the Fraser Institute's measure of economic freedom. None of the reports in their summary found that economic freedom is bad for growth. They do find some discrepancies as to whether both the change in economic freedom and the level of economic freedom are good for growth. For instance, Dawson (1998) finds that both the level and change of economic freedom are significantly correlated with economic growth, whereas De Haan and Sturm (2000) find that only the change in economic freedom is significantly correlated with economic growth. Doucouliagos and Ulubasoglu (2006) performed a meta-analysis in which they examined dozens of studies of the relationship between economic freedom and growth. Their analysis showed that there is a positive and robust relationship between economic freedom and economic growth, regardless of the sample of countries or measure of economic freedom. They also showed that economic freedom has a positive impact on physical capital formation, which generates economic growth.

    One difficulty in analyzing the relationship between economic growth and economic freedom is endogeneity. It is quite possible that as an economy grows, people put more value on economic freedom. Although econometric studies cannot prove that economic freedom causes growth, using Granger causality tests, a few studies have demonstrated that overall economic freedom precedes growth. Dawson (2003) shows that there is evidence that the overall level of freedom causes economic growth, but the changes in freedom appear to be determined jointly with economic growth. Looking at the components of the EFW index, Dawson gets mixed results. For example, property rights appear to cause economic growth, but government size appears to be caused by growth. Heckelman (2000) uses the Heritage Foundation's economic freedom index and finds through Granger causation tests that economic freedom precedes economic growth. One exception found by Heckelman is government intervention (government consumption and ownership), which may come after economic growth, which is consistent with the view of Wagner's law.

    This paper extends the literature that seeks to explain the empirically observed energy resource curse through institutional or...

To continue reading