Arellano, Manuel, and Olympia Bover. 1995. “Another Look at the Instrumental Variable Estimation of Error Component Models”, Journal of Econometrics 68, 29–51.
Auty, Richard M. 1997. “Natural Resources, the State and Development Strategy”, Journal of International Development, 9: 651–63.
Collier, Paul, and Anke Hoeffler. 2005. “Resource Rents, Governance, and Conflict”, The Journal of Conflict Resolution, Vol. 49, No. 4.
Feiss, Norbert M., and Dorte Verner. 2003. “Oil, Agriculture, and the Public Sector: Linking Intersector Dynamics in Ecuador“, World Bank Policy Research Paper 3094.
Iimi, Atsushi. 2007. “Escaping from the Resource Curse: Evidence from Botswana and the Rest of the World”, IMF Staff Papers, Vol. 54, No. 4.
Kauffmann, Daniel, Aart Kraay, and Massimo Mastruzzi. 2008. “Governance Matters VII: Aggregate and Individual Governance Indicators 1996–2007“, World Bank Policy Research Working Paper 4654.
Love, Inessa, and Lea Zicchino. 2006. “Financial Development and Dynamic Investment Behavior: evidence from Panel VAR”, The Quarterly Review of Economics and Finance, 46, 190–210.
Murshed, Mansoob. 2004. “When Does Natural Resource Abundance lead to a Resource Curse?” Environmental Economics Programme, Discussion Paper 04–01.
Ossowski, Ronaldo, Mauricio Villafuerte, Paulo Medas, and Theo Thomas. 2008. “Managing the Oil Revenue Boom: The Role of Fiscal Institutions”, IMF Occasional Paper 260.
Sala-i-Martin, Xavier, and Arvind Subramanian. 2003. “Ad dressing the Natural Resource Curse: An Illustration from Nigeria”, IMF Working Paper 03/139.
The author thanks Inessa Love and Lea Zicchino for the use of their panel VAR program.
See for instance, Sachs and Warner (1995), Ross (1999), Sala-i-Martin and Subramanian (2003), Murshed (2004), Iimi (2007), Leite and Weidman (1999), Isham et al (2004) and Bhattacharya and Ghura (2006).
One of the most comprehensive studies is the one by Sachs and Warner (1995). By examining 97 countries between 1971 and 1989, they show that states with a high ratio of natural resources exports to GDP in 1971 had abnormally slow growth rates even after controlling for initial per capita income, trade policy, investment rates, region, bureaucratic efficiency, terms of trade volatility and income distribution.
The Dutch disease refers to a situation where the non-natural resource tradable sector is being crowded out by a real exchange rate appreciation and/or by the resource pull effect (factor remuneration in the booming natural resource sector lure workers and capital away from the other sectors.
Manzano and Rigobon (2001) show that the 1980s debt crisis triggered by a significant reduction in commodity prices, can explain a large part of the negative effect of resource abundance on economic growth.
The Rentier Effect refers to the adverse impact of resource abundance on institutions’ quality. More specifically, it refers to a situation where because of their high earnings from natural resources, resource dependent countries have less need for tax revenues and are therefore relatively relieved of accountability pressures. Additionally, with the windfall of revenues, the governments can mollify dissents through a variety of mechanisms, including buying off critics, providing the population with benefits, infrastructure project while having the resources to pursue direct repression and violence against dissenters (Isham et al, 2004).
In a theoretical model, Leite and Weidman (1999) demonstrate that the opportunity costs of corruption are higher in labor-intensive industries rather than in capital-intensive sectors, suggesting that corruption is more likely in capital-intensive sectors such as the oil sector.
The Delayed Modernization and Entrenched Inequality effects refer to a situation where an elite that controls natural resources would resist industrialization or reforms that would diversify the economy because it fears could create several alternative sources of power that would compete over the natural resource revenues.
Auty (1997) argues that the type of the natural resources is what matters for growth. In his view, there is a greater chance of a vicious cycle of mismanagement, rent-seeking and conflict in countries, in which resources are concentrated and hence can be more easily expropriated (such as oil and minerals and unlike agriculture).
In their view, the link of the oil sector with the public sector indicates the capitalization of the public sector through rent from nationalized oil production while the link with the transportation and communication stems from subsidized combustibles.
The sample includes 23 oil exporting countries (Algeria, Angola, Azerbaijan, Bahrain, Cameroon, Chad, Congo, Equatorial Guinea, Gabon, Indonesia, Iran, Kazakhstan, Kuwait, Libya, Nigeria, Oman, Qatar, Saudi Arabia, Syria, Turkmenistan, United Arab Emirates, Venezuela and Yemen).
Oil intensity is calculated as the average share of oil GDP in total GDP (constant prices) over a period 1985–2008.
Low oil-intensity countries are defined as countries with an average (over the sample) weight of oil GDP in total GDP (in constant prices) below the sample’s median observation (11 countries), while countries with an average weight of oil GDP in total GDP above the sample’s median observation are defined as high oil-intensity countries (12 countries).
The indicator is based on 2007 data. It is measured in units ranging from about -2.5 to 2.5, with higher values corresponding to better government effectiveness.
The OBI assigns a score to each country based on the information it makes available to the public throughout the budget process. See http://www.openbudgetindex.org/index.cfm?fa=rankings.
Based on a cross-country analysis Iimi (2007) concludes that the degree of which natural resources affect growth depends on the level of governance.
It is reasonable to believe that in countries with high level of development, the institutional quality is, on average, higher than in less developed countries, see Isham (2004). The level of development is proxied by the income per capita in PPP terms.
According to the definition of the April 2009 World Economic Outlook (WEO). In addition, the sample includes Indonesia, which was an OPEC member until 2008, Chad and Cameroon. Advanced oil exporters such as Norway, Mexico and Russia were excluded from the sample as they are at different stage of development and policy implications may not be valid for them.
One of the main caveats in this approach is that it assumes that the country’s special characteristics are fixed over time.
This transformation preserves the orthogonality between the transformed variables and lagged regressors. The estimation uses lagged regressors as instruments and estimate the coefficient by GMM methodology.
Monte Carlo simulations are used to generate the confidence intervals.
The lag length was selected by using Akaike Information Criterion (AIC). Given the limited observations and the fact that data is annually, one, two and three lags were considered. The AIC results show that in the Low group, one-lag specification is slightly superior compared to other specifications while, in the High group, the three-lag specification has a significant lower AIC value compared to other specifications.
Due to lack of data, the analysis is focused on a shorter period (1992-2008).
Exchange rate data are available only from 1992.
In addition to the endogenous variables, the estimation also includes two dummies: PEACE for the postconflict period of 2002-08, which is characterized with greater macroeconomic stability; and Dum_94 to capture the substantial devaluation of the exchange rate in 1994.
Recent assessment of the real effective exchange rate (by CGER methodology) shows that the real exchange rate is broadly in line with its fundamentals.