Can oil-rich countries encourage entrepreneurship?

Abstract

This study provides the first empirical investigation to test one of transmission channels of resource curse, i.e. marginalized entrepreneurship activities. Our panel data analysis of 65 countries from 2004 to 2011 shows a negative and statistically significant association between oil rents dependency and entrepreneurship indicator. This finding is robust to control of other major drivers of entrepreneurship, unobservable country- and time-fixed effects and a different measurement of oil rents dependency. In addition, our main results show that government effectiveness among other dimensions of good governance has a statistically significant moderating effect in entrepreneurship–oil rents nexus.

Keywords::

• resource curse
• oil rents
• entrepreneurship
• governance
• business formation

Jel Classification::

• O13
• Q32
• M13

Acknowledgements

I would like to thank the Editor (Alistair R Anderson), two anonymous referees and participants at Doing Business Research Conference (Washington, DC, 2014) and International Conference and Project Meeting on Advanced Science and Technologies for Sustainable Development in Iran (STSD, Berlin, 2014) for useful comments and suggestions.

Notes

^1. The estimates of natural resources rents by the World Bank are calculated as the difference between the price of a commodity and the average cost of producing it. This is done by estimating the world price of units of specific commodities and subtracting estimates of average unit costs of extraction or harvesting costs (including a normal return on capital). These unit rents are then multiplied by the physical quantities countries extract or harvest to determine the rents for each commodity as a share of GDP. For example, oil rents are the difference between the value of crude oil production at world prices and total costs of production (http://www.data.worldbank.org/indicator/NY.GDP.PETR.RT.ZS).

^2. For more details, see http://www.finance.senate.gov/newsroom/chairman/release/?id = d9c1e0ca-6653-4312-b812-5870d6728926#_ftn1.

^3. Not all MENA countries are oil rich. On average, the share of oil rents in GDP from 2004 to 2012 in 5 from 19 MENA countries is zero. These countries are Israel, Jordan, Lebanon, Malta and Morocco. This ratio in Tunisia is 4%. However, in the remaining countries, oil rents have important weight in the economy (on average beyond 30%, reaching the maximum of 54% in Iraq and Libya). In oil-poor countries of the MENA region, however, we observe other sources of rents such as remittances (22% of GDP in Lebanon, 16% of GDP in Jordan and 7.5% of GDP in Morocco), and foreign aid. Future studies can shed more light on how such non-oil rents may shape the small business formation intensity.

^4. Kalantaridis and Fletcher (2012) explain the interaction between institutional change and entrepreneurship. Méndez-Picazo, Galindo-Martín, and Ribeiro-Soriano (2012) examine the empirical relationship between entrepreneurship and governance in a panel data analysis for 11 developed countries. Their results show a positive association between governance and entrepreneurship.

^5. Farzanegan (2009) suggests that about 90% of export smuggling in Iran involves oil-related products, explaining the role of energy subsidies which cause the large price disparity between Iran and her neighbours.

^6. Farzanegan (2013) explains how the economic and energy sanctions have affected the Iranian economy. Farzanegan and Raeisian Parvari (2014) also show that international oil markets are not affected significantly by Iranian oil sanctions.

^7. This is the sample in which we use oil rents.

^8. Note that our dependent variable focuses on business formation in the formal economy. Our interest in this study is also entrepreneurship in the formal economy. Thus our dependent variable may underestimate the total number of business formations by neglecting the informal businesses. More often measurement error in a dependent variable only weakens the model without introducing bias in either point or interval estimates (Baum 2006, 217).

^9. A similar approach to define entrepreneurship, i.e. private business formation rate can also be seen in Lee, Florida, and Acs 2004; Armington and Acs 2002; Reynolds, Storey, and Westhead 1994 among others.

^10. For more details, see http://www.doingbusiness.org/data/exploretopics/entrepreneurship.

^11. The natural resource variable covers oil, natural gas, coal, mineral and forest rents. Since oil rents play a significant role on the resource curse literature, we also focus more on this aspect of resource wealth.

^12.http://www.info.worldbank.org/governance/wgi/ge.pdf and Kaufmann, Kraay, and Mastruzzi (2010).

^13. The period of analysis is due to the availability of entrepreneurship data.

^14. We did also control for cost of doing business variables such as registration procedures, enforcement of contracts and startup costs. Due to their statistical insignificance we have not mentioned them in regressions.

^15. Alternatively we also use per capita oil rents (1 year lag) instead of oil rents as share of GDP to address the possible endogeneity of oil variable with respect to increasing private business formation.

^16. We have also estimated random effects instead of fixed effects. The results are similar to fixed effects. We use the Hausman test to find the most appropriate approach. The null hypothesis in Hausman test is that the coefficients estimated by the efficient random effects estimator are the same as those estimated by the consistent fixed effects estimator. In all models, we get a significant p-value (smaller than 0.05), providing more support for the use of fixed effects.

^17. Note that both oil and new business density are in logarithmic transformation.

^18. Results are available upon request.

^19. We have also tried an estimation in which all other dimensions of governance (political stability, regulatory quality, control of corruption, and voice and accountability) are controlled, besides government effectiveness (ge). We find that while ge and its interaction term with oil remain significant, none of the other dimensions of governance show a statistically significant effect on business formation.