The Effects of Majority State Ownership of Significant Economic Sectors on Corruption: A Cross-Regional Comparison
2008; Taylor & Francis; Volume: 34; Issue: 1 Linguagem: Inglês
10.1080/03050620701883579
ISSN1547-7444
Autores Tópico(s)Culture, Economy, and Development Studies
ResumoAbstract Given both corruption's and bureaucratic inefficiency's importance for development and good governance, understanding their causes is paramount. This paper argues that majority state ownership of most the most important economic sectors of a country results in higher levels of corruption and inefficiency. When political and managerial elites both own and manage the country's most important economic resources, they have greater incentives for corrupt or inefficient behavior. These elites use national resources at their disposal more for short-term personal and political goals than for long-term economic ones. This paper tests this hypothesis on a relatively underused, but often cited, data set from the 1980s. Using a cross-national, regression analysis, this paper finds that the best predictors a country's level of corruption or bureaucratic inefficiency are these: majority state ownership of significant economic sectors, levels of GDP per capita, levels of government spending, and levels of democracy. Other factors, such as common law heritage, percent of population that is Protestant, federalism, economic freedoms, or mineral/ oil exporting, were not consistent, significant predictors of either bureaucratic inefficiency or corruption. We also argue that Tobit may be the best estimation procedure for these data. KEYWORDS: corruptionbureaucratic efficiencymajority state ownershipinternational political economycross national regression modelTobitOLS1 Notes 1. I would like to thank Dennis P. Quinn, Jonathon W. Moses, Charles Murray*, and John Heilbrunn, as well as several anonymous reviewers, for their helpful comments and critiques. As always, any omissions or errors in the manuscript are the responsibility of the author. *Not the author of The Bell Curve. 2. This article does not attempt to give a comprehensive overview of work on corruption — whether definitions, causes, or consequences. However, each is addressed enough to gather variables and sufficient background for its work of an empirical test. It is intended as a test of general arguments on a little used, but promising, data set and new variable. However, do see citations for recent scholarship. Also, see Bibliography for some additional references. 3. See at http://wbln0018.worldbank.org/Network/PREM/PREMDocLib.nsf/58292AB451257BB9852566B4006EA0C8/3CFDF5E63560E5DD85256713000456B1/$FILE/annotate.htm 4. In fact, this lends support to the use his data in this analysis and an external prop for its validity. 5. Transparency International at www.transparency.de/ or at www.gwdg.de/∼uwvw/icr.htm 6. However, Treisman (2000) Treisman, Daniel. 2000. The Causes of Corruption: A Cross-National Study. Journal of Public Economies, 76(3) June: 399–457. [Crossref], [Web of Science ®] , [Google Scholar] does use Mauro's data as a control. 7. For problems with defining corruption, see Johnson (2004) Johnson, Michael, ed. August 2004. Crime, Law, and Social Change Vol. 42, August, No. 1 [Google Scholar]; Friedrich (1972) Friedrich, Carl J. 1972. The Pathologies of Politics: Violence, Betrayal Corruption, Secrecy and Propaganda, 15–24. New York: Harper & Row. Reprinted as “Corruption in Historical Perspective.” In A.J. Heidenheimer and M. Johnson, eds. (2002). Political Corruption: Concepts and Contests, New Brunswick [Google Scholar]; Gardiner (1993) Gardiner, John A. 1993. “Defining Corruption”. In Coping with Corruption in a Borderless World: Proceedings of the Fifth International Anti-Corruption Conference, Edited by: Punch, Maurice, Kolthoff, Emile, van der Vijver, Kees and van Vlit, Bram. 25–40. Deventer and Boston: Kluwer Law and Taxations Publishers. Reprinteo in A.J. Heidenheimer and M. Johnson, eds. (2002). Political Corruption: Concepts and Contests. New Brunswick [Google Scholar]; Philp (2002) Philp, Mark. 2002. “Conceptualizing Corruption”. In Political Corruption: Concepts and Contests, Edited by: Heidenheimer, A.J. and Johnson, M. New Brunswick: pp. 41–58. [Google Scholar]; and Leys (1965) Leys, Colin (1965). “What Is the Problem about Corruption?” Journal of Modern African Studies, Vol. 3, No. 2, pp. 215–230; Reprint in A.J. Heidenheimer and M. Johnson, eds. (2002). Political Corruption: Concepts and Contests, New Brunswick, pp. 59–73. “Library of Congress Country Studies Website.” S.V. various countries http://lcweb2.loc.gov/frd/cs/cshome.html [Google Scholar]. For other definitions of corruption, see A.J. Heidenheimer and M. Johnson, eds. (2002) Heidenheimer, A.J. and Johnson, M., eds. 2002. Political Corruption: Concepts and Contests, New Brunswick: Transaction Publishers. [Google Scholar]; Bardham (1997) Bardham, P. 1997. Corruption and Political Development: A Review of Issues. Journal of Economic Literature, 35(3): 1320–1346. [Google Scholar]; Nye (1967) Nye, J.S. 1967. Corruption and Political Development: A Cost-Benefit Analysis. The American Political Science Review, 61(2) June: 417–427. [Crossref], [Web of Science ®] , [Google Scholar]; Klitgaard (1988) Klitgaard, Robert. 1988. Controlling Corruption, Berkeley and Los Angeles: University of California Press. [Crossref] , [Google Scholar]; Treisman (2000) Treisman, Daniel. 2000. The Causes of Corruption: A Cross-National Study. Journal of Public Economies, 76(3) June: 399–457. [Crossref], [Web of Science ®] , [Google Scholar]. 8. For corruption as deviation from original form, see Genaux (2004 Genaux, Maryvonne. 2004. Social Science and the Evolving Concept of Corruption. Crime, Law, and Social Change, 42(1) August: 13–24. [Crossref] , [Google Scholar]) and Philp (2002) Philp, Mark. 2002. “Conceptualizing Corruption”. In Political Corruption: Concepts and Contests, Edited by: Heidenheimer, A.J. and Johnson, M. New Brunswick: pp. 41–58. [Google Scholar]. For ties to God and justice, see Genaux (2004) Genaux, Maryvonne. 2004. Social Science and the Evolving Concept of Corruption. Crime, Law, and Social Change, 42(1) August: 13–24. [Crossref] , [Google Scholar]; for view of ties to natural state à la Aristotle, see von Alemann (2004) von Alemann, Ulrich. 2004. The Unknown Depths of Political Theory: The Case for a Multidimensional Concept of Corruption. Crime, Law, and Social Change, 42(1) August: 25–34. [Crossref], [Web of Science ®] , [Google Scholar]. 9. See Gardner (2002) for this tricotomy. Also, Heidenheimer and Johnson (2002) Heidenheimer, A.J. and Johnson, M., eds. 2002. Political Corruption: Concepts and Contests, New Brunswick: Transaction Publishers. [Google Scholar], introduction. 10. Expectations upon office holders can change, especially over time, but from country to country as well. Also, office holders that are political appointees are often seen in a different light than ones who obtained their office through merit. For problems of establishing international norms, see Heidenheimer and Johnson (2002) Heidenheimer, A.J. and Johnson, M., eds. 2002. Political Corruption: Concepts and Contests, New Brunswick: Transaction Publishers. [Google Scholar], pp. 12–13; and Leys (1965) Leys, Colin (1965). “What Is the Problem about Corruption?” Journal of Modern African Studies, Vol. 3, No. 2, pp. 215–230; Reprint in A.J. Heidenheimer and M. Johnson, eds. (2002). Political Corruption: Concepts and Contests, New Brunswick, pp. 59–73. “Library of Congress Country Studies Website.” S.V. various countries http://lcweb2.loc.gov/frd/cs/cshome.html [Google Scholar]; Gardiner (1993) Gardiner, John A. 1993. “Defining Corruption”. In Coping with Corruption in a Borderless World: Proceedings of the Fifth International Anti-Corruption Conference, Edited by: Punch, Maurice, Kolthoff, Emile, van der Vijver, Kees and van Vlit, Bram. 25–40. Deventer and Boston: Kluwer Law and Taxations Publishers. Reprinteo in A.J. Heidenheimer and M. Johnson, eds. (2002). Political Corruption: Concepts and Contests. New Brunswick [Google Scholar]; and Scott (2002) Scott, James C. 2002. “Handling Historical Comparisons Cross-Nationally”. In Political Corruption: Concepts and Contexts, 3rd, Edited by: Heidenheimer, A.J. and Johnson, M. 123–138. New Brunswick, NJ: Transaction Publishers. [Google Scholar]. 11. And since we will use perception data, we can add as understood by international experts and businesspeople. 12. For several pieces (in French) on “Daily corruption” [La corruption au quotidian], see Politique Africaine No. 83 (Octobre 2001) 2001. La corruption au Quotidien. Politique Africaine, Octobre: 1–114. [Google Scholar], pp. 1–114. Most of the cases come from Benin, Niger, and Senegal. 13. See for example, Collier and Levitsky's (1997) Collier, David and Steven, Levitsky. 1997. Democracy with Adjectives: Conceptual Innovation in Comparative Research. World Politics, 49(3): 430–451. [Crossref], [Web of Science ®] , [Google Scholar] paper of democracies with adjectives. 14. See Johnson (2002) Heidenheimer, A.J. and Johnson, M., eds. 2002. “Measuring the New Corruption Rankings: Implications and Analysis and Reform”. In Political Corruption: Concepts and Contests, 865–884. New Brunswick: Transaction Publishers. (2002) [Google Scholar] for difficulties of measuring underground activities such as corruption. He also discusses issues of reliability and validity of such measures. 15. For description and links, go to: http://info.worldbank.org/governance/beeps/ 16. For a complete description of surveys commonly used to measure corruption, see Elliot (1997) Elliot, Kimberly Ann, ed. 1997. Corruption and the Global Economy, Washington, D.C.: Institute for International Economics. [Google Scholar], pp. 241–244. She also describes IRIS, ICRG, PERC, and others. Also, see http://www.transparency.org/cpi/2003/dnld/survey_sources2003.pdf 17. In 1995, these data had one sub-Saharan African country. In 1996, there were five; by 2001, they had 14; by 2004, there were around 30. 18. For more on corruption and its various measures, see Johnson (2002) Heidenheimer, A.J. and Johnson, M., eds. 2002. “Measuring the New Corruption Rankings: Implications and Analysis and Reform”. In Political Corruption: Concepts and Contests, 865–884. New Brunswick: Transaction Publishers. (2002) [Google Scholar] and Treisman (2000) Treisman, Daniel. 2000. The Causes of Corruption: A Cross-National Study. Journal of Public Economies, 76(3) June: 399–457. [Crossref], [Web of Science ®] , [Google Scholar]. Knack and Keefer's (1997) Knack, Steven and Keefer, Philip. 1997. Does Social Capital Have an Economic Payoff? A Cross-country Comparison. Quarterly Journal of Economics, 12(4): 1251–1288. [Crossref] , [Google Scholar] data would necessarily have majority state owned countries as riskier (and therefore more corrupt by definition) as most came into their ownership through appropriating foreign companies — which is not exactly corrupt from a nationalist perspective. 19. See http://www1.worldbank.org/publicsector/indicators.htm 20. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=405841#PaperDownload 21. Business International is now incorporated into the Economist Intelligence Unit. For more, see Mauro (1995) Mauro, Paolo. 1995. Corruption and Growth. The Quarterly Journal of Economics, August: 681–712. [Crossref], [Web of Science ®] , [Google Scholar]. 22. For other works on state ownership or its effects, see Jones (1982) Jones, Leroy P., ed. 1982. Public Enterprise in Less-Developed Countries, Cambridge: Cambridge University Press. [Crossref] , [Google Scholar], Nellis (1986) Nellis, John R. Public Enterprise in Sub-Saharan Africa. World Bank Discussion Paper No. 1. 1986. Washington, DC [Google Scholar], Shirley (1983) Shirley, Mary M. Managing State-Owned Enterprises. World Bank Staff Working Paper No. 577. 1983. Washington, DC [Google Scholar], World Bank (1989 World Bank. 1989. Sub-Saharan Africa: From Crisis to Sustainable Development, Washington, D.C.: World Bank. [Google Scholar], 1995) World Bank. 1995. Bureaucrats in Business: The Economics and Politics of Government Ownership, Washington, D.C.: World Bank. [Google Scholar]; Grosh and Mukandala (1994) Grosh, Barbara and Mukandala, Rwekaza S., eds. 1994. State Owned Enterprises in Africa, Boulder, CO: Lynne Rienner Publishers. [Google Scholar]; Tangri (1999) Tangri, Roger. 1999. The Politics of Patronage in Africa: Parastatals, Privatization, and Private Enterprise, Trenton, NJ: Africa World Press. [Google Scholar], and recently Jones Luong and Weinthal (2004) Jones Luong, Pauline and Weinthal, Erika. Combating the Resource Curse: An Alternative Solution to Managing Mineral Wealth. Paper for Annual Meeting of the American Political Science Association. September2–52004, Chicago, IL. [Google Scholar]. Also, government run marketing boards are not included as a litmus test of majority state ownership. 23. For logic or causes of such nationalizations see Rood (1976) Rood, Leslie L. 1976. Nationalisation and Indigenisation in Africa. Journal of Modern African Studies, 14(2): 427–447. [Crossref] , [Google Scholar]; Kennedy (1988) Kennedy, Paul. 1988. African Capitalism: The Struggle for Ascendancy, Cambridge: Cambridge University Press. [Google Scholar], Grosh and Mukandala (1994) Grosh, Barbara and Mukandala, Rwekaza S., eds. 1994. State Owned Enterprises in Africa, Boulder, CO: Lynne Rienner Publishers. [Google Scholar], Tangri (1999) Tangri, Roger. 1999. The Politics of Patronage in Africa: Parastatals, Privatization, and Private Enterprise, Trenton, NJ: Africa World Press. [Google Scholar], and Quinn (1999 Quinn, John J. 1999. “The Managerial Bourgeoisie: Capital Accumulation, Development, and Democracy”. In Postimperialism and World Politics, Edited by: Becker, David G. and Sklar, Richard L. 219–252. Westport, CT: Praeger. [Google Scholar], 2002) Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar]. For global perspective, see Krasner (1978 Krasner, Steven D. 1978. Defending the National Interest: Raw Materials Investments and U.S. Foreign Policy, Princeton: Princeton University Press. [Google Scholar], 1985) Krasner, Steven D. 1985. Structural Conflict: The Third World Against Global Liberalism, Berkeley and Los Angeles: University of California Press. [Google Scholar]. 24. For another discussion of the principal-agent dynamic, see North (1981) North, Douglass C. 1981. Structure and Change in Economic History, New York: W. W. Norton & Co. [Google Scholar]; for its use with state owned enterprises, see Jones Luong and Weinthal (2004) Jones Luong, Pauline and Weinthal, Erika. Combating the Resource Curse: An Alternative Solution to Managing Mineral Wealth. Paper for Annual Meeting of the American Political Science Association. September2–52004, Chicago, IL. [Google Scholar]. 25. The World Bank has held that state owned companies pollute more than others (1995). 26. Philip P. Pan, “Whistleblower Sentenced to Life in Prison in China.” Washington Post Foreign Service, Thursday, November 10, 2005; 12:45 pm; http://www.washingtonpost.com/wp-dyn/content/article/2005/11/10/AR2005111000987.html 27. Quinn (1999 Quinn, John J. 1999. “The Managerial Bourgeoisie: Capital Accumulation, Development, and Democracy”. In Postimperialism and World Politics, Edited by: Becker, David G. and Sklar, Richard L. 219–252. Westport, CT: Praeger. [Google Scholar], 1993) Quinn, John J. 1993. The Impact of State Ownership of Resources on Economic and Political Development in Sub-Saharan Africa. Ufahamu, 21(1 & 2): 60–79. [Google Scholar] also held that in sub-Saharan Africa (from 1966–1986), these countries were also less likely to be democratic as would other similar ones. 28. In fact, many early nationalizations were carried out precisely as a means of gaining control over formerly foreign-owned enterprises. See endnote 22. 29. Others emerged as a result of the threat from the first two. 30. For notion of fragments of accountability, see Sklar (1987) Sklar, Richard L. 1987. Developmental Democracy. Comparative Studies in Society and History, 29(4) October: 686–714. [Crossref], [Web of Science ®] , [Google Scholar]. 31. Nonetheless, Heilbrunn did not distinguish between countries with or without majority state ownershipof industries or mining; he was comparing state owned sectors to nonstate owned sectors within countries. Moreover, in his 2005 piece, he examines corruption in France linked to its previously majority state owned oil sector, though he does not stress the element of ownership. 32. See, for example, Ross's review essay (1999 Ross, Michael. 1999. The Political Economy of the Resource Curse. World Politics, 51(2): 297–322. [Crossref], [Web of Science ®] , [Google Scholar]). 33. Although most of her cases actually included majority state owned oil industries. 34. They looked at policy and not corruption. Importantly, the government of Botswana does not own a majority share of this sector (Quinn, 1999 Quinn, John J. 1999. “The Managerial Bourgeoisie: Capital Accumulation, Development, and Democracy”. In Postimperialism and World Politics, Edited by: Becker, David G. and Sklar, Richard L. 219–252. Westport, CT: Praeger. [Google Scholar], 2000 Quinn, John J. 2000. Economic Accountability: Are Constraints on Economic Decision Making a Blessing or a Curse?. Scandinavian Journal of Development Alternatives and Area Studies, 19(4) December: 131–169. [Google Scholar], 2002 Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar]). 35. I did consider using settler mortality rates from Acemoglu, Johnson, and Robinson (2001) Acemoglu, Daron, Simon, Johnson and James, Robinson. 2001. The Colonial Origins of Comparative Development: An Empirical Investigation. The American Economic Review, 91(5) December: 136–1401. [Crossref] , [Google Scholar] but the N was too small and too many European nations did not have data on mortality rates. 36. Also, Quinn (2002) Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar] uses a tipping point definition which is easier to operationalize prior to many partial and half hearted attempts at privatization throughout the third world (see Van de Walle, 2001 van de Walle, Nicolas. 2001. African Economies and the Politics of Permanent Crisis: 1979–1999, Cambridge: Cambridge University Press. [Crossref] , [Google Scholar]). 37. Due to missing data, the N is lower in some models. 38. This does not “prove” that the measure is good, but that it meets the standard of internal or convergent validation. See Manheim and Rich (1995 Manheim, Jarol B. and Rich, Richard C. 1995. Empirical Political Analysis: Research Methods in Political Science, White Plains, NY: Longman. [Google Scholar], p. 75). 39. This inversion has no effects on the magnitudes of the coefficients or the tests of significance (other than on the intercept). It only reverses the signs and makes the results more intuitive for readers not versed with this measure. [Tests were run both ways for certainty.] I thank the anonymous reviewer for the suggestion of inverting the dependent variable. 40. See http://www.gwdg.de/∼uwvw/ for the countries and ratings. However, only 45 of the 54 countries were included due to missing data. This measure does have two more years averaged in, but the data are used as a control and are not the primary data for testing. For sources, see http://www.gwdg.de/∼uwvw/hist-sou.htm 41. For the purposes of this analysis, a country has significant state-ownership of the economy when the government owns more than 50% or more of the following: (1) most major industries— usually resulting from explicit programs of nationalization or state investment, or (2) the major export producing sector—usually entailing either minerals or petroleum. In this latter case, to be included by this definition, a state has to own over 50% of the entire sector and not merely between 50.1 and 100% of one competing company within the sector. Moreover, for majority state-ownership of resources to have inward-oriented effects, it must include the means of production as well as distribution. Therefore, the existence of state-owned and operated agricultural marketing boards does not constitute a sufficient condition for an economy to be defined as state-owned: the same inward-oriented policy regimes are not necessarily associated with these institutions. (See Quinn 1999 Quinn, John J. 1999. “The Managerial Bourgeoisie: Capital Accumulation, Development, and Democracy”. In Postimperialism and World Politics, Edited by: Becker, David G. and Sklar, Richard L. 219–252. Westport, CT: Praeger. [Google Scholar], 2002 Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar]) Although agricultural state marketing boards may increase corruption, they were not found to have an effect of inward or outward orientation per se. Rather, state-ownership of export producing sectors refers to cases where the state has rights over the commodity itself, before it is marketed. Also to be considered a major export sector, it has to routinely comprise more than 30–40% of its exports, in value. 42. The most important sources were these: “The Library of Congress Country Studies Website,” at http://lcweb2.loc.gov/frd/cs/cshome.html; Quinn (1999 Quinn, John J. 1999. “The Managerial Bourgeoisie: Capital Accumulation, Development, and Democracy”. In Postimperialism and World Politics, Edited by: Becker, David G. and Sklar, Richard L. 219–252. Westport, CT: Praeger. [Google Scholar], 2002) Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar]; Rood (1976) Rood, Leslie L. 1976. Nationalisation and Indigenisation in Africa. Journal of Modern African Studies, 14(2): 427–447. [Crossref] , [Google Scholar]; Africa South of the Sahara, S.V. (various years and countries); and Kennedy (1988) Kennedy, Paul. 1988. African Capitalism: The Struggle for Ascendancy, Cambridge: Cambridge University Press. [Google Scholar]; and Investment Promotion Network at Pananet at www.ipanet.com. One quality control check for this variable comes from Gwartney, Lawson, and Block (1996) Gwartney, James, Robert, Lawson and Walter, Block. 1996. Economic Freedom of the World 1975–1995, Canada: The Fraser Institute. [Google Scholar]. They created an index of economic freedom and have one sub-category called government-operated enterprises as a share of the economy. This index runs from 0 to 10. Where it registered a two or less, the dummy variable automatically became a 1 to indicate majority state ownership, regardless of other coding schemes. Where it was an 8 or higher, the dummy variable would become a zero indicating less than majority state ownership. Otherwise, majority state ownership of an exporting sector or the most capital-intensive industries was arrived at through content analysis and country specific references. 43. As per a reviewer's suggestion, I also ran an interactive variable between Quinn's and Gwartney's. However, the Variance Inflation Factor gave a number of 27, which is quite a bit higher than the threshold of 4 (square root of less than 2) [or even of 10 or 20 that some people use]. When it was included instead of the other two measures, the results were nearly indistinguishable from the other equations. None of the variables that were significant at 95% became so and none became less so. I also interacted Quinn's variable with per capita incomes, levels of democracy, and government spending and none were significant while leaving in the dummy variable. 44. Once again, the index in inverted to convey a clearer conceptual idea. A subtraction of 10 minus the index would create the inverse of 0 through 10. 45. Unfortunately, this lowers the n, as Gwartney did not code as many (or the same) countries. Also, an interaction variable was included with both multiplied one against the other, but with the first two included the VIF numbers indicated extreme multicollinearity and it was excluded. 46. http://pwt.econ.upenn.edu/php_site/pwt62/pwt62_form.php RGDP chain series in 2000 constant prices, 1975–1980. Models were also run with HDI measures as well and the MSO finding remained strong. Since more data was available with GDP, this measure was preferred. In earlier versions, the year 1965 was used, but upon advice from reviewers, a date closer to the corruption period was chosen. But still we have a small lag between incomes and corruption, though there is a one-year overlap. Also, following the advice of a reviewer, I added the percentage of the GDP which came from agriculture, industry, manufacturing, and service as other measures of development (though I could not have all three agriculture, industry, and service in the same model as it introduced severe multicollinearity (VIF = 600 or so). With real GDP still included, the others were not significant, though the VIF was between 5–8 for several variables. With GDP out, either service was negative and significant or agriculture was positive and significant—as one would expect. When industry as dropped, only percent agriculture was significant (with industry, the VIFs were too high, but all four became insignificant). Nonetheless, MSO remained significant each time. However, the N fell from 61 to 46 given limits on data for all these sectors. Data derived from World Development Report 1986, Table 3. 47. See http://www.cidcm.umd.edu/inscr/polity/ 48. The overall average from 1972 to 1983 was used. Two countries were averages of shorter periods: Guinea Bissau included years from 1974–1983, and Angola includes the years 1975–1983. 49. Most data are single years taken from between 1980–1983. Most data come from the World Bank, The World Development Report (WDR) for 1986 and 1989 (for Africa). Other data come from the 1993 WDR, which were averaged from 1970 and 1991 and assumed to be representative for the early 1980s. Finally, for Botswana and Angola, data were taken from Africa South of the Sahara, 1984/85. 50. Although Spain is not coded as Federal, the law which effectively brought it into existence was not signed until 1984. 51. Other data not included are these: ICPSR as this source was highly collinear with per capita GDP —the Variance Inflation Factor (VIF) was 6.588. It was also collinear with political and civil rights at VIF at 4.929. VIF was higher when all three measures were included (anything over four calls significanceinto question). Also, Mortality of settlers (for the limited data obtained) was also highly collinear with GDP per capita with VIF well above 4. If the percent of protestant was not included in the Almanac, then Treisman's data were used. Also earlier versions of the paper included an African dummy variable and a measure for ethnolinguistic fractionalization. Neither was significant and each was removed on advice of referees. 52. The earlier data were run with SPSS checking for multicolinearity (in the OLS estimation procedure), and then they were run with SST correction for heteroskedasticity for both OLS and Tobit. 53. The Tobit procedure estimates a “normal” error structure for the countries near and at the lower limit of zero. It assumed that the data is censured and projects a new distribution beyond this limit with “normal” error structures so that they are more similar to the rest of the distribution projecting past the censored or truncated data. See Kmenta, 1986 Kmenta, Jan. 1986. Elements of Econometrics, 2nd, New York: Macmillan Company. [Google Scholar], pp. 560–566. The data also appear to be interval, so a Probit estimation is not called for. The changing of the dependent variable into a z-score was recommended by an anonymous reviewer. 54. For model #8 it was significant only at 93.6%. 55. In earlier versions of this paper, Singapore was excluded as an outlier, and the level of democracy was significant for each estimation procedure. However, standards for excluding outliers can be controversial, so it is no longer excluded. 56. The Pearson's correlation is .32 by themselves, and .22 in the larger matrix. 57. Ross looked at government owned timber industries in Malaysia, the Philippines, and Indonesia. Karl's most important case was Venezuela—which had majority state ownership. So too did Iran, Nigeria, Norway, and Algeria. Indonesia also had majority state owned oil, but the actual production was carried out in joint ventures that were owned 50% private and 50% government. Then the distribution was handled through the state-owned parastatals. 58. We are not suggesting that privatization is as a magical bullet or panacea or that doing it well is easy, just that it suggests itself from the findings. It would lead to an improvement of about 6 to 20%. Also, this might only be in the medium to long term as Quinn (2002) Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar] using a decision model showed that under rational choice models for Africa, most holders of power would be against privatization for political reasons, though politicians could potentially rig the game in their favor. Also, given the way majority state ownership is measured, there is no empirical reason to see benefits privatizing much below 50%,—no empirical data here so indicates. 59. The idea of “agencies of restraint” is from Collier (1999) Collier, Paul. 1999. “Learning from Failure: The International Financial Agencies as Agencies of Restraint in Africa”. In The Self-Restraining State: Power and Accountability in New Democracies, Edited by: ”, In Andreas Schedler, Larry, Diamond and Marc, F. Plattner. 313–332. Boulder, CO: Lynne Rienner. [Crossref] , [Google Scholar]. Economic accountability is from Quinn (2000 Quinn, John J. 2000. Economic Accountability: Are Constraints on Economic Decision Making a Blessing or a Curse?. Scandinavian Journal of Development Alternatives and Area Studies, 19(4) December: 131–169. [Google Scholar], 2002) Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar]. 60. The works of Gourevitch (1986) Gourevitch, Peter. 1986. Politics in Hard Times: Comparative Responses to International Economic Crises, Ithaca, NY: Cornell University Press. [Google Scholar], Gershenkron (1962) Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective, Cambridge, MA: Harvard University Press. [Google Scholar], and Hirshman (1968) Hirschman, Albert O. 1968. The Political Economy of Import-Substituting Industrialization in Latin America. Quarterly Journal of Economics, 12 February: 2–32. [Google Scholar] spring to mind as examples. 61. Only Zambia has privatized a major mineral sector in Africa. (See Quinn, 2002 Quinn, John James. 2002. The Road Oft Traveled: Development Policy and Majority State Ownership of Industry in Africa, Westport, CT: Praeger. [Google Scholar].) In Venezuela and Zimbabwe, governments have increased ownership in mining and petroleum. 62. http://www.itireports.com/cma_china.htm Also, Hu announced that “armaments, power generation and distribution, oil and petrochemicals, telecommunications, coal, aviation and shipping industries” must be under state control. http://www.chinadaily.com.cn/china/2006-12/19/content_762056.htm
Referência(s)