¹ See, for example, Schedler, Andreas, ‘The Menu of Manipulation’, Journal of Democracy, 13 (2002), 36–50CrossRefGoogle Scholar; Diamond, Larry, Developing Democracy: Toward Consolidation (Baltimore, Md.: Johns Hopkins University Press, 1999)Google Scholar; Linz, Juan and Stepan, Alfred, Problems of Democratic Transition and Consolidation (Baltimore, Md.: Johns Hopkins University Press, 1996)Google Scholar; Przeworski, Adam, Democracy and the Market (Cambridge: Cambridge University Press, 1991)CrossRefGoogle Scholar; O’Donnell, Guillermo and Schmitter, Philippe C., Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies (Baltimore, Md.: Johns Hopkins University Press, 1986)Google Scholar.

² E.g., Magaloni, Beatriz, Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico (Cambridge: Cambridge University Press, 2006)CrossRefGoogle Scholar; Diamond, Larry and Plattner, Marc F., Electoral Systems and Democracy (Baltimore, Md.: Johns Hopkins University Press, 2006)Google Scholar; Schedler, Andreas, Electoral Authoritarianism: The Dynamics of Unfree Competition (London: Lynne Rienner, 2006)Google Scholar; Levitsky, Stephen and Way, Lucan A., ‘The Rise of Competitive Authoritarianism’, Journal of Democracy, 13 (2002), 51–65CrossRefGoogle Scholar.

³ A few exceptions that look at the origins of term limits, implications of term limits or focus on third-term debates in small-N comparisons are Shugart, Matthew and Carey, John, Presidents and Assemblies: Constitutional Design and Electoral Dynamics (Cambridge: Cambridge University Press, 1992), pp. 76–93CrossRefGoogle Scholar; Carey, John, ‘Strong Candidates for a Limited Office’, in Scott Mainwaring and Matthew Shugart, eds, Presidentialism and Democracy in Latin America (Cambridge: Cambridge University Press, 1997)Google Scholar; Maltz, Gideon, ‘The Case for Presidential Term Limits’, Journal of Democracy, 18 (2007), 128–142CrossRefGoogle Scholar; Baker, Bruce, ‘Outstaying One’s Welcome: The Presidential Third-Term Debate in Africa’, Contemporary Politics, 8 (2002), 285–301CrossRefGoogle Scholar; Serrafero, Mario, Reelección y Sucesión Presidencial (Buenos Aires: Editorial de Belgrano, 1997)Google Scholar; Bamfo, Napoleon, ‘Term Limit and Political Incumbency in Africa: Implications of Staying in Power Too Long with References to the Cases of Kenya, Malawi, and Zambia’, African and Asian Studies, 4 (2005), 327–356CrossRefGoogle Scholar; Von Doepp, Peter, ‘Party Cohesion and Fractionalization in New African Democracies: Lessons from Struggles over Third-Term Amendments’, Studies in Comparative International Development, 40 (2005), 65–87CrossRefGoogle Scholar.

⁴ de Mesquita, Bruce Bueno, Smith, Alastair, Siverson, Randolph and Morrow, James, The Logic of Political Survival (Cambridge, Mass.: MIT Press, 2003)Google Scholar; Cheibub, Jose and Przeworski, Adam, ‘Accountability for Economic Outcomes’, in Adam Przeworski, Susan Stokes, Bernard Manin, eds, Democracy, Accountability, and Representation (Cambridge: Cambridge University Press, 1999), 222–251CrossRefGoogle Scholar; Goemans, Hein and Chiozza, Giacomo, ‘International Conflict and the Tenure of Leaders, Is War Still Ex Post Inefficient?’, American Journal of Political Science, 48 (2004), 604–619Google Scholar; Iqbal, Zaryab and Zorn, Christopher, ‘The Political Consequences of Assassination’, Journal of Conflict Resolution 20 (2008), 1–16Google Scholar.

⁵ Jefferson warned against the dangers of dictatorial takeover while Alexander Hamilton argued against restrictions on the grounds that they undermined policy continuity, among other things. See Hamilton, Alexander, Madison, James and Jay, John, The Federalist: A Collection of Essays, Written in Favour of the New Constitution, as Agreed upon by the Federal Convention, September 17, 1787 (Clark, N.J.: Lawbook Exchange, 2004 [1787])Google Scholar.

⁶ Quoted in Carey, John, ‘Policy Issues: The Reelection Debate in Latin America’, Latin American Politics and Society, 45 (2003), 119–133CrossRefGoogle Scholar, p. 121.

⁷ Geddes, Barbara, ‘Authoritarian Breakdown: Empirical Test of a Game Theoretic Argument’ (paper presented at the 95th Annual Meeting of the American Political Science Association, Atlanta, 1999Google Scholar). Also, see Hale, Henry, ‘Regime Cycles: Democracy, Autocracy, and Revolution in Post-Soviet Eurasia’, World Politics, 58 (2005), 133–165CrossRefGoogle Scholar.

⁸ Howard, Marc and Roessler, Philip, ‘Liberalizing Electoral Outcomes in Competitive Authoritarian Regimes’, American Journal of Political Science, 50 (2006), 365–381CrossRefGoogle Scholar.

⁹ Presidential regimes are those in which the effective head of government, a president, is elected for a fixed term directly (or indirectly). The president is effective in a sense that he is not dominated by a single ruling party or a military junta, and he cannot be discharged by a legislative no-confidence vote (Riggs, Fred, ‘Presidentialism vs. Parliamentarism: Implications for Representativeness and Legitimacy’, International Political Science Review, 18 (1998), 253–278CrossRefGoogle Scholar). In practical terms, presidential regimes are those coded as ‘Presidential’ in the Database of Political Institutions (Beck, Thorsten, Clarke, George, Groff, Alberto, Keefer, Philip and Walsh, Patrick, ‘New Tools in Comparative Political Economy: The Database of Political Institutions’, World Bank Economic Review, 15 (2001), 165–176CrossRefGoogle Scholar (version DPI 2004)), excluding military juntas, monarchies and one-party states, and those coded as ‘Indirect presidential’, excluding one-party states. This sample includes all presidents during 1960–2005 (I additionally coded for the 1960–75 period), thus covering the independence period in Africa and the third wave of democratization.

¹⁰ I collected comparative data on presidential tenure arrangements over the 1960–2008 period. The data detail the number of terms allowed, the term length, the duration of arrangement and various miscellaneous details. In this article, I use N×N to denote a particular arrangement: number of terms × term length (e.g., 1 × 5, 2 × 4 or N × 7). Table A1 in the Appendix details all presidential tenure changes in this period.

¹¹ Sklar, Richard, Onwudiwe, Ebere and Kew, Darren, ‘Nigeria: Completing Obasanjo’s Legacy’, Journal of Democracy, 17 (2006), 100–115CrossRefGoogle Scholar.

¹² Wilson, Catherine, ‘The Bolivarian Revolution According to Hugo Chavez’, Orbis, 52 (2008), 523–527CrossRefGoogle Scholar. Hugo Chavez lifted limits in a second referendum in February 2009.

¹³ Bienen, Henry and Van de Walle, Nicolas, Of Time and Power: Leadership Duration in the Modern World (Palo Alto, Calif.: Stanford University Press, 1991)Google Scholar, argue that term limits have little effect on longevity as they can be easily scrapped. Similarly, Bueno de Mesquita et al., The Logic of Political Survival, find that autocracies never implement term limits and, if they were to do so, those limits would not be enforceable.

¹⁴ Boston University News Release, ‘Six African Former Heads of State to Hold Summit’, News Conference at Boston University, 10 April 2003.

¹⁵ See, for example, the classical study by Adorno, T. W., Frenkel-Brunswik, Else, Levinson, D. J. and Sanford, R. N., The Authoritarian Personality (New York: Harper, 1950)Google Scholar, or the studies of Hitler, such as Brink, T. L., ‘The Case of Hitler: An Adlerian Perspective on Psychohistory’, Journal of Individual Psychology, 30 (1974), 23–31Google Scholar.

¹⁶ Bueno de Mesquita et al., The Logic of Political Survival; Downs, Anthony, An Economic Theory of Democracy (New York: Harper and Row, 1957)Google Scholar.

¹⁷ On time horizons and discounting, see Laibson, David, ‘Intertemporal Decision Making’, Encyclopaedia of Cognitive Science (London: Nature Publishing Group, 2003)Google Scholar; Price, Colin, Time, Discounting and Value (Oxford: Basil Blackwell, 2003)Google Scholar.

¹⁸ Bates, Robert, Prosperity and Violence: The Political Economy of Development (New York: W. W. Norton, 2001)Google Scholar; Easterly, William, Elusive Quest for Growth: Economists’ Adventures and Misadventures in the Tropics (Cambridge, Mass.: MIT Press, 2001)Google Scholar.

¹⁹ Havel, Václav, Toward a Civil Society: Selected Speeches and Writings, 1990–1994 (Prague: Lidove Noviny, 1994)Google Scholar.

²⁰ Acemoglu, Daron and Robinson, James A., Economic Origins of Dictatorship and Democracy (Cambridge: Cambridge University Press, 2006)Google Scholar; Diamond, Larry, ‘Economic Development and Democracy Reconsidered’, American Behavioral Scientist, 35 (1992), 450–499CrossRefGoogle Scholar; Epstein, David, Bates, Robert, Goldstone, Jack, O’Halloran, Sharyn and Kristensen, Ida, ‘Democratic Transitions’, American Journal of Political Science, 50 (2006), 551–569CrossRefGoogle Scholar; Lerner, D., The Passing of Traditional Society (New York: The Free Press of Glencoe, 1958)Google Scholar; Lipset, Seymour, ‘Some Social Prerequisites for Democracy: Economic Development and Political Legitimacy’, American Political Science Review, 53 (1959), 69–105CrossRefGoogle Scholar; Przeworski, Adam, Alvarez, Michael, Cheibub, José A. and Limongi, Fernando, Democracy and Development: Political Institutions and Well-being in the World, 1950–1990 (Cambridge: Cambridge University Press, 2000)CrossRefGoogle Scholar; Rueschemeyer, Dietrich, Stephens, Evelyn H. and Stephens, John D., Capitalist Development and Democracy (Chicago, Ill.: University of Chicago Press, 1992)Google Scholar.

²¹ Przeworski, Adam, ‘Democracy as an Equilibrium’, Public Choice, 123 (2005), 253–273CrossRefGoogle Scholar. In a related discussion on the stability of political regimes, democracies are seen to be more robust because they limit what is at stake in politics. See Przeworski, ‘Democracy and the Market’; Weingast, ‘Political Foundations of Democracy and the Rule of Law’; North, Douglass, Summerhill, W. and Weingast, Barry, ‘Order, Disorder and Economic Change: Latin America vs. North America’, in Bruce Bueno de Mesquita and H. L. Root, eds, Governing for Prosperity (New Haven, Conn.: Yale University Press, 2000), pp. 17–58Google Scholar.

²² Londregan, John and Poole, Keith, ‘Poverty, the Coup Trap, and the Seizure of the Executive Power’, World Politics, 42 (1990), 151–183CrossRefGoogle Scholar.

²³ North, et al. , ‘Order, Disorder and Economic Change’, p. 29Google Scholar.

²⁴ La Porta, Rafael, Lopez-de-Silanes, Florencio, Shleifer, Andrei and Vishny, Robert, ‘The Quality of Government’, Journal of Law, Economics, and Organization, 15 (1999), 222–279CrossRefGoogle Scholar; Ades, Alberto and Di Tella, Rafael, ‘Rents, Competition, and Corruption’, American Economic Review, 89 (1999), 982–993CrossRefGoogle Scholar; Treisman, Daniel, ‘The Causes of Corruption: A Cross-National Study’, Journal of Public Economics, 76 (2000), 399–457CrossRefGoogle Scholar.

²⁵ Transparency International, Global Corruption Report 2004 (London: Pluto Press, 2004), p. 13Google Scholar.

²⁶ ‘For the politically well connected, Zimbabwe offers a lifestyle unmatched in much of Africa. They are the ones who can obtain foreign exchange at official rates to import and sell at vast profit, smart Japanese vehicles in Harare’s burgeoning car dealerships’ (‘Zimbabwe teachers find earnings add up to 4.5 bananas a day’ (Financial Times, 17 March 2007), p. 7.

²⁷ Ades, and Di Tella, , ‘Rents, Competition, and Corruption’Google Scholar; Lamsdorf, Johann, ‘Causes and Consequences of Corruption: What Do We Know from a Cross-Section of Countries?’ in Susan Rose-Ackerman and Henry Luce, eds., International Handbook on the Economics of Corruption (Cheltenham, Glos.: Edward Elgar Publishing, 2006), pp. 3–52Google Scholar.

²⁸ van de Walle, Nicolas, African Economies and the Politics of Permanent Crisis, 1979–1999 (Cambridge: Cambridge University Press, 2001), p. 144CrossRefGoogle Scholar.

²⁹ Di Tella, Guido, Peron–Peron (Buenos Aires: Hyspamerica, 1983)Google Scholar.

³⁰ E.g., Penford-Becerra, Michael, ‘Clientelism and Social Funds: Evidence from Chavez’s Misiones’, Latin American Politics and Society, 49 (2007), 63–84CrossRefGoogle Scholar.

³¹ Way, Lucan, ‘The Real Causes of the Color Revolutions’, Journal of Democracy, 19 (2008), 55–69Google Scholar.

³² I would like to thank the anonymous reviewer for the valuable suggestion to distinguish magnitude from probability in the discussion of the value of political office.

³³ It is also likely that a ruler will monitor the experience of other ex-rulers. For example, Fidel Castro reacted very negatively to the 1998 arrest of General Pinochet, possibly worrying about his own fate. Reported by WSWS, 5 November 2005: www.wsws.org/news/1998/nov1998/cas-n05.shtml. It is hard to theorize what is the reference group for a leader and for how many years other rulers’ fates will remain relevant.

³⁴ Brownlee, Jason, ‘Hereditary Succession in Modern Autocracies’, World Politics, 59 (2007), 595–628CrossRefGoogle Scholar; Magaloni, , Voting for AutocracyGoogle Scholar.

³⁵ Hasson, Adam, ‘Extraterritorial Jurisdiction and Sovereign Immunity on Trial: Noriega, Pinochet and Milosevic – Trends in Political Accountability and Transnational Criminal Law’, British Columbia International and Comparative Law Review, 125 (2002), 125–158Google Scholar.

³⁶ VonDoepp, ‘The Problem of Judicial Control in Africa’s Neopatrimonial Democracies’; Independent (Zambia), ‘Former president held for corruption charges’, 27 July 2006.

³⁷ Baker, , ‘Outstaying One’s Welcome’Google Scholar.

³⁸ ‘I decided for myself from the very beginning: a third term was not an option. I am a realist. I did not want to make Ukraine into pariah state in the democratic world.’ Vremya Novostej, ‘Leonid Kuchma: in the personality contest Yushchenko had an advantage’, 16 March 2005.

³⁹ Kuzio, Taras, ‘Ukraine’s Orange Revolution: the Opposition’s Road to Success’, Journal of Democracy, 16 (2005), 117–130CrossRefGoogle Scholar.

⁴⁰ ‘Mugabe claims poll landslide has sent a message to Blair’, Observer, 3 April 2005.

⁴¹ Shevtsova, Liliya, Russia – Lost in Transition: The Yeltsin and Putin Legacies (Washington, D.C.: Carnegie Endowment for International Peace, 2007), pp. 81–96Google Scholar.

⁴² There are still more ingenious designs that are very difficult to account for empirically. For example, one strategy is to step aside and rule by proxy. Another strategy is to change the constitution to prolong the rule but step down anyway. This strategy can be chosen if the ruler does not intend to run for whatever reasons (health, age), but signals his intent to run in order to preserve stability and ensure that his chosen successor is well placed to succeed him. On the danger of announcing succession too early in dictatorships, see Tullock, Gordon, Autocracy (Boston, Mass.: Kluwer Academic, 1987)CrossRefGoogle Scholar. The succession in Azerbaijan in 2003 approximated this scenario, in which the outgoing, terminally-ill President Aliev cleared the constitutional obstacles to run for a third term but then withdrew in favour of his son, his designated successor. There were reports that President Aliev was actually dead while running for the presidency (Al-Ahram International Edition, ‘Like father, like son’, No. 650 (7–13 August 2003)).

⁴³ Goemans, Hein, Gleditsch, Kristian and Chiozza, Giacomo, ‘Introducing Archigos: A Data Set of Political Leaders’, Journal of Peace Research, 46 (2009), 269–283CrossRefGoogle Scholar.

⁴⁴ In cases where rulers modified their terms twice or three times, I include tenure up to the last, most significant extension. Alternative specifications that included time to the first extension produced similar results (but lost several data points).

⁴⁵ I exclude those who entered unlimited office and became presidents for life or extended their term length.

⁴⁶ Cingranelli, David and Richards, David, Human Rights DatasetGoogle Scholar, 2008 version.

⁴⁷ Henisz, Witold, ‘The Institutional Environment for Infrastructure Investment’, Industrial and Corporate Change, 11 (2002), 355–389CrossRefGoogle Scholar (POLCON 2005 Release).

⁴⁸ This is the same way that the issue is considered in constitutional design: first, what specific and reserved powers to instil, in a president and, second and independently, how to structure and/or limit presidential terms. Consider the French presidency without term limits of Chirac with the 0.7 level of constraints at entry into his office, for example.

⁴⁹ I would like to thank an anonymous reviewer for this suggestion. See Marshall, Monty and Jaggers, Keith, Political Regime Characteristics and Transitions, 1800–2004, Dataset Users’ Manual, Polity IV Project (College Park: University of Maryland, 2006)Google Scholar.

⁵⁰ Constitutional changes and Events are from Banks, Arthur, Cross-National Time-Series Data Archive, 1815–2003 (Binghamton, N. Y.: Arthur S. Banks, 2005)Google Scholar, CD-ROMs.

⁵¹ Honaker, James, King, Gary and Blackwell, Matthew, Amelia II: A Program for Missing Data (2007)Google Scholar. Observations are not missing at random as earlier years and poorer countries have more missing values. In the model for imputation, there were gaps totalling 23 per cent. This pattern makes it possible to employ imputation based on existing relationships in the observed data. The method enables the avoidance of bias caused by listwise deletion when observations are not missing completely at random.

⁵² Box-Steffensmeier, Janet and Jones, Bradford, Event History Modeling: A Guide for Social Scientists (Cambridge: Cambridge University Press, 2004), pp. 155–182CrossRefGoogle Scholar; Diermeier, Daniel and Stevenson, Randy, ‘Cabinet Survival and Competing Risks’, American Journal of Political Science, 43 (1999), 1051–1068CrossRefGoogle Scholar.

⁵³ Similar to the earlier regression, one can include frailty terms – latent random effects shared by rulers of their specific countries. However, the estimated frailty variance and the likelihood-ratio test of within-group correlation proved that the intra-group correlation was insignificant, perhaps due to the limited time frame.

⁵⁴ Wei, Lee-Jen, Lin, Danyu and Weissfeld, Lisa, ‘Regression Analysis of Multivariate Incomplete Failure Time Data by Modeling Marginal Distributions’, Journal of the American Statistical Association, 84 (1989), 1065–1073CrossRefGoogle Scholar. ‘Long’ if Strategies 9 and 10 are chosen, or if 3–8 for the second time, see section on robustness checks for more details.

⁵⁵ See Marshall, and Jaggers, , Political Regime Characteristics and Transitions, 1800–2004Google Scholar.

⁵⁶ Thus, Category 3 on the XCONST includes ‘Attempts by the executive to change some constitutional resrictions, such as prohibitions on succeeding himself, or extending his term, fail and are not adopted.’

⁵⁷ It is possible that the amount of foreign aid decreases when the ruler announces his or her decision to remain in power. I compared results obtained using the average foreign aid levels per tenure and per entry levels and found the results to be quite similar.

⁵⁸ The post-estimation tests, such as the link test or Hosmer and Lemeshow’s test, suggest that the models are correctly specified.

⁵⁹ I omit the irregular exits’ model because coups and assassinations are not the subject of this article. See the comprehensive analyses of various types of leaders’ exits in Goemans, Hein, ‘Which Way Out? The Manner and Consequences of Losing Office’, Journal of Conflict Resolution, 52 (2008), 771–794CrossRefGoogle Scholar.

⁶⁰ I follow Magaloni, Voting for Autocracy, items 32–42 coding of hegemonic party regimes.

⁶¹ Standard diagnostic tests were applied to examine model specification, such as the link test or the test based on Schoenfeld residuals.

⁶² King, Gary, Tomz, Michael and Wittenberg, Jason, ‘Making the Most of Statistical Analyses: Improving Interpretation and Presentation’, American Journal of Political Science, 44 (2000), 347–361CrossRefGoogle Scholar.

⁶³ Growth is in relative terms (1999–2007). Weisbrot, Mark and Sandoval, Luis, ‘Update: The Venezuelan Economy in the Chavez Years’, Center for Economic and Policy Research Working Paper (February 2008), available from www.cepr.netGoogle Scholar

⁶⁴ Long, Scott and Freese, Jeremy, Regression Models for Categorical Dependent Variables Using Stata (College Station, Texas: Stata Press, 2003)Google Scholar.

⁶⁵ It is possible that alternative models that account for president–party politics or small-case comparisons will complement the proposed explanation or uncover alterative mechanisms to explain the outcomes. Likewise, further model specifications could uncover how presidents update their horizons as their tenure progresses.

⁶⁶ Quoted in Cheibub, and Przeworski, , ‘Accountability for Economic Outcomes’, p. 234Google Scholar.

⁶⁷ Letter to Vermont State Legislature, 10 December 1807, quoted in Stein, Charles, The Third-Term Tradition: Its Rise and Collapse in American Politics (Westport, Conn.: Greenwood Press, 1943), p. 38Google Scholar.

⁶⁸ ‘£2.5m awarded to Mozambique’s Chissano’, Daily Telegraph, 24 October 2007.