Democracy and Development
Political Institutions and Well-Being in the World, 1950-1990
Adam Przeworski New York University
Michael R. Alvarez DePaul University
Jose Antonio Cheibub Yale University
Fernando Limongi University of Sao Paulo
§§i Cambridge
UNIVERSITY PRESS
Introduction
population, we return to study their effect on the growth of per capita incomes.
Because technical, typically statistical issues are entailed in many of the analyses, all the chapters include appendixes in which our methods are explained and additional tests arc offered. Two general appendixes follow the text. Appendix I presents a general model of selection that organizes the entire analysis. Appendix 11 is a listing of variables and the sources from which they were derived.
Wc will not reveal our conclusions here; an impatient reader can flip the pages. We will disclose only this: Whenever regimes do make a difference, lives under dictatorships are miserable. The ChurchiUian view may be not enough, but it is accurate. Democracies are far from perfect, but thev are better than all the alternatives.
Chapter One
Democracies and Dictatorships
Introduction
To study systematically the origins and the consequences of political regimes, we need first to determine what regime each country has had during each period of its history. This is a complicated and irredeemably controversial undertaking. Although "democracy" and "dictatorship" are concepts of everyday usage, the intuitions we associate with them are not always sufficient to determine that a particular country operated according to one or the other at a given time. Thus we must formalize these concepts in terms of rules that can be decisively and reliably applied to the observable aspects of national histories. And once we have such rules, we must apply them. Any such procedure necessitates going back and forth between conceptual analysis and historical observation. Closure is difficult to reach: In some cases, history simply has not generated the information called for by the rules, and we must accept the fact that there will be systematic errors. Other cases are so idiosyncratic that we must accept the fact that they have to be treated as random errors. Nevertheless, although no classification can be free of errors, the encouraging lesson is that independent endeavors by different researchers have led to very similar results.
Although several classifications of regimes, covering different periods and sets of countries, are now available, they can, in our -view, be improved-by-(l) a better-graundteg-m -politioaT-tlieoFyr(2)
exclusive reliance on observables, rather than on subjective judgments, (3) an explicit distinction between systematic and random errors, and (4) more extensive coverage. Thus, the purpose of this chapter is to develop a classification of political regimes guided by those objectives.
12
13
Democracies and Dictatorships
This chapter is organized as follows: The first section covers conceptual issues. The second section spells out the three basic rules we use to classify regimes. The third section focuses on the treatment of systematic errors and offers an additional rule that applies to a particular class of cases. The fourth section summarizes these rules and shows their effects on regime classification. The fifth section considers distinctions among democratic regimes and among authoritarian regimes. The sixth section enumerates some criteria we did not include. The seventh and eighth sections describe the results of applying our rules to historical observations of 141 countries between 1950 (or the year of independence) and 1990. Appendix 1.1 compares our approach with alternative measures. Appendix 1.2 lists the regimes by country and period. Appendix 1.3 provides some background information on regime transitions. Finally, Appendix 1.4 describes a somewhat smaller data set, our "short" data set, which is used in the remaining part of this book.
Democracy and Dictatorship
Our purpose is to classify the political regime observed in each country during each year either as a democracy or as a dictatorship, a term we use interchangeably with "authoritarian regime." Although we later distinguish different types of democracies and dictatorships, our basic classification is dichotomous.
Our general stance is minimalist. Perusing the innumerable definitions, one discovers that "democracy" has become an altar on which everyone hangs his or her favorite ex voto.1 Almost all normatively desirable aspects of political life, and sometimes even of social and economic life, are credited as definitional features of democracy: representation, accountability, equality, participation, dignity, rationality, security, freedom - the list goes on. Indeed, according to many definitions, the set of true democracies is an empty set. And from an analytical point of view, lumping all good things together is of little use. The typical research problem is to examine the relationships among them. We want to know if holding repeated elections induces govern-
1 To cite Macpherson (1966: 1): "Democracy used to be a bad word. Everybody who was anybody knew that democracy, in its original sense of rule by the people or government in accordance with the will of the bulk of the people, would be a bad thing - fatal to individual freedom and to all graces of civilized living. That was the position taken by pretty nearly all men of intelligence from the earliest historical times down to about a hundred years ago. Then, within fifty years, democracy became a good thing."
Democracy and Dictatorship
mental accountability, if participation generates equality, if freedom imbues political systems with rationality. Hence, we want to define "democracy" narrowly.
Three major distinctions dominate modern political thought concerning forms of government.2 Montesquieu's legacy (1995 [1748]) is the distinction between limited regimes and despotic regimes. Kelsen's contribution (1945), going back to Rousseau and Kant, was to distinguish between "autonomy" (systems in which norms are determined by those to whom they apply) and "heteronomy" (systems in which the legislators are distinct from those who are subject to the laws). Finally, Suhumpoter's innovation (1942) was to emphasize competition, or, to borrow Dahl's term (1971), which we prefer, "contestation," as the essential feature of democracy.
We focus on contestation. Our purpose is to distinguish between (1) regimes that allow some, even if limited, regularized competition among conflicting visions and interests and (2) regimes in which some values or interests enjoy a monopoly buttressed by the threat or the actual use of force. Thus "democracy," for us, is a regime in which those who govern are selected through contested elections. This definition has two parts: "government" and "contestation."
In no regime are all governmental offices filled by elections. Outside of classical Greece, generals, who are public officials, have never been elected. Judges rarely are. What is essential in order to consider a regime as democratic is that two kinds of offices be filled, directly or indirectly, by elections: the chief executive office and the seats in the effective legislative bod5'.
Moreover, in a democracy, the offices that are being filled by contested elections grant their occupants tho authority to exercise governance free of the legal constraint of having to respond to a power not constituted as a result of the electoral process. Thus, governmental responsibility either directly to voters or to a parliament elected by them is a defining feature of democracy. If we were to use only contested elections as the criterion, we would date the beginnings of democracy in many Western European countries to the period in which governments still were not autonomous, not independent of the crown, ~wtfHretaTrmrrrrWaTT^
process of democratization, and in Western Europe most often it was, we date the advent of democracy to the time of transfer of governmental responsibility from the crown (or a non-elected upper chamber)
* See Bobbio (1989: 100-25, and chap. 4).
14
15
1
Democracies and Dictatorships
to the parliament. Thus we date democracy in Great Britain to 1911 (Dahl 1990), and in Germany to 1919.
Until the early years of this century, the struggle for democracy concerned primarily suffrage. In most countries, democracy emerged only gradually, in a sequence of steps. Typically, what happened first was that legislatures, elected on a nonpartisan basis and under highly restricted suffrage, divided along partisan lines and began to contest elections on a partisan basis. Then there followed extensions of political rights, which were sometimes very gradual, as in Norway or Chile, and at times instantaneous, as in Finland in 1906. In contrast, in those countries that have only recently confronted the eventuality of having to establish democratic institutions, suffrage is not an issue: It is taken for granted that it will be "universal."3 Neither is governmental responsibility an issue 4 In the recent cases, the only focus of conflicts has been contestation: whether or not divergent political forces would be able to compete for governmental offices and to assume office if they won elections.
Contestation occurs when there exists an opposition that has some chance of winning office as a consequence of elections. We take Przeworski's dictum that "democracy is a system in which parties lose elections" (1991: 10) quite literally: Whenever in doubt, we classify as democracies only those systems in which incumbent parties actually did lose elections. Alternation in office constitutes prima facie evidence of contestation. In most nineteenth-century Latin American countries, the incumbent presidents, even if they could not be reelected, controlled their succession. We consider such regimes democratic only when that system no longer functioned, when the opposition won an election and assumed office.
Contestation entails three features: (1) ex-ante uncertainty, (2) ex-post irreversibility, and (3) repeatability.
By "ex-ante uncertainty" we mean that there is some positive probability that at least one member of the incumbent coalition will lose in a particular round of elections. Uncertainty is not synonymous with
3 The quotation marks are needed because the very notion of "universal" suffrage is a matter of convention. Suffrage was "universal" in hurope before the voting qualification was lowered from 21 to 18 years of age, and it is now "universal" at 18, rather than 16, 14, or 12. And "universal" is in turn defined relative to "citizen," itself a legally-regulated notion. Immigrants often do not have the right to vote in national elections even if they have lived in a country for a long time, and in several Western European countries they constitute more than 8% of the adult population.
4 Rxcept in Western Samoa.
16
Democracy and Dictatorship
unpredictability: The probability distribution of electoral chances is typically known. All that is necessary for outcomes to be uncertain is that it be possible for some incumbent party to lose.5 The best illustration of such uncertainty is the surprise expressed by an editorial in the Chilean right-wing newspaper, El Mercurio, in the aftermath of Salvador Allende's victory in the first round of the presidential elections of 1970: "No one expected that a marxist candidate could win elections through a universal, secret, bourgeois franchise." The franchise may have boon "bourgeois," the chances skewed, and the victory of a Marxist candidate may have been known to be unlikely. But it was possible. The eventual outcome was not certain ex ante.
This feature of democracies has practical consequences. Most people think that Argentina under President Arturo Illia (1963-66) was democratic, oven though the largest party in the country was prohibited from competing in the elections of July 1963. In turn, most agree that Mexico is not democratic, even though no party is legally banned from contesting elections. The reason is that Illia won narrowly, with 26.2 percent of votes cast, and he could have lost. In contrast, in Mexico it was certain that the Partido Revolucionario Institucional (PRI) would win.
By "ex-post irreversibility" we mean the assurance that whoever wins election will be allowed to assume office. The outcomes of elections must be irreversible under democracy even if the opposition wins. In 1929, the dictator of El Salvador, General Romero, announced that his country was about to join the family of civilized nations by celebrating the first free and honest election. He issued a decreto-ley that specified when the elections would take place, who would be qualified to vote, what the ballots would look like, when the polling places would be open, and so forth. The la.st point declared that "Army contingents will be stationed in the polling places in case the Opposition wins." 'Chat was not a democratic election.
A strong notion of uncertainty would require that there be some chance that the major member of a coalition might And itself out of office as the result of the next election: a weak notion would extend to any member of Ihe incumbent coalition. Under some con-
ditions. chaimes of minor partners may be considered.....jncnnsnqiiontial; Ttajy linriej- .
Christian Democratic domination may be a case in point. But in other cases, the electoral fates of minor parties can affect the orientation of a government: Israel and Germany are the relevant cases here. We have opted for the weak version: We consider any change of the governing coalition as alternation. Moreover, we do not want to imply that all alternations result directly from elections. In several countries - Poland between 1919 and 1926 is an example - a government has changed between opposing parties without an election intervening and the new government has remained in office as a result of the subsequent election.
MASARYKOVA UNIVERZITA
Fakulta sociálních studii 17
joštova 10 ___
602 00 B R N O CfiD
Democracies and Dictatorships
The practical consequence of this feature is to exclude sham elections as well as periods of liberalization. Liberalization is typically intended by dictatorial regimes to be a controlled opening of the political space. When it fails, that is, when the opposition does win, a clampdown sometimes follows. Hence, there is no certainty that the opposition will be able to celebrate its victory.
The final feature of contestation is that elections must be repeated. Whoever wins the current round of elections cannot use office to make it impossible for the competing political forces to win next time. Democracy, as Linz (1984) put it, is government pro tempore. All political outcomes must be temporary: Losers do not forfeit the right to compete in the future, to negotiate again, to influence legislation, to pressure the bureaucracy, or to seek recourse to courts. Even constitutional provisions are not immutable; rules, too, can be changed according to rules.
The practical consequence of this last feature is that we should reserve judgment about elections, since an electoral victory may serve only to establish an authoritarian rule. This has been true in several African countries following independence. Unless the losers are given political guarantees that their ability to contest future elections will be protected, the mere fact that elections have been held does not suffice to qualify the regime as democratic. Only if the losers are allowed to compete, win, and assume office is a regime democratic.
Throughout this discussion, wc have focused on democracy. We treat dictatorship simply as a residual category, perhaps better denominated as "not democracy." Our procedure is to establish rules that will disqualify a particular regime as democratic, without worrying about the nature of the regimes eliminated in this manner. Only then do we introduce some features that distinguish among different non-democratic regimes.
Operational Rules: Filling Offices by Contested Elections
Following Cardoso (1979: 38), as well as O'Donnell and Schmitter TT986T73)rby"Vegime'*"we~meafTthe systcfrToT'felaHons"Between the civil society and the state. A regime is a system of rules and practices that determine who has political rights, how they can be exercised, and with what effects for the control over the state. Hence, even if dictators succeed one another, the regime, in our sense of the term, remains the same as long as it remains a dictatorship. Thus, when General Viola
Operational Rules: Offices by Contested Elections
overthrows General Videla, we consider the entire period as one dictatorship, rather than as "Videla" and "Viola" regimes. And even in those rare cases when the basic democratic institutions are transformed without a break in legality, as in France in 1958, we still consider each of them as a single "democratic regime."
Democracy is a regime in which government offices are filled by contested elections. The first part of this definition is easy to operational-ize: it is relatively simple to observe which offices, if any, are filled as a result of elections. But whether or not these elections are contested, in the sense defined earlier, is not always apparent. The existence of more than one independent party is a sine qua non of contestation, hut it may not be sufficient.
We next specify the rules we use to classify regimes: first those that are applied to assess whether or not the relevant offices were filled via elections, then those that are used to assess whether or not elections were contested.1' Our rules are the following:
Rule 1: The chief executive must be elected.
The "chief executive" may be the president, the prime minister, or, in rare cases, a collegial body. Following Banks (1996), we define as the "chief executive" the occupant of the office formally designated as that of the head of government, thus excluding eminences grises: strongmen who effectively rule their countries but do not occupy a formal position.7
For a regime to be qualified as democratic, the executive must be directly or indirectly elected in popular elections and must be responsible only directly to voters or to a legislature elected by them. Indirect elections qualify as popular only if the electors are themselves elected. Elections by bodies that are themselves nominated do not qualify as popular elections.
Rule 2: The legislature must be elected.
The legislature can be a congress, an assembly, or a parliament. Only the lower house is considered. A constituent assembly that does not have ordinary legislative powers is not considered a legislature.
The information about elections is based on Banks (1996), revised and updated. Such as Deng Tsao Ping, whose only formal position for many years was president of the Chinese Bridge Association. Banks sometimes considers first secretaries of the Communist Party as chief executives, while at other times he takes presidents and prime ministers as chief executives in communist regimes. We could not discover what rules he used, and we took occupants of the forma.] office as chief executives.
18
19
Democracies and Dictatorships
Our rule is that the legislature must be elected for a regime to qualify as democratic.
Rule 3: There must be more than one party.
In some cases, there were no parties: Either there were no elections or elections were conducted but all political parties were banned. In other cases, there was only one party. We consider such regimes authoritarian.8
By "party" we mean an independent list of candidates presented to voters in elections. In communist Poland, for example, three parties and a number of Catholic groups were represented in the Scjm, but until June 1989 voters were offered only one list: a National Front or Patriotic Front or whatever front it was called at the moment. Hence, in cases where the share of seats held by the major party in the legislature was less than 100 percent we checked to see if there was more than one list in legislative elections. For example, although the ruling Vanguard of the Malagasy Revolution (Arema) did not control all the seats in the parliament in Madagascar after 1976, according to Freedom House (1992: 318), "Until March of 1990, when a High Constitutional Court decree permitted multi-partyism, political associations had to operate within the FNDR as the nation's sole legal political entity." The FNDR (National Front of the Malagasy Revolution) was thus the only list offered to voters - one party by our definition.
Applying this rule, we classified as dictatorships all regime-years during which legislatures were elected but parties were banned or during which a single party held 100 percent of the seats in the legislature (except for Jamaica and Trinidad and Tobago, as discussed later) or in which only one list was offered to voters in elections.
We also extended this rule to disqualify as democratic those regimes in which incumbents used an electoral victory to establish (1) non-party rule or (2) one-party rule or (3) a permanent electoral domination. This is called the "consolidation" rule.
Consolidation of non-party or one-party rule occurred whenever incumbents either banned all parties, or all opposition parties, or forced all parties to merge with the ruling one. If an incumbent party ^onsTJtMatMTlWin g 1 ts'toTOnrinnjffice^rOT rule, then the regime is considered to have been authoritarian from
8 Note that we do nut assume that the existence of political parties is a necessary condition for contestation; after all, most pre-1900 elections were non-partisan. We exclude from the democracies only those regimes in which all or all but one political parties were banned. Bernard Manin made us sensitive to this point
Operational Rules: Offices by Contested Elections
the moment at which the incumbent party assumed office. Note that we are not examining intentions: If they tried and failed, the regime is democratic.
We say that consolidation of incumbent rule also occurred whenever there was more than one party but at some time the incumbents unconstitutionally closed the legislature and rewrote electoral rules to their own advantages The entire period preceding the closing of the legislature during which the same party was in office is then considered authoritarian. In some analyses later in this book, we relax this rule in order to identify cases in which incumbents perpetuated an aulo-golpe. In these instances, regimes are classified as democracies until the year in which the autogolpe occurred, allowing us to separate the cases of transition to authoritarianism caused by the incumbent chief executive.10 These cases are listed in Appendix 1.3.
To understand how the consolidation rule was applied, consider Malaysia, a country whore three elections were held between independence in 1957 and 1969. The incumbents won absolute majorities of votes in the first two elections, but not in the third. They then declared a state of emergency, closed the congress, and changed the rules in such a way as to make this unpleasant experience unrepeatable. According to Ahmad (1988: 357), "the better showing by the opposition caused a temporary loss of confidence and even the conclusion by some in the ruling party that it had lost its mandate." The parliament was dissolved in 1969, a state of emergency was proclaimed, and a tough internal-security law, still in effect, was adopted. The constitution was rewritten to ensure that no more electoral defeats would occur. Ahmad (1988: 358) comments on this event: "What is more interesting about the conduct of elections as part of the democratic process, however, was probably the unstated notion that losing an election meant virtually total political defeat. Therefore 1969 served notice to the Alliance leadership that it might have to one day face the prospect of an electoral defeat. . . . The rules of the game of Malaysian democracy were therefore set for modification after 1969 because the
5 The mere act of dissolving the legislature is not sufficient to qualify as consolidation. In —many nam lBgtelatnrog in diwohrwl or suspended according-to extant cwwUluUuual
provisions: The 1975 Indian slate of emergency was duly approved by the two houses of the legislature. In some cases, notably Australia in 1975, the constitutionality of the dissolution has been dubious. Our rule is that if elections were immediately proclaimed and took place witliin the immediate future under the same rides, we did not treat such a dissolution as a breakdown of democracy. lu Note that what we observe are the outcomes of conflicts, rather than their initiation: hence, an inference is entailed in interpreting these results.
21
Democracies and Dictatorships
"Botswana" and the Alternation Rule
prospect of a zero-sum electoral result would be unacceptable if Malay political supremacy was not to be assured." As a result, "the fear of an electoral defeat has been diminished under the Barisari Nasional coalition concept. The parties that have not succumbed to the taste of power by joining the BN cannot pretend to be able to form the national government at any time in the foreseeable future."
In South Korea, President Chung-Hee Park held elections once and won enough votes; then he held them again and became dissatisfied with the result, closed the congress, and assumed dictatorial powers. Five years later, he reopened the congress under new rules. President Ferdinand Marcos in the Philippines won election twice and assumed dictatorial power when he could not amend the constitution to enjoy more terms.
Because in these cases we have prima facie evidence that incumbents were not prepared to yield office as a result of elections (although one could argue that although the Malays were not willing to do it in 1969, they might have been willing to accept defeat earlier), we classify these regimes as dictatorships.
Note that this part of the consolidation rule applies only up to the moment of the unlawful change of electoral rules. If a regime unconstitutionally changed the rules in its favor but subsequently yielded office under these new rules, then the regime is considered authoritarian up to the time of the openly dictatorial interregnum, and democratic subsequently. This is why we have classified the Joao Batista Figueiredo term in Brazil as democratic: Although his predecessor temporarily closed the congress and made it more difficult for the opposition to win, Figueiredo's successor, Tancredo Neves, won election against the candidate supported by the military under the same rules as Figueiredo.
Thus the "party" rule is that if(l) there were no parlies or (2) there was only one party or (3) the incumbents' term in office ended in the, establishment, of non-party or one-party rule or (4) the incumbents unconstitutionally closed the legislature and rewrote the rules in their favor, then the regime was a dictatorship. As shown later, the absence of legal opposition is the most frequent reason for classifying regimes as diciatorsKipsT ......
These three rules appear to us to be non-controversial, and they are easy to apply. The first thing we learned from applying them is that the great majority of cases, 91.8 percent of country-years, arc unambiguously classified by these three rules. There is, however, one particular class of regimes that cannot be classified one way or another.
22
"Botswana" and the Alternation Rule
Thus far wo have classified as democracies those regimes in which the chief executive and the legislature are elected in multi-party elections. But we do not know that all regimes that satisfy these criteria are in fact democracies.
Consider Botswana. Government offices in Botswana are filled by elections, more than one party competes, there is little repression, and there are no exceptional allegations of fraud. Hence, by the rules introduced thus far, Botswana should be considered a democracy, and indeed it is generally considered to be one. Yet the same party has ruled Botswana since independence, always controlling an overwhelming majority in the legislature. Thus, the question arises whether or not elections are held in Botswana only because the ruling party is certain to win them and whether or not the ruling party would yield office if it ever lost. These arc not moot questions: Looking into the future, a specialist on this country speculates that "the resulting conflict could well force the BDP [Botswana Democratic Partyl to choose between losing in parliamentary elections and abandoning elections as a method of leadership selection. Given the paternalistic attitude of the BDP from President Masire down, the latter choice would not be surprising" (Holm 1988: 208). Hence, if democracy is a system in which elections are held even if the opposition has a chance to win and in which the winners can assume office, then the observable evidence is not sufficient to classify Botswana one way or another.
Botswana is an ideal type: no constraints on the opposition, little visible repression, no apparent fraud. But the issue is more general. If the same party or coalition of parties had won every single election from some time in the past until it was deposed by force or until now, we cannot know if it would have held elections when facing the prospect of losing or if it would have yielded office had it in fact lost. We must thus decide which way to err: whether we prefer to commit the error of excluding from the democracies some systems that arc in fact democracies (type I) or the error of including as democracies some systems that are not in fact democratic (type II). Err we must; the ques-tion is which way.--------------------------.............---------------.......
In some cases, either antecedent or subsequent events have provided additional information. In the United Kingdom, we knew even before Blair's victory that the Conservatives had lost elections and allowed their opponents to assume office. In Japan, after a long tenure in office, the incumbent party finally lost elections and allowed the
23
Democracies and Dictatorships
opposition to assume office. Because this is the only information we have, we use it - not without a leap of inference - to conclude that these regimes are or were democratic. The same is true when we know that incumbents unconstitutionally prevented the opposition from winning elections or assuming office. In all these cases, we use this information retroactively. This clearly is not a very satisfactory solution: One might easily imagine that even if certain incumbents were willing to allow a peaceful alternation in office later on, they might not have been willing to tolerate it earlier; conversely, even if they suppressed the opposition later on, they might not have done so earlier. But this is the only information we have; wo cannot observe what might have happened. The only alternative would be to attempt to assess the degree of repression, intimidation, or fraud tor each election, but, in our view, such assessments cannot be made in a reliable way.
Japan is a paradigmatic case of a long tenure in office that ended with a lawful alternation. The LDP (Liberal Democratic Party) was in office continually until the 1993 election. Yet when the incumbents finally lost, they allowed the opposition to assume office. The same was true in Mauritius, the Bahamas, Barbados, Trinidad and Tobago, the Dominican Republic, Grenada, Jamaica, St. Lucia, and the Solomon Islands. In each of these countries the same party stayed in power for at least two terms, either following independence or the birth of democracy; yet eventually it lost an election and gave up office peacefully. We use this information retroactively: Whenever a ruling party eventually suffered an electoral defeat and allowed the opposition to assume office, the regime is classified as democratic for the entire period this party was in power under the same rules. Alternation thus overrides the party rule: In Jamaica, as well as in Trinidad and Tobago, at one point, one party controlled all the seats in the legislature. Yet it lost the subsequent election and relinquished office. We therefore consider these regimes to have been democratic even during the period of one-party rule.
We have already discussed cases in which incumbents, facing the prospect of an electoral defeat or having actually been defeated, unconstitutionally closed the legislature, introduced a state of emergency, and rewrote the rules in their own favor. In such cases, we evoke the rule —atetrrthe-eonsoMdatiorr-ofi^ regime
as authoritarian during the entire period before the openly dictatorial interregnum. In South Korea, President Park won the 1963 elections, and had he not instituted dictatorial rule nine years later, we would never have known whether or not he would ever have been ready to relinquish office. A Korean student of military politics (Se Jin Kim, cited
"Botswana" and the Alternation Rule
by Han 1988: 275) commented with regard to the 1963 elections that, "Park's victory was in fact a blessing for the future of democracy in Korea. Had the military lost, it can be safely assumed that the military would have ignored the electoral outcome and continued to rule even though such rule would have meant a total destruction of constitutionalism." But subsequently Park did close the congress and change the rules, and we use this information to infer that he would not have been ready to yield office during the preceding nine years.
Even when incumbents hold elections only because they expect to win, they sometimes make mistakes: They hold elections and lose. Then they have to decide whether to accept the popular verdict or override it. They can revert to post-election fraud: Anastasio Somo/a is purported to have said to his electoral opponent, "You poor s.o.b., perhaps you won the voting, but I won the counting," a recipe apparently applied by the Mexican PRI in 1988. Blatant fraud constitutes prima facie evidence that the incumbents were not predisposed to permit a lawful alternation in office. Or they can publish voting results and still not allow the opposition to assume office.
Yet, to return to Botswana, in some cases history has not been kind enough to provide even the information that we have for Japan or Malaysia: All we know is that the incumbents always win. Presumably, we would want to think that if Botswana is like Japan, it should be considered democratic, but if it is like Malaysia, it should be considered authoritarian. But we do not know if Botswana is like Japan or like Malaysia. Elections may be held in Botswana only because the ruling party is sure to win, but how are we to know what would happen if they expected to lose or in fact lost?
To provide more intuition, consider Turkey between 1950 and 1960, another period generally considered democratic. The Democratic Party (DP) came to power in 1950, holding 83.8 percent of the seats. It won in 1954 with 93.0 percent of the seats, and in 1957 with 69.5 percent, until it was ousted by the military in 1960. After the 1957 elections "the DP responded to its declining support by resorting to increasingly authoritarian measures against the opposition. . . . The last straw in the long chain of authoritarian measures was the establishment by the gov-
ernment party in Aprd 19bU ora^pluTuunehTarylionimitlee of inquiry to investigate the 'subversive' activities of the RPP [main opposition party]" (Ozbudun 1988: 200). Would the DP ever have yielded power peacefully had it not been deposed by force?
We choose to take a cautious stance, that is, to avoid type-II errors. While examining the histories of particular countries, we were
24
25
Democracies and Dictatorships
"Botswana" and the Alternation Rule
impressed that the dream of many political elites is to rule perpetually and to rule with consent: Politicians are just PRTstas by nature. The Mexican system has been the ideal for many politicians in Latin America and, until the defeat of the LDP, the Japanese system in Asia. Attempts at creating a hegemonic system, in which some or even all opposition would be allowed but the ruling party would not be threatened with losing office, have been made at various moments in Botswana, Gambia, Senegal,11 Argentina, Bolivia, Colombia,12 Ecuador, El Salvador, Guatemala, Honduras,13 Guyana, Bangladesh, Egypt,   Malaysia,   Pakistan,14   the   Philippines,15   South Korea,16
11 Coulon (1988: 154) writes about Senegal: "In 1978, a constitutional reform was adopted which put into place a system of 'controlled democracy.' The number of parties was
limited to three____The legislative and presidential elections of 1978 were a great
success for the Socialist Party (former LPS, the ruling party] (which received 81.7 percent of the ballots cast and 82 of the 100 scats in the Assembly) and for President Senghor personally (who won 82.5 percent of the vote).... It must be emphasized, however, that the elections were held in a tense climate and organized in a way that threatened the secrecy of the ballot."
"- In Colombia, Laureano Gomez was elected president with 83.8% of the vote and took office in August 1950. He continued tight censorship and increased repression against labor and violence against liberals and Protestants. He attempted to reform the constitution in order to impose a falangisl-corporatist framework, freeing Lbe presidency from most congressional constraints, centralizing power, and converting the senate into a corporatist body.
" In Honduras between 1965 and 1970, "the conservative, authoritarian civil-military government suppressed popular organizations and rigged the electoral machinery to assure National party victories in the 1965 and 1968 elections" (McDonald and Kuhl 1989: 113).
" In Pakistan, according to Hose (1988: 114-15), "Ayub's intention initially had been to establish a nonparty system but it quickly became clear that this would be counterproductive. Ayuh then moved to the opposite extreme, legalized virtually all parties thai applied, and formed his own party.... What was rather astonishing was [that] the 1962 constitution. Ay lib Khan's rather cleverly disguised authoritarian system, went along from 1961 to 1969 with no serious political challenges."
15 in the Philippines, according to Jackson (1988: 246), "die final structure created by Marcos was the Kilusang Bagong Lipuna (New Society Movement) or KBL. The KBL initially was not referred to as a political party, but was designed to selecl and eleel candidates to local, provincial, and national offices. KBL candidates, in an atmosphere of restricted press and speech, triumphed in the 1978 interim assembly elections as well as in the 1980 local elections. The degree of limited participation is indicated by
______Hip f«rt ih«i ihu 107(8 nppndllnn ma l«H fmm a       mil hy fnrmm SmMnr BmfcfMI
Aquino. The interim assembly contained only fourteen non-KBL members out of 200 members____President Marcos ran for reelection in June 1981 but his logical opponent, former Senator Aquino, was excluded by the constitution, which required all nominees to be at least fifty years of age (Aquino was forty-eight.) With virtually no opposition, President Marcos was reelected."
16 In South Korea, according to Han (1988: 268-9), "the new government was born with a democratic constitution and with the expectation that it would usher in democratic
26
Singapore, Turkey, Taiwan, and most likely other countries where the evidence is not that direct. Indeed, it seems that many dictatorships and some democracies are just failed attempts at creating a Mexico or a Japan: Sometimes the ruling party overdoes it, and the result is a naked dictatorship; sometimes the ruling party is forced to compromise, and the result is democracy.
Suppose that politicians want power but also want to be admired and adored. Ideally, they would hold office as a result of elections. Yet the hunger for power overwhelms other motivations: They prefer to remain in office by force rather than lose power. Incumbents have some notion of how likely it is that they will win the next election. If they think they will win, they hold elections. If liiey think they will lose, they do not. If these assumptions are correct, then the observed sample of regimes that hold regular elections is biased in favor of "democracies," that is, regimes that look like democracies, in the sense that they permit contestation and fill offices by elections, yet are not democracies in the sense that the opposition has a chance to assume office as a result of elections. Among the observed democracies, there are some that hold elections only because the opposition cannot win and some in which the opposition would not be allowed to assume office if it won. Hence, holding elections is not sufficient to classify a regime as democratic.
We thus need one more rule: alternation. This rule is applicable only to cases that qualify under the preceding three rules and in which in the immediate past the incumbents either held office by virtue of elections for more than two terms or initially held office without being elected. If all these conditions are satisfied and if the incumbents subsequently held but never lost elections, we consider such regimes authoritarian.
In making this decision, we are buttressed by an empirical observation. Among those cases in which alternation in office via elections did occur, except for some Caribbean islands, the share of seats of the incumbents was almost always smaller than two-thirds. Hence, the conditional probability that the seat share will be larger than two-thirds given that alternation occurs is very small. Because alternations via elections are generally less frequent than seat shares in excess of two-thirds, Bayes's rule iTnr^ffe^ThaTTEe '"coffdffionaT probability Thatfan
politics for South Korea. But tire Rhee government was determined to remain in power - for life - which required several constitutional changes, election rigging, and repression of the opposition. Rhee was able to establish his personal dictatorship by making use of the state power as exemplified by the national police."
27
Democracies and Dictatorships
Summary of Rules
alternation will occur given that seat share is larger than two-thirds is also very small: 8.8 percent.17 Countries in which one party wins an overwhelming share of seats are not likely to be democracies: this rule should classify regimes accurately about seven-eighths of the time.
Although there are some countries where a ruling party had been winning by very large margins and yet subsequently left office via elections, the striking finding is that we could have used this ex-post criterion to eliminate almost all the cases in which the same party continually held office. If we were to decide that a regime in which the ruling party always won more than two-thirds of seats was not democratic, Malaysia (where the share of seats after 1971 was always larger than 68.1) would fail by this criterion, as would Botswana (seats > 77.8), Egypt after 1976 (seats > 75), Gambia (seats > 69.7), Senegal after 1978 (seats > 83.0), South Africa (seats > 66.6), Mexico (seats > 72.2, or fraud in 1988), Guyana (seats started at 56.6, went to 83.1), and Singapore (seats = 74.5 in 1965, and all or all but one after 1968). In South Korea, the share of seats fell from 74.3 in 1967 to 55.4 in 1971, sufficient to prompt President Park to dissolve the congress, and when the legislature was opened again in 1973, the ruling party controlled 66.7 percent.
The cases distinguished by the alternation rule constitute systematic error. Those readers who prefer to err in the other direction can reclassify them (they arc marked with asterisks in Appendix 1.2). But an error is unavoidable.
Summary of Rules
For convenience, wc restate our rules. A regime is classified as a dictatorship during a particular year if at least one of these conditions holds:
Rule 1: "Executive selection." The chief executive is not elected (EXSE1.EC).
Rule 2: "Legislative selection." The legislature is not elected (LEGSELEC).
17 Our prior, the probability that during a random country-year the regime will be democratic, is 0.40. The probability that any election will end with more than two-thirds of seats going to one party is 0.38. The likelihood, the conditional probability that seats will be more than two-thirds given that a regime is democratic, is 0.126. Hence, by the Baycs rule, the posterior, the probability that a country will be democratic given that seats will be more than two-thirds, is 0.0877.
28
Rule 3: "Party." There is no more than one party. Specifically, this rule . applies if(l) there are no parties or (2) there is only one party or (3) the current, term in office ends In the establishment of non-party or one-party rule or (4) the incumbents unconstitutionally close the legislature, and rewrite the rules in their favor (PARTY. INCUMB).
Rule 4: "Alternation" (applies only to regimes that have passed the previous three rules). The incumbents will have or already have held office continuously by virtue of elections for more than two terms or have held offi.ee without being elected for any duration of their current tenure in office, and until today or until the time when they were overthrown they had not lost an election (TYPEII).
At the risk of repetition, we restate the relation between the "party" and the "alternation" rules. If at any time there are fewer than two parties, the regime is classified as a dictatorship. If there is more than one party, the question becomes whether or not there is a real possibility for the opposition to win and assume office. If incumbents lose elections at any time in the future, the regime is considered to have been democratic during their entire tenure in office. If incumbents repress the opposition at any time in the future, the regime is classilied as a dictatorship during their entire tenure in office up to this moment. If we observe neither consolidation nor alternation, we avoid type-II error and classify the regime as a dictatorship.
Finally, a word is needed about our timing rules. In all cases of regime transitions, we code the regime that prevailed at the end of the year, even if it came to power on December 31, as, for example, did Nigeria in 1983. Transitions to authoritarianism are signaled by a coup d'etat. Transitions to democracy are dated by the time of the inauguration of the newly elected government, not the time of the election. In the few cases, like that of the Dominican Republic in 1963, in which a democratic regime lasted six months (or in Bolivia in 1979, where the situation changed several times), the information about regimes that began and ended within the same year is lost.
Our data set covers 141 countries between 1950 (or the year of inde-
pendence) and T990. Altogether, ^during~ tTiis~pe7io^^prfbliri'd'''2"38"' regimes: 105 democracies and 133 dictatorships. They arc listed, by country and period, in Appendix 1.2. Of the total 4,730 years, countries lived 1,723 (36 percent) years under democracy and 3,007 (64 percent) under dictatorship (REG).
Table 1.1 shows the numbers of cases in the sample that were
29
Democracies and Dictatorships
Tabic 1.1. Distribution of observations by criteria used for classification as dictatorships
Executive selection: non-elected executive	1,313
Legislative selection:	940
No legislature	789
Non-elected legislature	151
Political party	2.250
No political party	651
One political party	1,599
Consolidation of incumbent rule	93
Alternation	389
Executive selection and political party	1,244
Executive selection and legislative selection	89
Legislative selection and political party	767
Executive selection, legislative selection,	731
and political party	
Total dictatorships	3,007
Total democracies	1,723
Total regime years	4,730
disqualified as democracies by each of the rules alone and by their combinations.
Distinguishing among Democracies and Dictatorships
Obviously, neither democracies nor dictatorships are all the same. Thus, further distinctions are required.
Given the recent popularity of this issue, we classified democracies as parliamentary, mixed, or presidential (INST). These types are defined as follows. Systems in which the government must enjoy the confidence of the legislature are "parliamentary"; systems in which the government serves at the pleasure of the elected president are "presidential"; systems in which the government must respond both to a TFgMffive^
In a parliamentary system the legislative assembly can dismiss the
18 This criterion coincides almost perfectly with the mode of selection of the government: by legislatures in parliamentary systems, by voters (directly or indirectly) in presidential systems. For a review of the differences, see Lijphart (1992J.
30
Distinguishing among Democracies and Dictatorships
government, whereas under a presidential system it cannot.19 Some institutional arrangements, however, do not fit either pure type; they are "premier-presidential," "semi-presidential," or "mixed," according to.different terminologies. In such systems, the president is elected for a fixed term and has some executive powers, but the government serves at the discretion of the parliament. These "mixed" systems are not homogeneous: Most lean closer to parliamentarism insofar as the government is responsible to the legislature; others (notably Portugal between 1976 and 1981) grant the president the power to appoint and dismiss governments (Shugart and Carey 1992).
Whereas we observed 105 democratic regimes, two countries changed institutional arrangements without breaking the continuity of democracy: Brazil twice and France once. Hence, altogether there were 108 democracies, as distinguished by institutional framework. Of these, 55 were parliamentary, 9 were mixed, and 44 were presidential. Of the total 1,723 years of democracy, countries spent 1,085 (63 percent) years under parliamentarism, 150 (9 percent) under mixed systems, and 488 (28 percent) under presidentialism.
To distinguish among dictatorships, we developed three alternative typologies and examined the relationships among them.
First, some dictatorships are "mobilizing," whereas others are "exclusionary." The former organize permanent political participation through a single or dominant party and regularly hold acts of popular mobilization that they call "elections." They require individuals to manifest loyalty to the regime by participating. The latter form, exclusionary dictatorships, may or may not hold elections, but they do not promote any kind of political participation by the masses. They only require that individuals not engage in acts oriented against the regime.
Operationally, we define as "mobilizing" those dictatorships that had at least one political party, and as "exclusionary" those that did not have parties (MOBILIZE). Altogether we observed 147 mobilizing and 127 exclusionary dictatorships. Of the 3,007 dictatorial years, 1,991 (66 percent) were spent under mobilizing dictatorships, and 1,016 (34 percent) under exclusionary ones.
19 The Chilean 1S91-1925 democracy does not fit this classification. Although it was popularly called "parliamentary," that was a misnomer. The Chilean lower house frequently censured individual ministers, but could not and did not remove the government or the chief executive, the president. In parliamentary systems, except for some early rare cases, the responsibility of the government is collective.
31
Democracies and Dictatorships
Second, we distinguish authoritarian regimes according to the number of formal powers: executives, legislatures, and parties (DIVIDED). Our intuition is derived from Machiavelli: Whenever decision-making is collective, there must exist some rules organizing the functioning of the government (Bohbio 1989). Hence, even if the legislature is a rubber stamp or the chief executive obeys dictates of the single party, the mere existence of such bodies means that there must, exist some formal rules allocating functions and specifying procedures. We are not claiming, as Kavka (1986) would, that divided governments are necessarily limited: Under dictatorship, some of these bodies may have no autonomous power and do not provide checks and balances. But the existence of rules distinguishes such regimes from those dictatorships in which the operation of government need not be organized by any formal rules.
The "powers" are always the chief executive and, if they exist, legislatures or parties. Hence, a "divided" dictatorship is one that in addition to the chief executive has a legislature or a party. "Monolithic" dictatorships have no legislatures and no parties. We observed 167 divided and 91 monolithic dictatorships, with 2,407 (80 percent) years spent under the former, and 600 (20 percent) under the latter.
Finally, another distinction, in the spirit of Montesquieu, is whether the dictatorship codifies and announces the rules it intends to apply to its subjects or governs without such rules (AUT). In the first case, rule is exercised "by fixed and established laws," whereas in the second case "a single person directs everything by his own will and caprice." "Bureaucracies" are dictatorships that have some internal rules for operating the government, at least rules regulating the competence of the chief executive vis-a-vis the legislature, and some external rules, namely, laws. To put it differently, bureaucracies are institutionalized dictatorships. Operationally, bureaucracies are dictatorships that have legislatures: Because all regimes have chief executives, the existence of a legislature implies that they must have some rules for regulating relations among different organs of government. It also indicates that rule is exercised by law, that is, that people can know the rules that dictators at least intend to enforce at a particular moment and, moreover, that these rules are universalislic in intent. In'turn, "autocracies" are despotic or, in the language of Linz (1975), "sultanistic" regimes, which have neither internal rules of operation nor publicly announced univcr-salistic intentions. Operationally, autocracies are systems in which there is a chief executive, and perhaps a single party, but no legislature. Yet some of the autocracies and bureaucracies that result from the applica-
32
What Wc Did Not Include
tion of these rules are, in fact, transitional regimes. We corrected for these regimes in the list presented in Appendix 1.2. Altogether, we observed 146 bureaucracies, which lasted 2,117 (70 percent) years, and 116 autocracies, during 890 (30 percent) years.
What We Did Mot Include
This conception of democracy in terms of contested elections for executive and legislative offices is clearly minimalist. Hence, it may be useful to make explicit at least some of the features that we did not consider when classifying regimes as democracies or dictatorships.
First, we do not include in our conception of democracy any social or economic aspects of a society. Many scholars (Weffort 1992) and, as surveys from many countries demonstrate, most citizens perceive social or economic equality as an essential feature of democracy. Yet the questions whether or not, on the one hand, contested elections tend to generate equality in the social or economic realm (Jackman 1974; Muller 1988) and, on the other hand, whether or not economic equality makes democracy more durable (Muller 1988) are just too interesting to be resolved by a definitional fiat. We prefer to define democracy narrowly and to study its causes and consequences.
Second, we do not think that "accountability," "responsibility," "responsiveness," or "representation" should be treated as definitional features of democracy. When Dahl (1971: 1) says that "a key characteristic of a democracy is the continued responsiveness of the government to the preferences of its citizens," or when Riker (1965; 31) asserts that "democracy is a form of government in which the rulers are fully responsible to the ruled," they mean either (1) that when and only when the government is responsive, the regime is democratic regardless of anything else, or (2) that if a system is democratic by some other criteria, then the government will behave responsibly. The standard way of thinking follows Dahl, who lists several conditions that are necessary and sufficient for governments to be responsive. And it is the presence of these conditions, not responsiveness, that defines a regime jg_dwr^CTatic: The staj£ment...that^'4-f~tto then governments will be responsive" is a theorem, not a definition. Moreover, this theorem is most likely false unless additional conditions are specified: First, the very notion of "responsiveness" or "accountability" is muddled, and second, probably only some otherwise democratic governments are "accountable" in any intuitive sense of this term (Przeworski, Manin, and Stokes 1999). Hence, the question whether or
33
Democracies and Dictatorships
not regimes characterized by freedom of opinion, widespread participation, and repeated elections are in fact responsive is best left open for investigation, rather than resolved by definition.
Third, whereas some degree of political freedom is a sine qua non condition for contestation, democracy cannot be sufficiently defined in terms of "liberties" or "freedom," or human rights, which underlie the Gastil (1980, 1990) or the Freedom House scales. The American conception of "freedom" perceives it as a condition, not as a predicate of actions: People are free, even if they never exercise their freedom. Thus U.S. citizens are free to form political parties; yet they almost never form them. They are certainly free to vote, yet about half do not. From our point of view, to paraphrase Rosa Luxemburg, the point is not to be free but to act freely. And acting freely in the political realm entails enabling conditions, institutional as well as social. Whereas democracy is a system of political rights - these are definitional - it is not a system that necessarily furnishes the conditions for effective exercise of these rights.20 Thus, assessing "freedom" or "liberty" without determining the conditions that enable its exercise can easily lead to ideologically motivated labels that measure only similarities to the United States, rather than the actual exercise of political rights.
Fourth, we do not include participation as a definitional feature of democracy, [n Dahl's conception of "polyarchy," both contestation and participation are necessary to classify a regime as democratic. Indeed, Dahl sets the participation threshold so high that by his criterion the United States would not qualify as a democracy until the 1950s. Vanhanen (1992) sets it lower, but still disqualifies as democracies those regimes in which elections are contested but participation is very limited. Yet we want to distinguish regimes in which at least some, but not necessarily all, conflicting interests contest elections. And empirical evidence from Western Hurope (Przeworski 1975) as well as from Latin America (Coppedge 1992) indicates that the distribution of votes among parties changes only slowly after each extension of suffrage, implying that even when suffrage is highly restricted, divergent interests are being represented. Moreover, we want to bo able to test theories about the effects of participation on the performance
and the durability of democracy (Huntington T96870'üonnell 1973;
20 Mueller's libertarian view, "political equality is something that evolves without much further ado when people are free" (1992: 988), should thus be contrasted with J. S. Mill's insistence that "high wages anil universal reading are the two elements of democracy" (quoted in Burns 1969: 290).
What We Did Not Include
Huntington and Nelson 1976). Using any threshold would produce a censored sample and a bias we prefer to avoid.21
Fifth, as long the chief executive and the legislature were elected in contested elections, we did not delve further into civil-military relations. Several distinctions coidd be made here. In some regimes that we classified as democracies, civilian institutions do not control the military, who in turn do not intervene in politics. In other democracies, civilian politicians use the threat of military intervention in strategic interactions among themselves ("praetorian politics"). Finally, in some democracies (Honduras and Thailand are prototypes), civilian rule is but a thin veneer over military power, exercised by defrocked generals. Yet as long as officeholders are elected in elections that someone else has some chance of winning, and as long as they do not use the incumbency to eliminate the opposition, the fact, that the chief executive is a general or a lackey of generals does not add any relevant information. Most generals who get elected only because they are generals are eliminated from consideration on the basis of other rules. Some probably sneak through; there is no measurement without error.
Finally, several countries have been ravaged by civil wars: F.l Salvador, Guatemala, Uganda, and Sudan are obvious cases. If a regime is a set of institutions that regulate the relationship between the civil society and the state, then there can be no regime where there is no state. To varying degrees, the very question whether a regime is or is not democratic turns out to be irrelevant. Voting typically occurs in only some parts of such countries, and legislatures frequently turn out to be ineffective.22 We considered excluding such periods altogether. Yet the degree to which a civil war disrupts the normal functioning of the political system is difficult to assess.
Thus, to repeat, our approach is minimalist. We want to be able to examine empirically, rather than decide by definition, whether or not the repeated holding of contested elections is associated with economic performance. Democracy, to repeat, is a regime in which some governmental offices are filled by contested elections. Whether or not
'-Note that it we were to use Dahl's participation threshold of 50% of adults to qualify countries as democratic, we would date Western European democracies quite late: in the case of Belgium or France, after World War II. The proportion of the population age 20 years or older who could vote in the 1946 election, the last one before women got the right to vole, in Belgium was 45.5%, of whom 90.3% voted, which yields a participation rate of 41.1%. In France in 1936. 40.1% of those age 20 and over could vote, and the turnout was 84.4S'o, implying a participation rale of 33.7%.
1 Most of Banks's coding for "ineffective legislature" was due to civil wars.
34
35
Democracies and Dictatorships
regimes in which the rulers are elected tend to generate social and economic equality, control by citizens over politicians, effective exercise of political rights, widespread participation, or freedom from arbitrary violence, as well as economic growth, high employment, low inflation, and public services, should be studied empirically rather than decided definitionally.
Stability and Change of Political Regimes
Of the 141 countries that we observed for at least some time between 1950 and 1990,23 73 were already independent in 1950, and 68 gained sovereignty after 1950 (Table 1.2).
Independence, as we know, came in waves. From 1950 through 1959, eight new countries entered the world of sovereign states: Oman, Laos, Morocco, Sudan, Tunisia, Ghana, Malaysia, and Guinea. The largest wave occurred between 1960 and 1968, when forty-one new countries appeared in the world (with the single largest expansion having occurred in 1960, when seventeen African countries gained independence). The remaining nineteen countries became independent between 1969 and 1981, with the largest additions occurring in 1971, when Bangladesh and some of the Persian Gulf states were created (Bahrain, Qatar, and the United Arab Emirates), and in 1975, with the independence of the last Portuguese colonies in Africa (Angola, Cape Verde Islands, and Mozambique) and of the Comoro Islands, Suriname, and Papua New Guinea. The last country to become independent prior to 1990 was Belize, in 1981.
Because our observations begin with 1950, and about half of the countries became independent only later, the aggregate distribution of regimes over the years depends on two factors: changes of regimes in the already existing countries and the entrance of new countries into the world (Table 1.3).
Given that the number of countries changed over time, one should
23 Because our observations end with 1990, the data set does not include any of the countries that resulted from the breakup of the Soviet Union (Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova. Tajikstan, Turk-'.....inettfetanvllkramerand-^
Macedonia, Slovenia). Czechoslovakia (Czech Republic and Slovakia), Ethiopia (Eritrea), and Somalia (Somaliland). The Soviet Union. Yugoslavia, Czechoslovakia, Ethiopia, and Somalia, however, are part of our data set. Moreover, because no economic data were available, the data set does not include Albania, Antigua and Barbuda, Bhutan, Brunei, Cuba, Cyprus, the Democratic People's Republic of Korea (North Korea), Dominica, Kiribati, Lebanon, Libya, Maldives. Namibia, St. Kitts and Nevis, St. Lucia, St. Vincent and die Grenadines, and Sao Tome and Principe.
36
■q
L
				O		
		tn co		\o		
				in		
		cm'				
		co				cm t-.
co	cm			n.		
		00	co"			m
CT-"	ö	cm				on o .. m
rH	rH	iC	cm"	co		CO ,
m"	on	c\]				Tf co „ cm
CO		in	GO	„ cm		r—
	r-	cm	co	i-H cm	cm	rH ^
o © O © O O ■
* CO CO CO 00
O h rt< K      h m
- oc
cm cm
CT-,
h- rH . cm ZO rH
wiflinwifäLTwujwln
° °-       10 ° 00 ° ° o 1
m in m in un LO \D \D KC v.
F- Is- Is- 1^
'O \0 \c o
N 00 h OQ co O ON CT^ O O O © rH
O ON ^ CO
ttcococororocococncoco
		m		
		CO		
- TO				
CO				
rH		^*		
- cm		\o		
cm CO				
		ö		
- oo				
o cm ^				cm 1-,
		in	'O	
- co	CO			in „ ON c
ON cm tJ-	rH		rH	
f1- iro GO	cm"	CO		- m cc . CO
.. cm CO „ . - rH in CO cm CO				
	n." CO	., cm rH cm	cm	- cm o . o rH ^ r-,
^  X   ff1, M \ß N ?y
co co co ^     rj- -HH
CO CO CO CO CO CO CO h~ r-, t^. h- h> [>_
J5 iCJ i£? i£? ^ 10 ^ « *o uo to
ONONCT, ONC-. G>ONO^OSONON
00 ON cm in ON O cm OOhhhNcvj
cm co un \o h- oo ^o o O vC sO ON ON ON ON ON Cn On
6t>
.a
60
a
■a
60
a
60
.3
- o
ro im .
r-   - h~ .   . xt< „ cow   . o ^ m in
mißNSSCO^mOf-NCCOOOHMfOrCHC^^
Table 1.3. Democracies and dictatorships in old and new countries
All
Old countries
New countries
Entering sample
Year	Total	Total	Dem	Die	Total	Dem	Die	Total	Dem	Die
1950	73	73	35	38	0	0	0	0	0	0
1951	74	73	34	39	1	0	1	1	0	1
1952	74	73	35	38	1	0	1	0	0	0
1953	74	73	35	38	1	0	1	0	0	0
1954	75	73	34	39	2	1	1	1	1	0
1955	75	73	34	39	2	1	1	0	0	0
1956	78	73	34	39	5	2	3	3	1	2
1957	80	73	34	39	7	2	5	2	0	2
1958	81	73	36	37	8	1	7	1	0	1
1959	81	73	37	36	8	0	8	0	0	0
1960	98	73	39	34	25	3	22	17	3	14
1961	101	73	39	34	28	4	24	3	1	2
1962	108	73	36	37	35	6	29	7	2	5
1963	109	73	35	38	36	5	31	1	0	1
1964	112	73	34	39	39	6	33	3	1	2
1965	115	73	33	40	42	7	35	3	0	3
1966	119	73	34	39	46	7	39	4	1	3
1967	120	73	33	40	47	6	41	1	0	1
1968	122	73	31	42	49	7	42	2	1	1
1969	122	73	31	42	49	5	44	0	0	0
1970	123	73	31	42	50	6	44	1	0	1
1971	127	73	32	41	54	6	48	4	0	4
1972	127	73	32	41	54	5	49	0	0	0
1973	128	73	31	42	55	6	49	1	1	0
1974	130	73	32	41	57	7	50	2	1	1
1975	136	73	33	40	63	9	54	6	2	4
1976	137	73	32	41	64	9	55	1	0	1
1977	138	73	31	42	65	9	56	1	0	1
1978	139	73	31	42	66	10	56	1	1	0
1979	139	73	34	39	66	11	55	0	0	0
1980	140	73	33	40	67	12	55	1	1	Ü
1981	141	73	33	40	68	12	56	1	1	0
1982	141	73	34	39	68	12	56	0	0	0
1983	141	73	37	36	68	11	57	0	0	0
1984	141	73	39	34	68	12	56	0	0	0
1985	141	73	40	33	68	11	_57____	_0____	_0-________	0
-4986—	-141-	16	42	31	68	13	55	0	0	0
1987	141	73	42	31	68	13	55	0	0	0
1988	141	73	44	29	68	14	54	0	0	0
1989	141	73	45	28	68	13	55	0	0	0
1990	141	73	48	25	68	12	56	0	0	0
Note: Sec Appendix 1.2 for country codes.
39
Democracies and Dictatorships
expect that the number of regime transitions would vary as well. As a null hypothesis, suppose that the probability of any democracy or dictatorship dying during a particular year is constant over time and is equal to the average rate over the entire period: 0.0261 for democracies and 0.0173 for dictatorships. The expected number of transitions from a given type of regime is then the product of the number of such regimes and their probability of dying. Comparison of the expected and the observed numbers of transitions suggests that until 1961 there were more transitions to democracy and fewer transitions to dictatorship than one would expect to happen by chance. The period from 1958 to 1973 was hostile to democracies; more of them died and fewer were born than one would expect. Finally, the post-1974 period was again favorable to democracies and hostile to dictatorships.
This periodization is due to Huntington (1991), according to whom (1) the "second wave" of democratization began in 1943 and ended in 1962, (2) the "second reverse wave" started in 1958 and ended in 1975, and (3) the "third wave" of democratization began in 1974. Note that Huntington refers to these patterns as "waves." If all he means is that the frequency of regime transitions was not the same during the entire period, then his observation stands. But were these "waves" in the sense that each transition made it more likely that another transition in the same direction would follow?
Consider first the seventy-three countries in our sample that were independent in 1950, when thirty-five of them (48 percent) were democratic. By 1960 the number of democracies among these countries increased to thirty-nine, only to fall to thirty-one by 1968. It was still thirty-one in 1978, after which it climbed back to thirty-nine in 1984 and to forty-eight in 1990. Hence, with regard to the "old" countries, those countries that were independent in 1950, our count roughly agrees with Huntington's analysis. But even among "old" countries, the waves depicted by Huntington are less general than they first appear. Fluctuations in the distribution of democracies among the countries that were independent in 1950 can be observed only in Latin America. In 1950, eight of the eighteen Latin American countries were democratic; that number went down to six in 1955, up to twelve in 1959, when- l reiiuiiiied urtil aJHTtlown agaThT» nWln~T976. ihe
number of democracies went up again after that, to reach fourteen in 1986, where it remained until 1990. Outside Latin America, the proportion of democracies remained relatively constant, around 48 percent, from 1950 through 1985 (Figure 1.1). Only after that year, as
40
Democracies and Dictatorships
a result of the democratization of Bangladesh, Pakistan, the Philippines, South Korea, Poland, Bulgaria, Czechoslovakia, and Hungary, did the number of democracies in the "old" countries outside Latin America start to increase: to 51 percent in 1985, to 55 percent in 1986, and to 62 percent in 1990 (Figure 1.2).
The story for countries that became independent after 1950 is entirely different. Three out of twenty-five (12 percent) newly independent countries were democratic in 1960; subsequently the numbers were seven out of forty-nine (14.3 percent) in 1968, ten out of sixty-six (15.1 percent) in 1978, and twelve out of sixty-eight (17.6 percent) in 1990. Hence, the proportion of democracies among these "new" countries grew slightly but steadily over the period. In turn, the decline of the aggregate proportion of democracies in the world during the 1960s was largely due to the emergence of new countries rather than to transformations of old ones. The "reverse wave" covering the 1960s was mostly due to the fact the number of countries increased dramatically in the 1960s, and a large proportion of the new countries entered the world as dictatorships. Thirty-two of the forty-one countries that became independent between 1960 and 1968 did so under authoritarian regimes.24 With a few exceptions, most of them remained authoritarian for the rest of the period. Hence, the waves of democratization and authoritarianism are at most limited to "old" countries, and particularly to Latin America.
Another way to examine whether or not regimes come and go in waves is simply to observe transitions by year. If transitions come in waves, we would expect that in each successive year more countries would transit in one direction, cresting at some peak, and then initiating the same pattern going in the other direction.
The annual frequency of transitions presented in Figure 1.3 does not show such a pattern. Between 1950 and 1961 (the tail end of Huntington's second wave of democratization) there were eleven transitions to democracy, but also nine to dictatorship (Table 1.4). Between 1958 and 197.3 (during the "reverse wave") there were twenty-five transitions to dictatorship, but also sixteen to democracy. Finally, between
24 The countries that emerged as dictatorships after independence are Algeria. Benin, Botswana, Burkina Faso, Burundi, Cameroon, Central African Republic, Chad, Gabon, Gambia, Ivory Coast, Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritania, Niger, Rwanda, Senegal, Swaziland, Tanzania, Togo. Uganda, Zaire, Zambia, Zimbabwe, Guyana, Singapore, Yemen Arab Republic, Western Samoa, and Kuwait. The countries that emerged as democracies are Congo, Mauritius, Nigeria, Sierra t.eone, Somalia, Barbados, Jamaica, Trinidad and Tobago, and Malta.
42
Stability and Change of Political Regimes
Table 1.4. Expected and observed numbers of transitions, by period
Periods	Transitions to democracy		Transitions to dictatorship	
	Expected	Observed	Expected	Observed
1950-1961	8.68	11	11.43	9
1958-1973	19.98	16	16.29	25
1974-1990	26.75	32	21.40	15
1974 and 1990 there were many more transitions to democracy, thrity-two, but still fifteen in the other direction. A simple statistical test is to correlate the annual numbers of transitions to democracy and to dictatorship. If regimes come in waves, then this correlation should be highly negative: During the years when there are many transitions in one direction, there should be few in the other, and vice versa. Yet the correlation is almost zero: -0.01. Moreover, the numbers of transitions in the two directions are not even autocorrelated: The annual autocorrelation of the number of transitions to democracy is -0.011, and to dictatorship -0.010.2:' Hence, however the waves roll, there is always an undertow pushing in the reverse direction. Huntington's oceanic metaphor is only that.
Dictatorships, on average, lasted longer than democracies. But because some regimes already existed in 1950 and some lasted beyond 1990, these averages could reflect the timing of their emergence as much as their ability to survive (Table 1.5).
Democracy is a phenomenon of the twentieth century. For that reason, in 1950 the democracies tended to be younger than the dictatorships. The average age of the thirty-four democratic regimes that existed then was 25.8 years. Only six of them - the United States,
A more complicated way of testing whether or not regime transitions come in waves is to analyze the autoregressive structure of the series using ARIMA identification methods. If the series are indeed cyclical, they should be at least of order 2, with a pair of imaginary roots. Because the series arc short, these tests are not highly reliable. Nevertheless, for all the three scries - transitions to democracy, transitions to dictatorship, and the difference between them - the tests indicate that they are of order 1. the seasonal component is negligible, and all the series pass the runs test. Hence, there is no evidence for cycles.
45
•
Democracies and Dictatorships
Table 1.5. Regime duration: average age in years at end of regime spell or last year of observation
All regimes
All
Democracy Dictatorship
Spells completed between 1950
and 1990 Spells in course by 1990 Age of regime as of 1950 Spells initiated in or after 1950 and completed by 1990 and in course by 1990
18.6
38.2
34.7 15.7
7.3
21.3
(97)« (141)
(70) (168)
(67) (101)
8.5 36.6 26.2 9.4 5.1 13.0
(45) (60) (34) (71) (33) (38)
27.4 (52)
39.4 (81)
42.7 (36)
20.3 (97)
9.4 (34)
26.2 (63)
Democracies
Parliamentary Mixed
Presidential
Spells completed between 1950
and 1990 Spells in course by 1990 Age of regime as of 1950 Spells initiated in or after 1950 and completed by 1990 and in course by 1990
11.2 (19)
41.9 31.2 10.7 4.5 15.0
(36) (21) (34) (14) (20)
4.5
28.8 7.0 9.7 4.5
16.7
(4)
(5)
(2) (7) (4)
(3)
9.4 (25)
24.4 20.2
8.2
5.3 11.2
(19) (11) 133) (17) (16)
Dictatorships
Spells completed between 1950
and 1990 Spells in course in 1990 Age of regime as of 1950 Spells initiated in or after 1950 and completed by 1990 and in course by 1990
Bureaucracy
20.9 (59)
Autocracy
9.3 (22)
26.0
42.1 12.7
7.7 19.3
(87) (28) (118) (67) (51)
13.9 37.7 7.3 6.3 11.7
(94) (8) (108) (87) (21)
1 Number of spells in parentheses.
France, Luxembourg, the Netherlands, Norway, and Switzerland36 -had been estabiished before 1900. In turn, the thirty-six dictatorships that existed in 1950 had an average ago of 43.8 years. Seventeen of them had been established prior to or in 1870: Ethiopia, Liberia, the
26 We extended the age of a regime back as far as 1870. All regimes, democratic or authoritarian, established before that date were recorded as having heen established in 1870. In Chile, democracy was first established in 1891, but there was a reversal in 1925.
Stability and Change of Political Regimes
Dominican Republic, El Salvador, Haiti, Honduras, Mexico, Nicaragua, Bolivia, Paraguay, China, Iran, Nepal, Thailand, Romania, Turkey, and the USSR (Russia).
However, democracies also lasted for shorter periods than dictatorships among the countries that did not exist prior to 1950. Of the 168 regimes that, were established in or after 1950, 67 had died by 1990. Thirty-three were democratic, and they lasted, on average, 5.1 years; 34 were authoritarian and lasted 9.4 years. Of the 101 remaining regimes, that is, the regimes that lasted beyond 1990, 38 were democracies, and 63 were dictatorships. The former, by 1990, had lasted 13 years, and the latter 26.2 years. Dictatorships, thus, tended to last longer than democracies, regardless of when they were observed.
During the 1950-1990 period, most countries each lived under a single regime. Of the 141 countries wo observed, only 41 experienced transitions between dictatorship and democracy. The remaining 100 countries never experienced regime transitions, and thus each ended the period with the same regime with which it was first observed (among these, 67 were dictatorships, and 33 were democracies). Seventeen countries had just one transition each, of which twelve were to democracy. The five countries where democracy gave way to dictatorships that lasted past 1990 are Laos, where, democracy fell in 1959, Congo in 1963, Sierra Leone in 1967, Somalia in 1969, and Sri Lanka in 1977. Countries that started the period of observation as dictatorships and then established democracies that lasted beyond 1990 are Colombia in 1958, Venezuela in 1959, the Dominican Republic in 1966, Portugal in 1976, Spain in 1977, El Salvador and Nicaragua in 1984, Bangladesh in 1986, Poland in 1989, and Bulgaria, Czechoslovakia, and Hungary in 1990."
Two countries began the period of observation as dictatorships, experienced brief democratic interludes, and became dictatorships again: Uganda (where democracy lasted for five years, from 1980 to 1984) and Indonesia (where democracy lasted for only two years, from 1955 to 1956). More typically, seven countries were democracies when wefiretj^seryedthem, went throtigk..QfterL.long--perwd-s-of--dietator-
ship, and returned to democracy. These are Grenada, Brazil, Chile, Ecuador, Uruguay, the Philippines, and Greece. Eight countries experienced three regime transitions: Nigeria, Panama, Suriname, and
We do not count East Germanv, which wc treat, as having dropped from the sample in 1990.
46
47
Democracies and Dictatorships
Stability and Change of Political Regimes
Myanmar started the period as democracies, became dictatorships, returned to democracy, and ended as dictatorships, whereas Bolivia, South Korea, Thailand, and Turkey began as dictatorships, experimented with democracy, returned to dictatorships, and became democracies again. Two countries had four transitions: Ghana and Pakistan. Sudan and Honduras had five, Guatemala and Peru had six, and Argentina, by far the record holder, had eight transitions between democracy and dictatorship.28
Thus, the regime histories of particular countries are highly heterogeneous. Most regimes, as we saw, lasted for a long time, with a majority of countries not experiencing any transition between democracy and dictatorship during the 1950-1990 period. Some countries alternated between dictatorship and democracy every few years: There would be a coup d'etat, and a dictatorship would be established: then, often following another coup d'etat, an election would be held and the democratically elected government would assume office, only to be overturned by yet another coup. In some countries this entire cycle occurred once during the period; in others it occurred twice, and in Argentina three times.
Moreover, systematic regional differences can be seen: Western Europe was predominantly democratic, and Eastern Europe was communist; in Africa, only Mauritius was democratic during its entire history; except for Israel, Middle Eastern countries were dictatorships; most of the Far Eastern countries, except Japan, were dictatorships; South Asian countries experienced some transitions; and many, but not all, Latin American regimes were highly unstable. Indeed, of the 97 transitions that occurred in the 141 countries between 1950 and 1990, 44 were in Latin America, which comprises eighteen countries29 (Table 1.6).
The fact that most countries each lived under the same regime for most of the time between 1950 and 1990 does not mean that their rulers or their political orientations or even their institutional frameworks remained the same. The democratic regimes
Appendix 1.3 lists countries by the number of transitions they have experienced ^(^ee^democracy..and.riic,taioj3j]ip_
1 The rale of transitions per country was highest in Latin America: 2.4. Latin America was followed by Southeast Asia. wrhere the rate was 1.57 transitions per country, and South Asia, where the rate was 1.2 transitions per country. In all other regions (including the countries of the Organization for Economic Cooperation and Development, the OECD), the rate of transitions per country was well below 1.
48
Table 1.6. Regimes and regime transitions by region
Kcgion
bub-Saharan Africa South Asia East Asia Snutbeast Asia Pacific Islands
Middle East and Vorth Africa
Latin America
Caribbean
Eastern Europe and
Soviet Union Industrialized countries Total
Years of democracy
Transitions to democracy
Years of dictatorship
Transitions to dictatorship
69 87
4 46 40 68
366 130
5
1,723
6 3 2 5 0 2 25
3 52
1.170 97 160 215 50 513 372 80 290
60 3.007
11
3
1 6 0
2 19
might be parliamentary, mixed, or presidential. Dictatorships, in turn, might be "bureaucracies," institutionalized regimes that promulgated laws, or "autocracies," regimes without any proclaimed rules. Using these distinctions, we observed 55 parliamentary, 9 mixed, and 44 presidential democracies, 146 bureaucracies, and 116 autocracies.
The staying power of democratic institutions was seen to be strong. During the entire period studied, democratic institutional frameworks were altered in only three instances: France in 1958, when the parliamentary system of the Fourth Republic gave way to the mixed system of the Fifth; Brazil in 1961, when presidentialism was replaced by a mixed system; and Brazil again in 1963, when presidentialism was restored after its overwhelming victory in a plebiscite held in January of that year. A few countries did change the institutional framework in their democracies after an authoritarian interregnum: Ghana, Nigeria, and South Korea replaced the parliamentary systems that had existed prior to their periods of dictatorship with presidential jryjrtmis once, _ __dmn£ura£y--w^-Heestet*edr-r^ was from parlia-
mentarism to a mixed system. Pakistan was the only country that went back and forth: from parliamentarism in 1950-1955 to a mixed system in 1972-1976 and back to parliamentarism again after 1988. All of the other seventeen countries that experienced at least one authoritarian
49
Democracies and Dictatorships
interlude went back to the type of democratic institutions that had existed before the authoritarian regime.3"
Authoritarian institutions, on the other hand, proved highly unstable. When we classify dictatorships according to the presence or absence of legislatures, we count 262 regimes, as opposed to 133 when we do not make any distinctions among them. There are, thus, 129 instances of openings and closings of legislatures (65 cases of closing, and 64 cases of opening). Again, a few countries account for a large proportion of the transitions from one type of authoritarianism to the other. Of the sixty-seven countries that remained under authoritarian regimes from 1950 through 1990, thirty-two experienced only one type of dictatorship: twenty-seven as bureaucracies and five as autocracies. The remaining thirty-five countries experienced seventy-six transitions, an average of 2.2 changes per country, from one type of authoritarianism to the other.31
Autocracies often emerge when democracy is overthrown and the legislature is temporarily or permanently closed: Of the forty-five cases of democratic breakdown, thirty-one resulted in this type of dictatorship. But autocracies can also emerge as a result of abortive attempts to liberalize bureaucratic dictatorships. Indeed, the cases in which an autocracy followed a bureaucracy were most frequent, suggesting that attempts at liberalization often fail: Of the eighty-seven instances in which bureaucratic regimes died, twenty-two ended in democracy, but sixty-five in autocracy.
Autocracy is not an easily sustainable form of authoritarianism. Only the four Persian Gulf monarchies (Oman, Qatar, Saudi Arabia, and the United Arab Emirates) wero autocracies during the entire period. In genera], autocracies do not last very long: As Table 1.5 indicates, the average duration of autocratic spells completed by 1990 was 9.3 years, and of those still in course in 1990 the duration was 13.9 years, compared with 20.9 and 26 years, respectively, for bureaucracies. Moreover, of all the regimes, democratic and authoritarian, autocracies are the ones at highest risk: During any year, an autocracy has a 10.56 percent chance of experiencing a transition to a different regime, which compares with 5.12 percent for presidential democracies, 4.11 percent
3,1 These countries are Sudan, Grenada, Guatemala. Honduras, Panama, Argentina, Bolivia, Brazil, Chile, Ecuador, Peru, Uruguay, Myanmar, the Philippines, Thailand, Greece, and Turkey.
31 Benin, Burkina Faso, Morocco, Kuwait, and Jordan were the most unstable authoritarian regimes according to this measure: They changed between bureaucracy and autocracy four, five, five, six, and six limes, respectively.
50
Stability and Change of Political Leadership
Tabic 1.7. Transitions between political regimes: parliamentarism mixed presidenuahsm, bureaucracies, and autocracies
Transition to:
Transition from:
Pari    Mix Pres
Pari Mix Pres Bur Aut
Total
0
1
10 21
32
64 78
Aut	Total	Number of years	Probability
12	19	1,085	0.0175
2	4	150	0.0267
17	25	488	0.0513
65	87	2,117	0.0411
—	94	890	0.1056
96	223	4,730	0.0471
for bureaucracies, 2.66 percent for mixed democracies, and 1.75 percent for parliamentary democracies (Table 1.7).
Stability and Change of Political Leadership
Rulers changed within each regime. By "rulers" we mean the chief executives, to whom we refer as "heads" of government, or simply "heads." These are presidents in presidential democracies, prime ministers in the parliamentary and mixed democracies, and whoever is the effective ruler in dictatorships. The latter sometimes can be designated explicitly as dictators, or they may opt for a variety of other titles: heads of military juntas, presidents, leaders of their ruling parties, executors of the state of emergency, or kings.
No changes of heads occurred during 3,927 years, one change occurred in 615 years, two changes in 101 years, three in 14, four in 3, and five in 2 years.32 Thus, altogether there were 881 changes of heads during the period we observed, once every 5.29 years (Table 1.8). Changes were more frequent in democracies than in dictatorships. Chief executives in democratic regimes were changed once every 3.48 years, with no significant difference between prime ministers (3.41 years when we combine parliamentary and ,mjxgjLl3a3nfisl3-LAntl _
" ™s acids lo 4,662 years. The difference from the total of 4,730 is due to the exclusion o Switzerland, Uruguay up to 1966, and Yugoslavia after 1980, each of which had a collective executive.
" ^?^ate'y• ^ fCrage iS 377 f0r prime mi"isters 1» Parliamentary- regimes and 2.03 tor prime ministers in mixed regimes.
51
Democracies and Dictatorships
Stability and Change of Political Leadership
Table 1.8. Distribution of changes of chief executives (HEADS) by regime'"
Number of changes of heads by year	Dem	Pari	Mix	Pres	Diet	Bur	Aut	Total
0	1,254	838	89	327	2,673	1,935	738	3,927
1	354	212	50	92	261	146	115	615
2	49	31	9	9	52	23	29	101
3	6	2	2	2	8	1	7	14
4	2	2	0	0	1	0	1	3
5	0	0	0	0	2	2	0	2
Total	1.665	1,085	150	430	2,997	2,107	890	4,662
No. of changes	478	288	74	116	403	205	198	881
Average duration	3.48	3.77	2.03	3.71	7.44	10.28	4.49	5.29
1 Excludes Switzerland, Uruguay until 1966, and Yugoslavia after 1981 because they had i leclive executives.
presidents (3.71 years). The difference, however, was large across types of dictatorships: Whereas in autocracies we observed one change of chief executive every 4.49 years, in bureaucracies we observed one change every 10.28 years, with an average for all dictatorships equal to 7.44 years.
Some incumbents experienced changes in their political regimes while in office. This happened during the tenure of sixty-eight chief executives; fifty-four of them survived one change of regime, nine survived two changes, three survived three changes, and one each survived five and six changes. Most of these changes were between different types of dictatorships and were due to the opening and closing of legislatures.34 A few, however, were from democracy to dictatorship or vice versa: for example, from presidentialism to autocracy in Uruguay under Juan Bordaberry in 1973, and from autocracy to pres-
34 The most extreme case is Jordan, where the legislature was closed for one year in
-19^6^-as-weH-as-during-1^74 19£4-anri again dur4ng-4-9&5-4-9&9--r<*presentmg six--
regime changes from bureaucracy to autocracy. Other cases of frequent changes of regimes without a change of chief executive are a.s follows: Morocco, where the legis lature was closed during 1963-1965, 1970-1972, and in 1978; Burkina Faso, where Sangoule Lamizana allowed an elective legislature to convene in 1970 only to close it in 1974 and reopen it in 1978: Laos, where Souvanna Phouma closed the legislature two times, in 1966 and 1974; and Nepal, where King Mahendra experimented with legislative bodies in 1959 and 1963.
52
Table 1.9. Average duration (in years) of chief executives' spells" bv regime '
Type of spell
.All
Censored6 Not censored
Regime change" No regime change
No regime change and not censored Democracies Parliamentary Mixed
Parliamentary and mixed
Presidential Dictatorships
Bureaucracies
Autocracies Regime change and not censored Regime change and censored No regime change and censored
Average	Maximum	N
5.8	44	858
8.1	39	138
5.4	44	720
13.3	39	68
5.2	44	790
4.9	44	671
3.7	23	395
3.9	23	242
2.3	7	57
3.6	23	295
4.2	12	96
6.6	44	276
8.0	36	167
4.5	44	109
11.8	38	49
17.2	39	19
6.6	31	119
" Continuous years of ruling by the same person. " Spells in course by 1990.
' poutical^me85™0 * ^      mclImbent chanKed the type of
identialism in Nicaragua under Daniel Ortega Saavedra in 1984. Once we take those two facts into consideration, we find that prime ministers in both parliamentary and mixed regimes had the shortest average tenure (3.6 years), that the durations of democratic presidents and autocratic rulers wore about the same (4.2 and 4.5jjiansj^sp^ctiydyj,.. and that the chieTexecutives in bureaucracies were the ones who lasted the longest (8 years) (Table 1.9).
A similar conclusion follows when we examine the rates of leadership turnover, defined as the annual number of changes in chief executive accumulated over the life span of a regime. As a benchmark, consider that the average turnover rate for all the countries we
53
Democracies and Dictatorships
Table 1.10. Leadership turnover rates by regime type"
Regime	Turnover rate
Democracy	0.164
Parliamentary	0.140
Mixed	0.271
Presidential	0.186
Dictatorship	0.073
Bureaucracy	0.047
Autocracy	0.190
All	0.103
0 Excluding spells in course in 1990.
observed was 0.17, somewhat less frequently than once every five years. Because, on the average, we observed each country for about thirty-three years, this turnover rate corresponds to an average of about six changes of chief executive per country.35 When we consider leadership turnover rates across political regimes, we find a similar rate only in democracies, where it is 0.16 (Table 1.10). The turnover rate is higher in autocracies and in mixed presidential democracies. These rates imply that an average democratic spell experiences about eighteen changes of leadership, whereas an average authoritarian spell experiences nine changes. Among democracies, the number of changes of heads is thirteen in the average parliamentary regime, and six in the average presidential regime.
Conclusion
These are, then, the basic facts about political regimes in the world between 1950 and 1990. Democracy is a system in which incumbents lose elections and leave office when the rules so dictate. Dictatorships -ar-e-a-rosiduai categor^lf-a-^aa^r^inw-«-n«^&mM-ratic, wo con-
sider it to be a dictatorship of one stripe or another. Moreover, we do not distinguish between dictatorships that succeed one another.
35 In this case, changes in chief executives were accumulated over the entire period during which we observed each country, regardless of political regime.
Appendix 1.1: Alternative Approaches
In most cases it is simple to apply this conception of democracy to ' classify the regimes that existed in the particular countries at the particular moments. All one needs to do is to observe whether or not the chief executive was elected, whether or not the legislature was elected, and whether or not there was political opposition. In some cases, however, history did not provide the necessary evidence: There was an opposition, officials were elected, but the same party always won. All one can do in such instances is to decide which error to avoid.
The resulting classification of regimes is not idiosyncratic. Whereas we were concerned to justify our approach theoretically and to ground the classification on observations, rather than judgments, our classification is almost identical with those produced by several alternative scales of democracy (see Appendix 1.1). Indeed, it seems that in spite of all their conceptual and observational differences, the various approaches yield highly similar classifications of regimes. Hence, there is no reason to think that the results that follow depend on the particular way regimes were classified.
In the chapters that follow, we first explain some of the patterns described here and then explore their consequences for economic performance and material well-being.
Appendix 1.1: Alternative Approaches
Conceptually, our scale is close to that of Bollen (1980), as well as that of Coppedge and Reinicke (1990). Bollen used four indicators: (1) whether or not elections were fair, (2) whether or not the chief executive was elected, (3) whether or not the legislature was elected, and (4) whether or not the legislature was effective. Coppedge and Reinicke coded answers to three questions: (1) whether or not elections presented voters with a meaningful choice, (2) whether or not the outcome was affected by significant fraud, and (3) whether all or some or no political organizations were banned. We used Bollen's second and third dimensions and Coppedge and Reinicke's third dimension. We did experiment with Banks's measure of legislative effectiveness, but found his assessments too uxreliable^IMs_cJe^
Iraud are even more frequent than its actual occurrence, and by all indications some fraud is a ubiquitous phenomenon in democracies. Screaming "Fraud!" is just part of the standard repertoire of democratic competition. Indeed, there are cases in which the opposition has withdrawn from the competition, claiming that the elections would
54
55
Democracies and Dictatorships
not be conducted fairly. We conclude that there is no way to assess the validity of such allegations in a standardized way. For example, the opposition decided not to contest the 1984 Nicaraguan elections, but some of its leaders later expressed regret about 1984 once they discovered that they had won the subsequent elections in 1990. Hence, although our approach is theoretically akin to those of Bollen and Coppedge and Reinicke, we have tried to the extent possible to avoid subjective judgments by relying only on observables. The Gurr (1990) measure in Polity II is conceptually somewhat different, because it considers the limited character of the government by coding "constraints on the chief executive." His assessments, however, are not easy to reproduce.
Although we have been careful to specify our understanding of democracy and to distinguish it from some rival conceptions, it appears that from a practical point of view alternative measures of democracy generate highly similar results. The dimensions used to assess whether or not and to what extent a particular regime is democratic seem to make little difference.36 To cite Inkeles (1990: 5-6), "the indicators most commonly selected to measure democratic systems generally form a notably coherent syndrome, achieving high reliability as measurement scales. ... A testimonial to the robustness of the underlying common form and structure of the democratic systems is found in the high degree of agreement produced by the classification of nations as democratic or not, even when democracy is measured in somewhat different ways by different analysts. .. . Thus Coppedge and Reinicke, following a quite independent theoretical model, end up with a scale of polyarchy which correlates .94 with Gastil's civil liberties measure for some 170 countries in 1985. Gurr's measure performs similarly in relation to Bollen's [and] his ratings of 118 countries circa 1965 correlate .83 with Bollen's measure and .89 with a score combining Gastil's separate measures of political and civil liberties for 113 countries in 1985."
Our measure is no exception. The Coppedge-Reinicke scale for 1978 predicts 92 percent of our dichotomous regimes, the Bollen 1965 scale predicts 85 percent, and the Gurr scales of Autocracy and Democracy „4er-j^gO-4-9g6.jemtiy-^fe€to-^-l-pefeeftt^T}ie-6asa-sealo of political-
liberties, covering the period from 1972 to 1990, predicts 93.2 percent
36 Note, however, that different measures appear to be biased in somewhat different directions. See Bollen (1993).
56
Appendix 1.1: Alternative Approaches
of our classification; his scale of civil liberties predicts 91.5 percent;. and the two scales jointly predict 94.2 percent of our regimes.37 Hence, our classification is by no means idiosyncratic. Different views of democracy, including those that entail highly subjective judgments, yield a robust classification.
The main difference between our approach and the alternatives is that we use a nominal classification, rather than a ratio scale. We believe that although some regimes are more democratic than others, unless the offices are contested, they should not be considered democratic. The analogy with the proverbial pregnancy is thus that whereas democracy can be more or less advanced, one cannot be half-democratic: There is a natural zero point. Note that Bollen and Jackman (1989) are confused: It is one thing to argue that some democracies are more democratic than others, but it is another to argue that democracy is a continuous feature over all regimes, that is, that one can distinguish the degrees of "democracy" for any pair of regimes.38
Bollen and Jackman (1989: 612) argue that difficulties in classifying some cases speak in favor of using continuous scales: "Dichotomizing democracy," in their view, "blurs distinctions between borderline cases." Yet why are there "borderline cases"? Suppose that wo have defined democracy and not-democracy, established operational rules, and found that some cases cannot be unambiguously classified by these rules. Does this mean that there are borderline cases and that democracy is thus "inherently continuous"? And should we stick the cases that cannot be unambiguously classified, given our rules, into an "intermediate" category, halfway between democracy and dictatorship? That view strikes us as ludicrous. If we cannot classify some cases given our rules, all this means is that we either have unclear rules or have insufficient information to apply them.
We have already seen that some "borderline cases" constitute sys-
liecause other scales are ordinal (and pretend to be cardinal), whereas ours is nominal, we use probit maximum likelihood to predict our classification on the basis of these scales,
' They also argue by assertion, referring to "the inherently continuous nature of the concept of political democracy* (1989: 612). claiming that "since democracy is con-
ceptualiy continuous, it is best measured" in continuous terms" (p. 612), and that "democracy is always a matter of degree" (p. 618). Hence, in their view, the "degrees of democracy" in Mexico, in Salazar's Portugal, and in Franco's Spain were different. How they decide that "democracy is conceptually continuous," whatever that moans, remains mysterious, but we are admonished that "it is important thai the measurement history of this construct not repeat itself" (p. 612).
57
Democracies and Dictatorships j
1
tematic error, whereas others bring random error. Systematic errors | can be treated by explicit rules, such as our "alternation" rule, and ^ their consequences can be examined statistically. There are some t regimes that cannot be unambiguously classified on the basis of all " the evidence produced by history. Because history produces a biased ~ sample of democracies - sampling is endogenous (Pudney 1989) - we must revert to counterfactual judgments. In such cases we must decide ; which error we prefer to avoid: classifying as democracies regimes that may not be democracies, or rejecting as democracies regimes that may in fact qualify. Yet, once this decision is made, the classification is unambiguous. Mexico is not a regime intermediate between democracy and dictatorship, not a "borderline case." It is a regime in which the ruling party allows some contestation but always wins: either a democracy or a dictatorship, depending in which direction one wants to err systematically.
In turn, some errors that are random with regard to the rules will remain, and we will have to live with them. But errors are errors, not "intermediate" categories. And there are no grounds to think that .; a finer classification would be more precise. A finer scale would generate smaller errors, but more of them, and a rougher scale would generate larger errors, but fewer of them. And if errors of larger mag- ; nitudo are less likely, the dichotomous scale will have a lower expected ■;■ error.
Suppose that the true nature of democracy lies on a ./-point scale,
j = 1.....J, but its measurement is subject to error. Let the unobserved
true score be £>T and the assigned value D, and let the probability of a j-point error be P(j) = Pr{\D - Dr\ = j} = a'. The reliability of the scale is then Pr{\D - DT\ = 0} = 1 - Pr(/). Assume that the distribution of the true observations is uniform. Then the expected value of the error will be
E(\D - J5t|) = £;.i Pr( j) * j * 2{J - ;),
where the first factor is the probability of an error of a given magnitude, the second factor is the magnitude, and the third is the number : of such errors. Assume, as an illustration, that the probability of making an error of magnitude 1 is a = 0.2, so that ?r{j = 0) = 0.75. Suppose that this is a Gastil scale, with seven points. Then the expected error for seven observations will be about 3.5. I
Now dichotomize this seven-point scale in such a way that if D < 4,
Appendix 1.2: Classification of Regimes
then the assigned score is D - 2.5 (which is the midpoint value for one regime), and if D > 4, then the assigned score is D = 5.5 (midpoint for' the other regime), so that each error costs three points on the seven-point scale. Let the probabilities of errors and the distributions of the true scores on the seven-point scale be the same. Then the expected value of the error is
E(\D - DT\) = '£j_1?nj)*3*2ld*j + (l - d)(J - j)l
where the last factor in each expression is the number of relevant errors (e.g., the only relevant one-point error is between 4 and 5, and there are two of them, misclassifying 4 as 5 or 5 as 4), and d = 1 if j < 4, and d = 0 otherwise. At a = 0.2, the expected error for seven observations of a dichotomous scale will be about 2.
Hence, there is less measurement error when a dichotomous scale is used. If the distribution of true observations is unimodal and close to symmetric, a more refined classification will have a smaller error, but in fact observations on all the polychotomous scales tend to be U-shaped, which advantages a dichotomous classification even more than our example with the uniform distribution.
In sum, wo think that our classification has some advantages. First, it is grounded in theory. Second, it is based exclusively on observed facts. Third, it separates cases subject to systematic error. Fourth, it contains less random error than polychotomous scales. Finally, it covers every year for 141 countries during forty-one years.
Appendix 1.2: Classification of Political Regimes, 1950-1990
At least some of the years for regimes marked with asterisks have been classified as bureaucracies on the basis of our "alternation" rule.
Country Regime Entry Exit
1. Algeria                              Bureaucracy 1962 1964
Autocracy 1965 1976
Bureaucracy 1977 1990
-2^-Angela-----Autocracy   -----1975-IWt
Bureaucracy 1980 1990
3. Benin                                  Bureaucracy 1960 1964
(continued.)
59
Country
Democracies and Dictatorships Regime Entry Exit
4. Botswana
5. Burkina Faso
6. Burundi
7. Cameroon
8. Cape Verde
9. Central African Republic
10. Chad
11. Comoros
12. Congo
13. Djibouti
14. Egypt
15. Ethiopia
Id. Gabon
17. Gambia
18. Ghana
Autocracy
Bureaucracy
Autocracy
Bureaucracy*
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy*
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy
Autocracy
Autocracy
Bureaucracy
Presidenti allsm
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy
Bureaucracy*
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy*
1965
1979 1990
1966 1960 1966
1970
1974 1978
1980
1962 1966 1982 1987 1960
1971
1973
1975 1960 1966 1987 1960 1975 1975
1978 1960
1963 1977
1979 1977 1950 1950 1957
1974 1987 1960
1978
1989
1990 1990 1965
1969
1973 1977
1979 1990 1965 1981 1986 1990
1970
1972 1990 1990 1965 1986 1990
1974 1990
1977 1990 1962 , 1976
1978 1990 1990 1990 1956
1973 1986 i 1990 1990
Appendix 1.2: Classification of Regimes Country
19. Guinea
20. Guinea-Bissau
21. Ivory Coast
22. Kenya
23. Lesotho
24. Liberia
25. Madagascar
26. Malawi
27. Mali
28. Mauritania
29. Mauritius
30. Morocco
31. Mozambique
32. Niger
33. Nigeria
Bureaucracy*
Bureaucracy
Autocracy
1965 1957 1965
(continued)
34. Rwanda
Regime	Entry	Exit
Parliamentarism	1970	1971
Autocracy	1972	1978
Presidentialism	1979	1980
Autocracy	1981	1990
Bureaucracy	1958	1983
Autocracy	1984	1990
Bureaucracy	1974	1990
Bureaucracy*	1960	1990
Bureaucracy	1963	1990
Bureaucracy	1966	1969
Autocracy	1970	1990
Bureaucracy	1950	1979
Autocracy	1980	1984
Bureaucracy	1985	1989
Autocracy	1990	1990
Bureaucracy*	1960	1971
Autocracy	1972	1976
Bureaucracy	1977	1990
Bureaucracy	1964	1990
Bureaucracy	1960	1967
Autocracy	1968	1981
Bureaucracy	1982	1990
Bureaucracy	1960	1977
Autocracy	1978	1990
Parliamentarism	1968	1990
Autocracy	1956	1962
Bureaucracy	1963	1964
Autocracy	1965	1969
Bureaucracy	1970	1971
Autocracy	1972	1976
Bureaucracy	1977	1990
Bureaucracy	1975	1990
Bureaucracy	1960	1973
Autocracy	1974	1990
Parliamentarism	1960	1965
Autocracy	~T9B5"	I978~
Presidentialism	1979	1982
Autocracy	1983	1990
Bureaucracy	1962	1972
	(continued)	
61
Country
35. Senegal
36. Seychelles
37. Sierra Leone
38. Somalia
39. South Africa
40. Sudan
41. Swaziland
42. Tanzania
43. Togo
44. Tunisia
45. Uganda
46. Zaire
47. Zambia .....
48. Zimbabwe
49. Bahamas
50. Barbados
51. Belize
Democracies and Dictatorships Regime Entry Exit
]
■
Autocracy
Bureaucracy
Bureaucracy*
Bureaucracy
Parliamentarism
Autocracy
Bureaucracy
Mixed
Autocracy
Bureaucracy
Bureaucracy*
Parliamentarism
Autocracy
Parliamentarism
Bureaucracy
Autocracy
Parliamentarism
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy
Bureaucracy
Aulocracy
Bureaucracy
Bureaucracy
Bureaucracy
Autocracy
Presidentialism
Autocracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy
Bureaucracy Parliamentarism Parliamentarism Parliamentarism
1973 1981
1960 1976
1961
1967
1968
1960
1969 1979 1950 1956 1958 1965
1969
1985
1986 1989 1968 1973
1978
1961 1960 1967
1979 1956
1962 1971
1980 1985
1960
1961
1963
1970
1964
1965 1973
1966 1981
1980 1990 1990 1990
1966
1967 1990
1968 1978 1990 1990 1957 1964
1968
1984
1985 1988 1990 1972
1977 1990 1990 1966
1978 1990 1990 1970
1979 1984 1990 1960 1962
1969 1990 1990
1990 1990 1990 1990
Appendix 1.2: Classification of Regimes
(continued) J
Country	Regime	Entry	Exit
52. Canada	Parliamentarism	1950	1990
53. Costa Rica	Presidentialism	1950	1990
54. Dominican Republic	Bureaucracy	1950	1961
	Autocracy	1962	1965
	Presidentialism	1966	1990
55. El Salvador	Bureaucracy	1950	1959
	Autocracy	1960	1960
	Bureaucracy*	1961	1983
	Presidentialism	1984	1990
56. Grenada	Parliamentarism	1974	1978
	Autocracy	1979	1983
	Parliamentarism	1984	1990
57. Guatemala	Presidentialism	1950	1953
	Bureaucracy	1954	1957
	Presidentialism	1958	1962
	Autocracy	1963	1965
	Presidentialism	1966	1981
	Bureaucracy	1982	1985
	Presidentialism	1986	1990
58. Haiti	Bureaucracy	1950	1985
	Autocracy	1986	1989
	Bureaucracy	1990	1990
59. Honduras	Bureaucracy	1950	1955
	Autocracy	1956	1956
	Presidentialism	1957	1962
	Autocracy	1963	1964
	Bureaucracy*	1965	1970
	Presidentialism	1971	1971
	Autocracy	1972	1981
	Presidentialism	1982	1990
60. Jamaica	Parliamentarism	1962	1990
61. Mexico	Bureaucracy*	1950	1990
62. Nicaragua	Bureaucracy*	1950	1970
	Autocracy	1971	1971
	Bureaucracy*	1972	1978
	Autocracy	1979	1983
	Presidentialism	1984	1990
63. Panama	Presidentialism	1950	1950
	Bureaucracy	1951	1951
I
(continued)
63
Democracies and Dictatorships
Appendix 1.2: Classification of Regimes
Country	Regime	Entry	Exit	1	Country	Regime	Entry	Exit
	Presidentialism	1952	1967	-.	72. Guyana	Bureaucracy*	1966	1990
	Autocracy	1968	1977		73. Paraguay	Bureaucracy*	1950	1990
	Bureaucracy	1978	1990		74. Peru	Bureaucracy	1950	1955
64. Trinidad & Tobago	Parliamentarism	1962	1990			Presidentialism	1956	1961
65. United States	Presidentialism	1950	1990			Autocracy	1962	1962
66. Argentina	Presidentialism	1950	1954			Presidentialism	1963	1967
	Autocracy	1955	1957			Autocracy	1968	1979
	Presidentialism	1958	1961			Presidentialism	1980	1989
	Autocracy	1962	1962			Bureaucracy	1990	1990
	Presidentialism	1963	1965		75. Surinaine	Parliamentarism	1975	1979
	Autocracy	1966	1972			Autocracy	1980	1986
	Presidentialism	1973	1975			Bureaucracy	1987	1987
	Autocracy	1976	1982			Mixed	1988	1989
	Presidentialism	1983	1990			Bureaucracy	1990	1990
67. Bolivia	Bureaucracy	1950	1950		76. Uruguay	Presidentialism	1950	1972
	Autocracy	1951	1955			Autocracy	1973	1984
	Bureaucracy*	1956	1963			Presidentialism	1985	1990
	Autocracy	1964	1978	j	77. Venezuela.	Autocracy	1950	1951
	Presidentialism	1979	1979			Bureaucracy	1952	1958
	Autocracy	1980	1981	j		Presidentialism	1959	1990
	Presidentialism	1982	1990	1	78. Bangladesh	Autocracy	1971	1971
68. Brazil	Presidentialism	1950	1960	■ >s		Bureaucracy	1972	1974
	Mixed	1961	1962	3		Autocracy	1975	1978
	Presidentialism	1963	1963	J*		Bureaucracy*	1979	1981
	Bureaucracy	1964	1967			Autocracy	1982	1985
	Autocracy	1968	1969	Í		Presidentialism	1986	1990
	Bureaucracy	1970	1978		79. China, People's Republic (PR)	Autocracy	1950	1953
	Presidentialism	1979	1990	j		Bureaucracy	1954	1990
69. Chile	Presidentialism	1950	1972	J	80. India	Parliamentarism	1950	1990
	Autocracy	1973	1989		81. Indonesia	Autocracy	1950	1954
	Presidentialism	1990	1990			Parliamentarism	1955	1956
70. Colombia	Bureaucracy	1950	1953			Bureaucracy	1957	1959
	Autocracy	1954	1957			Autocracy	1960	1970
	Presidentiaksm	1958	1990	1		Bureaucracy	1971	1990
71. Fcuador	Presirientinlism	J1950	1962	i	82. Iran		-4950—	1960
						—i * lA-i -^Ha-ui4±j--eirtyy-----■—		
	Autocracy	1963	1967			Autocracy	1961	1962
	Bureaucracy	1968	1969	i i		Bureaucracy	1963	1983
	Autocracy	1970	1978			Autocracy	1984	1990
	Presidentialism	1979	1990	I	83. Iraq	Autocracy	1950	1950
(continued)
(continued)
64
65
Democracies and Dictatorships
Country
Regime
Entry Exit
84. Israel
85. Japan
86. Jordan
87. South Korea
88. Laos
89. Malaysia
90. Mongolia
91. Myanmar
92. Nepal
Bureaucracy
Autocracy
Bureaucracy
Parliamentarism
Parliamentarism
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy*
Parliamentarism
Bureaucracy
Autocracy
Bureaucracy
Prcsidcntialism
Parliamentarism
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Bureaucracy*
Parliamentarism
Autocracy
Parliamentarism
Autocracy
Bureaucracy
Autocracy
Bureaucracy
Aiitnfirarcy____________
93. Pakistan
Bureaucracy Autocracy Bureaucracy Parliamentarism
1951
1958 1980 1950
1952 1950
1966
1967 1974
1984
1985
1989 1950
1960
1961
1972
1973 1988 1954
1959
1966
1967
1974
1957 1969 1971 1950 1950
1958
1960
1962 1974 1988
1990 1950
1959
1960 1963 1950
1957 1979 J 1990 1990 1990
1965
1966 1973
1983
1984
1988 1990
1959
1960
1971
1972 1987 1990
1958
1965
1966
1973 1990 1968 1970 i 1990 1990
1957
1959
1961 1973 1987
1989
1990
1958
1959 J
1962 i
1990 -i 1955
Appendix 1.2: Classification of Regimes
Country		Regime	Entry	Exit
		Bureaucracy	1956	1957
		Autocracy	1958	1961
		Bureaucracy*	1962	1971
		Mixed	1972	1976
		Autocracy	1977	1984
		Bureaucracy	1985	1987
		Parliamentarism	1988	1990
94. Philippines		Presidenti alism	1950	1964
		Bureaucracy	1965	1971
		Autocracy	1972	1977
		Bureaucracy	1978	1985
		Presidentialism	1986	1990
95. Singapore		Bureaucracy*	1965	1990
96. Sri Lanka		Parliamentarism	1950	1976
97. Syria		Bureaucracy*	1977	1990
		Bureaucracy	1950	1960
		Autocracy	1961	1969
		Bureaucracy	1970	1990
98. Taiwan		Bureaucracy*	1950	1990
99. Thailand		Bureaucracy	1950	1956
		Autocracy	1957	1968
		Bureaucracy	1969	1970
		Autocracy	1971	1974
		Parliamentarism	1975	1975
		Autocracy	1976	1976
		Bureaucracy	1977	1982
		Parliamentarism	1983	1990
100. Yemen Arab Bepublic		Autocracy	1967	1977
101. Austria		Bureaucracy	1978	1990
		Parliamentarism	1950	1990
102. Belgium		Parliamentarism	1950	1990
103. Bulgaria		Bureaucracy	1950	1989
		Parliamentarism	1990	1990
104. Czechoslovakia		Bureaucracy	1950	1989
		Parliamentarism	1990	1990
105. Denmark		Parliamentarism	1950	1990
106. Finland		Mixed	1950	1990
107. France		Parliamentarism	1950	1957
		Mixed	1958	1990
(continued)
(continued)
Democracies and Dictatorships
Country
108. East Germany
109. West Germany
110. Greece
111. Hungary
112. Iceland
113. Ireland
114. Italy
115. Luxembourg
116. Malta
117. Netherlands
118. Norway
119. Poland
120. Portugal
121. Romania
122. Spain
123. Sweden
124. Switzerland
125. Turkey
126. United Kingdom
127. Soviet Union
128. Yugoslavia
129. Australia
130. Fiji
131. New Zealand
132. Papua New Guinea
133. Solomon Islands
134. Vanuatu
68
Regime
Entry Exit
Bureaucracy
Parliamentarism
Parliamentarism
Autocracy
Bureaucracy
Parliamentarism
Bureaucracy
Parliamentarism
Mixed
Parliamentarism
Parliamentarism
Parliamentarism
Parliamentarism
Parliamentarism
Parliamentarism
Bureaucracy
Mixed
Bureaucracy Mixed
Bureaucracy
Autocracy
Parliamentarism
Parliamentarism
Presidentialism
Bureaucracy*
Parliamentarism
Autocracy
Parliamentarism
Parliamentarism
Bureaucracy
Bureaucracy
Parliamentarism
Bureaucracy
Autocracy
Parliamentarism
1950 1950 1950 1967 1971 1974 1950 1990 1950 1950 1950 1950 1964 1950 1950 1950 1989 1950
1976 1950 1950
1977 1950 1950 1950 1961 1980 1983 1950 1950 1950 1950 1970 1987 1950
1990 1990 1966 1970 1973 1990
1989
1990 1990 1990 1990 1990 1990 1990 1990 1988 1990
1975 1990 1990
1976 1990 1990 1990 1960 1979 1982 1990 1990 1990 1990 1990 1986 1990 1990
Parliamentarism Parliamentarism Parliamentarism
1975 1978 1980
1990 1990 1990
(continued)
Appendix 1.3: Basic Data, about Regime Dynamics Country
135. Western Samoa
136. Bahrain
137. Kuwait
138. Oman
139. Qatar
140. Saudi Arabia
141. United Arab Emirates
Regime	Entry	Exit
Autocracy	1962	1978
Bureaucracy*	1979	1990
Autocracy	1971	1972
Bureaucracy	1973	1974
Autocracy	1975	1990
Autocracy	1961	1962
Bureaucracy	1963	1975
Autocracy	1976	1980
Bureaucracy	1981	1985
Autocracy	1986	1990
Autocracy	1951	1990
Autocracy	1971	1990
Autocracy	1950	1990
Autocracy	1971	1990
Appendix 1.3: Basic Data about Regime Dynamics
Asterisks indicate cases classified as regime transitions according to the regular coding rules.
(A) Transitions to Dictatorships by Incumbents Country Year
Cameroon
Central African Republic Chad Djibouti Gabon Ghana Kenya Lesotho Malawi Rwanda -Siena Leone Uganda Zambia
1963 1962 1962 1982 1967 1972*
1969
1970 1966 1965
T967* 1970 1973
. (continued)
69
Democracies and Dictatorships
Country	Year
Zimbabwe	1980
Ecuador	1970
Uruguay	1973*
Bangladesh	1975
South Korea	1972
Malaysia	1969
Pakistan	1956*
Philippines	1972
Turkey	1980*
(B) Countries by Regime Type and the Number of Transitions to Authoritarianism Experienced by 1950
Democracies in 1950 that had experienced no transition to authoritarianism by then:
Canada
Guatemala
Panama
United States
Brazil
Ecuador
Uruguay
India
Israel
Myanmar
Pakistan
Philippines
Sri Lanka
Belgium
Denmark
France
Iceland
Ireland
Luxembourg
Netherlands
Norway
Sweden
Switzerland
United Kingdom
70
Appendix 1.3: Basic Data about Regime Dynamics
Australia New Zealand
Dictatorships in 1950 that had experienced no transition to authoritarianism by then:
Egypt
Ethiopia
Liberia
South Africa
Dominican Republic
El Salvador
Haiti
Honduras
Mexico
Nicaragua
Bolivia
Paraguay
China (PR)
Indonesia
Iran
Iraq
Jordan
South Korea
Mongolia
Nepal
Syria
Taiwan
Thailand
Hungary
Romania
Turkey
Soviet Union
Saudi Arabia
Democracies in 1950 that had experienced at least one transition to authoritarianism by then:_______________......................______________
Costa Rica
Argentina
Chile
Austria
Finland
71
i
L
Democracies and Dictatorships
West Germany
Greece
Italy
Dictatorships in 1950 that had experienced at least one transition to authoritarianism by then:
Colombia Peru
Venezuela
Bulgaria
Czechoslovakia
East Germany
Poland
Portugal
Spain
Yugoslavia
(C) Regime, Transitions, by Country NO TRANSITIONS (100 cases) Dictatorships (67 cases)
r
1
Algeria	1962-1990
Angola	1975-1990
Benin	1960-1990
Botswana	1966-1990
Burkina Faso	1960-1990
Burundi	1962-1990
Cameroon	1960-1990
Cape Verde	1975-1990
Central African Republic	1960-1990
Chad	1960-1990
Comoros	1975-1990
Djibouti	1977-1990
Egypt	1950-1990
Ethiopia	1950-1990
Gabon	1960-1990
Gambia	1965-1990
Guinea	1958-1990
Guinea-Bissau	1974-1990
(continued) 1
I
72
Appendix 1.3: Basic Data about Regime Dynamics
Dictatorships (67 cases, cont.)
Ivory Coast 1960-1990
Kenya 1963-1990
Lesotho 1966-1990
Liberia 1950-1990
Madagascar 1950-1990
Malawi 1964-1990
Mali 1960-1990
Mauritania 1960-1990
Morocco 1956-1990
Mozambique 1975-1990
Niger 1960-1990
Democracies (33 cases)
Rwanda
Senegal
Seychelles
South Africa
Swaziland
Tanzania
Togo
Tunisia
Zaire
Zambia
Zimbabwe
Haiti
Mexico
Guyana
Paraguay
Bahrain
China (PR)
Iran
Iraq
Jordan_
Kuwait
Malaysia
Mongolia
Nepal
Oman
1962-1990
1960- 1990 1976-1990 1950-1990 1968-1990
1961- 1990 1960-1990
1956- 1990
1960- 1990
1964- 1990
1965- 1990 1950-1990 1950-1990
1966- 1990 1950-1990 1971-1990 1950-1990 1950-1990 1950-1990
-1950=1990--.
1961- 1990
1957- 1990 1950-1990
1950- 1990
1951- 1990
Democracies and Dictatorships
Democracies (33 cases, cont.)
Qatar	1971-1990
Saudi Arabia	1950-1990
Singapore	1965-1990
Syria	1950-1990
Taiwan	1950-1990
United Arab Emirates	1971-1990
Yemen	1967-1990
East Germany	1970-1988
Romania	1961-1989
Soviet Union	1950-1990
Yugoslavia	1950-1990
Fiji	1970-1990
Western Samoa	1962-1990
Mauritius	1968-1990
Bahamas	1973-1990
Barbados	1966-1990
Belize	1980-1990
Canada	1950-1990
Costa Rica	1950-1990
Jamaica	1962-1990
Trinidad & Tobago	1962-1990
United States	1950-1990
India	1950-1990
Israel	1950-1990
Japan	1952-1990
Austria	1950-1990
Belgium	1950-1990
Denmark	1950-1990
Finland	1950-1990
France	1950-1990
West Germany	1950-1990
Iceland	1950-1990
Ireland	1950-1990
Italy	1950-1990
Luxembourg	1950-1990
Malta	1964-1990
Netherlands	1950-1990
Norway	1950-1990
(continued) |
Appendix 1.3: Basic Data about Regime Dynamics
Democracies (33 cases, cont.)
Sweden 1950-1990
Switzerland 1950-1990
United Kingdom 1950-1990
Australia 1950-1990
New Zealand 1950-1990
Papua New Guinea 1975-1990
Solomon Islands 1978-1990
Vanuatu 1980-1990
ONE TRANSITION (17 cases)
To Dictatorship (5 cases)
Congo
Sierra Leone Somalia Sri Lanka
1963
1967
1968 1977
To Democracy (12 cases)
Laos 1958
Dominican Republic 1966
El Salvador 1984
Nicaragua 1984
Colombia 1958
Bangladesh 1985
Venezuela 1959
Bulgaria 1989
Czechoslovakia 1989
Hungary 1989
Poland 1989
Portugal 1975
Spain 1976
TWO TRAf^STT«JNS-(-9-eases)
Die -> Dem -> Die (2 cases)
Uganda Indonesia
74
75
Democracies and Dictatorships
Dem -» Die -> Dem (7 cases)
Grenada
Brazil
Chile
Ecuador
Uruguay
Philippines
Greece
THREE TRANSITIONS (8 cases)
Dcm -» Die -> Dem -> Die (5 cases)
Nigeria
Panama
Suriname
Myanmar
Thailand
Die -> Dem -» Die -> Dem (3 cases)
Bolivia South Korea Turkey
FOUR TRANSITIONS (2 cases)
Ghana Pakistan
FIVE TRANSITIONS (2 cases)
Sudan Honduras
SIX TRANSITIONS (2 cases)
Guatemala Peru
EIGHT TRANSITIONS (1 case) Argentina
76
Appendix 1.4: The "Short" Data Base
Appendix 1.4: The "Short" Data Base
Because the economic data are not available for all the countries and years described earlier, we shall be working with a somewhat smaller data set. As the earliest year for which we have data on per capita income is 1950, our observations on the rate of economic growth begin in 1951. Moreover, because the patterns of economic development for countries that rely for most of their income on oil are sui generis, we excluded six countries in which the ratio of fuel exports to total exports in 1984-1986 exceeded 50 percent.39 These limitations delineate what we call our "short" data base.
The basic patterns that have been described remain unchanged as we move to the smaller data set. Overall, we lose 604 observations, 171 in the six excluded oil-producing countries, and the rest where the economic data are not available. The most significant losses are concentrated in East Asia (31.7 percent), the Pacific islands (30 percent), and Eastern Europe (36.9 percent). Because of data unavailability, we lose 78 years of democracy (4.5 percent) and 355 years of dictatorship (11.8 percent).
In the end, thus, the data set with which we work in the rest of this book contains observations for 1,645 years of democracy (1,022 of parliamentary democracies, 147 of mixed democracies, and 476 of presidential democracies) and 2,481 years of dictatorship (1,812 of bureaucracies and 669 of autocracies), for a total of 4,126 observations. They compose 99 spells of democracies (or 50 of parliamentarism, 9 of mixed systems, and 43 of presidontialism) and 123 spells of dictatorships (or 133 of bureaucracy and 98 of autocracy). This yields thirty-nine transitions from democracy to dictatorship, and forty-nine from dictatorship, to democracy.
M These countries are Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates.
77
J