Referendum and the Choice between Monarchy and Republic in Greece

Download 271,79 Kb.
Date conversion11.09.2018
Size271,79 Kb.
  1   2   3   4
Referendum and the Choice between Monarchy and Republic in Greece
Forthcoming in

Constitutional Political Economy

DOI 10.1007/s10602-009-9078-4

George Tridimas

Drawing on the 1974 Greek parliamentary elections and constitutional referendum which abolished the monarchy, the paper employs a spatial decision model to examine the strategic use of a parliamentary election and a non mandatory referendum by an agenda setter. The parliamentary election bundles two issues, the right to form a government and the right to choose the form of state. This implies that the election campaign efforts to attract votes are different from the campaign efforts to win an election for government and a separate referendum for the form of state. The choice of the agenda setter turns out to depend on his costs of campaign efforts in the different contests relative to those of the opposition, his benefits to be gained from winning the different contests, his comparative electoral appeal and the probability that the referendum secures his favourite outcome.

Key words: Greece; monarchy versus republic; non-required referendum; parliamentary elections; campaign effort; constitutional revision.
JEL classification: D7; N4
March 2009

G. Tridimas.

University of Ulster,

Newtownabbey, Co. Antrim, BT37 0QB, UK


Referendum and the Choice between Monarchy and Republic in Greece
Constitutional changes have often been ratified by holding a referendum. A long line of scholarship has inquired the comparative benefits of democratic decision making directly by the voters in a referendum and indirectly by elected representatives in parliament. There seems to be a wide consensus in the literature that deciding an issue of public interest by a referendum confers important benefits to the citizenry by involving the voters directly in the process of decision making. Typically, the existing volume of work considers the referendum as an alternative to the election for representatives, who decide the issue of public interest. The present paper extends and complements this line of inquiry by examining why an agenda setter chooses to call an election to decide some issues and a non mandatory referendum to settle some other issues, even though they could have been decided by the elected representatives.
In its analysis the paper draws on the experience of Greece, where the question of the form of the state, monarchy or republic, was put to a referendum six times in the period 1920-1974, while all other questions of policy and constitutional arrangements have always been decided by politicians elected in parliamentary elections (when not under authoritarian rule). The monarchy in Greece had a turbulent time, with four kings out of a total of seven deposed during the period 1832-1974. None of the six referendums about the monarchy was in any way constitutionally mandated, they were all held at the discretion of the incumbent governments. Both theory and practice have shown that a referendum may significantly change the result of the legislative process of representative democracy. Calling a referendum therefore, especially when there is no constitutional requirement to do so, may imply significant political risks and trade offs for a political ruler who acts as the agenda setter. What determines the choice of the agenda setter to call a referendum? This question is the subject of the present work.
The experience of the sixth referendum on the monarchy in Greece in 1974 is instructive: In 1974 after the collapse of the dictatorship there were two popular votes, first, a parliamentary election to produce a democratic government and vote in a new constitution; second, a referendum to either restore the monarchy or proclaim the country as a republic. Contrary to its predecessors, the pro-republic outcome of the 1974 referendum and the 1975 constitution which followed have proved long lasting and sufficiently flexible to allow orderly alternation of political parties in office and two constitutional revisions in accordance with the existing provisions.
The paper analyses the choice of the agenda setter to call a parliamentary election and to hold a non mandatory referendum to determine different issues of public interest. Even though the focus is on Greece, the question of how the agenda setter chooses to combine different voting mechanisms addresses a more general issue and as such, it is hoped, of wider interest.
The paper is structured as follows. By way of background, Section I summarises the history of the monarchy in Greece; it describes and compares the referendums (including fraudulent ones) that took place on its status and gives a detailed account of the 1974 referendum and the constitutional arrangements that followed it. Section II offers a selective survey of previous work on the referendum including a comparison to representative democracy, while Section III reviews earlier studies which also examined the use of the referendum as a means to advance the interests of an incumbent government. Section IV develops the theoretical argument of the paper. It examines the choice of the political ruler – agenda setter in a spatial decision framework where the expected outcomes of an election and a referendum depend on the utility from winning a vote, policy preferences and the effort expended to attract the support of the electorate. Comparison of the net expected payoff under the two mechanisms identifies the conditions which lead the agenda setter to call a referendum in order to advance his interests. The analysis in the paper is positive and concentrates on the choices of the agenda setter which maximise his expected utility leaving aside normative questions of whether an issue should be decided by a referendum on some normative criteria. Section V concludes.
I. Brief history of the monarchy and referendums about the monarchy in Greece

1 From monarchy to republic
In a monarchy the head of state is hereditary and holds the position for life, while in a (democratic) republic it is elected and serves for a limited term. The monarchy in modern Greece had a turbulent and troubled history which had been inextricably linked to both domestic and foreign policy developments. 1
After the War of Independence against the Ottoman Empire, Greece was recognized as an independent state in 1832 and Otto Wittelsbach, a Bavarian prince, was chosen as king. Otto ruled as an absolute monarch until 1843 when he was forced to grant a constitution 2 and was eventually deposed in 1862. In 1863 George Glücksburg, a Danish prince, was chosen as king and a new constitution came into effect in 1864, by which Greece was proclaimed a “crowned democracy”, a democracy with a monarch. 3 George I was assassinated in 1913 by a madman and was succeeded by his son Constantine I. The king insisting that Greece stayed neutral during WWI clashed with the elected prime minister, E. Venizelos, who argued for Greece to join the Entente. The rift precipitated a deep national schism dividing the country into two bitterly opposed camps of royalists and anti-royalists (also called Venizelists). Bowing to very strong pressures Constantine I gave up the throne in 1917 but without formally abdicating and went into exile with his eldest son, George. He was succeeded by his second son Alexander.
After the untimely death of Alexander in 1920, and the electoral victory of the royalist parties in the same year, a referendum resulted in a 99% majority for the return of Constantine. Even though the actual size of the majority was manipulated, it appears that there was sufficient support for his return to win in a fair ballot. 4 It bears noting that the referendum was not about the constitutional arrangement as such, but about the person to sit on the throne. However, in the aftermath of Greece’s heavy defeat in Asia Minor by Turkey in 1922, the military staged a coup which brought down the pro-royalist government and forced Constantine to abdicate. He was succeeded by his son George. With mounting social and economic problems5 the anti-royalist current grew so strong that in 1923 George II was forced to go abroad on ‘leave of absence’ and in March 1924 the anti-royalist government passed a parliamentary resolution abolishing the monarchy. A referendum in November 1924 recorded a 30% vote against the monarchy and 70% vote in favour of a republic. 6
The Republic, whose constitution was ratified by the parliament in 1927, was short lived. 7 The 1935 election (not contested by Venizelist parties after a failed putsch of Venizelist officers) resulted in a huge pro-royalist majority, which promptly restored the monarchy. A new referendum was held which returned a blatantly rigged vote of 98% in favour of the monarchy. A year later, in 1936, dictatorial rule was imposed with the approval of the king. In 1941 after the German army overran the Greek forces King George II left Greece. In March 1946 and in the middle of a savage civil war between the nationalists and the communists, parliamentary elections were held. With the leftists and the Communist party abstaining the alliance of the royalist parties gained a substantial parliamentary majority against the liberals. The nationalist government then held another referendum in September 1946 about the constitutional issue which resulted in a 68% vote in favour of the monarchy. That vote too was marred by confirmed allegations of fraud. Historians argue that although the monarchy did not enjoy much support there was widespread opposition to a communist government and the monarchy was seen as a security against a communist take-over. 8 King George II returned, but died a few months later in 1947 and was succeeded by his brother Paul. The civil war ended in 1949 and a new constitution came into effect in 1952. A period of economic progress followed under a rightwing government led by C. Karamanlis, but tensions between him and the king also developed.
On King Paul’s death in 1964 the throne passed to his son Constantine II. In turn, he clashed with the centrist PM, G. Papandreou, resulting in the resignation of the latter in 1965. In April 1967 the “colonels” staged a military coup and suspended the constitutional order. Following an unsuccessful counter-coup in December 1967 Constantine II fled the country and took up residence in London. After suppressing a mutiny led by navy officers, the military regime abolished the monarchy in 1973. A referendum followed to legitimise the change. As it was held under military law and characterised by widespread electoral irregularities, the reported 20% vote against the monarchy is discredited. 9 The junta collapsed in July 1974 amid mounting political and economic problems and the threat of war against Turkey and C. Karamanlis formed a government of national unity. The right-of-centre government which emerged after the elections, led again by Karamanlis, held a new, this time clean, referendum about the form of the state which recorded a 69% vote against the monarchy. In 1975 a new constitution was promulgated introducing a “presided” parliamentary republic form of government, a term coined to signify the extent of the prerogatives of the president of the republic, elected by the parliament, with the power to dissolve parliament, veto legislation and call a referendum. The constitution also stipulated that the articles which determine the form of state are not subject to revision. Since then the constitutional position of the republic has not come under questioning.
Table 1 summarizes the time line of the form of the Greek state and details the six referendums regarding the monarchy. There is little doubt that the role of the monarchy was controversial and it often became the source of political and constitutional instability.

Table 1: Form of State and Monarchy Referendums in Greece

Form of State

& Reigning Monarch



Referendums on the Monarchy


% Votes For





1828 1832


1832 1924

House of Wittelsbach

Otto (1815-1867)


Absolute ruler until 1843. Constitutional ruler from 1843. Deposed in 1862

House of Glücksburg



George I (1845-1913)

Constantine I (1868-1923)


Forced to flee in 1917

after dispute with PM

Alexander (1893-1920)









Constantine I (1868-1923)


Abdicated in 1922 after military defeat in Asia Minor

George II (1890-1947)


Forced to take ‘leave

of absence’ in 1923









1924 1935









1935 1973

George II (1890-1947)


Fled in 1941

after defeat by Nazis








Paul (1901-1964)


Constantine II (1940-)


Fled in 1967

after failed counter-coup against the military regime






Republic proclaimed




30.82 *




Republic proclaimed

* This is the only referendum whose result was not contested by the losing side.

Figures in brackets show the number of votes

Sources: Author’s compilation. The exact figures for the referendum votes are as follows: Pantelis, Antonis, Stefanos Koutsoumpinas and Triantafyllos Gerozisis. Texts of Constitutional History (in Greek) Vol. I for 1920 and 1924 and Vol. II for 1935 and 1946; Clogg (1987) for 1973 and Rizospastis Newspaper, 10-12-1974 for 1974

Of all the referendums held about the monarchy none was constitutionally required, they were all held at the discretion of the incumbent government. The 1920, 1924, 1935 and 1946 referendums were preceded by parliamentary votes on the question of monarchy / republic and in all cases the referendum result turned out to simply endorse the result of the parliamentary vote. Of course, as normal democratic politics had been suspended, no parliamentary vote preceded the 1973 referendum of the military government, which imposed its own constitutional arrangements, nor was any doubt that the junta would have allowed a result rejecting its proposals. The 1974 plebiscite was different. It was the first referendum on the constitutional question to pass the test of free and fair vote under universal suffrage (women in Greece were enfranchised in 1956) and it was the only referendum whose result was not contested by the losing side. In addition, crucially from the viewpoint of the present inquiry and contrary to all the earlier referendums on the monarchy, it was not preceded by a parliamentary vote, but it was called as the exclusive mechanism to decide the constitutional question at a time when a parliamentary election was also held to settle other issues of public policy and pass a new constitution.

2 The 1974 referendum and its constitutional ramifications
It is useful to detail both the political context and the conduct of the 1974 referendum. After the collapse of the military regime in July 1974, the transitory national unity government of conservatives and centrists led by C. Karamanlis announced that parliamentary elections would be held on 17 November 1974 to be followed by a referendum on the monarchy on 8 December 1974. In addition to its standard legislative powers, the new parliament would also revise the constitution subject to the referendum verdict. However, the convocation of the parliament was postponed until 9 December, one day after the referendum. 10 Note that the form of the state, monarchy or republic, is a constitutional question which in principle may be separated from the person of the monarch. However, in truth the only way for the monarchy to continue was for the Glücksburg dynasty to remain on the throne (as opposed to any other royal house); as a result the constitutional issue and the choice of the particular person could hardly be separated.
The parliamentary election was won by Karamanlis whose party polled a post WWII record of 54.4% of the popular vote and secured a huge parliamentary majority (220 seats out of 300). A large part of the supporters for the conservative government came from the pro-royalist camp favouring the restoration of the monarchy and the return of Constantine. The royalists also counted a number of government MPs and cabinet ministers amongst their supporters (although the exact number is not known), but, significantly, the PM, C. Karamanlis, was not amongst them. Voters in the referendum were called to choose between “Uncrowned” and “Crowned”. 11 During the campaign the pro-royalist and the anti-royalist sides were given the opportunity to present their cases freely and were afforded equal state-television coverage. Officially, neither the political parties nor their leaders participated in the referendum campaign. 12 Nor did King Constantine campaign in person, as he remained in London, but his videotaped messages were broadcasted. 13 In the end and as already said, 69.18% of the voters voted for “Uncrowned”.
What stands out is how C. Karamanlis used his powers of agenda control: There was a general election to produce a government with the task of deciding public policy and the constitutional arrangements, but the form of state, monarchy or republic, was decided not by that parliament but by the referendum vote. Thus, the chosen course of elections and referendum separated the issue of selecting a government from the issue of deciding the form of the state, while it joined the issue of normal legislative politics with writing (large parts) of the constitution. Moreover, the king was not in any way encouraged to return after the collapse of the dictatorship, which removed him from office in the first place.
Students of public choice would recognize that the “improvisation” of the 1974 arrangements is not atypical of the “episodic event” of constitutional drafting which follows the collapse of a dictatorial regime (or other major events), where the revision of the constitution seldom conforms to the relevant legislation for the amendment of the constitution. 14 Moreover, the drafting of the constitution by a sitting legislature violated what Mueller (2005) has called the “first law of constitutional writing, that is, those who are likely to hold office under the new constitution should not be involved in its writing lest they draft a constitution to serve their interests and not those of the citizens”, (p. 65, emphasis in the original). 15
To close this historical account it is noted that the constitution of 1975 was hardly the final word of constitutional writing as it was subsequently revised in 1986 by the then socialist government. Amongst other stipulations, the revisions removed the discretionary powers invested in the hands of the president to dissolve parliament, veto legislation and call a referendum (which then became the exclusive domain of the cabinet), so that the form of the state is now described as a parliamentary republic. A further revision took place in 2001 which introduced new individual rights and organizational rules, but did not otherwise change the structure and functions of the organs of the state. Contrary to all pre-1975 constitutional revisions, the amendments of 1986 and 2001 were carried out according to the provisions for amending the constitution, that is, within the constitutional order rather than by violating it, indicating that the settlement of 1974/1975 has established a certain degree of constitutional stability which was lacking beforehand.
II. Theoretical analysis of the use of referendum
1. Classification of referendums
In analysing the referendum process and its effect on collective choice scholars have first drawn attention to the institutional aspects of a referendum, see for example Butler and Ranney (1978), Mueller (1996), Hug and Scarini (2000) and Tsebelis (2002) and the literature therein. Summarising, referendums can be classified on three dimensions; namely (a) whether their results are binding or consultative; (b) the subject matter of the vote, that is, what exactly is decided, and of specific importance for our purpose whether the vote is on a constitutional issue (a rule on how to make policy) or a post-constitutional issue (a policy, given the constitutional rules); and (c) the initiator, that is, the political player who calls the referendum. Purely consultative referendums are rare in practice and probably uninteresting, because when voters know that policy making is not bound by the referendum outcome their incentives to engage in the collective choice process are questionable. Referendums to amend constitutions are common in democracies. Typical constitutional issues relate to national sovereignty, the division of powers in a polity and the adoption of fundamental rights. Non constitutional issues regard “bread and butter” policy questions like deciding tax and spending programmes, with Switzerland and the USA at the state level offering the best known examples. The third dimension is more complicated. A referendum is required when the constitution mandates so, for example, to approve a constitutional revision, or when a country has to cede part of its sovereignty to a supranational organisation like the European Union. Alternatively, a designated political actor may trigger off a referendum, even though it is not required by the constitution. There are three types of non mandatory referendums. (a) When the government calls one even though it can pass legislation through a vote in the parliament. This is known as a plebiscite or a passive referendum. The Greek referendums discussed above belong to this category. (b) When the opposition to the government, parliamentary or an organisation or interest group outside the parliament, initiates a referendum on a policy proposed by the government. (c) When the opposition, parliamentary or extra-parliamentary, initiates a referendum on a policy proposed by the opposition itself. Examples of the latter are the popular initiatives, where a specified minimum number of signatories can force a vote on their proposed policy and the recall referendum by which voters may remove an official from office.
2. Arguments for and against the use of referendum
An enormous and still growing literature has developed to examine how referendums affect the quality of the outcomes of collective choice, the quality of the democratic process and the gains they confer to citizens from the act of participation in policy making – for recent informative surveys see Matsusaka (2004) for a book long survey or for shorter treatments see amongst others Feld and Kirchgäessner (2000), LeDuc (2002) and Matsusaka (2005a) and (2005b).
The main arguments in favour of referendums are: (1) Legitimacy of outcome. It is well known that the outcome of a referendum, which aggregates all votes simultaneously, may differ from the outcome of a parliamentary vote, which sums votes separately at the constituency level, a result known as “the referendum paradox”. 16 Since the referendum outcome reflects decisions made directly by the citizens, it has greater legitimacy than the decisions made by their representatives. Specifically, the referendum process introduces popular preferences, typically approximated by the median voter of the electorate, as an additional veto player in the process of determining policy and may break the agenda control power of politicians to decide public policy issues; see Matsusaka (2004) for a thorough study of the evidence. (2) Accuracy of outcome. Decisions made directly by voters reflect more accurately their preferences in comparison to decisions made by the representatives of citizens, who may find opportunities to advance their own private interests, because of free rider problems in monitoring their behaviour. This is the principal – agent problem which bedevils political representation, where the voters–principals delegate significant discretionary powers to politicians–agents to decide and implement policy. (3) Focus on the issue. A referendum forces citizens to focus on the issue to be decided to the exclusion of other issues, while parliamentary elections are fought on broader and vague platforms, which bundle issues. This may improve the incentives of voters to be informed about the issue at hand and participate actively in the process of collective choice. Moreover, this largely resolves problems of logrolling, where legislators may trade votes in one issue to secure a favourable vote in another issue combining unrelated issues and resulting in inefficient outcomes. In addition, the referendum process may lead to better outcomes than decision making by representation when value issues are decided, like abortion limits, gay rights, stem cell research, etc, when there may not be an obvious “right” or “wrong” decision (Matsusaka, 2005a and b). (4) Further, direct democracy increases citizen involvement in politics and knowledge of the issues which are debated. This pedagogic effect may even have a bigger impact on citizens and democracy than the substantive on policy making (for a book treatment of this argument, see Smith and Tolbert, 2004). Better educated and more participating voters are able to better evaluate politicians and, once more, force them to be better office-holders. (5) Citizen satisfaction. Irrespective of whether voters are on the winning side of the referendum, the very act of direct participation in the policy making confers benefits to the voters (see Frey, Kucher and Stutzer 2001 and the literature cited therein).
On the other hand some of the most common arguments advanced against referendums include: (1) Lack of sophistication by voters. Voters are not able to fully understand the complicated issues involved in a referendum, while elected politicians who specialise in developing expert knowledge in questions of public policy are better informed and thus better equipped to choose policy (see Kessler, 2005, for a recent formal treatment and critical comparison). However, the latter advantage is sharply diminished if politicians are asymmetrically informed, in the sense that they are better informed on the issues dealt by the legislative committees where they serve, or when they may attend to the narrow interests of their own constituencies to the exclusion of other nation-wide considerations. In addition, recent theoretical and empirical research has shown that voters with little political information can use information shortcuts to cast their votes the same way that they would if they had richer information, see Lupia (1994) and Lupia and McCubbins (1998). (2) Discontinuity of choice. A referendum which takes the form of binary choice, yes-or-no, deprives the electorate of the opportunity to consider a richer set of possible outcomes. (3) Inability to account for the intensity of preferences. Voting by legislative bodies can take into account not only the direction of preferences on an issue but also their intensity. This allows for finely balanced choices which recognise trade-offs and make compromises on contentious questions. (4) Abuse of minorities. Referendums decided on a simple majority rule are less suitable for societies divided along ethnic or religious lines where a particular community has an inbuilt majority. Constitutional arrangements favoured by the majority community will be adopted but they will not necessarily produce stable settlements. Then power sharing with inviolable guarantees for the rights of minorities may turn out to be more durable, with Northern Ireland offered as the best example of this case (see Bogdanor 1994). However, this objection is more an argument against the use of the simple majority rule and less an argument against the use of the referendum as a mechanism of collective choice. (5) Abuse of the majority. 17 Under conditions of asymmetric information, well informed and well endowed interest groups hijack the referendum process, and especially that of popular initiatives, and force governments to adopt policies that serve the objectives of the interest groups at the expense of the majority of citizens. This risk, however, is much less pronounced in referendums regarding constitutional issues; see Matsusaka and McCarty (2001) for a critical assessment. (6) Deterioration of the democratic process. By removing decision making from politicians referendums weaken their authority and consequently that of representative government which ultimately weakens support for the democratic process. This argument is of obvious relevance in cases where authoritarian rulers under restrictions on freedom and rigged ballots exploit the referendum process to give the pursuit of their objectives a semblance of democracy. On the other hand, Matsusaka (2008) offers a subtler and previously unrecognized argument in favour of direct democracy, as well as some suggestive evidence. Namely, by deciding some issues directly and therefore removing them from the domain of elected officials, the referendum enables voters to sanction politicians more effectively even on the issues that remain under their control forcing them to improve their performance and be more faithful agents. 18
  1   2   3   4

The database is protected by copyright © 2016
send message

    Main page