Portuguese Emigrants’ Political Representation: The Challenges of the External Vote

Over the last few decades, emigrants all over the world have gained expanded voting rights. Despite the normative debates about this issue, there are few empirical studies on why states decide to implement external voting and how electoral systems perform. This chapter seeks to fill this gap by looking at the Portuguese case. Our study suggests that a combination of political and socio-economic factors explains the implementa tion of external voting. On the other hand, the interests of political parties and the low level of civil society engagement are key factors in the failure of both electoral reforms and attempts to overcome the shortcomings of external voting.


Introduction
The participation of emigrants in national political systems has become an increasingly important topic (Bauböck 2005;International IDEA 2007;Lafleur 2013) Indeed, most consolidated democracies recognise the right of emigrants to vote in national elections (Blais et al. 2001;International IDEA 2007), and there has been a growing tendency to broaden the political rights of emigrants over the last few decades, particularly in the electoral arena (Spiro 2006;Bauböck 2007;Lafleur 2013). 1 We refer to external voting rights as 'the active and passive voting rights of qualified individuals, independently of their professional status, to take part from outside the national territory in referenda or in supranational, national, subnational, or primary elections held in a country of which they hold citizenship but where they permanently or temporarily do not reside' (Lafleur 2013: 31). 2 Although there are various theoretical and normative studies on external voting (Barry 2006;Bauböck 2005;Lafleur and Martiniello 2009;López-Guerra 2005;Rúbio-Marin 2006;Spiro 2006;Lafleur 2013), empirical work suggests that the development of external voting depends more on contextual factors rather than on normative arguments (Tager 2006;Rhodes and Harutyunyan 2010;Lafleur 2011). In other words, while existing theoretical and normative studies are concerned particularly with the legitimacy of the external vote, they fail to consider why and how it is implemented. Moreover, there is a scarcity of work on actual voting practices and on the (positive and negative) impacts of electoral system designs on emigrant voting.
Where studies of the external vote exist, they focus mostly on countries that are undergoing processes of consolidation or where external voting is a very recent phenomenon (International IDEA 2007;Collyer and Vathi 2007;Smith 2008). Finally, most of the studies concern with the effects of external voting focus on the inclusion of emigrants in one specific country (Escobar 2007;Itzigsohn and Villacrés 2008). Overall, then, existing work tends to neglect the question of the broader 'success' of voting abroad, in particular as a means of improving political representation and participation.
In this chapter, we will examine the adoption and evolution of external voting in Portugal. We will also evaluate its success and identify the reasons behind the lack of reforms to improve the practice of out-of-country voting. We believe that this case study is worthy of examination for several reasons. First, Portugal was one of the first countries to extend voting rights to emigrants. The external vote was recognised in Portugal following the establishment of democracy in 1974, with the election of representatives of the emigrant community in the legislative elections of 1976. Second, Portugal has been characterised by significant emigration flows. According to the most recent data collected by the the Portuguese Emigration Observatory (Observatório da Emigração) based on the latest estimates by the United Nations, in 2015 the number of emigrants born in Portugal surpassed 2.3 million. About 22% of Portuguese nationals live outside the country (Observatório da Emigração 2017). However, if emigrants' descendants are factored in, the Portuguese communities abroad consist of about five million people. According to an estimate based on data from 2017, there are 1.375 million Portuguese nationals residing abroad with a potential right to vote. 3 There is also another peculiarity that makes the Portuguese case particularly interesting. This is related to the fact that Portugal, in contrast to other European countries, did not make the transition from an emigration to an immigration country.
As several studies have noted (Peixoto 2007;Pires et al. 2010), although emigration flows have experienced different cycles and some discontinuities, the number of Portuguese citizens that moved abroad has remained relatively high -even during the decades with highest immigrant inflows. The financial and economic crisis and the austerity policies that hit the country in 2011 led to a huge increase in emigrant outflows: a peak of around 120 thousand departures in 2013, followed by a progressive descent, to around 100 thousand in 2016 (estimates by Pires et al. 2017). Moreover, the fact that the main political parties have remained the same over the democratic period allows us to concentrate on the specificities of electoral rules and actors' interests. Finally, it is worth noting that there are few systematic studies about the Portuguese external vote and how electoral reform may improve the quality of emigrant representation. This is generally an understudied phenomenon and the Portuguese case is not an exception. There are few studies about Portuguese emigrants and their relationship with home-country politics -especially about their electoral participation -and those that exist tend to be essentially descriptive (Malheiros and Boavida 2003;Lobo 2007). Our study intends to fill this gap through research that explores the adoption and implementation of emigrants' political rights, as well as examining how electoral reform may improve the quality of emigrant participation and representation (Lisi et al. 2015;Belchior et al. 2017).
Why did Portugal enfranchise its citizens abroad? What are the characteristics of external voting in Portugal? How have these voting rights evolved? What are the main shortcomings of external voting and the problems/controversies associated with the practice of out-of-country suffrage? An analysis of the Portuguese experience provides an opportunity to answer these questions and to contribute to the debate about the emigrant vote. Moreover, an in-depth case study may shed more light on how the external vote can promote higher levels of inclusion and participation among emigrants, greater respect for the rule of law, and fairer political representation. Last but not least, a qualitative analysis is also well-suited to examining the factors that inhibit the improvement of emigrant representation and potential reforms of the electoral system design.
This study offers a detailed analysis of the development of external voting and the influence of the electoral system on the relationship between the emigrant community and elected representatives. The main argument is that external voting has displayed significant shortcomings that have prevented the fair political representation of Portuguese emigrants. Moreover, it will be argued that, although the debate on external voting reforms has been a constant feature of the political agenda, the Communities (DGACCP), embassies, consular services and the Directorate General of Internal Administration (latest data available in November 2017). Citizenship is a prerequisite for participating in elections from abroad. As in Italy, the Portuguese nationality law stipulates that a child born to a Portuguese parent is automatically a Portuguese citizen. Since 2017, grandchildren of Portuguese nationals abroad can also request Portuguese citizenship. conflicts between different political parties' interests have made it difficult to find a solution to improve voting methods for out-of-country electoral districts.
This study relies on extensive empirical material based on different types of sources on external voting. In particular, we draw on academic literature, the press, political parties' records, parliamentary and emigrant associations' documents, official statistics on electoral participation, as well as semi-structured interviews with members of parliament (MPs) and consultation with public officials. 4 The following section examines the full range of external voting procedures, highlighting the advantages and disadvantages of each method, and the scope for reforming electoral laws. The third section describes the origin and development of emigrant voting rights in Portugal, while the fourth evaluates the success of external voting. The fifth focuses on recent debates about electoral reform. The conclusions summarise the main findings and suggest lines of research that are worth exploring in the future.

The External Vote in Theory: Electoral Laws and Procedures
With the expansion of external voting policies, the scholarly literature has paid growing attention to the factors that led to this worldwide development and to the implications of external voting for the functioning of contemporary democracies. Within this research we can identify three main explanations that may help us understand the reasons behind the external vote's implementation. The first is based on procedural and institutional arguments and it emphases the importance of international and domestic norms that have promoted emigrants' rights to participate in home country elections. As Lafleur has noted (2013: 33-36), most of these international instruments -such as treaties and conventions -do not require states to adopt external voting policies. However, two factors have been particularly important for the implementation of expatriate voting. The first is the influence that European institutions have had in recent decades, namely through the documents drawn up by the Council of Europe and the European Union. The second is related to the constraints imposed by constitutional norms. As in the case of the electoral system at the domestic level, the constitution may include principles or norms -such as the need for approval through a referendum or the requirement of an absolute or qualified majority -that make it difficult to implement the external vote. 4 The interviews were carried out between January and April of 2010. We interviewed one representative from each Portuguese party with parliamentary representation, with the exception of the CDS-PP (who did not respond to the request): José Cesário (PSD), Paulo Pisco (PS), Helena Pinto (BE) and António Filipe (PCP). We obtained relevant details and clarifications from officials of the following public institutions: the Electoral Administration of the Directorate-General of Internal Administration (AE-DGAI), the National Commission for Elections (CNE), the Commission for the Registration of Portuguese Voters Abroad (COREPE) and the Council of Portuguese Communities (CCP).
The second explanation is based on an economic perspective. This suggests that the implementation of external voting has been motivated by dependence on emigrants' remittances. Yet comparative studies have found that states with a low level of remittances have also implemented the external vote (Collyer and Vathi 2007). Finally, several authors have pointed out the importance of political factors (Tager 2006;Rhodes and Harutyunyan 2010;Lafleur 2011;2013). According to this third explanation, political parties' interests and the dynamics of competition may help us understand the development and content of external voting legislation. Moreover, emigrant associations and lobbies can also play an important role in the debate on emigrants' enfranchisement. As several studies have shown (Rhodes and Harutyunyan 2010;Lafleur 2011), we need to consider which combination of variables can better explain the adoption of external voting in a particular country and its development over time.
The implementation of external voting is linked to several methods and operational procedures. The electoral norms that regulate voting mechanisms for citizens residing abroad can be included in national constitutions, but they are more often part of non-constitutional or administrative legislation (Nohlen and Grotz 2007: 67-68). One of the most important issues in an electoral law is how it converts votes into mandates. As regards the emigrant vote, there are usually two possibilities: one is to establish a number of districts abroad, which serve to convert votes into mandates; another is to distribute emigrant votes among existing national districts and then to convert them according to the electoral rules used in the national territory. As Nohlen and Grotz (2007) underline, each method has a distinct logic: while the first emphasises the extra-territorial nature of the emigrant vote, the second places greater emphasis on being situated within the national territory as a requisite for suffrage. The first method also allows the political actors to control the impact of emigrants' votes on the final results, while the second is more uncertain in terms of parties' electoral performance.
More often than not, emigrant votes are counted within national districts (Nohlen and Grotz 2007: 70;Collyer and Vathi 2007). In such cases, external votes are apportioned to national electoral districts and included in the vote count of the district that was the emigrant's last place of residence before leaving the country. There are exceptions to the last place of residence rule, however: external votes may be counted for the electoral district corresponding to the capital city, for instance, or for those districts with a total number of voters below the national average. When these other counting methods are used, it is harder to assess the nature of the relationship between emigrant voters and MPs, and the impact of the external vote on final election results.
The choice of administrative procedure is also crucial for the implementation of the external vote (Thompson 2007). There are generally four alternative voting methods for citizens residing abroad: (1) the personal vote; (2) the proxy vote; (3) the postal vote, and (4) the electronic vote. All these different methods present both advantages and disadvantages. In the personal vote an elector must attend at a polling station in person -normally in a diplomatic or consular building -in order to cast his/her vote. This mechanism usually ensures the highest level of transparency but it requires emigrants to travel to vote, which may be particularly difficult in very large countries (such as the US, Brazil or China) or where the network of diplomatic or consular representation is less dense.
The postal vote has logistical advantages and its costs are usually relatively low, but it does not offer the same level of security and confidentiality as the personal vote. In Italy, where this method has been adopted, the widespread use of a 'preferential voting' system in the 2006 elections has been cited as an example of how this system cannot guarantee voting freedom and secrecy (Tarli Barbieri 2007: 139). Another disadvantage of this procedure is that it can be overly lengthy, and may not work well in countries where the postal services are unreliable or inefficient. This has been observed in the Portuguese case.
By contrast, the proxy vote is easy to organise and implement and relatively inexpensive. Voting by proxy means that an elector can appoint someone to vote for him/her at his/her polling station. Each elector needs to tell his/her proxy which candidate(s) to vote for. The biggest problem with proxy voting is that it cannot ensure transparency and secrecy, and thus calls into question the principles of fairness and equality (Lafleur 2013: 22). Finally, the electronic vote also minimises costs and it can ensure rapid results and greater mobility. However, it has logistical and security-related disadvantages and it can raise issues of voter equality (Braun 2007). Spain is one of the few European countries (along with Austria, France, the Netherlands and Switzerland) to have implemented pilot projects to allow emigrants to cast ballots through the Internet. According to the 'Voting From Abroad Database' collected by IDEA (International Institute for Democracy and Electoral Assistance), only 8 countries out of 216 currently use this method, and in Europe only Switzerland has adopted it. 5 Overall, the choice of method depends not only on the electoral formula and system of voter registration, but also on the geographical distribution of emigrants, as well as the cultural and political traditions of the national community. It should be noted that some countries allow a variety of voting methods to take account of host country peculiarities and to ensure that emigrants are as well represented as possible.
According to the IDEA database, as of 2017 there are circa 155 countries that grant their citizens residing abroad the right to vote. Among these, 75 provide for the conventional personal vote only, which is cast at polling stations set up especially for elections. There are 22 countries that provide only for the postal vote, one with only the electronic vote and another five allow only proxy voting. Finally, 39 countries -18 of them European states -provide for two or more voting methods.
This brief outline of external voting systems shows the main characteristics of external voting policies. However, as Nohlen and Grotz (2007) suggest, it is necessary to consider not just legal factors, but also political factors related both to the logistics and organisation of elections (political-procedural factors), as well as to party strategies and the dynamics of the party system (political-institutional fac-tors). It is also important to take into account how successfully electoral laws enable the political representation of emigrants. Only with an understanding of the limits and constraints on emigrants' representation can reform proposals be improved.
How can we assess the success of the external vote? In order to evaluate how well electoral mechanisms represent citizens living abroad, it is important to consider three main dimensions. The first is related to the fairness and freedom of elections. Since Dahl's seminal contribution (Dahl 1971), there is a consensus that democratic elections imply the lack of significant fraud and coercion and that competition between diverse social and political groups must occur transparently and must be unbiased. The second important dimension is the equality of representation. The focus here is on whether emigrants have adequate representation compared with citizens living in the home country. Therefore, this dimension can easily be evaluated by considering the degree of proportionality of the external vote. The third element used in our study is based on the level of participation. Several scholars have argued that the impact of electoral systems on citizens' mobilisation is a fundamental aspect of the quality of contemporary democracies, not only because it fosters civic engagement and guarantees the transmission of people's interests, but also because it boosts government responsiveness (Powell 2000;Norris 2003;Beetham and Landman 2008). In the following, we will examine the implementation and development of external voting policies in Portugal and we will use these three criteria to evaluate the shortcomings of emigrants' representation and the prospects for reforming expatriate voting.

The Adoption of the External Vote in Portugal
Before examining the experience of external voting in Portugal, it is crucial to have some idea of the importance and characteristics of Portuguese emigration. Portugal has been a country of emigration since the second half of the nineteenth century. Between 1850 and 1975 two main cycles characterise the move abroad of Portuguese emigrants. The first occurred between 1850 and 1930 and was dominated by a transatlantic flow, mainly towards Brazil. The second cycle started after World War II, lasted until the end of the authoritarian regime in 1974, and concerned mostly European countries (France, Germany and Switzerland). In this period, and in particular from the mid 1960s until 1973, a massive flow of Portuguese nationals left the country. The third cycle began at the start of the democratic period and lasted until 2000. As noted earlier, Portuguese emigration has remained relatively high even during the period of economic growth experienced after access to the European Community. A fourth cycle of Portuguese emigration began in the 2000s, increasing in 2011 with the 'Great recession' and reaching its peak in 2013. Today Portuguese emigrants are distributed across more than 140 different countries worldwide.
During the authoritarian regime  there was little interest in the enfranchisement of Portuguese nationals living abroad. Salazar's New State (Estado Novo) did not explicitly recognise citizens' rights to emigrate and always attempted to control emigration flows for both political and economic reasons (Santos 2004;Pereira 2009Pereira , 2014. The dictatorship adopted an ambivalent emigration policy based on the discretionary powers assigned to the Emigration Board (Junta de Emigração) and on ill-defined regulations. On the one hand, the authoritarian regime officially restricted emigration in order to strengthen the colonial empire (in particular after the beginning of the colonial wars) and to retain control over the labour force with the aim of attaining the economic objectives of the state, particularly in agriculture and industry. On the other hand, emigration was encouraged not only because it served the nationalist component of the rhetoric elaborated under Salazar's dictatorship, but also because it limited the impact of the economic crises, especially after the regime's failed attempts to modernise the Portuguese economy (Malheiros and Boavida 2003).
Examining this emigration policy adopted by the New State helps us to understand the strong association between Portuguese nationals living abroad and rightwing political forces. As Table 4.1 shows, the Social Democratic Party (Partido Social Democrata, PSD) has always gained the most votes in out-of-country districts with an average share of 44%. 6 Its strongholds are countries outside Europe, Beyond the legacy of the authoritarian regime, it is worth emphasising that after the 'Carnation Revolution' of 25 April 1974, the move abroad of important personalities and political groups linked to the authoritarian regime enhanced the traditional dominance of conservative parties among Portuguese emigrants. In addition, the low levels of education and the low economic and professional background of those living and voting abroad have also benefited right-wing parties. 7 Consequently, the right has traditionally displayed a higher level of support for the introduction of external voting, whereas left-wing forces have been more skeptical.
Overall, Portugal's experience of the legalisation of the vote for emigrants was marked by two key moments (Malheiros and Boavida 2003: 467). The first was the passage of the first electoral law in 1976 during the transition to democracy, while the second crucial moment came in the 1990s, when emigrants' electoral rights were expanded to include participation in European and then presidential elections.
External voting rights for Portuguese emigrants were established with the first electoral law adopted after the fall of the authoritarian regime. This was a consensual issue among political parties and experts involved in the development of the first electoral law, which was used to elect the Constituent Assembly on 25 April 1975. The reasons for allowing Portuguese emigrants to exercise their voting rights were twofold. The first was related to the evolution of domestic politics following the regime change and consisted in a willingness to establish equal political rights, as well as opportunities for the effective participation and inclusion of all citizens. As the experience of other recent democracies shows, democratic transitions are often associated with the establishment and implementation of emigrants' political 7 It is worth noting that the second generation of Portuguese emigrants displays some important differences compared to the first generation: while the first emigration cycle was based on male rural workers who left their country permanently, the second wave was more balanced in terms of gender and included a higher proportion of emigrants engaged in the industrial sectors (Baganha 2003: 148-150). In addition, the move abroad was more temporary and the exchanges with the home country more dense. This partially explains why the PS has improved its vote share in European countries in recent decades. rights. 8 The second argument was based on the importance of emigrants for the Portuguese economy. Remittances represented a good share of the country's GDP between 1950 and 1970 and has subsequently been a crucial source of revenues for many families, especially in the uncertain period following the fall of the dictatorship. As other studies have emphasised (Calderón Chelius 2003), this argument has often been used to strengthen loyalty towards the home state and to stimulate ties between emigrants and their home country, especially in poor states (such as Mexico or Cape Verde).
Yet the electoral rights of Portuguese citizens living abroad initially only concerned legislative elections, as the military retained control over the (s)election of presidential candidates during the first decade of democracy. Apart from that, two main factors limited the use of external voting elections beyond legislative contests, namely for the direct election of the President and for national referenda. The first was the threshold of a two-thirds majority imposed by the Constitution in order to implement the electoral law reform. The second was related to the different interests of the main political parties. Under the Portuguese constitution, a consensus between the two main parties was needed in order to allow emigrants to vote in presidential elections and referenda. As a consequence, the expansion of voting rights was significantly influenced by the strategic calculations of the different political actors involved.
The first important change of the external vote that expanded emigrants' political rights occurred with the approval of Law 14/87, which allowed emigrants living abroad who do not opt to vote in another European member state to vote in European elections (see Table 4.2). 9 Moreover, the Council Directive 93/109/EC (6 December 1993) permitted European citizens to choose whether to vote for European elections in their state of residence or in their home country. The first time that Portuguese emigrants elected MPs to the European Parliament was in 1987. 10 In general, the PS has performed better in these elections than in legislatives ones, mainly because Portuguese citizens living outside Europe were not allowed to vote until the 2009 European elections. However, as in legislative elections, small parties have obtained a decreasing proportion of votes and turnout has also declined, with a strikingly low emigrant participation rate of 2.9% registered in the 2009 European elections, followed by a 2.1% registered in the 2014 elections.
The second major expansion of external voting concerned the election of the President of the Republic. Traditionally, the position of the right has been to defend the participation of emigrants in presidential contests, while the left has opposed it. 8 See, for example, several Eastern European and Latin American countries that introduced external voting in the 1990s (International IDEA 2007). The case of Portuguese ex-colonies such as Cape Verde, Mozambique and São Tomé and Principe are also of relevance here, as all adopted external voting after the establishment of the democratic regime. 9 The Maastricht Treaty approved in 1991 also established the right of emigrants to vote and run as candidates in European and local elections in their country of residence. 10 These elections took place in Portugal and Spain only because of their entry into the European Community in 1986. As a rule, the PSD has consistently supported broadening the emigrant vote since the 1980s as well as the adoption of the postal vote (Cruz 1998: 204). As mentioned above, this is because of the broad support that it has traditionally received from extraterritorial districts (Lobo 2007). Those who have opposed broader voting rights have argued that reform will unfairly increase the impact of the emigrant vote on election results. Emigrants could be decisive in a presidential election if that election is based on a majoritarian system, particularly in highly competitive contests. To this was added the argument that there were technical problems with the implementation of the external vote based on the personal vote, therefore calling into question the legitimacy of the final result (Malheiros and Boavida 2003: 476). In addition, the fact that Portugal's nationality law is based on jus sanguinis criteria was an argument used to imply that even people with very weak ties to Portugal can obtain citizenship. Finally, another argument deployed by those opposing reform was emigrants' lack of knowledge about national political life and the distance that separated them from the political parties, a problem that would be aggravated in 'second order' elections, in which levels of participation by the national electorate have been traditionally very low.
Despite the initial criticism of the socialists, the implementation of external voting for presidential elections became possible through an agreement between the PS and PSD on electoral system reform. In particular, the 1997 constitutional reform (fourth revision) made it possible to introduce single-member districts, corresponding to the (unfulfilled) expectation of the adoption of a mixed electoral system (Cruz 1998;Martins 2004;Sampaio 2009). With regard to external voting, this constitu- Note: in parentheses the year of the first election that applied the new regulation tional change made it possible for emigrants to vote in presidential elections and national referenda. In fact, the drafting of the electoral law of 2000 (Organic Law 3/2000), which originated with the fourth constitutional revision, was a key turning point in terms of broadening the right to vote of Portuguese citizens living abroad. The first time that this law was applied was in the 2001 presidential elections. How can we explain the strategic convergence between parties with regard to the expansion of emigrants' voting rights? Some important developments related to the domestic political scene have to be emphasised. First, the consolidation of democracy seems to have had a significant impact on the scope for electoral reform (Malheiros and Boavida 2003: 477). Institutional and governmental stability allowed the parties to adopt a more open position and to favour the expansion of emigrants' voting rights. Second, the party system has evolved and become more bipartisan, with the two main parties winning the bulk of total votes (Jalali 2007). Finally, from the 1990s onwards, there was growing recognition of the emigrant community's socioeconomic importance to Portugal. This led to growing support among the political parties for emigrants' right to political equality (interviews with PSD and PS MPs), and the government reinforced its commitment to the communities abroad with the creation in 1996 of a consultative and representative body, the Council for Portuguese Communities (Conselho das Comunidades Portuguesas). Emigrant associations have also been able to increase their political visibility, especially in longstanding destination countries where various Portuguese generations coexist such as France and the USA.

Evaluating the External Vote in Portugal
After considering the adoption of external voting and the expansion of emigrants' political rights, this section assesses the technical characteristics of voting procedures and evaluates their performance over the democratic period. Due to the lack of comprehensive reforms (Freire et al. 2008), voting abroad and its successive reforms have been characterised by several problematic features, especially in terms of voting procedures. As this section demonstrates, these shortcomings have significant implications for making emigrants' voting rights effective.
The Constitution approved in 1976 established the principle of proportionality so as to better reflect social pluralism and more effectively represent the different political forces (Cruz 1998). This law has a general character and applies equally to citizens living both inside and outside the national territory. In order to achieve this objective, the electoral formula was also 'constitutionalised' through the adoption of the d'Hondt method. The constitutionalisation stipulates that electoral reforms require two-thirds approval by the legislature, making it harder to alter the formula used to convert votes into mandates.
How proportional is the external vote in Portugal compared to the national territory? Generally speaking, the Portuguese electoral system has been considered relatively proportional (according to the Gallagher index, the average disproportionality between 1976 and 2015 was 4.8). The average district magnitude is 10.5 (for the 2015 elections) but there are variations among the districts. 11 Apart from a small number of large districts (in Lisbon and Oporto), which have a higher level of proportionality, Portugal is otherwise split up into either average-sized or small constituencies (Martins 2004;Jalali 2007;Freire et al. 2008).
As regards the delimitation of districts, the commission charged with drafting the electoral law established two extraterritorial districts for citizens living abroad. Since 1976, these districts have elected four MPs: two for the European and two for the non-European host countries. 12 The European and extra-European districts are the smallest compared with the national ones (with two MPs each), and very disproportional (Martins 2004;Martins and Mendes 2005). In practice, the high disproportionality of the extraterritorial districts means it is impossible to elect representatives from the smaller political parties; only the two largest parties are able to do so. 13 Given the number of Portuguese emigrants living in the two extraterritorial districts and the low number of MPs elected in these districts, the principle of proportionality was not strictly respected. This decision may be explained by the fear that Portugal's large emigrant community would be in a position to elect a large proportion of MPs, and this would be unfair since emigrant voters do not live in their home country (Lobo 2007: 83-84).
As regards voting procedures, the laws that regulate the external vote establish different voting methods: while in legislative elections emigrants vote by post (and have done so since 1976), in presidential elections they can only vote in person. In this case voters have to attend the polling station nearest to their place of residence in order to cast their vote. The Organic law 3/2000 also establishes a three-day voting process. As for the postal vote, voters have to fill in their ballot, which they receive 2 or 3 weeks before the election, and must then post it to the embassy or consulates in the host country, from where it is delivered to the Ministry of Interior where the votes are counted (Decree Law 95 C/76, article 8). The implementation of this voting procedure has been problematic, with several cases of illegal practices, especially where local notables or party structures were able to organise the vote collectively (in a 'syndicate vote') 14 . Moreover, the fact that the procedure adopted for legislative elections is different from presidential contests makes the practice of the external vote incongruent.
Besides the inequality between citizens in the home country and external electors and the lack of transparency, one of the main problems of voting abroad is related to emigrants' steady electoral demobilisation from the 1980s onwards. Indeed, turnout levels in legislative elections outside the national territory have declined significantly: while the rate of emigrant participation in the first post-transitional legislative elections in 1976 was around 82%, for the 1987 elections it declined to 26.5%, and in 2009 to 15.2%, while in 2011 the level of participation was 16.4%, and in 2015 the figures reached a historical low at 11.7%. 15 Although the trend towards increasing rates of abstention also characterises participation patterns within the national territory, it is worth noting that the intensity of this phenomenon in extraterritorial districts was significantly stronger than in domestic electoral districts and it also began considerably earlier. As a consequence, from the end of the 1980s national political forces and emigrant representatives began to discuss the need to reform the existing law to stimulate greater levels of participation and equality. Despite the lack of empirical studies, the causes of the decline in electoral participation in extraterritorial districts are to be found not only in voting procedures, but also in registration mechanisms (see below) and in the role played by political and civil society actors in mobilising voters.
Another problem associated with the low levels of participation concerns eligibility criteria. Overall, all citizens above the age of 18 can vote, a policy that is internationally widespread (Blais et al. 2001). It is important to note that unlike citizens residing in the home country, emigrants are not obliged to register, mainly because it is hard to monitor electoral registration abroad (Malheiros and Boavida 2003: 471). For external voting, registration occurs mainly at consulates, embassies or other designated centres, and it can be done any time up to 60 days before an election. Given the dispersion of Portuguese emigration and the weaknesses of the consular network, a substantial proportion of citizens living abroad does not register. For example, around 242,852 Portuguese citizens resident abroad registered for and were entitled to participate in the 2015 legislative elections, while a 2017 estimate put the total number of Portuguese nationals living abroad at 1.375 million 16 . This means that in practice, the eligibility criteria established for Portuguese nationals living abroad did not favour the mobilisation and integration of the Portuguese emigrant community.
Contrary to the electoral law for legislative elections, voting abroad for the presidential election presents more exclusive rules in terms of eligibility. According to 15 Figures of emigrants' turnout are based on the proportion of citizens living abroad who are enfranchised and who cast a ballot in home country elections (Bauböck 2007(Bauböck : 2398(Bauböck -2399. 16 No such estimate was available for 2015. The estimate is based on data provided by Portuguese government institutions, the Secretary of State for Portuguese Communities, the Directorate-General of Consular Affairs and Portuguese Communities (DGACCP), embassies, consular services and the Directorate General of Internal Administration (latest data available in November 2017). the fourth constitutional reform, voters must prove the existence of 'effective ties to the national community' (art. 121). The law approved in 2005 clarified this requirement by establishing that citizens can be held to have effective ties with the national community when: (a) they live in a country of the European Union or in an officially Portuguese-speaking country but ceased living in Portugal less than 15 years ago; (b) they live in any other country but ceased living in Portugal less than 5 years ago; or (c) they have gone to Portugal and stayed there for at least 30 days over the last 5 years, and can prove they can speak Portuguese (Law 5/2005 of 8 September 2005). Overall, the regulations regarding proof of effective ties to the community are not very clear and the criteria have been interpreted rather broadly (Mendes and Miguéis 2005: 11-12).
As for the relationship between MPs and emigrants, representatives of external districts have emphasised that it is impossible to represent the interests of citizens living in four different continents. Electoral campaigns have been concentrated in a small number of locations with a higher number of Portuguese residents (Paris, Brussels, São Paulo, Newark and New Bedford). The costs associated with organising abroad, the shortage of activists and human resources, and the low potential benefit in terms of votes amplify dramatically the isolation of regions where party penetration is less robust or nonexistent, and emigrant voters tend to remain less informed. A question mark hangs over the role of organisations representing Portuguese emigrants. On the one hand, they are expected to build community and encourage participation in home country politics among the local emigrants. On the other, their own existence as representatives may render voting in home country elections less necessary. By stimulating political engagement and awareness, these associations may activate the political rights of emigrants in their host country.
Our interviews with PSD and PS MPs indicate that electoral districts for the external vote need to be reformed in order to improve emigrants' representation. However, there is no consensus on the best way of strengthening MP-emigrant links. The disagreements are related not only to the number of districts (one or more), but also to the number of MPs that emigrants should elect. This debate, however, has been linked to the change of the electoral law that has long dominated the agenda on the reform of the political system (Cruz 1998;Freire et al. 2008;Freire 2019). Although electoral reform has been a recurring theme in debates about how to improve the political system, the external vote has not been a centrepiece of electoral reform proposals put forward to date. While there is convergence between parties about the need to maintain the degree of proportionality, issues related to the proximity between voters and elected representatives -specifically the type of electoral district (single-member vs multi-member) and preferential voting -have been highly contested (Martins 2004;Sampaio 2009).
It should be noted that both issues have been addressed from an exclusively national point of view, even though they are more problematic for the extraterritorial districts, especially due to the impact of the (weak) ties between emigrants and their representatives on turnout. The main reason for this is probably that external electoral districts have had a historically negligible impact on election results. Political parties and specialists alike have tended to neglect this dimension, not only because competition within these districts is limited but also because electoral outcomes are highly predictable. The fact that all MPs elected in extraterritorial districts between 1975 and 2015 belonged to the PSD or PS, with only one MP from the CDS elected in 1985 (see Table 4.1), has had a negative impact on the level of competition, as well as on parties' efforts to mobilise and inform Portuguese emigrants.
Two main conclusions emerge from this evaluation of external voting. First, there have been several problems in the implementation of emigrants' voting rights, especially with regard to fairness, political equality and turnout. The design of the electoral system for the external vote has limited the effective representation of emigrant communities, mainly due to the inconsistencies of (and controversies about) voting methods and the relationship between MPs and citizens. Second, political factors have played a crucial role in the debate about how best to represent this community, as well as in blocking electoral reform. This can be seen not only in the 'politicisation' of external voting, but also in the confrontation between the parties of the left and right, and the strategic calculations underlying the stances adopted by the political parties. It is noteworthy, for example, that the PS supported extra-European emigrant voting in presidential elections only when its demand for the personal vote was acceded to, while the PSD criticised this procedure when the PS proposed the same voting procedure for the parliamentary elections. Overall, the implementation of the external vote has displayed significant shortcomings, which have been brought to light by the incongruent reforms implemented during the democratic period.

Improving External Voting: The Recent Debate
The debate on the electoral law for external voting emerged again when the PS put forward a proposal in 2007 to adopt the personal vote not just for presidential but also for legislative elections (Draft Law 562/X). As mentioned earlier, there had already been a debate about voting procedures for citizens residing abroad when the legislation on presidential elections had been approved. On that occasion, left-wing parties supported the personal vote, while the PSD and the CDS-PP defended more permissive procedures, considering other voting methods such as postal voting.
The main argument in favour of the change proposed by the PS was that the personal vote would guarantee greater rigour and transparency. José Lello, the Socialist MP who authored the draft law, referred explicitly to vote fraud in Brazil during the 2005 legislative elections as an example of the problems with the existing electoral law. The underlying argument was that the existing law made it easier for parties to organise 'vote syndicates' and thereby distort the free choice of emigrant voters (Público, 23 September 2008). But there was another argument based on uniformity of methods, as it was deemed unacceptable that there should be two different procedures governing the external vote. Finally, José Lello highlighted that one of the problems with the postal vote was that it forced voters to choose a candidate long before the elections took place, when the electoral campaign was only just taking off. This meant it was harder for Portuguese citizens living abroad to be informed about and gain a clear picture of party stances and candidates' programmes. The socialists underlined that the high number of null votes was evidence of this problem: null votes accounted for around 8% of votes cast in the 2005 elections, compared with less than 1% for the previous presidential election. In 2015 elections, null votes accounted for circa 11% of votes.
Several parties and emigrant associations highlighted the logistical and organisational challenges presented by the new voting method, which reduced the electoral participation of Portuguese citizens living abroad. Indeed, the reform of the electoral law had been preceded by a decision, approved on 15 March 2007, to restructure the consular network, which led to the closure, merging and downgrading of various consular representations. Moreover, the Socialist government also decided to reduce the cost of dispatching electoral campaign material to emigrant communities. These measures produced public contestation among emigrant associations and lobbies, particularly in France and the US.
Despite the opposition of the main parties of the right and of the emigrant associations, the Socialist majority succeeded in getting the proposal approved in December 2008 with the support of all left-wing parties, while the PSD and CDS-PP voted against. However, President Cavaco Silva (of the right) vetoed the new law, on the grounds that not only was this reform not politically timely (it was an election year), but it also encouraged the electoral demobilisation of emigrants since the consular network would be unable to meet the logistical demands. The main criticisms expressed by the president echoed those of the parties of the right, namely that it would become harder for Portuguese emigrants to find polling stations. In his message to the parliament justifying his veto, Cavaco Silva highlighted the fact that the personal vote was associated with lower levels of participation as compared to the postal vote. A final issue he raised in his statement was the difficulty of finding alternative places to cast votes that would guarantee transparency and the inviolability of votes.
Whether or not to adopt the electronic vote was a key issue in the electoral reform debate. Indeed, José Lello criticised this method when presenting the Socialist draft law for two reasons: first, because there was evidence that the electronic voting system was not secure or transparent, as some European countries have shown; 17 and second, because inequality among citizens would increase, since the Internet was not widespread among the Portuguese emigrant communities.
The presidential veto made it harder to approve the law since a two-thirds parliamentary majority is required to pass organic laws. This impasse led the PS to give up on the reform, particularly in view of the upcoming electoral cycle in 2009. The reform of external voting was again at the centre of the political agenda in the period preceding the 2011 presidential elections. Given their disagreements on voting methods, the main parties adopted a 'minimalist' approach with the aim of implementing only minor and marginal changes. The new law (3/2010, 15 December) reduced the voting period from 3 to 2 days for presidential elections, and increased the number of voters required to establish a polling station (from 1000 to 5000). This law also abolished the requirement to have an effective linkage to the national community for those who have been living in Europe for more than 10 years or outside Europe for more than 15 years, thus increasing the inclusiveness of voting for citizens residing abroad. This reform was approved by the main governing parties (PS, PSD and CDS-PP), while both radical left parties either abstained (in the case of the Left Bloc, BE) or rejected (Portuguese Communist Party, PCP) the proposal, stating that it was too close to the election day.
During the XIII Legislature (inaugurated after the 2015 elections), another important change occurred with indirect consequences for the expansion of eligibility for the external vote. Decree Law n.° 71/2017, which revised the Nationality Law, was passed, allowing grandchildren of Portuguese emigrants to acquire Portuguese citizenship. 18 Finally, regarding eligibility criteria, in 2018 the government presented a proposal to implement the automatic registration of national citizens residing abroad. If it is adopted, this change will increase the number of registered voters abroad from circa 318,000 to 1.375 million voters (68% of new registered voters are resident in Europe, and 32% outside Europe) 19 . In addition, the debate on voting methods -in particular the adoption of electronic voting -is still ongoing. After the petition discussed by Parliament on this topic in 2016, the PSD proposed to give emigrants this possibility and the Socialist government agreed to implement a pilot to test this procedure 20 .

Conclusions
Emigrant voting rights throughout the world have increased over time. This has also happened in Portugal, which recognised this right during the transition to democracy, in time for the first democratic legislative elections. As in the Spanish case, the fall of the authoritarian regime gave the biggest impulse to the enfranchisement of 18 Criteria for granting nationality by effect of will to grandchildren of a Portuguese national: Individuals born abroad with at least one second-degree ancestor of the Portuguese nationality who has not lost this nationality, and who wish to be granted Portuguese nationality, shall meet the following requirements cumulatively: a) To declare that they want to be Portuguese; b) To have an effective connection with the national community; c) To register his/her birth in the Portuguese civil registry, after recognition of the connection to the national community (Decree Law n°71/2017). 19 The estimate is based on data provided by Portuguese government institutions, the Secretary of State for Portuguese Communities, the Directorate-General of Consular Affairs and Portuguese Communities (DGACCP), embassies, consular services and the Directorate General of Internal Administration (latest data available in November 2017). 20 See https://www.publico.pt/2017/06/04/politica/noticia/governo-pondera-modelopiloto-paratestar-voto-electronico-dos-emigrantes-1774550 Portuguese emigrants. In addition, the economic importance of remittances led political actors to implement emigration policies in order to sustain the allegiance of citizens living abroad towards their home country.
Two main problems affected the full inclusion and equality of emigrants' political representation: the lack of representativeness of the electoral system, on the one hand, and the limitations of and incongruencies posed by voting methods, on the other. These shortcomings also have important implications for the level of participation of Portuguese citizens living abroad, which has displayed a significant decrease over the last decades. These conclusions confirms previous findings (Smith 2008;Bolzman 2011;Tintori 2011) which indicate that the procedural and organisational aspects of electoral laws are a key to determining the success of the external vote.
It should be noted that political parties were crucial actors for determining the content of the external voting legislature and its (possible) reform. This work has shown that the main reasons underlying the difficulty of reforming have been not just the 'constitutionalisation' of electoral rules, but also the conflict between the main political parties over emigrant suffrage rights. Indeed, there has been a strong divergence among political actors along partisan lines, particularly as regards voting methods. Traditionally, while the PS and radical left parties (PCP and BE) have favoured the personal vote, the PSD and CDS-PP have been more open, backing the postal vote and, more recently, the adoption of electronic voting. Therefore, strategic considerations represent an important barrier against the improvement of external voting -a conclusion that confirms the findings from other empirical studies (Tager 2006;Rhodes and Harutyunyan 2010;Lafleur 2013). The limited impact of emigrants' vote on final electoral results, the low levels of civil society mobilisation and the characteristics of Portuguese emigration -in terms of both socio-economic background and geographical distribution -help explain why the reform of external voting has been blocked by political parties' interests. Finally, the main political actors did not recognise the autonomy of the external vote with respect to the domestic electoral system, relegating this problem to the general debate about the need to improve political representation at the national level.
This study has served to illustrate that there are limits on the representation of emigrants, namely in terms of participation, proportionality and the efficacy of voting methods. Therefore, the reform of external voting is an important step towards the improvement of the quality of contemporary democracies through a greater inclusiveness and equality among citizens. Moreover, the analysis of the activities of parties, MPs and other institutional actors with regard to emigrant communities provides further elements for understanding the relationship between emigrants and their representatives and patterns of mobilisation. ISCTE -Instituto Universitário de Lisboa (ISCTE-IUL), and financed by the Portuguese Emigration Observatory (Observatório da Emigração). The proofreading of this chapter was funded by the Portuguese Emigration Observatory (Observatório da Emigração). The authors would like to thank the DGAI, the CNE, the DGACCP, the CCP and the OEm for their support and cooperation. The authors would also like to thank also the anonymous reviewers and Imiscoe Series editors for their useful comments on a previous version.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.