Polity Size Matters? The Effect of Country Size on Politicians' Professionalization

Abstract

This article examines the effect of country size on the professionalization of politicians in six European micro-states and a large-scale democracy – Germany – since 1980. The article revisits an ongoing debate about the extent to which either country size or government size are causal factors in the individual professionalization process. Using an original dataset consisting of 6,940 parliamentary mandates – 2,809 individuals – in national parliaments, the article shows that country size is a determinant of the degree of politicians' professionalization. The article further demonstrates that political parties' gatekeeper role is the key causal mechanism explaining the influence of population size on politicians' professionalization.

Democratic politics in large-scale polities are characterized by the installation of a representative government in which the holding of public office has been increasingly professionalized. The academic literature explains how the enlargement experienced by contemporary political units, in terms of both the population's voting rights and government size, set up the perfect frame to develop institutional and individual professionalization.

The population size of contemporary political units has grown, affecting the nature of political deliberation and making direct democracy impossible, at the same time increasing the difficulty of dealing with societal problems (Reynaert 2012). Due to polity size and the implementation of universal suffrage, contemporary large-scale democracies need representative democracy and its professionalization in order to deal with the heterogeneous interests, ideologies and problems that exist in their societies. As they developed, large-scale democracies, of course, were not only bigger than old democratic city-states but also acquired new state functions (Marshall 1950). The main objective of the central state of protecting its population from foreign intrusion or domestic violence was no longer the only main function. New modern state welfare had to deal with economic insecurity and providing services and income on the basis of individual rights. In addition, governments increasingly began to take an interest in many other social issues such as public education, public health, length of working hours, and relations between employers and workers (Kuhnle and Sander 2010). Increased competencies created higher demands of public office holders, for more expertise in the policy process and for higher levels of investment in terms of personnel effort and time. In other words, nation states needed more staff and qualified, full-time, dedicated policymakers (Samuels and Shugart 2010). In sum, it seems that either country size or government size could be the explanatory factor for individual professionalization.

However, the micro-state literature claims that it is the (small) population size that, in reality, explains the absence of individual professionalization in smaller states. The main argument points to the effect of population size on a key determinant of individual professionalization: the political parties. The professionalization literature argues that individual professionalization is determined by the politician's ambition, the institutional structure of opportunities and party goals (Borchert 2003; Maestas 2000). However, even though the influence of political parties on political professionalization is well established in national politics in large-scale democracies, such influence is not ensured in smaller political entities such as micro-states. In fact, due to the high levels of social intimacy and direct forms of communication that exist in small-scale political units, the process of representative delegation through political parties becomes less necessary and more personalistic. Political parties in micro-states are less developed and have weak extra-parliamentarian organizations (Corbett and Veenendaal 2018), and so have lower capacity to monopolize the recruitment process. The political party's role as a gatekeeper thus seems to be a causal mechanism between the population size of the country and individual professionalization.

This article represents an attempt to disentangle empirically the effects of country size on the professionalization of parliamentarians. To do so, it compares six European micro-states – usually understood as those sovereign states with fewer than 500,000 citizens (Veenendaal and Corbett 2015) – with a large-scale democracy, Germany. In doing so, the article makes three important contributions to understanding the causes of politicians' professionalization. Firstly, it confirms the micro-state literature's argument that population size, and not government size, is the determinant for politicians' professionalization. Secondly, the article assesses the role of political parties as a causal mechanism that explains the influence of country size and individual professionalization. Finally, the analysis contributes to the micro-state literature by explaining how population size affects parties' function with regard to the recruitment and nomination of candidates.

Determinants of politicians’ professionalization in large-scale democracies

After the locus of democracy shifted from communes, cantons and city-states to the nation state, contemporary political units enlarged their population size and competencies. This created the perfect conditions to develop institutional and individual professionalization. The political professionalization of national politics thus became a contemporary characteristic of large-scale democracies (Musella 2014).

Consequently, individual and institutional professionalization began to be studied by political scientists. Different scholars underlined how most political offices at the national level were occupied by professionals in contemporary democracies (Borchert and Zeiss 2003). The literature thus generated a common wisdom concerning the negative consequences of the monopolization of political positions by professional politicians (Allen 2013; Azari 2017).

A professional politician is a person who works full time in politics, has a long period of experience in this field and obtains their main income through their political activity (Borchert 2011). Due to politicians' ambition to maintain their careers, individual professionalization creates accountability and representation problems (Maestas 2000). However, the process of individual professionalization is related not only to politicians' motivation by ambition but also to a concrete interplay between individuals' motivations, the institutional structure of opportunities and party goals (Borchert 2003; Maestas 2000).

As Jens Borchert (2011) underlines, professional politicians do not differ markedly from a practitioner in any other distinguished profession: they want to stay and rise to a higher position. Therefore, the ambition of the politician is a clue to understanding the professionalization process.¹ The salary – the first component related to the professionalization process (Squire and Moncrief 2019) – constitutes a great incentive to enter and continue political service in the legislature (Cotta and Best 2007). Monetary rewards allow legislators to focus their energies on their legislative responsibilities. In this way, politics becomes the individual's exclusive occupation.

Other important motivations for the professional politician are career maintenance and career advancement. As the sociology of professions illustrates, safeguarding one's career and the aspiration to hold various positions of increasing relevance (Astudillo and Martínez-Cantó 2019) are natural goals of any professional worker. In politics, the establishment of a salary and the ambition of career maintenance creates the frame to achieve greater political expertise (Thompson and Moncrief 1992).

Institutional particularities and political parties' characteristics also influence individuals' professionalization. Following increased public administration responsibilities (Borchert 2011),² public offices required a higher level of expertise and higher levels of investment in terms of time and personal energy (Carreras 2017). Higher workloads demanded institutional professionalization. In that sense, institutional professionalization created a structure of opportunities for the potential candidate – a concrete cursus honorum – to fulfil, and created selective incentives (salary) that led to an ambition and capacity to live off politics (Schlesinger 1966).

The literature also highlights how political parties became essential for the development of politicians' political careers (Samuels and Shugart 2013). After the implementation of universal suffrage, political parties became essential actors of representation. The increase in the number of voters made direct participation unfeasible; thus, political parties became politicians' instruments to win elections and articulate people's concerns. Furthermore, the number of potential voters made it impossible for a politician to conduct a political campaign by him/herself: politicians need an extra-parliamentary organization in order to reach potential voters (Aldrich 2011).

Though political parties became increasingly instrumental – seeking to win elections in order to implement policy changes – most political parties, at least in Western Europe, underwent a transformation in which maintenance of the organization itself became the intrinsic goal of the political party (Panebianco 1990). Consolidated parties, with a developed extra-parliamentary organization, evolved coordinated strategies (Scharpf 1997) aimed to preserve their power and maintain the institution (Panebianco 1990), effectively turning politicians into the agents of the extra-parliamentary organization. Politicians not only represent and serve citizens, they also work for and represent political parties (Maravall 2003). This process was reinforced because, at the same time – and in part due to the party organization's acquisition of value – citizens began to change their electoral behaviour. Citizens' votes moved from being candidate-centred to being party-centred.

Nevertheless, political parties still face so-called agency problems in the political recruitment process. Political organizations need politicians to follow their instructions to fulfil their intrinsic goals. So, to avoid adverse selection and moral hazard,³ institutionalized parties use their role as political gatekeeper to create controls to test politicians' ‘reliability’ (Samuels and Shugart 2010). As a result, if an individual aspired to certain political positions, they needed to demonstrate the skills designated for the required tasks and loyalty to the political party.

As David Samuels and Matthew Shugart (2010: 63) argued, political parties apply ex ante and ex post controls to test candidates. In this way, political parties subject candidates to an implicit interview process in which politicians compete to demonstrate their usefulness and reliability. As in any other profession, such an implicit interview is based on a selection criterion: previous experience. Institutionalized political parties prefer qualified politicians with full-time dedication and those who are socialized in the party's culture and practices (Hazan and Rahat 2001), all the more so when citizens vote for parties rather than independent candidates who are unlikely to be selected. In consequence, and in a bid to persuade ambitious politicians to be involved in politics and in the organization, political parties look for what incentivizes politicians and will encourage them to deploy the effort and time they need to spend on the political party (Schlesinger 1984). Thus, politicians' professionalization is also motivated by the candidate selection processes of the institutionalized political parties.

The effect of country size on the individual professionalization process

Academic studies – some explicitly, some implicitly – argue that small state size makes the process of professionalization harder. As Élodie Guérin and Éric Kerrouche (2008) illustrated in their investigation, the common wisdom in European local politics is that representatives typically have a voluntary and amateur character.

Authors like Herwig Reynaert (2012) have shown that the size of a polity and its heterogeneity affects the nature of deliberation and the degree of difficulty in dealing with society's set of problems. A larger population means a greater need for professional politicians in order to have a greater chance of responding effectively to heterogeneous interests, ideologies and problems.

One causal mechanism that is said to explain the country's effect on individual professionalization is related to the role of political gatekeeper. While the influence of political parties on political professionalization is well established in national politics in large-scale democracies, such influence is not ensured in smaller political entities. In fact, due to the social proximity of citizens and politicians in smaller political units, the representative delegation process through parties becomes less necessary.

Contemporary large-scale democracies need representative democracy; thus, political parties become the necessary link between society and the state (Katz and Mair 1995). As a consequence of such a monopoly of representation, political parties in this context become political gatekeepers, politicians becoming agents of the extra-parliamentary organization. Therefore, individuals aspiring to certain political positions need to hold various positions of increasing responsibility and authority in public administrations and within a party (Astudillo and Martínez-Cantó 2019).

The lower standards of professionalization in local politics emerge due to the existence of some sort of direct politics due to the close relationship between citizens and their representatives. The smaller scale allows politicians to organize and conduct their political campaigns by themselves, and makes the citizens' vote more candidate-oriented, regardless of the electoral system. Political parties, in this scenario, have less developed and weaker extra-parliamentary organizations, and have a lower capacity to monopolize the recruitment process.

The problem with this comparison between local and national politics is that local politics are different not only in terms of population size, but in competencies, too. The political competencies of a national government in a large-scale democracy make its political management more complex, requiring higher levels of institutional professionalization. In brief, the local-versus-national comparison gives rise to a covariance scenario in which it is not possible to determine whether it is population size or competencies that explain variation in individual levels of professionalization.

Aiming to resolve such questions, this article introduces an analysis of European micro-states. The article runs an analysis comparing the level of politicians' professionalization in European micro-states and a large-scale democracy, Germany. This analysis maximizes differences in terms of size and reduces the difference in institutional competencies that exist when comparing local and national politics.

Research design

The empirical analysis consists of two parts to study the extent to which country size has an effect on individual professionalization. First of all, we examine whether either country size or government size, or both, is a causal factor of individual professionalization. In this first analysis, and seeking to avoid the bias that the inclusion of Germany in the results can produce,⁴ the article runs an additional model to determine the effect of size on politicians' professionalization in European micro-states. The article goes on to examine the causal mechanism that explains the influence of country size on individual professionalization: the political parties' gatekeeper function.

This study has faced some important limitations in terms of available information. Although the literature concerning micro-states underlines the lower institutionalization of political parties and lower levels individual professionalization (Corbett and Veenendaal 2018), in practice it is not possible to run a systematic comparative and quantitative analysis between large state democracies and micro-states because of the lack of data. Alongside the omission of micro-states in the literature (Veenendaal and Corbett 2015), there is a shortage of comparative information about politicians' professionalization in large-scale democracies. The existing data on politicians' professionalization are usually based on case studies, and these are normally centred on qualitative analysis of political professionalization characteristics (Borchert and Zeiss 2003; Squire and Moncrief 2019). Consequently, there are no comparative indicators of politicians' professionalization in either micro-states or large-scale democracies. For that reason, we have created an original database for European micro-states and Germany. Germany was selected as a benchmark because it is the large-scale state paradigm of professionalization of politics in general (Weber 1958) and the exemplar of individual professionalization in parliaments. Germany was also selected due to the long-term party commitment present in politicians' biographies (Wessels 1997).

The database is composed of those politicians who have won parliamentary representation. As is common in the professionalization literature, this article focuses on politicians in national parliaments. Due to the availability of information and the years of elections, the exact periods for each country were: Andorra 1981–2019, Liechtenstein 1986–2017, Iceland 1995–2017, Malta 1982–2017, Monaco 1978–2018, San Marino 1983–2016 and Germany 1990–2017.⁵ In total, the article analysed 6,940 parliamentary mandates, or 2,809 individuals. The original dataset was built from information primarily obtained from parliamentary websites, newspapers, internal party documents and politicians' biographies.

Hypotheses

To understand the variation between Germany and the micro-states, along with the variance between micro-states, it is necessary to recognize the particularities that explain the presence of a difference in terms of politicians' professionalization.

Our first hypothesis aimed to test if country size is a determinant for individual professionalization:

Hypothesis 1a: Ceteris paribus, small democracies have a lower degree of politicians' professionalization than Germany.

Hypothesis 1b: Ceteris paribus, the smaller the population size within small democracies, the lower the degree of politicians' professionalization.

We also analyse the causal mechanism that explains the influence of country size on individuals' professionalization. Political parties in most European micro-states have a weak extra-parliamentary organization, low institutionalization, low professionalization, low membership and highly personalist politics (Corbett and Veenendaal 2018). Therefore, micro-states' political parties should have a lower capacity to monopolize the recruitment process.

However, institutionalized micro-state political parties are still interested in winning elections and also have the intrinsic goal of maintaining the extra-parliamentary organization. Such political parties will therefore keep using their role as a political gatekeeper, looking for well-known candidates, although the organization cannot ensure their reliability and technical skills. Because of the low membership of micro-state political parties (Corbett and Veenendaal 2018), political recruitment from the centre, based on selecting insiders, becomes harder. Additionally, due to the small pool of candidates who can demonstrate agency experience, political parties have to choose outsiders with vote-drawing ability and a supra-partisan public image (Samuels and Shugart 2010). In sum, micro-states, due to their personalistic politics and the lower capacity of political parties to monopolize political recruitment, have less control over the access of insiders to political positions. Thus, our second and third hypotheses aim to respond to the question raised by Jack Corbett and Wouter Veenendaal (2018) on how country size affects parties' functions with regard to the recruitment and nomination of candidates.

Hypothesis 2: Ceteris paribus, in small-scale democracies there is a lower likelihood that an MP has previous party experience than in Germany.

Finally, based on what the literature says about how size should affect the party condition as a determinant of politicians' professionalization, our third hypothesis is as follows:

Hypothesis 3: Those MPs with previous party experience will have a significantly higher degree of individual professionalization.

Main variables and operationalization

Main dependent variables

To measure the role of political parties as gatekeepers, this study uses a proxy that seeks to capture the extent to which political parties are capable of monopolizing the recruitment process and imposing their selection criteria. For that reason, we use previous experience in political parties. This variable captures the parliamentarian's experience in any relevant position in the extra-parliamentary organization, assigning a value of 0 to politicians with no previous experience in their political party and 1 to those with experience.

To measure the degree of politicians' professionalization, we use two different indicators to help capture the complexity of a concept such as individual professionalization. As Max Weber (1958) established, a professional politician is defined by two decisive factors based on time and monetary budget. Professional politicians differ from amateurs because they have long experience in politics, work full time in politics and their main income is obtained from political activity. Hence, the age at which the politician first took up public office is assumed to be a good indicator of the period dedicated to politics (Schlesinger 1994).

We considered the percentage of adult life dedicated to politics and the time spent (part-time or full-time) in politics. Thus, the index of individual professionalization includes the proportion of adult life involved in politics since the age of 21 years and the type of political dedication.⁶ The proportion is calculated by dividing the total number of years any public office was held by the total years of adulthood (Astudillo and Martínez-Cantó 2019). The proportion of adulthood involved in politics is divided by the type of political dedication, full-time (1) or part-time (2). It is considered that part-time and full-time reflect whether a politician is living for or living off politics.

Main independent variables

To measure the size of a country, this study first uses a dichotomous variable to capture the effect of micro-state political characteristics. The variable assigns a value of 0 to Germany and 1 to the European micro-states. For the models that analyse variation within micro-states, we use the population of each state as the country size, following Robert Dahl and Edward Tufte (1973). The population is logged to consider the difference in population between Malta (388,931 inhabitants) and Iceland (306,001 inhabitants), and the other European micro-states – Andorra (63,747 inhabitants), Liechtenstein (32,881 inhabitants), Monaco (32,638 inhabitants) and San Marino (27,692 inhabitants).

Secondly, like Frank Häge (2003), we use general government expenditure as a percentage of GDP to measure government size. In this way we avoid overstating the size of the public sector.

Finally, to analyse the causal mechanism that explains the influence of country size on individual professionalization, we use previous experience as the main independent variable.

Control variables

The control variables of both statistical analyses are based on factors that the literature argues have an effect on individual professionalization and political party gatekeeper function.

We control for gender due to the historical male monopolization of national politics in Western democracies (Verge and Astudillo 2019). As Tania Verge and Javier Astudillo (2019), define, female candidates are held to higher standards due to a gender-biased evaluation of candidates made by party selectorates. Thus, female candidates have to be exceptional to overcome social, structural and political barriers to office (Verge and Astudillo 2019). As a consequence, the expectation in this case is that, due to the power and desirability of national positions and the higher internal competition for the position, women will have a lower probability of being professionalized.

Education level is another important variable of the homogenization of the political elite. A the common trend in national politics is that the education level of politicians has increased in the last decade (Tronconi and Verzichelli 2007). Because of political parties' interest in well-educated candidates, we would expect professional politicians to be more highly educated. This study also looks at the previous professions of politicians. As Maurizio Cotta and Heinrich Best (2007) underlined, there is a predominance of professional backgrounds among the political elite, civil servants and professors.

In addition, the analysis includes institutional professionalization. As Peverill Squire and Gary Moncrief (2019) argued, the development of institutional professionalization facilitates individual professionalization. To measure the level of institutional professionalization, the Squire Revised Index is used. This index includes three main indicators: parliamentarians' salary, parliamentary staff and the sitting days of parliament. Due to a lack of information on the particular staff of each of the micro-states' parliaments, we use the alternative measure defined by Squire (2007), the parliamentary budget. In contrast to Squire's investigation, and because we are comparing different states, this article does not establish a parliament as a benchmark to calculate the index, and it standardizes each of the components of the index. In sum, institutional professionalization is calculated by summing each of the standardized components and then dividing it by three.

Finally, the analysis includes some control variables related to countries' characteristics. We include the age of the democracy, calculated following Carles Boix et al.'s (2013) instructions.⁷ The age of the democracy also influences political parties' gatekeeping function. Finally, the article includes an economic variable, the annual growth of national GDP (Rasmussen and Knutsen 2019).

Is country size a determinant? Individual and institutional professionalization of European micro-states and Germany

As a first way of assessing the effect of country size on politicians' professionalization, this article offers a comparative overview of the main political traits of the European micro-states and our macro-state, Germany.

Table 1 shows that there are important and significant differences between Germany and the European micro-states in relation to our dependent variables. Politicians in micro-states occupied their first public office when they were, on average, 39 years old, while politicians in Germany had access three years before. Furthermore, the individual professionalization index reveals that politicians in Germany have more than double the politicians' degree of professionalization in the micro-states. Finally, the table describes existing differences in terms of party service. In European micro-states, only 36.60% of parliamentarians had previous party experience, while 77.39% of Bundestag parliamentarians registered previous experience in their extra-parliamentary organization. This confirms Corbett and Veenendaal's (2018) thesis on the existence of less institutionalized political parties. The political characteristics of micro-states clearly have implications for the party service results.

Table 1. Descriptive Data of the Dependent Variables

Note: Except for ‘Age at first public office’ – which has equal variances – the differences have been calculated with Welch's t-test due to their unequal variances.

Table 2 shows how there are also substantial and significant differences in terms of institutional professionalization and its components. In general terms, micro-states register lower results for all components. For example, the mean annual salary of politicians in micro-states was $34,994.45 (purchasing power parity – PPP), while in Germany it was $96,603.8 (PPP). Micro-states registered 47.022 sessions annually compared to 61.62 sessions in Germany. Finally, the parliament's mean annual budget for micro-states was $6,617,331 (PPP), while the mean in Germany was $775,337,316 (PPP).

Table 2. Descriptive Data of the Independent Variables

Note: Except for ‘Politician annual salary’ – which has equal variances – the differences have been calculated with Welch's t-test due to their unequal variances.

However, despite such differences in political professionalization between the micro-states and Germany, Table 2 shows that there was a lesser, but significant, difference in terms of government size. While the micro-states registered 17.19% of general government consumption expenditure as a percentage of GDP, Germany spent 19.37%. There is covariation between all independent and dependent variables. It is not, therefore, possible to know which variable matters. It is necessary to introduce controls in a quantitative analysis to unravel the influence of such variables.

Statistical analysis

In the following statistical analyses, our unit of analysis is parliamentarians elected in each general election in our six micro-states and Germany. Given that this article treats each time an individual is elected as a parliamentarian as a separate observation and the same individual may be elected several times as a representative, the article has estimated a series of random effects panel regression models. In these panel models, the standard errors of estimates are corrected to consider repeated observations for each individual across legislative terms.

The effect of country size on politicians' professionalization

We built four models to test the effect of country size on two dependent variables: the age at which the politician joined parliament for the first time and the index of politicians' professionalization. We use the micro-state condition and the population size to capture the effect of country size on individual professionalization.

The results of our models in Table 3 and Table A1 (see Supplementary Material) indicate that country size is a determinant of having a higher degree of individual political professionalization). Table 3 identifies the micro-state condition as a negative determinant for politicians' professionalization. In Model 1, the micro-state condition negatively influences the age of the politician when first accessing public office. As Figure 1 shows, country size clearly affects the age of the politician on first holding public office. In Germany, a politician first held public office when he/she was 37.38 years old. In micro-states, on average, the mean age was 43.18 years. Model 2 (Table 3) and Figure 2 demonstrate that the micro-state condition subtracted 23.52 points on the individual professionalization index in comparison with Germany. Table 3 validates the expectation that politicians in political units with larger populations need to be full-time, to have high expertise and knowledge in order to have a greater chance of solving the state's heterogeneous interests, ideologies and problems.

Figure 1. Predictive Marginal Effect on Age First in Public Office

Figure 2. Predictive Marginal Effect on Individual Professionalization Index

Table 3. Determinants of Politicians' Professionalization

Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

The robustness check in Table A1 in the Supplementary Material shows that, even within extreme cases (European micro-states), population size remains a negative determinant of the professionalization of politicians. This provides a more nuanced lens on the ‘amateur’ character of micro-states and recognizes their variance, strengthening the argument that decentralization, and its associated emphasis on citizen participation, is a valid antidote to the increased professionalization of the political classes in larger states (Stoker 2006).

Interestingly, Table 3 and A1 reveal a non-significant effect of government size on three out of four of the models. Government size seems not to be a determinant for individual professionalization. In addition, the models in Table 3 and A1 reveal that other systemic and individual characteristics act as determinants for the professionalization of politicians. Both models show a positive effect of age on the age when the politician first took public office, indicating a possible cohort effect in one of the two models (Model 1, Table 3). Table 3 shows that university studies have a positive relationship with the degree of individual professionalization. The analysis confirms the homogenization of the political elite through increasing levels of education of politicians (Tronconi and Verzichelli 2007). Table 3 also indicates that some professions facilitate a higher degree of professionalization of politicians. The models in Table 3 suggest that trading or banking professions lead to politicians first entering public office at a younger age.

Finally, the models in Table 3 show that the age of the democracy is a determinant for individual professionalization. On the one hand, the older the democracy, the earlier its politicians first have access to public office. On the other hand, the results of the professionalization index show that the older the democracy, the higher the degree of individual professionalization. The results indicate that growth in the GDP of the country is significant in both models: the greater the growth in the country's GDP, the higher the degree of professionalization.

The effect of country size on politicians' professionalization through political parties' gatekeeper role

We examine the effect of country size on politicians' professionalization through political parties' gatekeeping role, by building three additional models. We begin by analysing the effect of country size on the previous party experience of parliamentarians. The second analysis tests the effect of previous party experience on politicians' professionalization proxies. We use the micro-state condition to capture the effect of country size on individual professionalization.

The first analysis (Table 4) demonstrates the negative effect of the micro-state condition on the political gatekeeper role of political parties. The statistical model reveals how being a micro-state reduces the probability that MPs have partisan experience, underlining the idea that population size affects the capacity to select from within the organization. Micro-state political parties select more outsiders with a vote-attracting ability and a supra-partisan public image (Samuels and Shugart 2010).

Table 4. Determinants of Previous Party Experience

Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

As previously stated, Germany may produce this effect not because of its size but due to another unknown factor. Thus, we replicate the analysis within European micro-states (see A2 in the Supplementary Material). The results underline the robust effect of size on the political gatekeeper role of political parties. The statistical model highlights how, even within micro-states, a larger population increases the probability that a politician has partisan experience. Hence, the results validate our second hypothesis, and that of the micro-state literature, which proposed that country size will have a positive effect on access to political positions. However, as Corbett and Veenendaal argued (2018: 13), European micro-states are part of a group of worldwide micro-state cases in which there is an institutionalized party system (such as the Caribbean islands, the Seychelles, Samoa and Tonga). Consequently, the scope of such results is restricted to states exhibiting this party-system characteristic. This article does not seek to address those micro-states that have not institutionalized their party system (the Maldives, São Tomé and Príncipé or Vanuatu) or those that do not have political parties (Kiribati, the Marshall Islands, Nauru or Tuvalu).

Table 4 reveals that there are other individual and systemic characteristics acting as determinants of greater experience in the extra-parliamentary organization of political parties. Regarding individual characteristics, the model indicates that the age of the politician and their profession are explanatory factors for having experience in the party. In relation to politicians' professions, the model reveals that being a manager is the profession that most facilitates previous experience in political parties' extra-parliamentary organizations. In terms of the age of the candidate, the model in Table 4 indicates that being older facilitates the probability of having previous experience in a political party. The result indicates an opportunity-cost problem. The opportunity cost of a well-established professional, normally older, to switch career is higher than for those who are at the beginning of a professional career. For that reason, older politicians who are present in the legislature have more interest in being in parliament and are therefore highly interested in demonstrating their reliability to a political party. Finally, as the literature argued (Mainwaring 1998), the analysis underlines how the age of the democracy is relevant in allowing political parties to control their political recruitment process.⁸

After determining that the micro-state condition hinders the selection of candidates with previous party experience, Table 5 tests whether having previous experience in political parties' extra-parliamentary organizations affects the process of politicians' professionalization. The results of our models in Table 5 reaffirm Samuels and Shugart's (2013) argument concerning the relevance of the partisan component in the professionalization process. As the models in Table 5 demonstrate, parliamentarians who register previous experience in their political parties had access to the parliament for the first time when they were younger (Figure 3). Therefore, despite the micro-states' particular characteristics, political parties in micro-states are still applying previous experience as a determinant for candidate selection. In fact, as shown in Figure 4, party service becomes a springboard to reach public office faster and to have more time in parliament. Such results confirm that politicians are not only agents of citizens but work for and represent political parties (Maravall 2003), giving rise to the potential existence of agency problems. In other words, because political parties use controls to test politicians' reliability (Samuels and Shugart 2010) and because professional politicians are motivated by the maintenance and advancement of their career, there is a conflict with their responsiveness and accountability to citizens.⁹

Figure 3. Predictive Marginal Effect on Age First in Office

Figure 4. Predictive Marginal Effect on Individual Professionalization Index

Table 5. Determinants of Politicians' Professionalization

Note: Standard errors in parentheses. * p < 0.05, ** p < 0.01, *** p < 0.001.

The models in Table 5, as with Table 3, also reveal that there are other systemic and individual characteristics acting as determinants for the professionalization of politicians. Firstly, both models confirm that the professionalization of the institution facilitates the development of higher individual professionalization (Squire and Moncrief 2019). In addition, and in relation to the individual characteristics that facilitate professionalization, the models in Table 5 reinforce the effect of most of the variables described in Table 3. Despite this, Table 5 points out how being female has a negative effect on the professionalization index. In other words, as was expected, women are less likely to be higher on the professionalization index due to the power and desirability of professionalized positions. Sadly, these results underline again how female candidates are held to higher standards because of the gender-biased evaluations of candidates made by party selectorates (Verge and Astudillo 2019). Even with such a difference, Table 5 underlines how university studies and being an administrator, trader or banker increase the likelihood of having a degree of individual professionalization. Additionally, the models in Table 5 reveal that the age of the politician has a significant and positive effect on individual professionalization proxies. Finally, it is shown that the age of the democracy remains significant and has a positive effect on the degree of individual professionalization.

Conclusions

This article addressed the debate concerning the role of country size in politicians' professionalization and one of the most cited causal mechanisms: the role of political parties. Politicians' professionalization is a characteristic of contemporary political systems whose consequences and justifications are highly debated in the media and academia, although the topic remains under-examined in relation to smaller polity units.

The academic literature suggests that population growth and the acquisition of new state competencies made contemporary political units seek higher-qualified, full-time, dedicated policymakers (Saafeld 1997). Therefore, we evaluated whether either country size or government is a causal factor of individual professionalization. The article analysed whether country size is a determinant for individual professionalization when government size is controlled for.

The first analysis demonstrates that the micro-state condition has a negative effect on individual professionalization. After controlling for government size, the analysis shows that country size has an effect, even if we focus just on European micro-states. Thus, the analysis shows that country size, and not government size, is a determinant for individual professionalization.

The article also assessed the underlying reason for the influence of country size on individual professionalization. We analysed the effect of country size on the political parties' role as a gatekeeper. The literature has established that political ambition, institutional structure and political parties are determinants for politicians' professionalization in large democracies. In small units, the role of political parties is still debated and under-studied (Corbett and Veenendaal 2018). Consequently, the article tested whether European micro-states maintain their role as a political gatekeeper due to the country size characteristics of their political parties. The analysis revealed that, effectively, size matters when it comes to monopolizing the political recruitment process.

Finally, we analysed whether political parties in micro-states remain a determinant of politicians' professionalization even though they have a lower capacity to monopolize public offices. The results of our models indicate that previous experience in a political party is always a positive determinant of having higher standards of individual political professionalization. These results confirm the hypothesis in the literature that political parties are a determinant of politicians' professionalization in both large-scale and small-scale democracies. Our article contributes to the micro-state literature by answering the question introduced by Corbett and Veenendaal (2018) as to how size affects parties' function with regard to the recruitment and nomination of candidates. In sum, this study helps to unravel the covariation between country size and competencies in relation to individual professionalization, and describes the causal mechanism of population size in politicians' professionalization.