What Does It Take to Get Elected in a Post-Communist Democracy?

The literature on post-communist democracies has traditionally suggested that organisational strength is considerably less important for electoral success than extensive media-based campaigns. Recent studies on party-level electoral dynamics, however, indicate that this might not be the case any longer. Building on these insights, this study goes beyond the party-level analyses of electoral success and failure by focusing on the electoral fortunes of individual candidates in a post-communist democracy. Using original data from the 2011 Estonian Candidate Survey, this article looks at the comparative impact of candidates’ campaign spending and the strength of their local party organisation, alongside other potentially relevant characteristics, on their likelihood of getting elected and vote share. The findings suggest that candidates’ electoral performance in Estonia is still first and foremost shaped by their own campaign spending. In addition, I find evidence that candidates fare better if they have prior local-level and national-level political experience, conduct more personalised campaigns, and are positioned higher up on their party’s district-level list.


Introduction
The role that money plays in politics has once again taken centre stage in public debates and become a major source of public disillusionment with politicians and politics more generally. One has to look no further than the expenses scandal in the United Kingdom or the Silvergate affair in estonia to find recent high-profile cases in advanced Western democracies as well as post-communist democracies that problematise politicians' handling of money. This is only to add to the long-standing concerns about the lobbying power of political donations, and the increasingly frequent calls to tighten up campaign finance regulations and limit how much candidates can spend on their electoral campaigns. 1 Consequently, it is important that we truly understand the complex role that money plays in contemporary politics, including its relevance in shaping electoral outcomes.
Whereas studies of electoral politics in advanced democracies have consistently shown that both campaign expenditure and party organisational strength are positively related to electoral performance, 2 the conventional understanding of post-communist politics suggests that party organisation is substantially less important for electoral success than sophisticated and expensive media campaigns. 3 This discrepancy is, however, being challenged by an emerging body of party-level literature, indicating that organisational strength might have become as important for parties' electoral success as campaign spending in post-communist democracies. 4 With the contemporary evidence on the relative role that money plays in post-communist democracies vis-à-vis party organisational factors still being sporadic, and deriving from party-level analyses, it is important to build on the existing insights by shifting the debate towards the political actors whom voters ultimately cast their votes for; that is, individual candidates. 5 This article studies the influence of candidates' individual-level campaign spending and their local party organisational strength on their electoral performance in the post-communist estonia. 6 It does so by linking these characteristics, alongside other factors that existing studies have shown to influence the success and failure of wouldbe MPs, with individual-level electoral results. I advance two core arguments. First, building on previous studies of electoral politics in advanced democracies, 7 I expect campaign spending to be positively related to candidates' vote share and likelihood of getting elected. Second, utilising insights from the general organisational theory in sociology and economics, 8 I expect those candidates to fare better who belong to parties with stronger local organisations in the district that they stand for election in. These factors should influence candidates' electoral performance even when controlling for the impact of other potentially relevant characteristics. I evaluate these arguments using an original 2011 estonian Candidate Survey, and find that the success and failure of parliamentary candidates is still best explained by how much they spend on their own electoral campaigns. The strength of their local party organisation, however, does not produce a significant individual-level effect visà-vis candidates' vote share or their likelihood of getting elected. In addition, I find that incumbent and challenger spending have similar effects on candidates' electoral performance, while it is those politicians who are incumbents, have past local-level political experience, conduct more personalised electoral campaigns, and are higher up on their party's district-level list who win more votes and are more likely to get elected.
These findings are important for two reasons. First, I demonstrate that the patterns associated with campaigning and campaign effectiveness in post-communist democracies continue to follow the traditional understanding that electoral outcomes in these countries are shaped, to a large extent, by campaign spending. Whereas an emerging body of party-level literature on the electoral success and failure of parties in post-communist democracies indicates that party-level organisational strength also matters, 9 a corresponding effect is not found on the more fundamental candidate level. The findings presented here indicate that the individual-level patterns of campaign effectiveness in post-communist democracies still do not quite mirror those associated with advanced democracies.
Second, these findings emphasise the continuing difficulties in incentivising politicians in post-communist democracies to contribute their time and effort to the development of local party organisations and, through that, closer and more extensive links with voters on the grassroots level. as money can win seats for would-be MPs, with the strength of their local party organisation seemingly unimportant, the electoral context simply does not create a need for politicians to invest in the longer-term party development. Therefore, it is unsurprising that the existing party structures remain quite centralised and, arguably, still not embedded in the underlying fabrics of the society. Whereas there were good reasons for top-down party formation in post-communist democracies, 10 far-reaching and active local party organisations are generally seen as desirable for stable and healthy democracy. 11 although the party system has become more stable over the last two decades in estonia, 12 it still appears necessary to "force" the limits on the role that money can play at elections through changing campaign funding regulations (e.g., cap campaign spending) in order to incentivise politicians to invest their time and effort in building influential local party organisations and, through that, closer and more permanent links with voters. This has not yet occurred naturally, at least when looking at the individual-level electoral dynamics.

Explaining Success at Parliamentary Elections
existing literature on elections and campaigning in advanced democracies has found several individual-level and contextual characteristics to influence electoral outcomes, with campaign spending and party organisational strength being among the more salient ones. It has consistently been shown that campaign spending is positively related to candidates' electoral performance, 13 while parties with strong local organisations tend to get more of their candidates elected. 14 Meanwhile, utilising insights from party system development in post-communist democracies, described as a top-down affair that saw no need for parties to build extensive organisations to win elections, 15 traditional understanding suggests that elections in these countries can be won with expensive campaigns and that party organisational strength has little or no electoral value. 16 There is a discrepancy in the perceived comparative relevance of campaign spending and party organisational strength in influencing electoral outcomes in post-communist and advanced democracies.
The conventional wisdom on how campaign spending and party organisational strength affect electoral outcomes in post-communist democracies is, however, being challenged. Recent studies by Tavits find no consistent evidence for the impact of campaign spending on electoral results across different post-communist democracies, 172 east european Politics and Societies and Cultures but show that parties with stronger organisations-defined as having extensive networks of branch offices, large membership, and professional staff-do consistently fare better. 17 also, Ibenskas shows that party membership organisations-measured through the number of delegates that parties are able to put forward to serve as members of electoral commissions and electoral observers-and campaign spending have roughly equal effects on the electoral persistence of political parties in Lithuania. 18 There are indications that party organisations might influence electoral outcomes in post-communist democracies more than traditionally perceived.
Whereas these studies have highlighted the need to revisit our interpretation of what shapes electoral performance in contemporary post-communist democracies, uncertainty remains about the comparative importance of campaign spending and party organisational strength in determining electoral success and failure as the emerging evidence is still rather sporadic and somewhat inconsistent. For example, the analysis by Tavits indicates that campaign spending is negatively related to parties' vote share in estonia, 19 which is highly inconsistent with our understanding of electoral politics, while Ibenskas focuses predominantly on electoral persistence rather than performance. 20 In addition, no study has yet, to my knowledge, utilised individual-level campaign spending measures alongside other potentially relevant characteristics to explain the electoral performance of individual candidates in a post-communist democracy. There is room to build on the existing studies, particularly at the time when our conventional understanding of electoral dynamics in postcommunist democracies is being challenged.

Valuing Short-Term Campaign Spending and Long-Term
Organisation Building a common feature of contemporary parliamentary elections is the growing ability of electoral campaigns to influence who gets elected. While campaigning is unlikely to influence some voters (e.g., party members), there is a widespread and growing rise in the number of late-deciders, swing voters, and in split-ticket voting. 21 With the potential of extensive campaign activities to galvanise more "last-minute" support and spending less than one's rival to have more detrimental effects on one's electoral chances, it is unsurprising that campaigning is becoming a highly sophisticated and expensive global business.
Candidates who spend more on their electoral campaign are able to print and distribute more leaflets, hire more staff to work on their campaign, pay for additional advertisement slots on TV and radio, develop a more professional-looking website, etc. These, and other campaign activities, are all potentially beneficial for raising candidates' profile, promoting their policy positions, and helping candidates to distinguish themselves from their fellow co-partisans. This latter point is particularly relevant in estonia where, given the use of open lists and large district magnitudes, candidates compete with their co-partisans as well as candidates from other parties. While no guarantee exists that candidates spend money wisely, negative campaign spending effects are unlikely given the increasingly professional nature of polling and campaigning. 22 In addition, the positive effect of any additional campaign spending is further aided by the supportive framework that is present for short-term preelection activities to influence the voting choices of a larger proportion of the electorate.
Hypothesis 1: Candidates' campaign spending is positively related to their likelihood of getting elected and vote share.
Whereas the short-term strategy of campaign spending should influence candidates' electoral fate, it is also likely shaped by the ability of their party to mobilise locallevel support for its district-level candidates at large. 23 The latter is likely influenced by the extent to which the longer-terms efforts to build strong local party organisations have succeeded. Stronger local party organisations are better at attracting and mobilising voters. as Tavits points out, parties with strong organisations tend to be more effective in reaching voters as they can have more immediate and frequent contact with more of the electorate in a more organised manner, while they are also more persuasive by appearing more competent and reliable. 24 In addition, greater local presence is likely to increase parties' awareness of local issues and improve their ability to better tailor their campaign efforts to the concerns of the district-level electorate or, at the very least, it adds credibility to the claim of being more in touch with local issues and public opinion. as such, it is reasonable to believe that parties with strong local organisations have a greater capacity to engage with and convince potential voters, and in doing so, are likely to increase the number of people who end up casting their ballot for a candidate of that party.
More specifically, a wider membership (1) increases the pool of loyal voters for all local party candidates in the district, (2) helps candidates to be more in touch with local public opinion, 25 (3) facilitates more personalised campaigning on candidates' behalf as party members are likely to take up an active role in promoting their party's candidates and be more willing to seek face-to-face contacts with the electorate through community outreach, and (4) is more likely to offer access to the full range of different minority groups in the district. at the same time, a more extensive network of branches (1) offers the structures necessary to better mobilise voters during elections, 26 (2) increases the permanent visibility of the local party in the district, and (3) increases support for party candidates by bringing the party closer to voters.

Hypothesis 2:
The strength of candidates' local party organisation is positively related to their likelihood of getting elected and vote share.

Using an Original Survey to Explain Individual-Level Electoral Success
I evaluate these arguments using an original 2011 estonian Candidate Survey. 27 as candidates in estonia, with the exception of independents, do not disclose personal campaign spending, 28 a survey of candidates offers an opportunity to collect unique information on their individual-level campaigns. Moreover, it provides information on candidates' political background that, together with their campaigning choices, can be linked to their electoral performance.
as part of data collection, all 789 candidates were approached. It was a postelection survey, carried out between May and June 2011. 29 The survey used a mixedmode design-postal and online-to minimise measurement error. 30 The final sample used in the following analysis, that is, the number of candidates for whom information on all explanatory variables was available, is 143 candidates. The sample appears reasonably representative. When using the Duncan index of dissimilarity on the distributions of two major characteristics-the district and the party list that the candidate stood for election in-within the full population of candidates and the sample used, it yields values of 0.16 and 0.20, respectively. 31 Moreover, the proportion of women among all candidates and those in the sample is very similar at 23 percent versus 26 percent, as is the candidates' mean age (47 vs. 48 years), and the proportion of successful candidates (13 percent vs. 18 percent).

Variables and Model Choice
Two parallel dependent variables are used in the study to capture the electoral performance of parliamentary candidates. 32 To start off, a simple binary measure of elected is used. all candidates who became MPs after the election were scored 1 and all candidates who did not were scored 0. In order to tease out even more variation in regard to candidates' electoral performance, a second dependent variable-vote share-is also used. It is measured as the percentage of district-level votes received by the candidate, ranging from 0 ("no votes") to 100 ("all votes"). 33 The inclusion of the latter is particularly important given the use of open lists and large district magnitudes in estonia, which can lead to considerable variations in the vote shares of both successful and unsuccessful candidates. 34 The first main explanatory variable in the analysis is candidate's campaign spending. 35 It is an individual-level measure, operationalised by dividing a candidate's selfreported campaign expenditure on his or her electoral campaign 36 by the mean campaign spending of all candidates in the same district, and then taking a natural logarithm of the obtained measure. 37 Two aspects should be noted about this operationalisation. First, a natural logarithm is used to prevent outliers from distorting the analysis and to capture the marginally diminishing returns produced by increases in campaign expenditure. 38 Second, a relative measure (i.e., how much a candidate spent relative to the mean campaign spending of his or her district-level competitors) is preferred to the absolute measure (i.e., how much a candidate spent) to address the endogenous nature of campaign spending. It is widely acknowledged that candidates' spending decisions are influenced by their expectations about votes. 39 although the problem of endogeneity is weaker in estonia, 40 the use of a relative measure will allow accounting for the context where the (mis-)fortune of one affects the (mis-)fortune of another. If spending can actually help candidates obtain more votes, then a candidate should outspend his or her direct rivals, with the extent to which he or she gains more votes and increases his or her likelihood of getting elected being influenced by the extent to which he or she outspends those rivals. The relative measure accounts for district-level dynamics and mitigates the endogenous nature of campaign spending. 41 In line with the theoretical approach, the second main explanatory variable in the analysis is the organisational strength of candidate's local party. Two separate indicators are used in parallel to measure the concept. 42 Organisational strength: members describes the number of party members in the district that the candidate stood for election in as the percentage of total district electorate, divided by the number of party candidates in the district. Since estonia has a rather high level of political party membership among post-communist countries, 43 an alternative indicator is also used to provide an additional robustness check for the effect that local party organisational strength has on candidates' electoral performance and increase the generalisability of the findings. Organisational strength: branches describes the number of municipallevel party branches in the district that the candidate stood for election in, divided by the total number of municipalities in the district and by the number of party candidates in the district. This operationalisation mirrors closely that of Tavits in her seminal study on electoral politics in post-communist europe, departing only by adding the "per candidate" element (i.e., dividing district-level party organisational strength by the number of party candidates in the district). 44 This is preferred given the individual-level nature of the study. It is reasonable to expect that local parties with equal organisational strength are more beneficial for their candidates when this strength is shared by fewer candidates, that is, the organisational strength does not get "diluted" as much as it would if more candidates could draw from it. 45 To control for rivalling explanations, five additional variables are introduced: three relate to candidates' political capital and two affect their electoral capital. Starting with the former, candidates' electoral performance is likely linked to how competent and in touch with local issues they are able to present themselves. Candidates who are incumbents are likely to fare better than challengers as they can claim to have a proven track record of working as an MP and tend to be more well known. 46 Incumbency is operationalised as a dichotomous variable, with incumbents coded 1 and challengers 0. at the same time, candidates who are local-level representatives are likely to fare better than those without such experience as they are able to claim greater knowledge of local issues and tend to be better known to the 176 east european Politics and Societies and Cultures district-level electorate. Candidates are, therefore, differentiated between those who have never been members of a local-level legislature, 0; those who have been members in the past, 1; and those who are members at the time of the election, 2, as part of the local political experience variable. Finally, candidates who are active members of their local party might benefit from being more involved in the local political scene. Local party membership ranges from 0 if a candidate has never been an active member of his or her local party organisation to 3 if he or she is an active local party member and officeholder. 47 Moving on to the two variables that relate to electoral capital, a variable called campaign aim is included to capture the substantive content of candidates' campaigns. Describing the self-perceived objective of a candidate's campaign, it ranges from 0, "to attract as much attention as possible for my party," to 10, "to attract as much attention as possible for myself." 48 given that estonia uses an open list system, it is reasonable to expect a positive relationship between campaign aim and electoral performance. Finally, candidates' placement on the district-level party lists is controlled for. although voters are required to vote for a specific candidate, it is likely that not all voters are informed enough to differentiate between all the candidates. It is reasonable to expect that some voters use list placement as a cue for a candidate's standing among the party's district-level candidates. as such, being higher up on a district-level party list is likely to improve one's electoral performance independent of other factors. List position is coded as a candidate's placement on his or her district-level party list, with list leaders coded as 1 and other candidates 2, 3, etc. based on their list placement. 49 Utilising the variables highlighted above, two sets of models are run to explain the electoral performance of parliamentary candidates. The variation in candidates' likelihood of getting elected is assessed via logit models with standard errors clustered by districts, and the variation in candidates' vote share is assessed via ordinary least squares models with standard errors clustered by districts. 50

Which Candidates Performed Better?
To determine how these characteristics shape the success and failure of parliamentary candidates, I begin by comparing the actual electoral performance of different candidates. Table 1 groups the candidates by shared characteristics, presenting how many of them got elected and their median district-level vote share. 51 It suggests preliminary support for hypothesis 1. Candidates who spend more do in fact fare better, with the percentage of candidates who got elected rising from 4.1 percent to 61.5 percent when comparing candidates who spent less than the district mean on their campaign to those who spent more than twice the district mean. The corresponding rise in these candidates' average vote share is from 0.4 to 2.5 percent. at the same time, no trend appears present when comparing candidates whose local party organisation is strong vs. weak. This initial evidence does not seem to support hypothesis 2. However, incumbents do perform better than challengers (72.7 percent vs. 13.6 percent got elected; 2.9 percent vs. 0.5 percent average vote share), and those who lead their party's district-level list have an advantage over those with a low list placement (42.9 percent vs. 5.6 percent got Finally, this initial evidence suggests that no significant difference exists between incumbent and challenger spending effects, and that electoral performance is not influenced by local party membership. although the initial analysis does not support hypothesis 2 when looking at all candidates, local party organisational strength might still be useful for some of them. Similarly, it is important to see if the positive effect of campaign spending (hypothesis 1) is consistent across different subsets of candidates. Table 2 divides candidates into various subsets by combinations of explanatory characteristics, presenting how many candidates got elected and their median district-level vote share. Note first that campaign spending has a positive effect on electoral performance when looking at all combinations of explanatory characteristics. at the same time, no subset of candidates appears to significantly benefit from strong local party organisations, offering further support to the idea that campaign spending (and not local party organisational strength) drives the patterns associated with candidates' electoral performance.

Explaining Electoral Performance
as already seen in Tables 1 and 2, candidates' electoral fortunes vary considerably with regards to their characteristics. Focus is now turned to going beyond the descriptive statistics. Table 3 presents the multivariate models that explain variation in candidates' likelihood of getting elected and their vote share. The findings presented in Table 3 confirm hypothesis 1 (campaign spending hypothesis), but not hypothesis 2 (organisational strength hypothesis). With regards to the former, positive and statistically significant coefficients of 1.73/1.93 and 0.40/0.39 show that candidates who spend more on their campaigns are indeed more likely to get elected and receive larger proportions of district-level votes. at the same time, no consistent evidence is found that candidates fare better when their local party organisation is stronger, regardless of whether looking at membership levels or the density of municipal-level branches. 52 These findings lend further support to describing the individual-level electoral patterns in the post-communist estonia as rather fluid and shaped less by the more formal organisational structures than those associated with advanced democracies.
Four of the control variables also have significant effects in the expected direction, increasing the confidence in the findings overall. Two relate to candidates' political capital (incumbency and local political experience) and two affect their electoral capital (campaign aim and list position). Candidates are likely to come across as more competent and be better-known if they are representatives in the national legislature, while members of the local-level representative body can additionally claim to be more in touch with local issues. as a result, it is unsurprising that incumbents perform better than challengers and those with local-level political  Note: Values within parentheses are standard errors; ***p < 0.01, **p < 0.05, *p < 0.1. experience do better than those without such experience (shown by coefficients of 2.76/3.48 and 0.87/0.90 for the former; 1.09/1.14 for the latter 53 ). In addition, candidates with more personalised campaigns fare better (shown by the positive coefficients of 0.47/0.50 and 0.09/0.09), which is unsurprising given that voters need to cast their ballot for an individual candidate. Similarly, candidates who are higher up on their party's district-level list perform better (shown by the negative coefficients of -0.35/-0.40 and -0.13/-0.14) as they benefit from voters who use list placement as a cue for differentiating between their preferred party's candidates. at the same time, local party membership does not appear to affect electoral performance, and no difference is found in the effects of incumbent and challenger spending.
To illustrate the effect sizes of the independent variables, Table 4 presents predicted values for candidates' likelihood of getting elected and vote share. For each effect, a particular characteristic is allowed to vary while others are held constant at their mean.
Note first that the effect size associated with campaign spending stands out when comparing the impact that minimum-to-maximum shifts in the explanatory characteristics have on candidates' predicted electoral performance. as Table 4 demonstrates, the probability of getting elected increases by 60 percent when comparing candidates who spend nothing on their campaigns to those who spend almost eight times the district-level mean, while the corresponding increase in the predicted vote share is 2.87 percent. These are the largest differences in candidates' electoral performance that are brought about as a result of changes in the explanatory characteristics. In addition, three control variables, one affecting candidates' political capital and two relating to their electoral capital, stand out. Regarding the former, incumbents have a 22 percent higher predicted likelihood of getting elected than challengers (37 percent vs. 15 percent) and are expected to receive 0.87 percent more of the district-level vote (2.06 percent vs. 1.19 percent). Regarding the latter, list leaders are 27 percent more likely to get elected than candidates with the lowest list placement (28 percent vs. 1 percent) and are expected to win 2.01 percent more of the district-level vote (1.87 percent vs. -0.14 percent). The respective advantages of conducting a candidate-centred as opposed to a party-centred campaign are 28 percent (35 percent vs. 7 percent) and 0.91 percent (1.77 percent vs. 0.86 percent). These findings lend further support to the understanding that individual-level electoral performance in post-communist estonia is driven by how much candidates spend on their own campaigns; expensive campaigns can indeed win elections for individual candidates. While some characteristics that relate to political and electoral capital are also relevant, their impact on the success or failure of individual candidates remains of secondary scope.

Conclusions
Whereas the conventional understanding of electoral patterns in post-communist democracies suggests that elections can be won by expensive media campaigns and the strength of party organisation is of little relevance in these countries, 54 there is a growing body of literature to indicate that, at least on party level, both factors are now substantially contributing to parties' success and failure. 55 However, with the latter insight deriving from party-level analyses, it is also important to assess the comparative role of campaign spending and party organisational strength in shaping the electoral fortunes of actors whom voters ultimately cast their votes for, that is, individual candidates.
Building on previous studies of individual-level campaign effects in advanced democracies and the more general organisational theory, I argue that short-term strategies (i.e., increased campaign spending) and long-term strategies (i.e., development of stronger local party organisations) should both contribute to parliamentary candidates' electoral success. Whereas candidates who spend more on their own campaigns have increased ability to raise their profile and "sell themselves," stronger local party organisations have greater capacity to raise support for all of its candidates. I test for these effects using original 2011 estonian Candidate Survey data, but find support for the former only. On individual level, electoral performance in post-communist estonia is still driven first and foremost by candidates' own campaign spending. Contrary to the theoretical expectation, I find no consistent evidence that candidates benefit from strong local party organisations. Instead, it is candidates' greater political capital (i.e., incumbency and experience in local-level legislature) and electoral capital (i.e., personalised campaign strategy and being high up on the district-level party list) that have impacts of secondary nature on their electoral performance. My findings contribute to our understanding of electoral dynamics in post-communist democracies in several ways. First, I show that individual-level electoral performance is still first and foremost influenced by how much candidates spend on their own campaigns. While some longer-term strategies that involve building up more permanent structures that could be called upon to support one's campaign do have positive effects, these relate to candidates' political capital, as opposed to the organisational strength of their local party, and bring about considerably weaker electoral benefits. From the perspective of candidates, it is still possible to approach elections as short-term processes and opt for the "smash-and-grab" strategy. as expensive campaigns, particularly those that are candidate-centred and promote candidates high up on the district-level party list, remain sufficient to get elected, the individual-level patterns of campaign effectiveness in post-communist estonia still do not quite mirror those associated with advanced democracies.
The dominance of short-term factors in shaping the success and failure of candidates standing for election also has implications for the development of more active and far-reaching local party organisations. as money can win seats for would-be MPs, and the support of their local party organisation is seemingly unimportant, the current context does not create the need for politicians to invest time and effort in party development. Strong local party organisations are, however, widely seen as desirable for stable and healthy democracy. 56 Particularly in the post-communist democracies, where party formation was a top-down affair, 57 developing local party organisations that are more prominent features of the societal fabric allows moving away from the centralised and elite-driven party democracy to a more participatory and grassroots democracy. This would likely contribute to the stability of the party system and promote political participation, but it should also enhance policy responsiveness and effectiveness through better awareness of public opinion and local context. as it stands, however, the patterns related to individual-level electoral performance offer little in terms of incentivising politicians to contribute their time and effort for such a shift.
So what can be done about it? In the context of electoral politics, the findings presented here suggest that it would be necessary to "force" limits on campaign spending through changing campaign finance regulations (e.g., cap campaign spending) in order to reduce the role that money can play in politics via its dominating impact on individual-level electoral outcomes. By limiting the extent to which candidates can rely on the short-term "smash-and-grab" strategy to get elected, they will have more incentives to think long term and develop a greater structural capacity to connect to the electorate. Strong local party organisations would be able to do exactly that by offering a closer connection with the electorate on the grassroots level through a more permanent and visible presence in the district.
There are of course many other aspects that can influence candidates' electoral performance: individual-level characteristics such their campaign spending strategy, and contextual factors like electoral rules. This study represents a first-cut empirical effort to assess the comparative relevance of individual-level campaign spending and local party organisational strength in shaping the success and failure of parliamentary candidates in a post-communist democracy. Therefore, it adds useful depth to our comparative understanding of contemporary electoral processes in advanced and post-communist democracies. at the same time, it leaves room for (and highlights the need for) expanding this research agenda. Future research should consider how different campaign spending strategies affect electoral performance, and apply a similar research design to other, and ideally multiple, post-communist democracies.