Votes,money can buy. The conditional effect of EU Structural Funds on government MPs’ electoral performance

Abstract

Members of Parliament are often thought to attract funding into their constituencies to increase their local support, and eventually, secure re-election. Nevertheless, there is only limited research focusing on the question whether or not government funding indeed affects the MPs’ electoral performances. I focus on pork barrel politics in a single country, namely Hungary, and use town-level European Union Structural Funds data between 2007 and 2010 to explain the electoral performance of single-member district MPs at the 2010 general elections. I find that increasing funding improves government MPs’ electoral performances and that the size of the effect is conditional upon the mayors’ party affiliations. In towns with government mayors, government MPs perform significantly better than in towns with opposition mayors. The results have consequences in the fields of the distribution of Structural Funds, the electoral connection between legislators and voters and the European Union’s contribution to regime legitimation.

Keywords

Elections Hungary legislators pork barrel Structural Funds

Introduction

The idea of ‘pork barrel politics’ refers to the distribution of government funds to locally organised projects. It is commonly applied by incumbent legislators to channel money into their constituencies with significant implications for the national budget. Thus, eventually, pork barrel politics affect both citizens living in the district the funds are allocated to, and those from other parts of the country, whose towns, as a consequence, are not subject to government funding. Additionally, pork barrel is considered a form of (allocation) responsiveness and thus is an important building block of the electoral connection.

The source of pork barrel spending varies across and within countries. Whilst government spending constitutes a considerable share of the funding, governments increasingly capitalise external sources such as European Union (EU) funds to either support districts where the government party has fared well in elections or to increase their vote share in marginal constituencies. In countries where the distribution of EU Structural Funds (SF) is organised by a central government agency, it is easier for governments to frame funding as pork barrel and enjoy its benefits during election times. Hence, in countries where pork barrel politics increase the government vote, the EU regional policy may have substantial effects on national politics and plays an important role in regime legitimisation.

Members of Parliament (MPs) are thought to attract funding into their constituencies to increase their local support, and eventually, secure re-election (Cain et al., 1987; Mayhew, 1974). Nevertheless, there is only limited research focusing on the question whether or not government funding, and particularly EU SF, indeed affect government MPs’ electoral performances. Although government MPs do not formally participate in funding allocations, they are interested in advertising winning projects as their own success. I demonstrate that the electoral support of government MPs is larger in towns with larger funding. There is, however, more to the story. Besides the MP, the mayor is another key actor in town-level politics, one who alters voter perceptions of who ‘brought home the bacon’, and thus whom voters reward at elections. I theorise that because voters are better able to credit the funding to parties instead of individual politicians, in towns with mayors affiliated with government parties (henceforth ‘government mayors’), the electoral reward of pork barrel politics for government MPs is larger than in towns with mayors affiliated with opposition parties (henceforth ‘opposition mayors’).

The article focuses on how the distribution of EU SF affected town-level election results in Hungary in 2010. The Hungarian case seems suitable for an investigation of this issue for several reasons. First, the single-member district (SMD) tier of the mixed-member electoral system ensures a one-in-one connection between the MP and the constituency, minimising the problem of ‘shirking’. To get a more nuanced picture, the article explains town-level electoral performance. As towns are nested within electoral districts, just as in the case of SMDs, the connection between the legislator and the town remains immediate. Second, there was a steep decline in government popularity before the 2010 elections. An anti-government political climate makes it very unlikely that government spending can increase government vote share. Thus, if we find evidence in support of a positive effect of funding on government MPs’ electoral performance at this particular election, it is very likely that we will find such effects under more balanced power relations. Therefore, Hungary at the 2010 general elections constitutes a least likely case (Gerring, 2006).

The article concludes that on the one hand, larger funding improves government MPs’ electoral performance, whilst decreasing the vote share of opposition MPs, and on the other hand, the effect size is conditional upon the mayors’ party affiliations. Government MPs perform significantly better in towns with government mayors as opposed to towns with opposition leadership.

Targeted spending and EU SFs

Targeted spending in the district is mostly viewed from a strategic perspective, in which an exchange between legislators (or parties) and constituents takes place. Parties and MPs procure local funding in the hopes that they are rewarded in the form of votes. As electoral strategies in general, ‘bringing home the bacon’ is also constrained by institutions, especially electoral rules (Denemark, 2000). Several scholars argue that because of a clear linkage of accountability, the incentive for pork barrelling is highest in SMDs as opposed to multi-member constituencies (Lancaster, 1986; Lancaster and Patterson, 1990; Stratmann and Baur, 2002), but open-list proportional representation is also shown to invite such strategies (Ames, 1995; Golden and Picci, 2008).

The literature on targeted spending as a strategy is vast. Many have shown that – either formally or informally – incumbents influence the distribution of funding (such as regional development grants, infrastructure investments, temporary grants or discretionally programmes) all over the world to improve re-election chances (Calvo and Murillo, 2004; Case, 2001; Castells and Solé-Ollé, 2005; Dahlberg and Johansson, 2002; John and Ward, 2001; Kneebone and McKenzie, 2001; Milligan and Smart, 2005). Governments, however, differ in their re-election strategies: Some support their core voter base (Costa-I-Font et al., 2003; Cox and McCubbins, 1986), whilst others try to win the swing vote (Bickers and Stein, 1996; Stein and Bickers, 1994).

However, is targeted spending an effective vote-gathering strategy? Do government MPs whose districts benefit from the funds perform better at the next elections? Although a majority of authors assume that legislators target the districts to increase their electoral support, surprisingly, the literature actually testing this claim is scarce and presents mixed results. On the one hand, in the case of the United States (US), Feldman and Jondrow (1984), for example, find no evidence of a relationship between government spending and the incumbents’ election results in the 1970s, whilst Stein and Bickers (1994) demonstrate a positive relationship only in the case of a politically attentive electorate in the 1980s. On the other hand, Alvarez and Saving (1997) and Levitt and Snyder (1997) present compelling evidence for a positive effect of funding on election results. Outside the US, Leigh (2008) demonstrates that targeted funding increased government performance at the 2004 Australian elections.

Whilst the literature overwhelmingly discusses the American case, European countries offer plenty of opportunities to observe pork barrel politics. Most importantly, governments of EU member states may not only decide on government spending but can support the constituencies by allocating supranational fiscal transfers, such as SF. SF, along with the Cohesion Fund, are a financial tool that is implemented to reduce regional disparities by balancing income and wealth inequalities across the EU. During the 2007–2013 programming period, the SF consisted of the European Regional Development Fund (ERDF) and the European Social Fund (ESF). SF are allocated to member states on a seven-year basis and are subject to application from a wide range of actors such as small and medium-sized enterprises, non-profit organisations, cultural organisations, churches, local and regional governments and state departments. Each country sets up its own infrastructure to distribute funding; but in each country, a coordinating authority organises and monitors the operation of implementing agencies. Although it could not be further removed from its original goals, in countries with a centralised agency to distribute the funds, governments may use SF to reward government-friendly districts (Medve-Bálint, 2017).

Several scholars have shown that the political leaning of the sub-national regions plays a significant role in the distribution of SF. Bouvet and Dall’erba (2010) and Kemmerling and Bodenstein (2006) argue that the political situation within a country influences funding allocation. Dellmuth and Stoffel (2012) find evidence in Germany that although the distribution of the SF is in accordance with their goal, electoral politics still weighs in. Bloom and Petrova (2013) show that in Bulgaria and Latvia, SF went to towns with a government-friendly electorate. Medve-Bálint (2015, 2017) demonstrates that in the case of Hungary and Poland, the regions’ political loyalty towards the central government plays a key role in funding allocation. Similarly, analysing the Hungarian data, Kálmán (2011) and Muraközy and Telegdy (2016) conclude that politically aligned municipalities receive a larger grant value than towns with opposition mayors.¹ These results support the claim that SF are parts of the governments’ vote-gathering strategies.

The literature is limited on the effects of intergovernmental fiscal transfers – such as SF – on electoral support. Veiga and Veiga (2013) find that grants directed to the municipalities pay off in votes in Portugal, whilst Muraközy and Telegdy (2016) show that EU funds influence close races at the Hungarian local elections. Even citizens with relatively low levels of political knowledge show increasing support for European integration as funding increases. With this in mind, it is expected that SF allocation influences voter choices at the national elections as well.

There is evidence in the comparative economic voting literature that voters hold governments responsible for economic performance and, therefore, governments benefit from increasing welfare at the elections (Lewis-Beck and Stegmaier, 2000, 2007; Nadeau and Lewis-Beck, 2001). Similarly, I argue that as government parties and MPs have greater influence over the distribution of SF, voters may attribute a town’s success in attracting the funds to the government party rather than to the opposition.

H1: The positive effect of the SF for legislators’ electoral performance is restricted to government MPs.

It has been shown that despite being heavily monitored, in East-Central Europe, EU funds increase institutionalised grand corruption and government favouritism that often reaches high-level politicians (Fazekas et al., 2013, 2014). Thus, it seems that politicians indeed weigh in at the distribution of EU funds. However, because of the informal character of their influence, it is difficult to take this aspect into account in a comprehensive analysis. But even if a government MP has no formal power in influencing the distribution of funds, she should benefit from them regardless. Legislators often frame successful projects as their own achievements by participating at opening ceremonies of projects funded from SF or simply by reporting on them in their personal newsletters or on their websites. Thus, government MPs may increase their vote share not only because they successfully lobbied for their constituencies, but because of the spillover effects of their parties’ success in directing funds to the district. Therefore, the first hypothesis should hold even if (a) the government MP was not the one who secured the funding, but other relevant government party actors and if (b) voters are unable to identify if it was the MP or another relevant government party actor who actively participated in the distribution of funds.

Investigating the effect of government spending on legislators’ town-level electoral performance, one has to take into account the role of the mayor in town-level politics. A majority of mayors in Europe are directly elected (Magnier, 2006), which gives them strong legitimacy (Heinelt and Hlepas, 2006) and ‘a capacity to foster cooperation and to mobilise support’ (Borraz and John, 2004: 115). Whilst they sometimes lack the resources that would enable them to carry out their tasks effectively, citizens have still high expectations of them. Although ‘[m]uch depends on the leaders’ ability to […] have direct relationship to the citizens’ (Borraz and John, 2004: 116), mayors are increasingly important figures in local politics. In Germany, for example, local councils generally avoid excluding the mayor’s party from the coalition majority (Debus and Gross, 2016). In the 1970s, Tarrow (1977) finds that mayors in France and Italy used their positions to invite government resources. There is further evidence that mayors, particularly in Southern Europe, frequently engage in pork barrel politics to mobilise support and strengthen their legitimacy (Borraz and John, 2004). Furthermore, the findings of Nicolescu and Gorcea (2005) suggest that in Poland, Estonia and Bulgaria, local political actors have a propensity towards maintaining a close relationship with the citizens.

To sum up, mayors are particularly visible actors on the local level who also have great potential in attracting financial support. Additionally, they are also very likely to benefit from such strategies at the next local elections. Luo et al. (2010) show that public goods investment increases the likelihood of village leaders to be re-elected in China. For Central and Eastern Europe, Muraközy and Telegdy (2016) find evidence not only that municipalities aligning with the national government receive more EU funds in Hungary, but also that larger funds result in a higher probability of re-election for incumbent mayors. Banaszewska and Bischoff (2017) show that the relationship between EU funds and the incumbent mayors’ vote shares is prevalent, but conditional upon certain town-level characteristics, such as the proportion of an EU-friendly and well-educated population.

The question arises whether voters are able to decide which politician (or which party) to reward if both MPs and mayors (or their respective parties) have the potential to compile funding. One may distinguish between two fundamentally different scenarios in this respect. First, when both the mayor and the MP are from a government party, voting for the government MP and thus rewarding the wrong political bloc has only a small chance. Second, in the case of cohabitation (i.e. when the MP is from a government party and the mayor is not), voters may find it difficult to decide which political bloc to reward for the funding. Does funding automatically increase the electoral chances of the government MP due to her party’s better access to funding allocation, or can the opposition mayor divert votes away from the government incumbent by capturing the limelight?

In such a case, assuming that voters lack the necessary information on who ‘brought home the bacon’, voting for the government MP instead of her opposition challenger holds the risk of rewarding the wrong political side. Under such unclear political responsibility, the electoral advantage of the government may not increase with government funds at such rate as it would in a case where the mechanisms of accountability are clear and simple (Anderson, 2000; Powell and Whitten, 1993; Whitten and Palmer, 1999). Thus, the government vote is expected to increase at a smaller rate than in a town with a government mayor.

H2: The positive effect of SF for a government MPs’ electoral performance is larger in a town with a government mayor than in a town with an opposition mayor.

The case of Hungary

To test the above hypotheses, I use Hungary as a case for three reasons. First, as the EU’s less developed regions receive a majority of the support, Central and Eastern European countries are amongst the top beneficiaries. Amounting to about 3% of the gross domestic product (GDP) in the 2007–2013 programming period, 5400 billion Hungarian Forint (HUF) (approximately 16.8 billion EUR) were allocated to Hungary (European Commission, 2009). The role of EU funds in Hungary’s economy is best illustrated with GDP growth data: Between 2006 and 2015, GDP growth would have amounted to 1.8% without the funds as opposed to the 4.8% of actual growth (with the EU funds), says a study by KPMG and GKI (Keszthelyi, 2017; KPMG, 2017). Furthermore, SF amounted to about 8% of the total national expenditure between 2007 and 2013 (KSH, n.d.). Second, during the period under investigation (2006–2010), SF were domestically distributed by a central government agency (National Development Agency); thus, funding allocation was under exclusive central government control. Therefore, the role of the government parties is clear in distributing funds. Third, as shown by Dellmuth et al. (2017), the comparative incentive offered SMDs over proportional representation also exists in the case of SF distributions. Hence, Hungary’s mixed-member electoral system, with its prominent (45.6% of parliamentary seats) SMD tier, is an ideal setting for testing the hypotheses of this article.

By 2002, Hungarian politics were dominated by the competition of two parties. In 2006, the conservative Fidesz (Fidesz–Hungarian Civic Alliance) and the socialist MSZP (Hungarian Socialist Party) won 59.8% of the seats in Parliament. The distribution of SMD mandates across parties was quite balanced: Fidesz brought in 32.4% of the seats, whilst MSZP gained 57.9%. Together with SZDSZ (the Alliance of Free Democrats), MSZP formed a government with a 54.4% majority. The popularity of the Socialists was then set to a downslide in September 2006 and continued until the 2010 elections. MSZP’s heavy defeat and Fidesz’s win in 2010 seemed extremely predictable. Eventually, MSZP received 15.3% of the seats (winning in only two SMDs), and Fidesz–KDNP won a majority that post-transition Hungary has not seen before (68%). The coalition parties together filled 173 SMD seats out of 176.

The context of the 2010 elections provides us, according to Gerring (2006), with a least likely case. Former MSZP voters reluctant to support the Socialists were quite unlikely to be persuaded to stay loyal with increasing targeted spending in their towns. Thus, if the analysis can confirm the effect of spending on government MPs’ electoral performance, it should make an even larger difference under more balanced power relations, and when former voters of the government are not alienated from their previous choice.

With regards to local politics, Hungary had local elections in the fall of 2006, when – as an early sign of the Socialist downfall – only 4.5% of the towns elected government mayors. 7.4% of the towns with government MPs elected government mayors, whilst this percentage was 2.5 in the case of towns with opposition MPs.²

As discussed above, SF are distributed by a central government agency. MPs and mayors do not have a formal say in the allocation of funds; however, they are able to attract funding to their districts and towns through informal networks, which are mostly based on their party affiliations, and their positions within their parties.³ SF typically fund small and medium-sized enterprises with the promise of launching innovative projects, and research and development projects, as well as projects that create jobs. Civil organisations, schools and local governments are also among the main beneficiaries of SF, with a special focus on local infrastructure development projects. To a lesser extent, under certain operative programmes, SF are allocated to ministries as well as big enterprises to fund investment projects (Medve-Bálint, 2018).

Four studies in particular look at the connection between the political leaning of a town and the distribution of SF in Hungary. Kálmán (2011) and Medve-Bálint (2015, 2017) show that politics play a crucial role in the allocation of SF: Towns with government-friendly leadership are more likely to benefit from SF than towns leaning toward the opposition. Muraközy and Telegdy (2016) demonstrate that the government’s strategy pays off at local elections: More money for the town results in more votes for the mayor. The current study follows in their footsteps, but instead of explaining the results of local elections, it focusses on national election outcomes and tests how the local and national levels interact with the amount of SF in determining electoral results.

Data and variables

To test the hypotheses of the study, I use town-level SF and electoral data complemented with publicly available information on SMD MPs. As SF and electoral information are observed on the town level, the same MP appears in the dataset multiple times depending on how many towns the given electoral district consists of. The number of towns in an SMD ranges from 1 to 102 (mean = 18.1, SD = 20.7). Thirteen cities⁴ including multiple constituencies, 37 towns with no incumbent MP at the time of the 2010 election, towns with retiring MPs (562), and those with MPs elected at by-elections sometime between 2006 and 2010 (77) are excluded from the dataset. This leaves us with about 2400 towns in the sample, which covers 76% of the Hungarian municipalities.

Dependent variable

The dependent variable of the analysis is the SMD MP’s first-round vote share in the town at the 2010 election relative to the MP’s constituency-average vote share

D V_{ij} = \frac{Votes %_{ij}}{\frac{\sum_{i = 1}^{k} Votes %_{ij}}{# Town s_{j}}}

where i stands for towns within the district and j denotes SMDs.

This approach has two advantages over alternative methods. First, this way, we can test if larger funds result in larger support for the government MP whilst factoring out variables influencing the SMD-level support of the MP (i.e. district-level characteristics and most of the MP-related features). This arguably decreases omitted variable bias. If the value of the dependent variable is greater than 1, the MP performs better in the town compared with the district average, whilst values smaller than 1 suggest poorer performance. Hence the variable does not give an absolute measure of electoral performance, but the deviation from the SMD-level support. The models, therefore, show the magnitude with which the funding diverts the MP’s town-level vote share away from her overall popularity in the district. Second, the dependent variable is normally distributed, whereas the distribution of raw town vote share is bimodal, inviting modelling strategies more difficult to interpret (for the distribution of the variables, see the Online appendix). The results presented in the article are robust (see the Online appendix) to alternative measures of the dependent variable, such as the raw town-level vote share and the incumbent’s vote share relative to her SMD-level vote share.

Independent variables

The study has three key independent variables: (a) the per capita grant value of the SF allocated to the town, (b) a dummy variable indicating whether the MP is a representative of a government party and (c) another dummy variable indicating whether the mayor of the town officially supports a government party.

Starting with the value of SF, all EU countries are obliged to release the list of beneficiaries and the amount of support they receive from the SF. Given the information on project locations, the list of beneficiaries issued by the Prime Minister’s Office is aggregated to the town level. I use the natural logarithm of the total awarded value of SF per capita (million HUF) allocated to Hungarian towns between 2007 and 2010. The SF data includes both the ERDF and ESF. The second and third independent variables are dummies and measure the government affiliation of the MP and the mayor of the town. After the 2006 election (i.e. the start of the period under investigation), MSZP and SZDSZ formed a government coalition. In April 2008, SZDSZ opted out of the coalition but continued to support the minority government of MSZP from opposition. Thus, I considered both MSZP and SZDSZ as government parties. However, following its crisis before the 2010 elections, SZDSZ did not nominate SMD candidates in 2010, and therefore, all government MPs in the sample are from the Socialist Party. This circumstance is noteworthy, because as the literature on economic voting argues, government parties do not benefit from the voters’ evaluation of the economy the same way: The party of the prime minister is more likely to be rewarded than its coalition partner(s). Given that in the 2010 elections, only the prime minister’s party nominated candidates in SMDs, the coalition setup of the government is not relevant from the viewpoint of this study. Roughly 42% of the towns are represented by government MPs. The same logic is used to gauge the effect of the mayors’ party affiliation. Here, 86.3% of the mayors were not supported by any of the parliamentary parties. About 6% was officially connected to government parties (here, SZDSZ is included because of its mayors), whilst 7.5% enjoyed the support of the opposition.

Control variables

To obtain the net effect of the independent variables, several controls are included. The lagged dependent variable measures the MP’s vote share in the town relative to her average vote share in the constituency at the previous election (2006). There is considerable concern about the inclusion of the lagged dependent variable, because it has been shown that it drives the model to underestimate the effects of other independent variables (Achen, 2001). Keeping this in mind, I control for the lagged variable for an important reason. Lagged dependent variables are often used to absorb the effects of unmeasured variables, thus decreasing omitted variable bias (see, for instance, Burkhart and Lewis-Beck, 1994). Since a difference-in-difference design is not applicable because of data availability, including the lagged DV appears the best available option to minimise bias.

Furthermore, I control for the 2006 electoral margin in the district to measure the expected fierceness of the competition. I include a variable that takes 1 if the MP is also the mayor of the town and 0 if not. The MP’s vote share in the town at the previous election (2006) is included to sort out the effects of potential government strongholds.⁵ Town-level variables control for the legal status of the town (0 = town, 1 = village), total tax revenue in the year prior to the election (2009, logged), total municipal spending in 2009 (logged) and the share of the registered unemployed in the population of working age in 2009. Finally, to capture the economic development of the region, the county (megye)-level GDP is taken into account relative to the country-average. I do not control for MP-level socio-demographics (such as age, gender and tenure) because there is no reason to suspect that they influence the town vote share relative to the SMD-average town vote share. However, taking into account the possibility that results are drawn by a few talented individuals, I also ran the models with MP fixed effects (see the Online appendix).

Analysis

Ordinary least squares (OLS) regressions test the effect of the per capita amount of SF allocated into the town on the electoral performance of the incumbent MP.⁶ Standard errors are robust and are clustered by SMD. Models have been checked for multicollinearity and outliers. Outliers do not appear influential, as removing them from the observations does not change results. The residual term does not correlate either with any of the independent variables or their linear combination, limiting concern over omitted variable bias.

The results of Model 1 in Table 1 align with the first hypothesis, namely that only government MPs benefit from targeted spending. Figure 1 helps interpret the interaction between the per capita value of SF and the MP’s party affiliation. The dashed line shows how the town-level electoral performance of the government MP increases with government spending. The negative slope of the solid line indicates that larger government spending results in a weakening opposition.

Figure 1.

The effect of SF per capita on town-level electoral performance.

Table 1.

OLS models explaining town-level electoral performance relative to SMD-average town vote share.

	Model 1	Model 2
	Coef. (SE)	Coef. (SE)
Main effects
Structural Funds (logged)	–0.001 (0.001)*	–0.001 (0.000)
Government MP	–0.059 (0.013)***	–0.050 (0.013)***
Government mayor	0.142 (0.028)***	0.158 (0.049)***
Interactions
Structural Funds (logged) * Government MP	0.006 (0.002)***	0.004 (0.001)**
Structural Funds (logged) * Government mayor		–0.009 (0.004)***
Government MP * Government mayor		–0.339 (0.152)**
Structural Funds (logged) * Government MP * Government mayor		0.044 (0.013)***
Control variables
Lagged dependent variable	0.788 (0.084)***	0.782 (0.085)***
Electoral margin	0.084 (0.066)	0.092 (0.064)
MP is a mayor	–0.009 (0.004)*	–0.011 (0.004)**
Vote share in 2006	–0.209 (0.096)**	–0.207 (0.097)**
Village	–0.043 (0.017)**	–0.034 (0.017)*
Total revenue in 2009 (logged)	0.008 (0.005)*	0.007 (0.004)*
Total spending in 2009 (logged)	–0.018 (0.007)***	–0.017 (0.007)**
Unemployment in 2009	–0.000 (0.101)	–0.008 (0.104)
Relative GDP in the region	0.000 (0.000)	0.000 (0.000)
Constant	0.520 (0.073)***	0.517 (0.073)***
N	2418	2418
F	32.51***	30.04***
R²	0.316	0.324

OLS: ordinary least squares; SMD: single-member district; MP: member of parliament; GDP: gross domestic product.Note: Entries are unstandardised linear regression coefficients. Cluster-robust standard errors in parentheses. Standard errors are clustered by electoral district (incumbent MP). ***p < 0.01, **p < 0.05, *p < 0.1.

To test the second hypothesis, Model 2 of Table 1 includes the government status of the mayor into a three-way interaction⁷ along with SF value and MP government status. In the case of government MPs (right panel of Figure 2), no matter the mayor’s party affiliation, the effect of SF value is positive on electoral performance. Nevertheless, a steeper dashed line indicates that if the mayor is officially supported by a government party, voters reward government MPs to a greater extent than they do in towns with opposition mayors, where the linkage of accountability is confused (solid line). These results confirm the second hypothesis.

Figure 2.

The effect of per capita SF on town-level electoral performance across the MPs’ and the mayors’ party affiliations.

On the other hand, the mayor’s party affiliation also affects how many votes opposition MPs lose with increasing government funding (see the left panel of Figure 2). In towns with a government mayor (dashed line), opposition MPs fare significantly poorer than in towns where the mayor is supported by the opposition (solid line).

For a substantive interpretation of the results, Table 2 presents eight combinations of SF per capita and the mayor’s party affiliation. Figures show that whilst with zero funding, government MPs score lower in towns with government mayors (14.8%) than in towns with opposition mayors (18.4%) – probably because of the declining government popularity – increasing the funding to an average level increases their electoral performance by 8.6% points in towns with government mayors. Interestingly, in comparison with towns with average funding, there is almost no electoral gain in doubling the amount of funds even in towns with government mayors. Similarly, halving the funds would not result in a substantial vote loss at the next elections.

Table 2.

The predicted vote share of government MPs across varying levels of funds and the mayors’ party affiliations.

Structural Funds	Opposition mayor	Government mayor
Zero	18.4	14.8
Average/2	18.9	23
Average	19	23.4
Average × 2	19.04	23.9

MPs: members of parliament; SMD: single-member district.Note: Entries are percentages. Calculations are based on the results of Model 2 of Table 1. The average value of SMD-average town vote share is taken into account as baseline.

A similar picture emerges if one looks at how one unit change in the value of per capita SF changes electoral performance. In a town with average support for the MP, increasing the funds by 10,000 HUF per capita (approximately 31 EUR) does not change expected vote share at all, even with a government mayor. However, in a hypothetical town with a starting value of 0 for funding and a government mayor, the same amount of change results in a 6.6 percentage-point increase in the government MP’s vote share. This growth is even larger (9.9% points) if we consider SMDs where government MPs perform best (28% SMD-average town-level vote share). Contrarily, in towns with opposition mayors, the increase in vote share is negligible independent of the starting value of funding. These figures suggest that in towns with a relatively low level of funding, an additional 10,000 HUF per capita increase achieves a larger increase in the government MP’s vote share than in towns where the level of funding is high (see the Online appendix for more details). Consequently, if governments want to improve the electoral performance of their MPs, they should opt for an even distribution of government funds instead of focusing larger amounts of money into a few constituencies.

Conclusions

In this study, I tested whether the allocation of EU SF affects the town-level electoral performance of Hungarian SMD MPs. I used SF allocation data between 2007 and 2010 and found that funding improves government MPs’ electoral performances, whilst decreasing the vote share of opposition MPs and that the effect size is conditional upon the mayor’s party affiliation. In a town with a government mayor, the government MP performs significantly better than in a town with an opposition mayor.

The findings have consequences in three broader areas. First, on the one hand, offering true incentives to vote gathering, successful pork barrel politics invite additional targeted funding in areas with a friendly political climate, and therefore, it may re-direct funding from less developed regions. This is in conflict with the designated goals of EU SF. On the other hand, however, it is also true that giving money to towns with either opposition MPs or opposition mayors does not strengthen the opposition. Therefore, developing regions may be beneficiaries of SF regardless of their political leaning without sabotaging the government’s electoral goals. Consequently, SF still have the potential to decrease regional disparities. Second, from the perspective of local representation, pork barrel politics creates the electoral connection. MPs become interested in lobbying for their districts. However, in connection with the distribution of SF, this is true only for government MPs, and thus local representation improves only in government-friendly areas. Third, as EU funds increase government support at elections, these funds may unintentionally contribute to the legitimisation and strengthening of the regime. With the recent emergence of Eurosceptic governments in Eastern Europe, EU funds may also strengthen anti-EU sentiments by helping these governments further stabilise their positions.

Although the findings of this study are robust to a number of changes in the models, some limitations have to be taken into account. Most importantly, the challengers of incumbent MPs are not included in the analysis. The electoral results of incumbents are also affected by the characteristics and actions of the challengers and their capacity to attract funding. Second, the connection between the funding and electoral results may not be as direct as assumed. Analysing the case of Brazil, Samuels (2002) and Ames (1995) point out that targeted funding may not influence voters directly, but indirectly, through local campaign financiers. Third, the analysis focusses only on one electoral term. However, to provide more convincing evidence, further research should look at how changes in SF over time affect the electoral performance of the same legislator. Fourth, there is no information about who actually ‘brought home the bacon’: The MP, the mayor, the challenger or someone else. We also do not have any information on the type and intensity of advertising that accompanied the funding. It is very likely that voters react to funding from the same agent differently depending on the visibility of the project. Nevertheless, even with these limitations in mind, the results strengthen the literature’s presumption on whether pork barrel politics works and raise questions about the effect of EU funds on regime support.

Supplemental Material

Supplemental material for Votes, money can buy. The conditional effect of EU Structural Funds on government MPs’ electoral performance

Supplemental Material for Votes, money can buy. The conditional effect of EU Structural Funds on government MPs’ electoral performance by Zsófia Papp in European Union Politics

Footnotes

Acknowledgements

The author is a recipient of the János Bolyai Research Scholarship. I thank Gergő Medve-Bálint for graciously allowing me to use his data on Structural Funds, and his advice on EU funding schemes, and Bastien Chabé-Ferret for checking the replication code. I also thank Veronika Patkós, Márton Bene, Martin Gross, Inger Baller, Hiroki Kubo, Christopher Kam, Rosario Aguilar, Monika Banaszewska and Attila Bartha for their valuable comments to earlier versions of the article. The four reviewers of this article and the editors of European Union Politics did a tremendous job, for which I am very thankful. Finally, I would like to thank László Lovász, president of the Hungarian Academy of Sciences and the chief officers of the Hungarian Academy of Sciences for persisting in their negotiations with the Hungarian government to make sure that academic freedom prevails in times when autonomies are scarce.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The author(s) received financial support from the National Research, Development and Innovation Office under the grant PD115747.

Notes

Supplemental Material

Supplemental material for this article is available online.

References

Achen

(2001) Why lagged dependent variables can suppress the explanatory power of other independent variables. In: Annual meeting of the political methodology section of the American Political Science Association, Los Angeles, CA, 2001.

Alvarez

Saving

(1997) Deficits, democrats, and distributive benefits: Congressional elections and the pork barrel in the 1980s. Political Research Quarterly 50(4): 809–831.

Ames

(1995) Electoral rules, constituency pressures, and pork barrel: Bases of voting in the Brazilian congress. The Journal of Politics 57(2): 324–343.

Anderson

(2000) Economic voting and political context: A comparative perspective. Electoral Studies 19(2–3): 151–170.

Banaszewska

Bischoff

(2017) EU funds and the prospect of re-election: Evidence from local elections in Poland. In: EU Cohesion Policy in Eastern and Southern Europe: Taking Stock, Poznan, 2017.

Bickers

Stein

(1996) The electoral dynamics of the federal pork barrel. American Journal of Political Science 40(4): 1300–1326.

Bloom

Petrova

(2013) National subversion of supranational goals: ‘pork-barrel’ politics and EU regional aid. Europe-Asia Studies 65(8): 1599–1620.

Bodenstein T and Kemmerling A (2012) Ripples in a Rising Tide: Why Some EU Regions Receive More Structural Funds than Others. European Integration Online Paper 16: 1--24.

Borraz

John

(2004) The transformation of Urban Political Leadership in Western Europe. International Journal of Urban and Regional Research 28(1): 107–120.

10.

Bouvet

Dall’erba

(2010) European regional structural funds: How large is the influence of politics on the allocation process? Journal of Common Market Studies 48(3): 501–528.

11.

Burkhart

Lewis-Beck

(1994) Comparative democracy: The economic development thesis. The American Political Science Review 88(4): 903–910.

12.

Cain

Ferejohn

Fiorina

(1987) The Personal Vote: Constituency Service and Electoral Independence. Cambridge: Harvard University Press.

13.

Calvo

Murillo

(2004) Who delivers? Partisan clients in the Argentine electoral market. American Journal of Political Science 48(4): 742–757.

14.

Case

(2001) Election goals and income redistribution: Recent evidence from Albania. European Economic Review 45(3): 405–423.

15.

Castells

Solé-Ollé

(2005) The regional allocation of infrastructure investment: The role of equity, efficiency and political factors. European Economic Review 49(5): 1165–1205.

16.

Costa-I-Font

Rodriguez-Oreggia

Lunapla

(2003) Political competition and pork-barrel politics in the allocation of public investment in Mexico. Public Choice 116(1/2): 185–204.

17.

Cox

McCubbins

(1986) Electoral politics as a redistributive game. The Journal of Politics 48(2): 370–389.

18.

Dahlberg

Johansson

(2002) On the vote-purchasing behavior of incumbent governments. American Political Science Review 96(1): 27–40.

19.

Debus

Gross

(2016) Coalition formation at the local level: Institutional constraints, party policy conflict, and office-seeking political parties. Party Politics 22(6): 835–846.

20.

Dellmuth LM (2011) The cash divide: the allocation of European Union regional grants. Journal of European Public Policy 18(7): 1016--1033.

21.

Dellmuth

Stoffel

(2012) Distributive politics and intergovernmental transfers: The local allocation of European Union structural funds. European Union Politics 13(3): 413–433.

22.

Dellmuth

Schraff

Stoffel

(2017) Distributive politics, electoral institutions and European structural and investment funding: Evidence from Italy and France. Journal of Common Market Studies 55(2): 275–293.

23.

Denemark

(2000) Partisan Pork Barrel in parliamentary systems: Australian constituency-level grants. The Journal of Politics 62(3): 896–915.

24.

European Commission (2009) European Cohesion Policy in Hungary. European Commission. Available at: https://ec.europa.eu/regional_policy/sources/docgener/informat/country2009/hu_en.pdf (accessed 26 November 2018).

25.

Fazekas

Chvalkovska

Jiri

, et al. (2014) Are EU funds a corruption risk? The impact of EU funds on grand corruption in Central and Eastern Europe. In: Mungiu-Pippidi

(ed) The Anticorruption Frontline. The ANTICORRP Project. vol. 2. Berlin: Barbara Budrich Publishers, pp.68–89.

26.

Fazekas

King

Tóth

(2013) Hidden depths: The case of Hungary. In: Mungiu-Pippidi

(ed) Controlling Corruption in Europe. vol. 1. Berlin: Barbara Budrich Publishers, pp.74–82.

27.

Feldman

Jondrow

(1984) Congressional elections and local federal. American Journal of Political Science 28(1): 147–164.

28.

Gerring

(2006) Case Study Research: Principles and Practices. New York: Cambridge University Press.

29.

Golden

Picci

(2008) Pork-Barrel politics in Postwar Italy, 1953–94. American Journal of Political Science 52(2): 268–289.

30.

Hainmueller J, Mummolo J and Xu Y (2018) How Much Should We Trust Estimates from Multiplicative Interaction Models? Simple Tools to Improve Empirical Practice. Political Analysis 27(2): 163--192.

31.

Heinelt

Hlepas

(2006) Typologies of local government systems. In: Bäck

Heinelt

Magnier

(eds) The European Mayor. Political Leaders in the Changing Context of Local Democracy. Wiesbaden: Springer, pp.21–42.

32.

John

Ward

(2001) Political manipulation in a majoritarian democracy: Central government targeting of public funds to English Subnational Government, in space and across time. The British Journal of Politics and International Relations 3(3): 308–339.

33.

Kálmán

(2011) Derangement or development? Political economy of EU structural funds allocation in new member states – Insights from the Hungarian Case.

34.

Kemmerling

Bodenstein

(2006) Partisan politics in regional redistribution: Do parties affect the distribution of EU structural funds across regions? European Union Politics 7(3): 373–392.

35.

Keszthelyi

(2017) Hungary’s economy heavily depends on EU funds, study finds. Budapest Business Journal, 30 March. Available at: https://bbj.hu/economy/hungarys-economy-heavily-depends-on-eu-funds-study-finds_130880 (accessed 4 July 2019).

36.

Kneebone

McKenzie

(2001) Electoral and Partisan cycles in fiscal policy: An examination of Canadian Provinces. International Tax and Public Finance 8(5–6): 753–774.

37.

KPMG (2017) A magyarországi európai uniós források felhasználásának és hatásainak elemzése a 2007-2013-as programozási időszak vo-natkozásában. KPMG. Available at: www.palyazat.gov.hu/magyarorszagi_europai_unios_forrasok_elemzese# (accessed 26 November 2018).

38.

KSH (n.d.) A központi költségvetés kiadásai (2005). KSH. Available at: www.ksh.hu/docs/hun/xstadat/xstadat_eves/i_qse007a.html (accessed 17 January 2019).

39.

Lancaster

(1986) Electoral structures and pork barrel politics. International Political Science Review 7(1): 67–81.

40.

Lancaster

Patterson

(1990) Comparative pork barrel politics: Perceptions from the West German Bundestag. Comparative Political Studies 22(4): 458–477.

41.

Leigh

(2008) Bringing home the bacon: An empirical analysis of the extent and effects of pork-barreling in Australian politics. Public Choice 137(1–2): 279–299.

42.

Levitt

Snyder

(1997) The impact of federal spending on house election outcomes. Journal of Political Economy 105(1): 30–53.

43.

Lewis-Beck

Stegmaier

(2000) Economic determinants of electoral outcomes. Annual Review of Political Science 3(1): 183–219.

44.

Lewis‐Beck

Stegmaier

(2007) Economic models of voting. In: Dalton

Klingemann

H-D

(eds) The Oxford Handbook of Political Behavior, pp. 518--537.

45.

Luo

Zhang

Huang

, et al. (2010) Village elections, public goods investments and pork barrel politics, Chinese-style. The Journal of Development Studies 46(4): 662–684.

46.

Magnier

(2006) Strong mayors? On direct election and mayors’ entrepreneurship. In: Bäck

Heinelt

Magnier

(eds) The European Mayor. Political Leaders in the Changing Context of Local Democracy. New York: Springer, pp.351–374.

47.

Mayhew

(1974) Congress: The Electoral Connection. 2nd ed. New Haven, CT: Yale University Press.

48.

Medve-Bálint

(2015) Converging on Divergence: The Political Economy of Uneven Regional Development in East Central Europe (1989-2014). Central European University, Budapest.

49.

Medve-Bálint

(2017) Funds for the wealthy and the politically loyal? How EU funds may contribute to increasing regional disparities in East Central Europe. In: Bachtler

Berkowitz

Hardy

, et al. (eds) EU Cohesion Policy. Reassessing Performance and Direction. London: Routledge, pp.220–240.

50.

Medve-Bálint

(2018) The cohesion policy on the EU’s eastern and southern periphery: Misallocated funds? Studies in Comparative International Development 53(2): 218–238.

51.

Milligan

Smart

(2005) Regional Grants as Pork Barrel Politics. ID 710903, SSRN Scholarly Paper. Rochester, NY: Social Science Research Network. Available at: https://papers.ssrn.com/abstract=710903 (accessed 30 May 2017).

52.

Muraközy

Telegdy

(2016) Political incentives and state subsidy allocation: Evidence from Hungarian municipalities. European Economic Review 89: 324–344.

53.

Nadeau

Lewis-Beck

(2001) National economic voting in U.S. presidential elections. The Journal of Politics 63(1): 159–181.

54.

Nicolescu

Gorcea

(2005) Approaches to local representation in Poland, Estonia, and Bulgaria. In: Soós

Zentai

(eds) Faces of Local Democracy. Comparative Papers from Central and Eastern Europe. Budapest: Open Society Institute, pp.207–242.

55.

Powell

Whitten

(1993) A cross-national analysis of economic voting: Taking account of the political context. American Journal of Political Science 37(2): 391–414.

56.

Samuels

(2002) Pork barreling is not credit claiming or advertising: Campaign finance and the sources of the personal vote in Brazil. The Journal of Politics 64(3): 845–863.

57.

Stein

Bickers

(1994) Congressional elections and the pork barrel. The Journal of Politics 56(2): 377–399.

58.

Stratmann

Baur

(2002) Plurality rule, proportional representation, and the German Bundestag: How incentives to pork-barrel differ across electoral systems. American Journal of Political Science 46(3): 506–514.

59.

Tarrow

(1977) Between Center and Periphery: Grassroots Politicians in Italy. New Haven: Yale University Press.

60.

Veiga

(2013) Intergovernmental fiscal transfers as pork barrel. Public Choice 155(3–4): 335–353.

61.

Whitten

Palmer

(1999) Cross-national analyses of economic voting. Electoral Studies 18(1): 49–67.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.37 MB