How does politics affect the allocation of public investment?

Public finance theory has argued that public sector intervention in the economy is primarily motivated by the principles of efficient allocation of resources, equal distribution of wealth, and stabilisation of economic activity over the business cycle (Musgrave 1959). However, research conducted over the last three decades has frequently demonstrated how political considerations have seriously influenced the allocation of public expenditures to different jurisdictions, leading to pork-barrel and undermining the capacity of public resources to generate greater social welfare (e.g. Besley and Coate 1998, Johnston 1977, Margolis 1968).

Our research tests empirically the relationship between electoral results and regional public investment spending in Greece for the period between the restoration of democracy (17 November 1974) until the outbreak of the economic crisis in 2009.

Why Greece?

The political context

Greece represents a very interesting case to study the politicisation of public investment. Like other southern European countries, such as Portugal or Spain, it witnessed massive political and socioeconomic transformations in the last five decades. The two most important ones have been the transition from dictatorship to democracy and membership of the European Union. The fall of the dictatorship in 1974 led to the creation of the Third Hellenic Republic and ushered in one of the most pluralist and progressive periods in modern history. The Greek political panorama soon became a two-party system: the right-wing New Democracy Party (ND), also known as the Liberal Party, and the left-wing Pan-Hellenic Socialist Movement (PASOK), also known as the Socialist Party, alternated in government throughout the whole period of analysis (1974–2009). The two main parties garnered the support of the vast majority of the electorate, sometimes jointly amassing in excess of 80% of the votes cast (Lyrintzis and Nikolakopoulos 2004). The establishment of an electoral system that favoured the party coming first – the so-called ‘reinforced proportionality’ system which awards extra seats in Parliament to the winning party – delivered stable majorities which enhanced political stability, but also marshalled plenty of opportunities for graft and clientelism. A system designed to guarantee stable governments also created incentives for the governing party to make sure it came first in national elections, as single-party-governments were the only outcome of the electoral system in Greece until the outbreak of the 2009 crisis.

Members of Parliament (MPs) were elected with an open-list system and voters could express preferences for party and candidates. In a Parliament of 300 seats, 151 MPs secured an overall majority. MPs where elected in 56 constituencies. Forty-nine out of the 56 constituencies coincide with prefectures – the NUTS 3 geographical level – and these did not change during the period of analysis. Electoral constituencies returned MPs in proportion to their population and the number of MPs by constituency could be adjusted in response to changes in population. Athens B was the largest constituency by population, electing 42 MPs in the 2004 elections. Eight small constituencies elected only one MP each.

In a country with a very centralised administration and with strict voting discipline in Parliament, incumbent governments always held a monopoly over decision-making on the geographical allocation of public investments. MPs of the governing party thus acted as the main conveyors of local political interests in Athens. We paid particular attention to the role of the incumbent government in decisions about the geographical allocation of public investment.

Public investment

The distribution of public investment in Greece was highly centralised. Almost 75% of total public investment expenditure was administered by National Ministries. Of the remaining 25%, 10% was devolved to the regions (prefectures) and 15% to the municipalities. The geographical allocation of public investment did not rely on any particular formula, making the geographical distribution of public funding extremely open to political bargaining.

Our analysis considers only the percentage of public investment that is regionally identifiable. This represents on average 55% of the total and includes the EU structural assistance to Greece, alongside the national contribution to EU co-financed projects, as well as projects of purely national interest and funding (Rodríguez-Pose et al. 2012). Public investment was very unevenly distributed across Greek regions (Figure 1). The territorial allocation of funds had the aim of achieving greater regional economic development, although this objective may have been tainted by politics. Decisions by central government regarding the distribution of funds may have, to a greater or lesser extent, followed political rather than socioeconomic criteria. As such, public investment may capture political influence and pork-barrelling, as these two factors concern the provision of public and collectively consumed goods and services. Pork-barrelling may have affected some types of investment (e.g. road infrastructure building) to a greater extent than others (e.g. expenditures in social well-being). Nevertheless, the general perception is that the whole Greek payments system was subject to political manipulation.

Figure 1. Regional distribution of public investment expenditures per capita and GDP per capita

Findings

The results of our analysis demonstrate the important role politics has played in the geographical distribution of public investment in Greece. In the 34 years before the outbreak of the crisis, public expenditure was closely correlated to electoral results. The two main Greek political parties when in office followed a ‘loyal’ voter model to the detriment of a ‘swing’ voter model – they rewarded those territories that returned them to office, rather than favouring marginal territories which may have brought them more votes and seats in future elections. When in office, the two main parties showered funds to those regions which a) delivered the greatest number of votes and MPs and b) where the distance between them and the main opposition party was greatest. There is, by contrast, no evidence that areas of opposition party strength (i.e. ‘hopeless’ areas for the governing party) or contested areas (i.e. ‘marginal’ or ‘swing’ regions) benefitted in any way from pork-barrel politics. In general pork-barrel investments ended up in regions where the two main parties had a strong grip and often monopolised the number of deputies returned. Single-seat constituencies were the main beneficiaries of the system, and received significantly higher shares of public investment per capita relative to multi-seat regions. Finally, pork-barrel was reinforced after the re-election of an incumbent government. Changes in the political orientation of the government after national elections resulted in a decline of the political allocation of public investment.

Overall, the analysis points to the importance of politics for the geographical allocation of public investment per capita in Greece. Our results highlight that there is little evidence that public investment followed criteria aimed at reducing the pervasive territorial inequalities in the country, but plenty more of political meddling and pork-barrel. The strong political polarisation in a country with relatively weak institutions, and the political rewards associated with an electoral system aimed at securing political stability, left the two main political parties in the country with significant room to reward those territories that returned them to office.

This article is published in collaboration with VoxEU. Publication does not imply endorsement of views by the World Economic Forum.

To keep up with the Agenda subscribe to our weekly newsletter.

Author: Yannis Psycharis is a Professor of Regional Economic Analysis at the Department of Economic and Regional Development, Panteion University of Social and Political Sciences. Andrés Rodríguez-Pose is a Professor of Economic Geography, London School of Economics; Research Fellow, CEPR. Vassilis Tselios is an Assistant Professor of Regional Economics at the Department of Planning and Regional Development, University of Thessaly.

Image: Restoration experts stand inside the temple of the Parthenon. REUTERS/John Kolesidis