The EU spends €130 billion per year, equivalent to roughly 1% of the gross national income of its member states. One of the primary expenditure items is regional policy. A substantial share of the regional policy budget consists of subsidies to firms located in poor regions. The purpose is to attract firms to lagging regions based on the assumption that this will reduce regional inequalities, while ensuring that policy is consistent with overall economic growth and efficiency. In times of shrinking national budgets, it is fair to ask to what extent regional policy actually achieves this purpose, and if not, why.
It is a consensus in the field of political economy that policy is not the outcome of a sole decision made by a welfare-maximising social planner, but instead determined by the interaction of politicians, interest groups, and voters. Moreover, there are grounds for believing that inefficient policies are chosen because they serve the interests of politicians or social groups with political power. Subscribing to this view, this article reviews and draws out the policy implications of the small but emerging literature on the political economy of regional policy and industrial location.
The literature takes as its starting point the observation that governments tend to favour rural regions to an extent that seems out of proportion with their population size.1
According to the explanation put forward below, political-economy factors bias regional policy in favour of economically smaller regions (as measured by population size). This reverses the agglomeration process that follows from traditional models of the new economic geography, which treat regional policy as exogenous and predict that all industry concentrates in economically larger regions when trade is liberalised (see, for example, Krugman 1980). As will be seen, this reversed agglomeration process has implications for total regional welfare and the design of regional policy.
The first paper to incorporate political economy considerations in the study of regional policy and industrial location is Robert-Nicoud and Sbergami (2004). They argue that the population in economically larger regions is more widely spread out along the ideology dimension than the population in economically smaller regions. Therefore, politicians will find that a given subsidy level can buy more votes when the subsidies are distributed to firms based in smaller regions, where there are relatively more ideologically neutral voters, willing to swing their vote for changes in regional policy – so called swing voters. For this reason, smaller regions attract more industry than what is predicted by the traditional models of the new economic geography.
Do regions of different economic size also differ in terms of political participation? If so, what implication does this have for regional policy and industrial location? These are valid questions given the evidence that voters who participate in elections at higher levels are more likely to be represented in the policymaking process (see, for example, Martin 2003 and Strömberg 2004).
In Wiberg (2011), I show that the electorate in smaller regions gains relatively more from firm subsidies. Since the welfare level is lower in smaller regions, where fewer firms are located and consumer prices consequently are higher, a policy-induced increase in total regional welfare increases the welfare level more in smaller regions measured in relative terms. This makes the electorate in smaller regions reward regional policy with a relatively higher voter turnout compared to the electorate in larger regions. Vote-maximising politicians therefore allocate relatively more firm subsidies to smaller regions, since this increases the number of votes to be gained in elections. This, in turn, attracts more industry to smaller regions than what follows from new economic geography models with no voting over regional policy.
Besides characteristics of the electorate, the electoral system itself may have an impact on regional policy formation and industrial location. Indeed, advances in comparative political economy link policy choices to specific features of the electoral competition under majoritarian and proportional elections (see, for example, Persson 2002 and Wiberg 2014a).
In Wiberg (2014b), I show that smaller regions obtain more firm subsidies under majoritarian voting than with proportional representation. Intuitively, majoritarian elections increase competition between vote-maximising politicians by focusing it into pivotal districts in regions that host the electorate with the least average ideological bias. Since the majoritarian election outcome therefore is more sensitive to policy, smaller regions, which contain more swing voters, obtain a regional policy closer to their optimum than under the proportional representation.2 Thus, firms locate at relatively higher rates to smaller regions under a majoritarian system to take advantage of the more abundant government support.
The prediction that smaller regions receive more subsidies with majoritarian voting is supported by cross-country data for 44 countries over the period 1993-2007. In Wiberg (2014b), I show that countries with majoritarian representation on average spend about 40% more on government support to rural regions than countries with proportional elections.
Data suggests that elections as well as lobbying have important effects on the formation of policy in democracies. Specifically, lobbying groups representing geographic interests play a key role in the policymaking process in most societies (see, for example, Cadot et al. 2006). For example, in the EU, the system of structural policies has empowered member regions by allowing them to take part in the bargaining and decision-making phase of the allocation process of structural funds.3 Capturing features of this process, in Wiberg (2010) I analyse industrial location when rent-seeking firms compete over government subsidies.
In Wiberg (2010), I argue that the per-firm subsidy is larger in smaller regions than in larger regions; a dollar increment of the total subsidy is worth more to the fewer firms located in a smaller region, since less of it will be diluted by the number of firms sharing the subsidy than what would be the case if it ended up in a larger region, where there are relatively more firms competing for the rent. Therefore, firms in smaller regions are willing to expend relatively more resources to acquire such an increment. Once again, this shifts regional policy in favour of smaller regions, which makes them attract more firms compared to the case with no rent seeking.
Welfare and policy implications
Electoral concerns and rent seeking will bias regional policy in favour of smaller regions. This political bias implies that all industry concentrates in smaller regions when inter-regional trade costs are sufficiently low; when the movement of goods gradually becomes unrestricted between regions, firms become increasingly indifferent to the market-access advantage of producing in regions with more consumers (since firms are indifferent to location when trade costs are sufficiently low). Therefore, for some level of trade costs, the benefit to firms of a policy directed towards subsidising production in smaller regions outweighs the market-access advantage of larger regions.
This is illustrated in Figure 1, which shows how the share of industry in a larger region, denoted s*, changes as trade barriers are reduced and regional policy is determined by electoral concerns or rent seeking. s* takes the value of 1 when all industry locates to the larger region and 0 when all firms concentrate in the smaller region. φ is a measure of the freeness of inter-regional trade, where 0 corresponds to infinite trade barriers and 1 represents free trade. Figure 1 also displays as a dashed curve the location outcome that follows from the traditional models of the new economic geography, denoted s. As seen in the figure, when the political-economy factors determine regional policy, firms concentrate to the smaller region as trade is liberalised, while the traditional models lead to a shift of production to the larger region.
Figure 1. The share of industry for different degrees of openness
What does this mean for welfare? As I show in Wiberg (2014b), a regional policy that results in more firms located in smaller regions than in larger regions, where more consumers reside, leads to lower total regional welfare (i.e. the sum of welfare in all regions). Since there are more consumers in larger regions, and thus more individuals in total who stand to lose from higher expenditures on inter-regional imports when the agglomeration of firms takes place in smaller regions, total welfare decreases if there are relatively more firms located in smaller regions.
Hence, the policy bias introduced by electoral concerns and rent seeking, which favours smaller regions, lowers total welfare.
Moreover, this bias reduces total welfare more under majoritarian voting than under proportional elections, since the majoritarian regional policy results in more firms located in regions with fewer consumers.
However, it is important to note that there might be other goals behind regional policy besides overall welfare. For instance, some might want policy outcomes to be more equitable across regions. Unfortunately though, the goals of regional policy are often described in terms of economic growth and efficiency. For a more transparent policy discussion, politicians should be better at explaining these goals and their trade-offs. And, correspondingly, economists should be better at explaining the costs of regional policy.
Cadot, O, Röller, L and A Stephan (2006), “Contribution to productivity or pork barrel? The two faces of infrastructure investment”, Journal of Public Economics 90, pp. 1133-1153.
Homburg, S (1997), “Ursachen und Wirkungen eines Zwischenstaatlichen Finanzausgleichs”, in Oberhauser, A. (ed.) Fiskalföderalismus in Europa (Berlin: Duncker and Humblot).
Krugman, P (1980), “Scale Economies, Product Differentiation and Pattern of Trade”, American Economic Review 70, pp. 950-959.
Martin, P (2003), “Voting's Rewards: Voter Turnout, Attentive Publics and Congressional Allocation of Federal Money”, American Journal of Political Science 47, pp. 110-127.
Persson, T (2002), “Do Political Institutions Shape Economic Policy?”, Econometrica 70, pp. 883-905.
Robert-Nicoud, F and F Sbergami (2004), “Home-Market vs. Vote-Market Effect: Location Equilibrium in a Probabilistic Voting Model”, European Economic Review 48, pp. 155-179.
Strömberg, D (2004), “Radio's Impact on Public Spending”, Quarterly Journal of Economics 119, pp. 189-221.
Wiberg, M (2010), “Location Equilibrium with Endogenous Rent Seeking”, Journal of Economic Geography 9, pp. 869-887.
Wiberg, M (2011), “Political Participation, Regional Policy and the Location of Industry”, Regional Science and Urban Economics 41, pp. 465-475.
Wiberg, M (2014a), “Comparative Trade Policy”, Review of International Economics 22, pp. 410-421.
Wiberg, M (2014b), “The Comparative Political Economy of the Location of Industry”, Canadian Journal of Economics, forthcoming.
1 For example, GNP per capita had no effect on transfers within the EU in 1992, while population size was significantly negatively related to the distribution of regional support (Homburg 1997).
2 The argument that smaller regions contain more swing voters draws on Robert-Nicoud and Sbergami (2004).
3 EU member states and the European Commission first negotiate over national shares of the structural funds budget. Once a negotiation outcome has been reached, the national governments bargain with their respective regions over subsequent allocations.