سیاست های منطقه ای در یک فدراسیون می تواند بهره وری اقتصادی را بهبود بخشد؟

In the European Union and in many federal and non-federal countries, the central government pays subsidies to poor regions. These subsidies are often seen as a redistributive measure which comes at the cost of an efficiency loss. This paper develops an economic rationale for regional policy based on economic efficiency. We consider a model of a federation consisting of a rich and a poor region. The economy is characterized by imperfect competition in goods markets and unemployment. Firms initially produce in the rich region but may relocate their production to the poor region. We show that a subsidy on investment in the poor region unambiguously increases welfare if labour markets are competitive. If there is unemployment in both regions, the case for regional subsidies is weaker.

In many countries, governments spend a considerable amount of public funds on infrastructure and private investment subsidies in poor regions. Europe provides several examples for this type of regional policies. The European Union (EU) uses the so-called structural funds to subsidize public infrastructure spending in regions where per capita income is below 75% of the EU-wide average. Furthermore the cohesion fund aims at supporting economic development and growth in poor member countries. In 2003, the EU spent approximately one third of its entire budget on regional policies.1 After Eastern enlargement of the EU, the regional component in EU policy is likely to become even more important. Similar policies can be observed at the national level in many EU countries. For instance, between 1991 and 1999, Germany spent approximately 30 billion Euros per year (appr. 1.5% of GDP) to subsidize private investment and public expenditure on infastructure in Eastern Germany.2 The case of the Italian Mezzogiorno provides another notorious example where investment in poor regions is subsidized by the central government. Similar examples can be found in many other countries. A common feature of regional policies is that they are designed to raise investment in poor regions by either i) directly granting subsidies for private investment or ii) improving the regional infrastructure. This observed pattern of regional policies is not easily explained in economic terms.3 For example, the “new economic geography” literature suggests that, due to increasing returns to scale, regional agglomeration is a key engine for economic development (Krugman, 1991). In this framework it is difficult to explain why a country tries to raise investment in poor regions rather than further supporting agglomeration in successful areas. One potential argument is that regional imbalances in economic development might induce migration from poor to rich regions which is considered to be harmful. For instance, an inflow of households into rich regions may be detrimental because of crowding externalities. In this case, regional subsidies may be used to induce inhabitants of poor regions to stay at home. However, this explanation for regional policies is hardly convincing. From an economic perspective, the best way to reduce migration to rich regions is to tax migrants or to pay direct subsidies for staying in the poor regions. But observed policies often enhance the mobility of households in poor regions. For example, the EU has considerably reduced barriers to mobility across countries and subsidizes human capital investment in particular in poor regions, which also increases household mobility. Another example is the case of Germany, where the government offers considerable subsidies for unemployed people who move from the East to the West. This suggests that the main motive behind regional policies is not the objective to reduce migration flows. Distributional considerations provide another rationale for regional policies. Subsidies on investment in poor regions may offer an instrument to redistribute income between rich and poor regions. There is a growing literature dealing with the redistributive impact of regional policies (see e.g., Martin, 1999). This literature argues that there is a trade-off between economic growth and regional income equality so that redistributive regional subsidies come at a cost in terms of economic growth and efficiency. The present paper develops an economic rationale for regional polices based on economic efficiency, rather than distributive considerations. We consider a model of two regions. In the rich “West”, monopoly suppliers produce goods which can be exported to the poor “East”. But since wage costs in the East are lower than in the West, firms may decide to close their factories in the West and produce in the East. Since firms face different fixed setup costs for a plant in the East, an equilibrium emerges where some firms stay in the West while other firms shift their production to the East. It turns out that, in this equilibrium, the number of firms producing in the East is inefficiently low if labour markets are competitive in both regions. The key reason for this inefficiency is that firms do not take into account the impact of lower production costs on the welfare of consumers. Our model thus provides a simple rationale for regional subsidies based on economic efficiency. This rationale for regional subsidies is questioned, though, if we assume that labour markets in both regions are characterized by wage rigidities and unemployment. In this case, too many firms may invest in the East because they neglect that the reduction of labour demand in the West reduces the welfare of Western workers. We also analyse the role of regional policies in the case where regional governments and a central government coexist. If too few firms invest in an underdeveloped region, it is natural to ask why the local government in this region should not be able to correct for this distortion. We show that regional governments in both regions do take into account positive (and negative) externalities of local investment. But the government of the poor region also has incentives to tax investment in the region because the firms are owned by residents of the other region. Therefore, the decentralised policy equilibrium is again characterized by too little investment in poor regions, so that the case for federal regional policy holds in this case as well. The rest of the paper is organized as follows. Section 2 presents the basic structure of the model. In Section 3, we analyze the role of regional investment subsidies assuming that there are no regional governments. Section 4 considers the case with autonomous regional governments. Section 5 concludes.

In this paper, we have analysed the efficiency of firm location and the case for regional policies in a model of a federation consisting of a rich and a poor region. Our model includes two sources of market failures: imperfect competition in markets for consumer goods and wage rigidities which give rise to involuntary unemployment. Both may distort the location decisions of firms under laisser faire. If we abstract from the labour market distortion by assuming that labour markets are competitive, the number of firms locating in the poor region is unambiguously too low and the optimal regional policy subsidizes investment in this region. The same result arises if there is unemployment only in the poor region. But if both regions suffer from unemployment, the number of firms moving to the poor region may be too high. Unemployment as a justification for regional investment subsidies is thus problematic if unemployment exists in high and low income regions. Empirically, this is precisely the situation in the European Union, where many high income countries do suffer from unemployment. If regions have sufficient autonomy to levy taxes and pay subsidies or transfers, an optimal allocation again requires federal subsidies to the poor region. In this case, the poor region does not subsidize investment sufficiently because these subsidies increase profits accruing to households living outside the region. The federal subsidy neutralises this tax exportation effect. Our model thus provides an economic rationale for observed regional policies which is based on economic efficiency, rather than distributional arguments usually emphasized in the literature. Of course, it should be taken into account that our analysis abstracts from several aspects which are relevant for regional policy. An important issue we do not discuss is the role of regional infrastructure. A significant part of regional subsidies support the development of the local infrastructure. This may also be seen as a way of attracting firms, but better transport infrastructure may also discourage local production since transport and commuting costs are reduced. Another problem we do not discuss is the choice between labour and capital subsidies. In most cases, regional subsidies take the firm of capital subsidies and may therefore distort the industry structure in favour of capital intensive industries.11 Finally, our analysis does not apply to cases where regional policies are linked to the decline of specific sectors or industries which dominate certain regions, such as, for instance, the steel and coal industry or the textile industry.