How Effective Are Place-Based Industrial Policies?

Place-based policies are government initiatives designed to stimulate economic growth in specific, typically underdeveloped regions. These strategies encompass public infrastructure investment, reduced business payroll taxes, governmental backing for private loans, and incentives for private capital to invest in research and development. One aim of these policies is to ameliorate long-standing inequality in economic opportunities across different regions.¹

For their paper, “Micro- and Macroeconomic Impacts of a Place-Based Industrial Policy,” Philadelphia Fed Economic Advisor and Economist Enghin Atalay and his coauthors, Ali Hortaçsu, Mustafa Runyun, Chad Syverson, and Mehmet Fatih Ulu, analyzed the direct and indirect impacts of a place-based policy introduced by the Turkish government. They examined whether targeted subsidies increased economic activity for firms and industries that were newly eligible to receive subsidies, the extent of the “spillover” benefits for targeted firms’ customers and suppliers, and the overall impact of the policy on reducing regional wage inequality in the short term and the long term.

As in many other countries, Turkey’s internal differences in income per capita are large. In 2020, gross domestic product (GDP) per capita in Istanbul, in northwest Turkey, was almost five times greater than it was in the southeast city of Şanlıurfa.² “While some of the spatial differences in income per capita [in Turkey] reflect variation in workers’ human capital,” the authors write, “a large portion is due to inequality in economic opportunities.”

To address these regional disparities, the Turkish government introduced a new program of place-based subsidies in 2012. The program’s two largest elements were, first, investment tax credits and, second, rebates for employers’ and employees’ mandatory social security contributions.³ The government provided these subsidies to firms in certain industries, primarily in the agriculture, mining, manufacturing, and wholesale sectors. It divided the country into six “subsidy regions,” grouped mainly according to their average income per capita. For firms in covered industries, the program was more generous in less developed regions, which (as it happens) were mostly provinces in the southeast of the country. (Nevertheless, the country’s regions are far from isolated from each other, the authors note, with substantial trade and migration linkages across regions.)

In the first component of their analysis, Atalay and his coauthors considered the microeconomic impact of the subsidy program, estimating how economic activity evolved for firms newly eligible to receive subsidies. They find a significant increase in economic activity in the industry-province pairs that were the target of the government subsidies. More precisely, they show that a 5 percentage point increase in the investment tax credit subsidy rate — roughly equivalent to the difference in subsidies received among eligible-industry firms in the wealthiest and poorest subsidy regions — was associated with a 16.2 percent increase in firm revenues, an 8.7 percent increase in employment, and a 3.3 percent decrease in marginal costs.

The authors also measured the indirect benefits (positive spillovers) of the subsidy policy, identifying the extent to which subsidies accrue to the suppliers and customers of subsidized firms. They find meaningful spillovers: A 5 percentage point increase in the share of a firm’s customers and suppliers who received the subsidy corresponds to an increase in revenues of 0.7 percent, an increase in employment of 0.6 percent, and a decrease in marginal costs of 0.1 percent.

In the second step of their analysis, Atalay and his coauthors used a dynamic general equilibrium model with many industries and regions to assess the aggregate implications of the government’s subsidy program.⁴ Using this model, the authors explored the various channels through which subsidies can dissipate across industries and locations, benefitting not only the areas that the government targeted but also those that it did not.

The authors show that the subsidy program likely had a modest impact on reducing regional income inequality in the short term and an even smaller impact in the long term. They attribute the limited impact to domestic trade flows and migration: First, many of the suppliers and customers of heavily subsidized firms are in the richer western part of the country; some of the benefits of the subsidies increase economic activity there. Second, the reforms induce migration toward the subsidized regions, especially in the long run. This increase in the number of workers in the less developed parts of the country mitigates the inequality-reducing direct effects of the subsidy reforms.

Although this study focused on a specific Turkish policy, many of the authors’ conclusions will likely apply in other environments. Domestic trade and migration are important within other countries, so the spillover effects identified in their analysis will likely be relevant elsewhere. Additionally, the authors note that the slow response of migration to real-wage differences is a common empirical pattern across countries. So, the short- and long-term impacts of place-based policies may differ as well.

However, their analysis doesn't explore all the potential policy-related consequences. It doesn't, for example, address the costs of funding the subsidies. Nor does it consider the possibility that it may be costly (or impossible) for the government to ensure that firms don’t game the system, receiving subsidies without following through with their proposed investment projects. These are important considerations for future research.