اثر تجارت خارجی و آزادسازی سرمایه گذاری در تمرکز مکانی فعالیت های اقتصادی

I examine the varying responses of countries to foreign trade and direct investment liberalization on spatial concentration of their economic activity by taking into consideration moderating factors such as their market size and level of economic development. I argue that liberalization increases the concentration under normal conditions but large market size, and underdevelopment can disperse economic activity. Using data from 168 countries for the period of time starting in 1980s, I found support for all hypotheses.

In the past several decades, the world experienced large scale liberalization, when many countries reduced their international trade and investment barriers with their natural partners through regional trade agreements and as part of the global GATT/WTO obligations. While the general trend across the countries has been toward lower barriers, the pace and timing has been different. While some countries, such as those in the European Union, started the liberalization process fairly early, others came in much later after observing its positive results in Europe. In developing countries, Structural Adjustment Programs of the IMF and the World Bank included reductions in trade and investment barriers as part market reforms. The effect of these liberalization efforts on the spatial concentration of economic activities has not been very clear. Of particular interest is whether liberalization leads to an increase in the concentration of economic activity or disperses it to a larger geography, and the underlying factors affecting this relationship. Unfortunately, empirical literature on this subject is very limited and the results are too industry- or country-specific to make any generalizations. Only a handful of studies examined individual industries or particular industries in several countries, such as Hanson (1997) on the implications of NAFTA on Mexico, Storper, Chen, and De Paolis (2002) on European economies, analysis of Spain by Tirado, Paluzie, and Pons (2002), He, Wei, and Xie (2008) on China, Sanguinetti and Volpe Martincus (2009) on Argentina's manufacturing industry, and Sjoberg and Sjoholm's (2004) work on Indonesia's manufacturing industry. While the empirical study of Sjoberg and Sjoholm (2004) on Indonesia's manufacturing industry finds that concentration did not decrease as a result of liberalization, Hanson (1997) finds that NAFTA led to a less-concentrated spatial distribution in Mexico as firms found it more profitable to locate along the border to the US rather than old industry centered in Mexico City. In other words, results in these limited-scope empirical analyses have been inconclusive and in support of both arguments. Similarly, in the theoretical literature, one can find arguments on either side. Briefly speaking, some argue that liberalization reduces concentration since under protectionism, companies tend to locate closer to main domestic markets and that strategy changes with liberalization (Behrens et al., 2007, Henderson, 1982 and Krugman and Livas, 1996). Others suggest that expansion of international trade and investment primarily favors existing industrial centers, leading to higher concentration (Haaparanta, 1998, Paluzie, 2001, Rauch, 1991 and Yeboah, 2000). After conducting an extensive literature survey, Brulhart (2011) finds that the theoretical outcome depends on the modeling choices. He concludes that whether liberalization leads to concentration or dispersion of economic activity depends on each country's specific characteristics. This article aims to contribute to both the empirical and theoretical literature on this subject. Since the existing theoretical literature on the implications of foreign trade and investment liberalization on the concentration of economic activity is inconclusive, one of this article's key contributions is contextualizing this relationship by considering how the impact of liberalization on concentration is affected under the presence of moderating factors. In particular, this article considers country specifics such as market size and level of economic development. Further, the empirical analyses on this subject suggest that the result of liberalizations might have been affected differently by such specifics affecting the industries and/or countries analyzed. The mixed results in these empirical studies emphasize the need for a comprehensive study on this subject covering multiples of countries’ overall economy over an extended period of liberalization. The intention of this paper is to fill this gap with a study covering the experiences of 168 countries, which provides sufficient diversity in terms of country specifics. This study covers the time period since the 1980s – a period of extensive liberalization efforts of different intensity across countries with a range of different effects on spatial concentration. Using this comprehensive dataset in terms of country coverage and time period, I measure and analyze how the concentration of economic activity has been impacted by trade and investment liberalization. I also identify the role of country specifics such as market size and economic (under)development. I believe that from a policy perspective, discriminating between these two sets of outcomes and identifying the factors under which dispersion or concentration occurs would be of considerable interest.

This study aims to explore the implications of trade and investment liberalization on the spatial distribution of economic activity within a country. There are contradicting theoretical arguments and empirical studies involving only several countries or industries suggesting that liberalization can lead to both a concentration of economic activity at existing major centers and dispersion of activity across many locations in a country. Hence, existing research presents a gap on whether trade and investment liberalization increases or decreases the concentration of economic activity universally for all countries and in all industries, and more importantly identifying conditions under which a concentration or dispersion occurs is still left largely unexplored. I extend the current research by carrying out a comprehensive analysis of more than 100 countries over a period of time that exceeds 30 years to explore these conditions. 5.1. Main arguments and findings The main hypotheses of this study are that after moderating for some factors that cause heterogeneity in the results, reduction in barriers to free trade and investment results in an increase in the concentration of economic activity in existing centers. In this study, I moderated for the size of the market, and the level of economic development, as potential factors that may generate varying results. I hypothesize that countries that are less developed and small in terms of market size, tend to be more concentrated initially. Further, the process of liberalization may yield different results for developing countries in comparison to developed, and for small markets in comparison to large. I hypothesize that liberalization increases concentration of economic activity more for developed countries and for smaller countries relative to developing and large countries, respectively. Using panel data from 168 countries over a period of time starting with the 1980s to 2010 for most countries, I empirically tested the above hypotheses. I used population movements in administrative regions to track changes in the volume of economic activity during liberalization, and used a concentration measure based on Herfindahl index. The analyses showed support for the argument that both liberalization of trade and that of investment leading to higher concentrations of economic activity after moderating for other factors. It provided further statistically significant evidence that countries with smaller markets tend to have more concentrated economic activities and that liberalization leads to more concentration in smaller countries relative to larger ones. The results were statistically significant and supported the hypotheses on the impact of liberalization on concentration for developed countries relative to less developed ones. Accordingly, economic activity is less concentrated in developing countries, and liberalization increases concentration more for developed countries. Briefly speaking, this study advances the international business literature on the impact of liberalization on agglomeration. It introduced two factors, market size and the level of economic development that can affect multinationals’ decisions on locating in existing centers of economic activity or elsewhere, thus affecting the level of agglomeration in a country. 5.2. Policy implications Location decisions are probably one of the most important decisions for multinationals interested in engaging in international trade with or making a direct investment in a country. It affects their access to markets, key resources and strategic assets. It also has implications on their efficiency and thus their competitiveness. Further, it can have strategic consequences such as the potential for engaging in multipoint competition. Should they locate in existing industrial centers and benefit from agglomeration or should they locate elsewhere to avoid costs associated with agglomeration? Benefits could be very significant such as availability of labor resources with diverse skill sets, access to large markets, benefits associated with proximity to suppliers, customers, and to other firms, network effects, knowledge spillovers, proximity to critical facilities such as transportation hubs, technology parks or universities, and to government agencies, logistical benefits, and cost reductions due to economies of scale. Likewise, costs resulting from agglomeration cannot be ignored. These include congestion leading to an increase in prices and rents, pollution affecting the ability to find key labor resources, and a shortage of labor leading to increases in costs. This study examined the experiences of countries at which point the costs associated with agglomeration exceeded its benefits and made the above suggestions with regard to the level of economic development and market size based on historical decisions of existing multinationals engaged in international trade or investment in these countries. Accordingly, multinationals are drawn to existing centers of economic activity for relatively developed and smaller countries. When these countries open up for international business through liberalization of foreign trade and investment, the result is a further increase in the concentration of economic activity in their few economic centers. These have important policy implications for the governments of such countries. There are a few preparations they can do prior to liberalization to reduce the costs associated with further agglomeration. They would need to take precautions against increasing pollution and congestion. Precautions should also include efforts to reduce potential inflation due to increases in prices and rents by financial institutions including central banks. Governments in such countries should invest in improving the infrastructure in their bigger cities to handle increases in its use, as well as investing in education in support of the development of technology centers. Such centers of economic activity are also likely to draw migrant workers from smaller cities, as they see their population depleted. Necessary social measures should be in place to facilitate the adaptation of rural populations into urban culture. Further, the local governments should consider reducing their services and expenditures to avoid budget deficits as they will most likely see their revenues fall. Central governments should consider financial assistance and guidance to local governments in such situations. For larger and less developed countries, agglomeration is not a major problem. Countries with large markets tend to have either multiple centers of economic activity, and less developed countries do not have a sufficient level of economic activity to make its concentration a serious concern. Further, liberalization has a much smaller concentrating impact as the market size increases. In fact, as the results of this article suggest, it may even have a reducing effect on the level of concentration for significantly large markets. In such situations, local governments should be prepared for migration out of large cities, rural development and provision of necessary services in larger geographies. Note that this study does not imply that globalization will lead to a few mega cities. Theory suggests that trade and investment liberalization increases the rate of growth and economic development. Hence, the size of markets will increase, and that increase will eventually reduce the rate of agglomeration in big cities, as the results suggests. 5.3. Limitations and further research The empirical results of this study should be interpreted within its limitations. The analysis used proxies for a couple of the important variables. One is the measurement of the dependent variable, concentration of economic activity, and the other is the level of economic development. Income, GDP per person to be more precise, is used in this study to capture the latter. The relationship between income and economic development is generally positive and the variables are highly correlated for most countries. But, there are several countries, where their level of income reflects more the level of output predominantly in certain sectors, such as natural resources, than their level of economic development. Composition, size and maturity of individual industries need to be taken into account more carefully in future research. This study also assumed that population changes would reflect the changes in the level of economic activity. While there is a reasonable link between the two, variations in technology across countries and even throughout time for a country might complicate the comparisons. Countries that have higher levels of economic development tend to have technologies that are not as labor-intensive relative to those in less developed countries. Hence, when economic activity picks up in a developing country, population increase will follow, but the increase in population will not be as high as in developed countries. Further, as countries experience technological advancement as time goes by, implying that they become less and less labor-intensive, increases in population will be less and less reflective of the amount of economic activity. This actually makes it more intriguing why the results show that liberalization leads to more concentration for developed countries than developing countries as the correlation between population and economic activity gets weaker for higher levels of economic development. I propose that future research should explore possible explanations for this result. While it may be the case that the underlying assumptions that led to this hypothesis, in particular lack of transportation infrastructure, may not be valid, this only explains why concentration would not increase more for developing countries in comparison to developed countries. Also, future research should look into repeating the same exercise in this study using all countries that provide regional economic data, rather than population data, to see if the same results are observed. Additionally, some of the results might have been affected by dispersal of economic activity from the largest city to the next largest city or to second tier of cities in other administrative regions. There is evidence to suggest that the former of these two forms of dispersal may happen first, and the latter coming into its own at later stages of development (Henderson et al., 2001). Last but not least, the Herfindahl index used also has its own limitations. Researchers should investigate what could be a better measure of changes in the level of economic activity.