Do They Promote Economic Development or Lead to Inequality
Published in the Southeastern Geographer
By J. M. Lane
Abstract
Foreign-trade zones (FTZs) are restricted-access sites where domestic and foreign goods are stored, manufactured, or assembled. Products foreign bound from an FTZ do not pay duties, tariffs, or ad valorem state taxes, yet are considered domestic goods. These zones are outside of US Customs and Borders Protection (CBP) jurisdiction and are considered foreign territory. FTZs are intended to promote economic development, provide a competitive advantage to US firms, and improve access to foreign markets for US manufacturing firms. However, while these zones can positively impact local economies, they may attract investment away from underserved regions, thereby exacerbating spatial inequality. This paper analyzes the spatial relationship between the number of FTZs, median household income, unemployment rates, income growth rates, and the number of manufacturing firms by county in the Southeastern United States. Results from this study find that counties with FTZs in the Southeast have significantly higher economic output than counties without access to FTZs.
Introduction
Place-based economic policies are used by policymakers to promote economic growth in certain locations by attracting industries such as manufacturing, technology, and distribution. An early example of place-based economic policy, foreign-trade zones (FTZs) were designed to increase local economic output, promote foreign commerce, and improve national productivity (Bolle and Williams 2012). Foreign-trade zones are not a new phenomenon to the United States and were discussed for decades preceding their creation. Several proposals made their way to Congress after a series of studies were performed by the Department of Commerce on the effectiveness of free ports in Copenhagen and Hamburg. Previous attempts at creating free ports were voted down by a Congress fearful of the impact of trade liberalization on domestic production. After completing these studies on European free ports, economists met together with prominent politicians and eventually passed the Foreign-Trade Zone Act of 1934 (McCalla 1990).
The early version of the act allowed imported goods to be sorted and transported without taxation. Four zones existed across the US in the beginning of operations and the number stayed relatively low for several decades (Easterling 2016). Closely guarded and regulated by U.S. Customs and Border Protection (CBP) authorities, original zones were required to have barbed wire fencing of at least 6 gage wiring and meshing could not be wider than 2 inches. Tax laws were not applied to these early zones; however, goods traveling within each zone were inventoried directly by CBP. In a legal sense, FTZs were, and still are, considered international territory while simultaneously existing within the US border (Orenstein 2011). Today, FTZs act as physical locations used for storage, manufacturing, assembly, and/or exhibition for foreign and domestic goods (Foreign-Trade Zone Board 2018).
When determining the location of FTZs, geography is important for siting these zones (Bolle and Williams 2012). The Foreign-Trade Zone Act outlines specific rules and guidelines for the creation of FTZs, giving a minimum of one FTZ within 60 miles of a port of entry (19 U.S.C. § 81b). Zones were limited based on proximity to ports of entry until the Boggs Amendment created subzones (19 U.S.C. § 81c). Public corporations, city councils, or county governments can apply for subzones in areas not located within proximity to a port of entry (15 C.F.R. § 400).
Due to the importance of manufacturing and scientific research on the economy of the Southeastern United States (Graves and Kalafsky 2017), it is important to analyze the economic impact of place-based policies such as FTZs on the surrounding population. According to McGilvray (2017), the Southeast currently has 56 active zones in 52 counties across a total of 924 counties in the region. These zones produce as much as $42 billion in exports and employ approximately 149,500 people. If FTZs are a catalyst for export-led growth, these zones could also be a factor for economic development in urban areas throughout the Southeastern United States. Under the current economic paradigm, Kalafsky and Graves (2018) argue that there is a significant relationship between metropolitan economic growth and exports in the Southeast. However, the same relationship does not exist in rural areas outside of the purview of large metropolitan regions (Kalafsky and Graves 2018). Because FTZs are designed to boost exports and re-exports (Bolle 1999), it is important to analyze the possible link between the location of FTZs and economic development in the Southeast.
FTZs are spatial by nature and cannot be considered without locational context. FTZ policies provide tax benefits to certain firms over others, giving those businesses a competitive advantage over other firms that are unaware of these policies (Kanellis 1995). Because some firms receive benefits over others and FTZs are unevenly distributed amongst major urban centers, these zones may exacerbate spatial inequality that already exists between urban and rural populations in the Southeastern United States. By attracting investment to major urban centers throughout the Southeast, FTZs can lead to highly uneven distribution of incomes, employment, and manufacturing.
The original intent of FTZ legislation was to promote re-exports and encourage foreign trade (S. Rep No. 905, 73rd Cong. 1934). Subsequent legislation was passed attempting to stimulate manufacturing (Bolle and Williams 2012). Additionally, the Foreign-Trade Zone Board (2018) explicitly states that FTZs are designed to promote local development, provide a competitive advantage for United States businesses, and attract investment in the domestic economy. Have FTZs been successful at promoting local development and manufacturing or have they influenced spatially uneven economic development? This study is designed to analyze the relationship between FTZ locations and the spatial distribution of income, income growth rates, unemployment, and manufacturing firms on the county scale in the Southeastern United States. This paper finds evidence that FTZs may have an impact on uneven development patterns across the Southeast.
Literature Review
Foreign-trade zones (FTZs) and other place-based policy measures have become a major factor in regional development in recent years. Many of these policies are designed to promote manufacturing, research and development, and increase regional income. The original intention of FTZs was to promote exports and boost the domestic economy during the Great Depression by removing customs on materials used in production of goods for re-export. Lawmakers argued that FTZs would attract foreign freighters to United States’ port infrastructure as stopping points between long distance shipments (Kanellis 1995). The broad research question guiding this project is whether FTZs accomplish their intended goals. Research to date has largely been concerned with issues such as the geographic distribution of FTZs, benefits to firms, and the impact of FTZs on exports and employment.
Geographies of FTZs
Free ports and FTZs are designed to loosen restrictions on trade in certain areas and promote regional/local development. According to McCalla (1990), these zones are created and sited by political bureaucracies, but exist outside the control of these governing institutions. Congress created FTZs to stimulate economic development by relaxing, and in some cases removing governmental regulations that would impede legitimate commerce. In theory, these zones should promote local development, thereby boosting the national economy. Thoman (1952) argued that because FTZs promote localized development, these zones are attracted to populated areas and are unevenly distributed. This author discussed the early development of FTZs and their distribution, stating that Atlantic and Pacific coastal regions were the largest beneficiaries to duty free zones. Of course, this was not limited to the early 1950s. According to McGilvray (2017), the vast majority of FTZs exist in states with access to coastal waterways; of the 195 total active zones throughout the United States, 96 are in Texas, Florida, California, New York, and Washington.
The growth of FTZs across the United States can have an impact on the spatial development patterns as they may promote dense clustering in major urban areas. Place-based policies are designed to promote the agglomeration of similar industries and promote collaborative research and development (Koo 2005). In addition to agglomeration, Bobonis and Shatz (2007) argue that FTZs and similar initiatives may also attract foreign direct investment. These authors also find that the agglomeration effects of FTZs do not hold to state boundaries and often spill over into neighboring counties and states. Similarly, according to Head et al. (1999), states that created FTZs earned significantly higher growth in foreign direct investment than states not adding FTZs. This study finds that FTZs have a substantial impact on industrial agglomeration, specifically from Japanese manufacturers, and the elimination of FTZs by states would significantly decrease regional investment.
What impact do FTZs have on the geographic distribution of non-manufacturing jobs in the US? Ghosh et al. (2016) used recent data to determine that FTZs significantly impact short-term and long-term non-manufacturing growth in FTZ zip codes and neighboring zip codes. Results from this study found that zip codes with FTZs and neighboring zip codes had higher growth in non-manufacturing jobs than other zip codes. While this study found a relationship between FTZs and non-manufacturing job growth, it did not consider the relationship between FTZs, income, total unemployment, and manufacturing firms.
Benefits of FTZs to Individual Firms
Most of the early research on FTZs focused on the potential of FTZs to reduce operational costs for US corporations. Many of these studies emphasized ways that companies and states could take advantage of FTZ policy to promote economic growth. (McDaniel and Kossack 1983, Robles and Hozier 1986, Cornwell 1989, Ferguson 1989, Tansuhaj and Jackson 1989). Authors have argued that FTZs can give a competitive advantage to US firms “and help attract direct investment to state and local economic development projects” (Robles and Hozier 1986, p 53). This article discussed the necessary management framework that companies needed to establish so that they could benefit from FTZ law.
Other authors described the types of goods that could be used by FTZ firms and provided a list of benefits. According to Fergusen (1989), these zones could help to facilitate the shipment of goods in and out of FTZ firms. This paper performed a cost/benefit analysis on firms of different sizes, concluding that larger firms benefited while costs often increased in smaller firms. McDaniel and Kossack (1983) designed a quantitative model for determining the financial benefits of FTZs on individual firms. This paper found that firms importing products benefited over firms that were exporting to other countries. Other scholars were less objective in their approach to FTZs. Beeman and Magill (1988) provided instructions on how firms could benefit from FTZs, stating that “FTZs can help reduce the weakening competitive position of many U.S. firms” (p 17). This article advertised the use of FTZs as a cost-saving venture.
Do FTZs meet original intent?
After FTZs became adopted on a larger scale, researchers began to investigate the effects of these zones on the larger population. By the 1990s, FTZs were evolving into something very different from their original intent. Some research focused on the impact of FTZs on exports and their overall effectiveness within a societal context (Kanellis 1995, Mathur and Ajami 1995, Hazari and Sgro 1996, Mathur and Mathur 1997). Evidence began to emerge showing interesting policy implications, leading Mathur and Ajami (1995) to conclude that FTZ administrators should promote the use of FTZs by manufacturing companies to reduce costs on duties and tariffs. This study found that FTZ firms have more advanced technology, more exports, and lower costs than non-FTZ firms; this is especially true for firms with 10 to 15 years of experience in exports and imports.
While FTZs showed significant impacts on company profit margins and marginal costs, evidence began mounting against the effectiveness of FTZs for reducing unemployment and increasing income levels. Arguing against reports by the Foreign-Trade Zone Board about the benefits of FTZs on the general population, Kanellis (1995) concluded that it was impossible to assess the effects of FTZs on the population at large. Other problems arise throughout the FTZ approval process. If a firm pursues a grant of approval from the Foreign-Trade Zone Board, competitors have sixty days to contest the approval. After sixty days, FTZ status is approved and no notifications are given to nearby competitors, giving FTZ firms a competitive advantage over non-FTZ firms. Other scholars have concluded that FTZs have fallen short of stated goals. Profits may increase for FTZ firms but no noticeable increase in exports occur (Mathur and Mathur 1997).
With the introduction of manufacturing in the zones post-1950, FTZs no longer acted as duty free ports of entry and exit. Firms could use these zones to manufacture products for re-export and avoid tariffs placed on raw materials. With subsequent amendments to the Foreign-Trade Zone Act, the legislative intent of FTZs has gradually shifted toward an acceptance of increased imports. In fact, imports were specifically mentioned in an amendment to the law in the 1990s (Bolle 1999). Others have argued that rather than promote exports, these zones may have promoted the growth of imports. However, by promoting the development of centrally located distribution sites to alleviate costs, FTZs aid in facilitation and transportation of foreign and domestic goods (Seyoum and Ramirez 2012).
Other scholars have evaluated the effectiveness of FTZs at promoting exports. According to Seyoum (2017), FTZs are necessary for US firms to remain competitive; but companies working within the boundaries of these zones focus mainly on US markets, producing few goods with export in mind. If a car manufacturing firm imports a radio into its FTZ, it does not have to pay duties or tariffs. That firm can then install the radio into a car and sell the car to US consumers without paying customs, duties, or local ad valorem taxes. Mathur and Mathur (1997) found that FTZs were often added to locations after exports increased, therefore having a negligible impact on exports. These authors argued that a feedback loop existed between FTZs and exports: when exports increased, FTZs were added and exports continued to increase. Other authors took a more critical view. Orenstein (2011) argued that FTZs were created to avoid protectionist policies by wealthy corporations while remaining close to the market. FTZs allow companies to manufacture products within the borders of the United States, sell those items under the guise of ‘Made in the USA,’ while avoiding taxes that their competitors are required to pay.
Spatial development in the Southeast
The purpose of the present study is to evaluate the possible impact of FTZs on economic development at the county scale within the Southeastern United States. For this reason, it is important to investigate literature on spatial development patterns in the Southeast. A recent study by The Brookings Institution’s Hamilton Project found that Southeast ranks lower in per capita income than other regions (Shambaugh and Nunn 2018). In fact, poor economic output has come to define the Southeast as a region. James (2010) used global and local Moran’s I to evaluate clustering of low per capita income and used the local results to define the borders of the economic South. Results from this study reveal that the economic borders defining the Southern economy do not follow conventional state boundaries.
While the Southeast has lower economic output than the rest of the US, there is a great deal of spatial heterogeneity. Differences between rural and urban incomes have steadily increased. According to Graves and Kalafsky (2017), the Southeast made several strides to catch up to the national average income but saw a decline beginning in 2013. While some states are growing considerably, other states such as Louisiana, Arkansas, and West Virginia remain some of the poorest states across the United States. Economic growth in the Southeast is unevenly distributed within state boundaries as well. Kalafsky and Graves (2018) argued that major metropolitan areas such as Miami, Atlanta, Orlando, Jacksonville, and Tampa are experiencing the highest growth in exports, foreign direct investment, and income while rural areas are losing ground. In 2016, these cities had at least one FTZ, exported $11.8 billion in goods out of these zones, and employed as many as 13,250 people in FTZ industries (McGilvray 2017). Exports from FTZs accounted for 18 percent of total exports value in these cities combined within the same year (International Trade Administration 2019).
I contend that the location of FTZs in major metropolitan areas throughout the Southeast may have a stronger relationship with economic development than previously suggested, thereby exacerbating spatial inequality between rural and urban areas. The purpose of this study is to analyze the spatial relationship between FTZs, income, unemployment, and manufacturing firms on the county level to better understand the changing economic landscape in the Southeast. Results from this study reveal a significant correlation between the location of FTZs and the spatial arrangement of median household income, unemployment rates, and manufacturing firms at the county scale.
Methods
There are 195 active and 65 inactive FTZs throughout the United States and Puerto Rico. The scope of this study is the Southeastern United States. Currently, there are 56 active FTZs across 10 states in the Southeastern United States; Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee, Virginia, and West Virginia were included in this study. These are the states that constitute the Southeastern Division of the American Association of Geographers (SEDAAG). There are 924 counties and 83,727 manufacturing firms in the Southeast, all of which were included in this analysis. Data concerning median household income, unemployment rates, and income growth rates were gathered from the U.S. Census Bureau (2017); data regarding FTZs were gathered from the Foreign-Trade Zone Board (McGilvray 2017); data concerning manufacturing firms were gathered from Infogroup (2017). These variables were included to examine the major differences in income levels, employment, and manufacturing between counties in proximity to FTZs and counties not in proximity to FTZs.
Data were compiled into a workable database using Microsoft Excel and converted to .csv format for usage within ESRI ArcMap 10.6. Statistical analyses were performed using Microsoft Excel version 1908, ESRI ArcMap 10.6, and GeoDa 1.4.6. All county data were presented in ArcMap as polygon vectors and were downloaded from the U.S. Census Bureau in geodatabase format. Foreign-trade zones were presented in point file vector format. All business data used in this study are licensed by Infogroup and contained within the Business Analyst Software. This dataset contains spatial and attribute data from approximately 13 million businesses in the United States, including industrial classification codes, total sales volume in US dollars, and number of employees (ArcGIS Online 2018).
After business point files were added to the map, standard industrial classification (SIC) codes between values of 200,000 and 400,000 were selected and a new point vector file was created using the select tool. A firm’s SIC code represents its type of industry. SIC codes used in this study represent manufacturing firms. After the new file was created, the newly created manufacturing point file was clipped and points that fall within the boundary of the Southeastern United States remained. These manufacturing point files were spatially joined with the county polygon vectors so that data could be analyzed along with income and unemployment data.
An optimized hotspot analysis on the county scale was used to explore any clustering of high median household income, low unemployment rates, high income growth rates, and manufacturing firms near FTZs. The Optimized Hot Spot Analysis tool in ArcMap uses the Getis-Ord Gi function to measure spatial autocorrelation of certain variables across space. Spatial autocorrelation refers to a phenomenon where objects close to each other have values more similar than those that are further away, also known as Tobler’s Law (O’Sullivan and Unwin 2010). Tobler’s Law states that “everything is related to everything else, but near things are more related than distant things” (Tobler 1970, p 234). This commonly occurs with income (i.e. wealthy counties are more likely to be near other wealthy counties). The Getis-Ord Gi function is a local statistic that creates a visual output showing local hot spots of similar high values and cold spots of similar low values, if any exist. For each location i, the local value is represented by:
G_i (d)=(∑_jw_ij (d)x_j 〗)/(∑_(j=1)^nx_j ) and i≠j (1)
where x_j is the values of the attributes in location j and w_ij (d) represents the weights within the spatial weights matrix (O’Sullivan and Unwin 2010). This formula was used in each location and a choropleth map was created showing the results of the analysis.
After the Getis-Ord Gi models were analyzed, spatial regression models were used at the county level to evaluate the potential relationship between median household income, unemployment rates, income growth rates, number of manufacturing firms and access to an FTZ. Spatial regression is used to analyze the relationship between two or more variables while considering distance as a variable. Spatial regression also tests for the existence of spatial autocorrelation using spatial lag models. The following formula depicts this method:
Spatial Lag Model: y=ρWy+Xβ+ε (2)
“where y is a N [population size] by 1 vector of observations on the dependent variable, Wy is the corresponding spatially lagged dependent variable for weights matrix W, X is a N [population size] by K matrix of observations on the explanatory (exogenous) variables, ε is a N [population size] by 1 vector of error terms, ρ is the spatial autoregressive parameter, and β is a K by 1 vector of regression coefficients” (Anselin and Bera 1998, p 246). In this study, spatial weights using first order queen contiguity were included so that the model would consider potential spillover into neighboring counties. First order queen contiguity refers to counties that touch the sides and corners of a county in question (O’Sullivan and Unwin 2010).
In order to test the differences in median household income, income growth rates, unemployment rates, and number of manufacturing firms between counties near FTZs and counties not near FTZs, counties were grouped into control and sample counties. Sample counties include counties with an FTZ and counties queen adjacent to counties with an FTZ. Control counties are the counties that remain (Figure 1). A one-tailed two-sample difference of the means t-test was used to determine if the variables in sample counties were significantly higher than variables in the control counties. A one-tailed two-sample t-test is used to determine whether the difference between two sample groups and is directional in nature. The following formula is used to calculate the two-sample difference of the means t-test:
t=(〖X̅〗_1-〖X̅〗_2)/σ_(〖X̅〗_1-〖X̅〗_2 ) (3)
where 〖X̅〗_1 represents the mean of the sample group, 〖X̅〗_2 represents the mean of the control group, and σ_(〖X̅〗_1-〖X̅〗_2 ) is the standard error of the difference of the means (McGrew and Monroe 2009).
Results
Results from the Getis-Ord Gi model set show significant clustering of high median household income around FTZs located in major metropolitan regions of Northern Kentucky, Northeastern Virginia, Central and Southern North Carolina, Central Tennessee, North-Central Georgia, and the East Coast of Florida. Significant clustering of low unemployment occurred around FTZs in Western Kentucky, Northern Kentucky, throughout Virginia, and Central Tennessee. No significant clustering of income growth rates occurred around FTZs except in Jacksonville, Florida. Significant clusters of manufacturing firms occurred around FTZs in Central North Carolina, North-Central South Carolina, North-Central Georgia, and peninsular Florida (Figure 2).
Results from the SRM set show significant results in three of the four variables included in this analysis (Table 1). Model 1 found a significant positive linear relationship between median household income and access to FTZs (p-value = 0.0008). Counties with FTZs and counties adjacent to those counties had significantly higher median household income. These results explain 59 percent of the variation within model 1. Results from model 3 (unemployment rates) show a significant negative relationship, provided a p-value of 0.0326, and explained 37 percent of the variation in the model. A significant positive linear relationship also exists between the number of manufacturing firms in a county and FTZs (model 4, p-value < 0.0001). Model 4 had the highest r-squared value, explaining 60 percent of the variation.
Table 1
Model Averaged Parameter Estimates of foreign-trade zones, median household income, income growth rates, and number of manufacturing firms in each county within the Southeastern United States using spatially weighted regression (SRM). Spatial lag method using first order queen contiguity weight matrices were included in all SRMs (Sources: Infogroup 2017, McGilvray 2017, U.S. Census Bureau 2017. Generated using GeoDa).
Table 2
Model Averaged Parameter Estimate for two-sample t-tests comparing control counties to sample counties. Sample counties have an FTZ or are adjacent to a county with an FTZ. Control counties are those that do not have direct or indirect access to an FTZ (Sources: Infogroup 2017, McGilvray 2017, U.S. Census Bureau 2017. Generated using Excel).
Model 3 had no significant result, therefore the null hypothesis failed to be rejected. The following is the list of resulting formulae from the significant models in the SRM set:
Median Household Income (2016): ŷ=10335.4+3548.03x (4)
Unemployment Rate (2016): ŷ=3.08-0.67x (5)
Manufacturing Firms (2016): ŷ=15.59+552.3x (6)
According to resulting formulae, as a county or neighboring county adds one FTZ, median household income increases by $3,548.03, unemployment rate declines by 0.67 percent, and the number of manufacturing firms increases by ~552. While these results show a significant relationship, they do not suggest causation. Due to the amount of variation explained in the three significant models presented here, it must be concluded that there are other variables that also impact median household income, unemployment rates, and number of manufacturing firms at the county level in the Southeastern United States.
Results from the two-sample t-tests reveal significant differences between three of the four variables included in this portion of the study. Median household incomes were significantly higher in sample counties than control counties. Unemployment rates were significantly lower in sample counties than control counties. Results from the t-tests also showed significantly higher number of manufacturing firms in sample counties than control counties (Table 2). No significant difference was present between income growth rates in sample and control counties. While these results do not provide a complete picture of FTZs in the Southeast, they show that is a relationship exists between the location of FTZs and economic development.
Analysis
The statistical analyses performed in this study provide valuable information on the spatial relationship between FTZs and economic development at the county level in the Southeastern United States. While the results from this analysis are limited to the borders of Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee, Virginia, and West Virginia, there is a broader context to be considered here. Foreign-trade zones exist across the contiguous United States, Alaska, Hawaii, and Puerto Rico. This study shows that there is a significant spatial relationship between the existence of FTZs and economic development at the county level in the Southeast, a relationship that should be explored in a larger spatial context.
The Getis-Ord Gi analysis revealed significant clustering of high median household income, low unemployment rates, and manufacturing firms in counties near FTZs in major urban centers across the Southeastern United States. Results from this analysis confirm earlier assertions by Graves and Kalafsky (2017) that the spatial distribution of income and other economic factors are unevenly distributed in the Southeastern United States. Similarly, Kalafsky and Graves (2018) argued that urban regions such as Atlanta are developing at a much faster rate that rural areas throughout the Southeast. Results from the Getis-Ord Gi analysis substantiate these claims and further demonstrate that Louisville, Nashville, Charlotte, Jacksonville, Raleigh/Durham, Northern Virginia, and Peninsular Florida are economically outperforming other regions in the Southeast.
Once the Getis-Ord Gi function provided validation for the existence of spatial autocorrelation of median household income, real unemployment rates, and manufacturing firms on or near counties with FTZs, spatial regression was used to determine the nature of any relationship that might exist between FTZs and these variables. Spatial regression results show that a significant spatial relationship exists between counties with FTZs and adjacent counties, median household income, unemployment rates, and number of manufacturing firms. These results imply that the addition of a foreign-trade zone in a county may have a positive economic impact. Results from the t-tests show that counties with access and proximity to FTZs had significantly higher income, lower unemployment rates, and more manufacturing firms than control counties. This analysis confirms results from the spatial regression models. Due to corresponding results in both model sets, it has been concluded that FTZs had the strongest relationship with median household income, unemployment rates, and number of manufacturing firms.
The spatial regression models and t-tests reveal that FTZs have a significant impact on regional development patterns, thereby dispelling assertions made by Kanellis (1995) that it would be impossible to evaluate the effectiveness of FTZs on meeting legislative intent. Spatial regression and t-test results show that spillover can occur across state boundaries, confirming conclusions from Bobonis and Shatz (2007). Spatial regression and t-test results in this study also found that FTZs have an impact on manufacturing firms, lending credence to claims by Ghosh et al. (2016) that FTZs affect the growth of non-manufacturing firms.
Results from this analysis show a generally positive and significant relationship between economic development and access to FTZs. However, it is cautioned that none of the tests performed in this study prove that FTZs cause economic growth. Because of a lack of historical spatial data, more calculations could not be performed. For this reason, FTZs cannot be viewed as a driver of economic development. Instead, FTZs can be viewed as one economic policy measure that can have an impact on development patterns in the Southeast.
Discussion and Conclusion
Foreign-trade zones are part of a larger attempt to reduce duties and taxes in certain geographic locations while continuing to tax other areas at a higher rate. Other examples of place-based economic policies used across the Southeast are science industrial parks (SIPs) and economic and technological development zones (ETDs). The Research Triangle in North Carolina, located between Durham, Raleigh, and Chapel Hill, is an example of how place-based policies often overlap (Easterling 2016). The Research Triangle is categorized as a SIP; however, some of the manufacturing and warehousing facilities are considered part of FTZ 93. Therefore, some of these industries are receiving benefits from two types of place-based economic policies. According to McGilvray (2017), As much as $250 million in untaxed merchandise flowed through FTZ 93 in 2016, including $25 million worth of exports. This FTZ employed between 2,001 and 2,500 people in 2016 and contained firms such as Merck Sharp & Dohme and Revlon Consumer Products. What about other small- and medium-sized firms located in smaller peripheral communities? These businesses are not receiving the same benefits.
Place-based policies such as FTZs positively impact industries within these zones, but spillover into adjacent regions is often inconsistent. Surrounding communities are not reaping the benefits (Ham et al. 2010). Scholars have argued that place-based economic policies are designed specifically to encourage agglomeration and spillover (Koo 2005). Policy experts promoting this strategy argue that it promotes national growth, thereby making the benefits worth the costs. This view fails in a very important way by assuming that national economic growth is worth the price of an increase in spatial inequity. According to Martin (2008), while agglomeration theory proposes that place-based economic policies promote development on a local scale in a more efficient way than statewide economic policy, little evidence has been provided to back such a claim.
In many ways, place-based policy has a negative impact on statewide economic welfare because it pulls capital away from poor rural communities by attracting it to wealthy urban regions. According to Lee and Rodríguez-Pose (2016), these policies are designed to cluster high-tech manufacturing industries in urban areas and promote economic development, but rarely help alleviate poverty. These policies have failed to trickle down to poor neighboring rural communities. Head et al. (1999) argue that “states which offered foreign trade zones, job-creation subsidies, and low taxes received significant increases in investment” (p 216). Firms agglomerate near similar industries and port infrastructure in order to reduce intermodal costs and take advantage of FTZ benefits. Place-based policies such as FTZs have a positive impact on economic development in major metropolitan areas and attract potential investment away from smaller communities without existing industry and infrastructure.
Due to the popularity of place-based economic policies designed to promote agglomeration economies, many communities across the Southeast have been forgotten. Rural communities across the southeast are seeing much higher poverty rates than urban areas where most FTZs are located (U.S. Census Bureau 2017). According to Graves and Kalafsky (2017), the Southeast is seeing some economic growth in urban areas but “outside of a few pockets of prosperity the Southern economy is evolving along a path that fails to enhance the welfare of much of its population” (p 124). This statement is confirmed by analyses performed in this study. While FTZs have provided income and jobs to certain locations, they have not alleviated poverty. Instead, these zones may have attracted investment away from poverty stricken rural areas.
Place-based economic policies can be harmful to the state as a whole and may obscure a much deeper problem. While FTZs have been effective at promoting agglomeration, there is no evidence that these zones have helped to improve the economic welfare of communities other than those in direct contact with a zone or subzone. This study reveals that there is a spatial link between FTZs and economic development, which may exacerbate existing spatial inequality across the Southeast. Because the analysis performed in this study does not prove causality, it is perfectly plausible that FTZs are attracted to urban centers of higher income. Even this possibility is intriguing. According to the Foreign-Trade Zone Board (2018), current efforts to provide “zone access as part of [local communities’] economic development efforts” are intended to promote “investment in the domestic economy” (p 6). If FTZs encourage local economic growth and agglomeration, investment is attracted to these locations and away from other areas. However, if FTZs are being granted in locations that are already wealthy, then the Foreign-Trade Zone Board is granting FTZs to boost the economy in wealthy metropolitan areas without considering the impact of spatially unequal tax policies on rural communities.
Policy Recommendations and Future Research
Place-based policies were designed to promote agglomeration and attract capital to certain regions. While many policymakers have used place-based economic policy to promote regional and national economic growth, evidence reveals that it negatively impacts national and regional welfare (Martin 2008). FTZs are just another example of place-based policies falling short of intended goals. Results from this paper lend credence to a general approach to economic development in the Southeastern United States. Current policy is leading to higher concentration of wealth in certain areas in the Southeast (Graves and Kalafsky 2017, Kalafsky and Graves 2018) and FTZs are integral in the distribution of wealth across the region. According to Clark and Doussard (2019), strategies such as FTZs and other industrial growth policies have led to uneven economic development across space.
It is my contention that the federal government should extend equal benefits to all manufacturing firms throughout the United States, thereby allowing all firms to benefit from reduced taxes. According to Kanellis (1995), smaller firms may not be aware of the benefits that their competitors are receiving. The extension of FTZ benefits to all manufacturing firms would place all United States based firms on the same taxation level. By providing an unfair competitive advantage to favorable corporations, FTZs have proven to be a tax shelter for a minority of firms and have negatively affected the distribution of wealth throughout the Southeastern United States. A general approach will allow small firms in poor communities to receive the same benefits that larger competitors currently receive.
While findings from this study reveal that FTZs have an impact on economic development in the Southeast, these findings do not prove causation. Future research should address the spatial impact of FTZs on economic development across the United States using spatial regression and geographically weighted regression. Akin to methods performed by Ghosh et al. (2016), future research should analyze short-term and long-term effects of FTZs on local development. These methods will help clarify questions that remain unanswered in this study.
Biographical Sketch
Jesse M. Lane is a Ph.D. Candidate in the Department of Geography, Environment, and Sustainability at the University of North Carolina at Greensboro in Greensboro, North Carolina, 27412. His research interests include economic geography, economic development policy, spatial inequality, political economy, and geopolitics. Email address: jmlane2@uncg.edu.
References
15 C.F.R. § 400
19 U.S.C. § 81a-81u
Anselin, L., and Bera, A.K. 1998. Spatial dependence in linear regression models with an introduction of spatial econometrics. In Handbook of applied economic statistics, edited by Ullah, A., and D.E.A. Giles, 237–289. Marcel Dekker.
ArcGIS Online 2018. ESRI Demographics. ArcGIS. Available at https://doc.arcgis.com/en/esri-demographics/data/business.htm.
Bobonis, G.J., and Shatz, H.J. 2007. Agglomeration, adjustment, and state policies in the location of foreign direct investment in the United States. The Review of Economics and Statistics 89 (1): 30–43.
Bolle, M.J. 1999. U.S. foreign-trade zones: Current issues. Washington, DC: Congressional Research Service, Library of Congress.
Bolle, M.J., and Williams, B., R. 2012. U.S. foreign-trade zones: Background and issues for Congress. Washington, DC: Congressional Research Service, Library of Congress.
Clark, J., and Doussard, M. 2019. Devolution, disinvestment and uneven development: US industrial policy and evolution of the national network for manufacturing innovation. Journal of Regions, Economy and Society 12 (2): 251–270. https://doi.org/10.1093/cjres/rsz009.
Cornwell, T.B. 1989. Foreign‐Trade Zones in the United States: A Longitudinal Management Perspective. International Marketing Review 6 (6): 42–52. https://doi.org/10.1108/eum0000000001526.
Easterling, K. 2016. Zone. In Extrastatecraft: The power of infrastructure space, 25–69. London, UK: Verso.
Ferguson, W. 1989. How foreign trade zones can reduce costs. Management Decision 27 (5): 12–16. https://doi.org/10.1108/00251748910132557.
Foreign-Trade Zone Board 2018. US foreign regulation (FTZB): Foreign-trade zones in the U.S. 2018th ed. Washington, DC: The Law Library.
Foreign-Trade Zone Board 2019. What is a foreign-trade zone? U.S. Foreign-Trade Zones Board, 2019. Available at https://enforcement.trade.gov/ftzpage/info/zone.html.
Ghosh, S., Reynolds, C.L., and Rohlin, S.M. n.d. The spillover effects of United States foreign trade zones. Applied Economics 48 (43): 4112–4130. https://doi.org/10.1080/00036846.2016.1153787.
Graves, W., and Kalafsky, R., V. 2017. Evolving growth prospects? Assessing the state of the South in the global economy. Southeastern Geographer 57 (2): 112–130.
Ham, J.C., Swenson, C., İmrohoroğlu, A., and Song, H. 2010. Government programs can improve local labor markets: Evidence from state enterprise zones, federal empowerment zones and federal enterprise communities. Madison, WI: Institute for Research on Poverty. pp.1–45.
Hazari, B.R., and Sgro, P.M. 1996. Free trade zones, tariffs and the real exchange rate. Open economies Review 7 (3): 199–217.
Head, C.K., Ries, J.C., and Swenson, D.L. 1999. Attracting foreign manufacturing: Investment promotion and agglomeration. Regional Science and Urban Economics 29 (2): 197–218.
Infogroup 2017. Marketing & analytical research. ESRI Business Analyst.
International Trade Administration 2019. Metropolitan Data: Exports. International Trade Administration, 2019. Available at http://tse.export.gov/metro/SelectReports.aspx?DATA=Metro.
James, R. 2010. Where is the South? Using beta convergence to define a fuzzy region. Southeastern Geographer 50 (3): 346–365. https://doi.org/10.1353/sgo.2010.0002.
Kalafsky, R., V., and Graves, W. 2018. Exports and growth: Learning from the case of southern U.S. metropolitan areas. The Professional Geographer 70 (3): 383–394.
Kanellis, W.G. 1995. Reining in the Foreign Trade Zones Board: Making foreign trade zone decisions reflect the legislative intent of the Foreign Trade Zones Act of 1934. Northwestern Journal of International Lay & Business 15: 606–635.
Koo, J. 2005. Technology spillovers, agglomeration, and regional economic development. Journal of Planning Literature 20 (2): 99–115.
Lee, N., and Rodríguez-Pose, A. 2016. Is there trickle-down from tech? Poverty, employment, and the high-technology multiplier in U.S. cities. Annals of the American Association of Geographers 106 (5): 1114–1134.
Martin, R. 2008. National growth versus spatial equality? A cautionary note on the new ‘trade‐off’ thinking in regional policy discourse. Regional Science Policy & Practice 1 (1): 3–13.
Mathur, L.K., and Ajami, R. 1995. The role of foreign-trade zones in U.S. exporting. Multinational Business Review 3 (1): 18–26.
Mathur, L.K., and Mathur, I. 1997. The effectiveness of the foreign-trade zone as an export promotion program: Policy issues and alternatives. Journal of Macromarketing 17 (2): 20–31. https://doi.org/10.1177/027614679701700203.
McCalla, R. 1990. The geographical spread of free zones associated with ports. Geoforum 21 (1): 121–134.
McDaniel, W.R., and Kossack, E.W. 1983. The financial benefits to users of foreign-trade zones. Columbia Journal of World Business 18 (3): 33–41.
McGilvray, A. 2017. 78th annual report of the Foreign-Trade Zones Board to the Congress of the United States. Washington, DC: Foreign-Trade Zones Board, U.S. Customs and Border Protection. pp.1–157.
McGrew, J.C., and Monroe, C.B. 2009. An introduction to statistical problem solving in geography. 2nd ed. Long Grove, IL: Waveland Press, Inc.
O’Sullivan, D., and Unwin, D.J. 2010. Geographic information analysis. 2nd ed. Hoboken, NJ: Jon Wiley & Sons.
Orenstein, D. 2011. Foreign-trade zones and the cultural logic of frictionless production. Radical History Review (109): 36–61. https://doi.org/10.1215/01636545-2010-014.
Robles, F., and Hozier, Jr., G.C. 1986. Understanding foreign trade zones. International Marketing Review 3 (2): 44–54. https://doi.org/10.1108/eb008305.
Seyoum, B. 2017. US foreign trade zones and import intensity examining determinants of import intensity in US foreign trade zones. European Business Review 29 (1): 103–122. https://doi.org/10.1108/EBR-06-2016-0072.
Seyoum, B., and Ramirez, J. 2012. Foreign trade zones in the United States: A study with special emphasis on the proposal for trade agreement parity. Journal of Economic Studies 39 (1): 13–30. https://doi.org/10.1108/01443581211192080.
Shambaugh, J., and Nunn, R. 2018. Place-based policies for shared economic growth. Washington, DC: The Brookings Institution.
S. Rep No. 905, 73rd Cong., 2nd Sess. 2 (1934).
Tansuhaj, P.S., and Jackson, G.J. 1989. Foreign trade zones: A comparative analysis of users and non-users. Journal of Business Logic 10 (1): 15–30.
Thoman, R. 1952. Foreign-trade zones of the United States. American Geographical Society 42 (4): 631–645.
Tobler, W.R. 1970. A computer movie simulating urban growth in the Detroit region. Economic Geography 46: 234–240.
U.S. Census Bureau 2017. 2016 Tiger geodatabases. Tiger geodatabases, 2017. Available at https://www.census.gov/geo/maps-data/data/tiger-data.html.
Kommentare