Measuring the direct impact of special economic zones at the firm level – and especially the impact on the domestic enterprises - presents a challenge, due in part to a lack of data, and in part due to the difficulty in finding appropriate enterprises that can serve as “controls” with which to make valid comparisons. Indirect measures can suggest the impact of zones on exports, investment, employment and spillovers on firms outside zones, yet even these assessments are scarce. So far, the major source of information on the zones’ global impact is International Labour Organization’s EPZ study (Boyenge, 2007). As of 2007, it estimates, EPZs created about 68 million direct jobs and US$851 billion worth of exports, accounting for about 41 percent of global exports (FIAS, 2008).
However, the global picture is quite heterogeneous. Based on disaggregated studies, the impact of SEZs in driving economic and private-sector development seems to be quite mixed across countries and regions.
Successful SEZs are able to attract large numbers of multinational companies and domestic firms, and to make great contributions to business investment, employment generation and economic development. In China, estimates show that the national-level SEZs (including various industrial parks) account for more than 30 million jobs and about 22 percent of national GDP, 46 percent of foreign direct investment, and 60 percent of exports (Zeng, 2010).
An analysis of panel data of 270 cities at the prefecture level (a jurisdiction between county and province in China) over 23 years finds that the introduction of a major zone in a city in the post-reform years led to an average increase in the GDP level of 12 percent, with the effect depending on the type of zone. The long-term (cumulative) effect of an SEZ could be an increase of about 20 percent in the GDP level (Alder, S., et al., 2013). Another analysis (Wang 2013) of 321 prefecture-level cities between 1978 and 2008 shows: a) on average, an SEZ program increases the level of per-capita FDI by 21.7 percent, and the growth rate of FDI by 6.9 percentage points; b) the SEZ program generates significant agglomeration economies; it increases the technological progress of the earlier treated municipalities by 1.6 percentage points compared to the later ones; and c) the average wage of workers in the treatment group increases 8 percent more than in the control group. (Over the evaluation period, the cost of living rises by 5 percent.)
However, SEZs in China have experienced mixed results with regard to export diversification. For instance, Schminke and Van Biesebroeck (2013) compare ETDZs in China with Science and Technology Industrial Parks (STIPs). They find significant differences between the two types. Firms locating in an ETDZ achieve much higher export values, driven by higher volumes of trade and numbers of destinations. Firms locating in a STIP perform best on quality dimensions, fetch higher export prices, and have more success exporting to high-income countries (Schminke and Van Biesebroeck, 2013).
Elsewhere in Asia, Johansson and Nilsson (1997) find that SEZs have positive impacts on exports in Malaysia, Mauritius, and Sri Lanka. Their findings indicate that SEZs are more likely to have a positive impact on exports when countries adopt outward-looking, export-oriented policies at the national level to promote their SEZs. In particular, Johansson and Nilsson (1997) highlight the case of Malaysia’s “catalyst effect” on potential domestic exporters by filling an “ideas gap” in the market. In essence, by the early 1990s in Malaysia, SEZs attracted the right mix of foreign investors who brought along knowledge on how to master production, marketing and distribution of export goods. This knowledge translated into spillover effects on the local market, and stimulated domestic firms to enter the export market and to increase their production of export goods, the authors contend. The Philippines has experienced tremendous economic growth through its eco-zones, which focus on agro-industry, tourism, recreation, commerce and financial services (the environmental aspect of zones will be addressed in section IV). Eco-zones’ share of national merchandise exports increased from 22 percent in 1995 to 76 percent in 2003, and eco-zones’ share of national foreign direct investment increased from 30 percent in 1997 to over 81 percent in 2000 (FIAS, 2008).
Evidence from many parts of the world suggests the potential of the SEZ phenomenon to spur growth and employment. Examples include: The more than 7,000 firms at the Jebel Ali Free Zone in Dubai currently employ 170,000 people, or 13 percent of Dubai’s workforce. By 2012, the Aqaba Special Economic Zone in Jordan had attracted $18 billion in investment and generated 10,000 jobs. Eight zones in Bangladesh attracted 412 firms that have made investments totaling $2.6 billion, and have employed 350,000 people (IFC, 2016). In the Dominican Republic, employment in industrial-free zones rose from 500 in 1970 to almost 200,000 in 2007, and in Costa Rica, the EPZ share of manufactured exports jumped from less than 10 percent in 1990 to 55 percent in 2003, with export items changing from mainly apparel and textile products to modular circuits and other electronic components (FIAS, 2008). In Madagascar, a $165 million World Bank project contributed to a tenfold increase in the stock of formal enterprises and a near seven-fold increase in the number of formal jobs in the growth poles of Fort Dauphin and Nosy Be, despite a complex political environment. The project successfully integrated a number of reforms to improve the business climate and job creation around these growth poles (World Bank Group, 2016).
However, the impacts of SEZs are not uniform, as analyses of various outcomes show.
Export impact and diversification
In some countries, SEZs have not positively affected exports. Johansson and Nilsson (1997) assert that countries that fail to eliminate trade restrictions, and fail to adopt export-oriented strategies are less likely to experience positive impacts on exports. They highlight examples from Mexico and the Dominican Republic. For example, the Dominican Republic developed a rationale for SEZs and created what many considered to be successful SEZs with catalyst effects, increased employment and high levels of productivity (Rhee et al., 1990); nevertheless, SEZs in the Dominican Republic did not have a significantly positive impact on exports. The country continued to practice import substitution policies and maintained a series of trade barriers, which Rhee et al. argue stunted the impact of SEZs on exports. SEZs in the Dominican Republic today continue to be largely isolated from the rest of the economy (Carneiro et al., 2015).
In South Asia, Aggarwal (2005) and Aggarwal et al. (2008) assess the impact of SEZs on export diversification but find SEZs have mixed results. For instance, in 2008, after a 40-year-long record with SEZs, India’s SEZ exports represented only 5 percent of overall exports; by contrast, in a short span of time, SEZ exports accounted for nearly one-fifth and one-third of exports in Bangladesh and Sri Lanka, respectively (Aggarwal, et al., 2008). At the time of these findings, India was undertaking a major expansion of its SEZ policy; nevertheless, recent studies, including a report by the Comptroller and Auditor General’s office in India, continue to highlight the mixed success of India’s SEZ policy (CAG, 2014).
The roles SEZs have played in export diversification have varied by country, across sectors and products, as Aggarwal et al. (2008) also highlight. Substantial exports from SEZs have been observed in some sectors that were already outward-oriented before SEZs appeared in the market – thus making export performance difficult to attribute directly to the presence of SEZs; the information technology in Southern India provides a case in point. However, in other instances, such as the garment industry in Bangladesh, more directly observable linkages emerge between the creation of SEZs and increases in export productivity.
Industrial upgrading and technology transfer
Assessments about the role of SEZs in industrial upgrading and technology transfer are mixed. Certain assessments have suggested that skill levels in zone workforces have not significantly increased over time. The share of skilled labor in the maquila workforce, for example, increased only slightly from 6.6 percent to 7.2 percent in 1988-1998 (Sadni-Jallab and Blanco de Armas, 2002). Other analyses have suggested the opposite. Substantial evidence shows that SEZs have played an important, catalytic role in the industrial upgrading and technology transfer in the newly industrialized East Asian economies, especially in South Korea, Taiwan (China), Malaysia and the Philippines, where significant industrial upgrading has occurred in the electronics sector located mainly within industrial zones (Lall, 2000). The Philippine Economic Zone Authority has documented substantial rise in skills levels in the Philippine eco-zones where major activities have shifted from production to design and R&D (FIAS, 2008).
Firm-level performance
Little is known about the relationship between SEZs and firm-level performance. Few studies have attempted to explore the relationship closely. Aggarwal (2005) asks how cost-reduction measures found in zone-related policies influence firm performance. Focusing on export diversification, and exclusively on EPZs, Aggarwal (2005) conducts firm-level surveys in SEZs in Bangladesh, India, and Sri Lanka. Her primary-survey analysis is based on three types of cost-reduction measures that EPZs provide to firms: infrastructure, location, and regulatory/governance measures. Given that she has no counterfactual to assess firms outside of zones, she is unable to successfully model how belonging to an SEZ impacts firm performance. Nevertheless, in a subsequent paper, Aggarwal et al. (2008), using the same data, the authors note: “Although data limitations make it challenging to evaluate the institutional characteristics and trade performance across zones and countries, the comparative assessment attempted in this study yields several findings. In particular, zone location, access to and quality of infrastructure, and the governance structure of the zones seem to influence the performance of SEZs” (p. 234).
In an older analysis of firm-level performance in zones, Rhee et al. (1990) survey foreign and domestic firms as well as joint ventures operating within EPZs in the Dominican Republic. Their primary focus is on the relationship between foreign and domestic firms, and the impact of EPZs on domestic productivity and labor market outcomes. The authors find that foreign firms have a positive impact on local export supply in the Dominican Republic. Foreign firms acted as catalysts to initiate exports of a wide range of products that were also adopted by domestic firms. The authors also highlight knowledge transfer and vocational skills that foreign firms were able to supply to domestic firms within EPZs. Like the Aggarwal research, the primary constraint of this paper stems from its exclusive focus on in-zone firms, without comparison with the performance of non-zone firms.
Labour-market outcomes
A considerable amount of research has looked at the relationship between SEZs and various labor-market outcomes, ranging from the influence of SEZs on job creation to working environment conditions and unionization. Cirera and Lakshman (2014) draw a sample of 59 such studies to explore the relationship between SEZs and employment, wages, and labor conditions. They focus specifically on freedom of association, health and safety, and working hours in developing countries. Their findings highlight very disparate and mixed outcomes associated with this research. In terms of unionization, Cling et al. (2005) and Glick and Roubaud (2006) find more unionization in EPZs than in the private sector outside in Madagascar, although the authors compare different sectors of activity. Both ILO (1988) and Sen and Dasgupta (2008) find very similar unionization rates between firms inside and outside the EPZs in some Asian countries in the early 1980s and in India in 2004–2006. Ver Beek (2001) finds less unionization inside the EPZs in Honduras in 1998, while Zohir (2001) suggests that unionization is banned inside Bangladesh’s EPZs, but not outside. Therefore, excluding those cases where unionization is legally prohibited in EPZs, it’s hard to conclude whether restrictions on union rights can be attributed to EPZs or to general failures in domestic labor institutions.
While significant health and safety issues in EPZs have been documented in the literature – ranging from anecdotal evidence to more robust studies – the few studies reviewed comparing workers inside and outside the EPZs show conflicting results. Liberato and Fennell (2007), using a survey in the Dominican Republic, find that working in EPZs negatively affects health and increases the likelihood that women will be hospitalized, but meanwhile, they find that EPZs also improve health outcomes in the household through better health benefits or the use of preventative medicine. Two further studies (Guendelman and Silberg, 1993; Hovell et al., 1988), comparing women working in EPZs and in other activities in Tijuana, Mexico, find no morbidity differences between women working in the maquiladora EPZ and other sectors. Furthermore, the evidence reviewed does not provide any clue on whether the cases of higher incidence of illness in the zones are explained by the concentration of sectors with more illness prevalence (Cirera and Lakshman, 2014).
