Investing in Low-Wage Jobs Is the Wrong Way to Reduce Migration

Amid today’s global immigration crises, wealthy governments are seeking ways to keep out newcomers. Border security measures (typically more sophisticated than U.S. President Donald Trump’s wall) are often first on the list. But other ideas are gaining adherents as well. In the European Union and elsewhere, policymakers are betting that investment and trade will reduce migration from less developed countries. Put money into export-oriented jobs where the would-be workers are, the theory goes, and fewer will leave in search of opportunities elsewhere.

Despite its surface appeal, the proposition is rooted more in political expedience than in evidence. The U.S. and Mexican governments touted the North American Free Trade Agreement, or NAFTA, in part as a way to decrease Mexican migration to the United States. Instead, the number of Mexicans heading north to work soared in the 15 years after the agreement came into force in 1994. Although neoclassical economics once advanced the view that moving production to countries with low wages would keep workers there from migrating—the idea that capital and labor are substitutes—today there is no consensus in economic theory or empirics that supports the idea that increasing foreign investment and trade will decrease emigration, at least in the short term. (There is, however, broad agreement that migration increases trade—a different issue entirely.) In fact, the jobs that result from trade are likely to fund increased emigration for a decade or more, as opposed to stemming it.

Nonetheless, two new international agreements, both endorsed by the U.N. General Assembly last year, affirm the trade and investment approach. The jobs that result from trade are likely to fund increased emigration for a decade or more, as opposed to stemming it.The Global Compact for Migration recommends increased foreign direct investment and new trade preferences for less developed nations as a way to limit migration out of those countries. The Global Compact on Refugees encourages preferential trade agreements in order to open jobs to refugees in countries where they first seek shelter, with a similar goal.

More concretely, in the refugee context, the European Union, the World Bank, and other institutions have invested billions of dollars in turning the idea into reality. In 2016, these actors implemented the proposal by signing a refugee compact with the Jordanian government. Its principal goal was to employ Syrian refugees in Jordan, primarily in garment factories in export processing zones, in order to keep them from departing for European shores. That same year, Britain, the EU, and the World Bank reached an agreement with the Ethiopian government on a smaller jobs compact for refugees in that country.

The situation in both Jordan and Ethiopia today raises serious concerns about the “trade as migration policy” theory.

The Jordan compact promised 200,000 formal jobs for Syrians, of which 150,000 would be in designated industrial export zones. As an incentive, the EU reduced tariffs on goods produced at factories in Jordan with a minimum number of refugee employees. Despite hundreds of millions of dollars and countless hours invested in achieving these goals by the Jordanian government, other nations, global lenders, and local and international NGOs, only a handful of Syrians are employed in the zones today. Instead, the 50,000 or so Syrians who hold valid work permits largely remain in agricultural and construction jobs in the informal economy, where they already labored before they were legally allowed to work.

The U.N. Development Program, the World Bank, and other key players argue that it will just take more time before Jordanian export manufacturing becomes a core source of refugee jobs. They characterize the obstacles in Jordan as technical ones—matching workers with employers, reducing the distances refugees must travel to work, covering training and child care needs. All of these problems, they maintain, can and will be addressed in the coming years.

This position ignores a critical issue: Garment work pays too little and demands too much to be a viable option for most Syrian refugees in Jordan. Most workers in the Jordanian garment industry are female labor migrants from India, Sri Lanka, and Bangladesh. They come alone, live on-site, and remit their earnings to countries where they go much further than in Jordan. The compact was based on the idea that if these migrants worked well as stitchers in Jordan, then Syrians—especially women—would too. Unlike migrants, however, most Syrians have spouses and children with them and live in Jordanian cities. What they earn must cover high Jordanian prices for housing, food, and other needs. Syrian women also need enough time left in the day to take care of the needs and expectations of their family members. Like Jordanians themselves, they cannot afford to work in garment manufacturing.

Poverty-level wages and long hours in the garment industry do not have a technical fix—they are structural features of the global apparel business model. Wage levels are a primary driver of brand decisions about where to manufacture garments. Long and unpredictable hours, likewise, are a response to the fast-fashion nature of the sector, where data from global inventories demands new styles at a breakneck pace. Were apparel firms in Jordan to raise wages and limit hours of work, they would no longer be globally competitive.

Perhaps Jordan is the wrong proving ground for the theory that export manufacturing can deter migration. As a middle-income country with no relevant natural resources, it has never been a natural site for low-wage garment manufacturing. Jordan has a garment industry only because of a deal it struck with the United States in the context of negotiations over the 1994 Jordan-Israel peace treaty.

Ethiopia would seem to be a better test case: It is a less-developed country with significant cotton and leather industries, a comparatively low cost of living, and where the government has invested heavily in the garment export industry as a driver for development.

The Ethiopia jobs compact initially promised the creation of 100,000 new jobs in Ethiopian industrial parks, of which 30,000 would be reserved for refugees. Implementation was delayed until the Ethiopian Parliament passed a new Refugees Proclamation, which only happened in January 2019. But with the Jordanian experience in mind, the World Bank recently surveyed Eritrean, Somali, Sudanese, and South Sudanese refugees in Ethiopia about their skills and job interests. The survey found that refugees were largely uninterested in manufacturing opportunities, which in the Ethiopian government’s flagship Hawassa Industrial Park pay $1.25 to $1.60 per day (compared with the World Bank’s international poverty line of $1.90 per day). Interestingly, two separate new studies by University of Chicago economists and a Cambridge University industrial policy specialist find that many Ethiopians themselves do not see garment manufacturing as desirable employment, using it instead as a gap filler between more lucrative opportunities in the informal sector.

From the employers’ perspective, too, refugees in Ethiopia are a poor fit for the job. Garment employers at Hawassa seek to hire women between the ages of 18 and 25, ideally unmarried and childless, with 8th- to 10th-grade educations. There is very little overlap between those requirements and the demographic characteristics of Ethiopia’s 1 million refugees. These realities raise serious concerns about garment manufacturing for export as a way to support refugees in Ethiopia. With these results in mind, the World Bank and the Ethiopian government have reconfigured the agreement to refer to 30,000 formal economic opportunities for refugees, a figure no longer tied to export manufacturing.

To be clear, it is cause for celebration that some wealthier countries have finally begun to take on a small part of the financial burden borne by the poorer nations that host the vast majority of the world’s refugees. In addition, all players now also understand that programs to create economic benefits for refugees must help host nation citizens as well or more. This, too, is good.

Furthermore, it’s not hard to understand why factories in industrial parks have emerged as the favored vehicle for refugee jobs compacts, given the interests of the parties engaged in compact negotiations. The plan is ideologically consistent with the free market and trade-as-development commitments of donor governments and the World Bank. It offers European politicians a way to suggest that migration can be limited by humanitarian-sounding means. It promises host governments significant funding along with the hope that refugees can be employed in enclaves, limiting competition with natives and thus curtailing political opposition.

But the lowest-wage export manufacturing jobs are the wrong place to invest all of this faith and money. Whether better opportunities for natives and refugees will come from other tradable sectors remains to be seen. If they emerge, they are still likely to spur migration for at least a decade. Over the long term, developed countries hoping for less immigration must support decent work in migrant-origin countries, not a seat at a sewing table at any cost. Unless refugees and would-be migrants can build lives with dignity where they are, they will continue to seek ways to move on.