July 19, 2012 | Contribution by
GLADYS LOPEZ-ACEVEDO By Gladys Lopez-Acevedo, Senior Economist, Poverty Reduction and Equity Department, the World Bank.
At the end of 2004, the Multifibre Arrangement (MFA) was finally phased out, and with it, 30 years of quota restrictions by industrial countries on textile and apparel imports from developing countries. As expected, this phase-out led to a substantial reallocation of production and employment worldwide. In 2008, 70 percent of global apparel exports came from developing countries, making the sector a critical engine of growth. It often provides entry into formal employment for the unskilled, the poor, and women, owing to the relatively low technology and the labor-intensive nature of the work.
In the post-MFA world, what exactly has happened to the apparel value chain—all the activities that cover a good’s production from conception to end-use. It’s a topic that few studies have examined (Gereffi and Frederick 2010; Frederick and Gereffi 2011; Staritz 2011). For that reason, the World Bank recently undertook a study (see “The Promise and Peril of Post-MFA Apparel Production”) to explore the effects of the MFA removal on employment, wages, inter-industry changes, working conditions, and related government policies. Case studies were conducted in nine countries: Bangladesh, Cambodia, Honduras, India, Mexico, Morocco, Pakistan, Sri Lanka, and Vietnam.
The results show that post-MFA, some countries gained while others lost. But contrary to expectations, it wasn’t just China that benefitted. What mattered was how countries prepared for the phase-out and if they continued innovating and upgrading in the value chain – such as shifting to higher-value goods or “modernizing” production techniques.
First, export gains in the apparel industry were not simply owing to industry relocation from higher-wage countries to lower-wage ones – domestic policies mattered too. Global buyers try to maximize profits by minimizing costs, so it makes sense that production shifts in a labor-intensive industry might have been driven primarily by wage differences across countries. However, the study shows that only about a third of the variation in cross-country changes in apparel exports was driven by wage differences. The rest came from domestic policies targeting the apparel sector, ownership type, and functional upgrading of the industry.
Second, changes in exports usually, but not always, matched changes in wages and employment. This finding shows that using exports as a metric of “success” for helping the poor isn’t sufficient. While rising exports correlated with rising wages and apparel jobs in the large Asian countries, rising exports coincided with fewer apparel jobs in Sri Lanka. In Mexico and Honduras, both exports and apparel jobs fell, but Mexico – unlike Honduras – was able to absorb these workers into other industries. If the country is moving into a more advanced manufacturing, moving out of apparel may be a sign of economic development.
Third, the changes in the “apparel premium” were predictable: rising (in most cases) in the winners and falling in the losers. This matters because the apparel premium over other industries is the critical component of wages that helps lift workers from poverty (De Hoyos et al., 2008, and Robertson et al, 2009). So not only are job opportunities generally lost when exports contract and gained when exports expand, but one of the key features that made these “good” jobs—the wage premiums—is also lost (or gained), representing a double impact for workers in developing countries.
Fourth, the countries with larger increases in apparel exports – such as India, Bangladesh, Vietnam, and Pakistan – promoted apparel sector upgrading, while those that didn’t ended up with smaller increases or even falling exports (like Honduras). Although this may imply that upgrading facilitates competitiveness, upgrading doesn’t always boost jobs or wages. In fact, the opposite might be true (as in Sri Lanka, where upgrading coincided with more exports but fewer jobs, particularly for women). Moreover, some countries (such as Cambodia and Bangladesh) expanded employment while staying at lower levels of the value chain. Because upgrading often requires workers with more advanced skills, less labor demand affects mostly low-skilled poor; hence the need for targeted workforce development.
Fifth, better labor conditions and worker treatment made a difference. In an industry driven by buyers sensitive to importing countries’ reputation, a concern for labor conditions and worker treatment may be more than a labor rights issue. Better working conditions may also give a competitive advantage, as occurred in Cambodia.
A major policy concern is that the opportunities for pro-poor apparel production may be short-lived and are highly sensitive to changes in the global economic environment. Thus, for economies that are gaining, it is important to know who is benefitting and what conditions are necessary for the poor to benefit from the policy change. If the poor aren’t benefitting, policy makers need to ask why and to determine what, if anything, might be done to help the poor capture some of the gains.