By Sherry Miao
For more than a century, the dominant model of economic development followed a predictable sequence: countries moved workers out of agriculture and into manufacturing, and industrial productivity drove rising living standards. This framework shaped decades of economic theory and policy advice. Yet in many developing countries today, particularly across South Asia and Sub-Saharan Africa, this pattern is no longer unfolding in the same way. Workers are leaving farms, but factories are not absorbing them. Instead, many are entering consumer-facing services such as retail, hospitality, transportation, and personal care. Fabrizio Zilibotti, Tuntex Professor of International and Development Economics at Yale University, and Michael Peters, Associate Professor of Economics at Yale University, presented new research on service-led growth in East and West Africa, while discussant Doug Gollin, Jason P. and Chloe Epstein Professor of Economics at Tufts University, reflected on what these findings mean for development economics and structural transformation more broadly.
Consumer services are driving growth in Africa, and may be more productive than previously assumed
Using data from rapidly growing economies in East and West Africa, Zilibotti and Peters found that most workers leaving agriculture moved into services rather than manufacturing. Approximately 15 percentage points of employment shifted into services, compared to only 3 percentage points into manufacturing. The services absorbing labor were largely local and consumer-facing, including retail, restaurants, hospitality, and transportation. A key question in the research was whether this shift reflected real productivity growth or simply higher consumer demand as incomes rose elsewhere in the economy. Using a structural model combining household-level and macroeconomic data, the researchers found evidence that productivity growth in consumer services was genuinely strong, in some cases even higher than in tradable sectors. This finding challenges long-standing assumptions that manufacturing is the primary source of economic dynamism and productivity growth.
The benefits of service-led growth are meaningful, but unevenly distributed
While the research highlights the economic potential of services, the gains are not evenly shared. The researchers found that urban households benefited significantly more from service-sector productivity growth than rural households, which relied more heavily on gains from tradable goods sectors. Wealthier households also captured a disproportionate share of the welfare improvements compared to poorer households. These findings suggest that service-led growth does not automatically produce inclusive development outcomes. The researchers identified several potential drivers of future productivity gains in services, including urban agglomeration effects, reductions in labor market frictions, and technologies such as mobile banking that improve efficiency even in informal markets.
Development economics may need to rethink its focus on manufacturing
Discussant Doug Gollin argued that the paper’s broader significance lies in how it challenges traditional assumptions about structural transformation. Development economics has long treated manufacturing as uniquely capable of generating productivity, scale, and prosperity, while services were often viewed as residual or low-productivity sectors. Gollin suggested that countries may not need manufacturing specifically, but rather some combination of productivity growth, employment generation, and market integration across sectors. He also emphasized the importance of connecting rural and urban economies and expanding participation in exchange-based markets. Overall, the discussion suggested that service-led growth should be taken more seriously as a viable pathway to economic development.
