Technical Network on Poverty Analysis (THINK-PA)

Webinar challenges conventional wisdom on farm productivity and land reform


On 21 September 2023, the FAO Technical Network on Poverty Analysis (Think-PA) held the webinar “Are there too many farms in the world? Labour market transaction costs, machine capacities, and optimal farm size”, where esteemed economists Andrew Foster and Mark Rosenzweig shared their research findings that question established beliefs regarding farm productivity, land reform, and the future of agriculture. The virtual event brought together a diverse audience keen to understand and explore the implications of their research.

The presentation began with the illustration that the smallest farms in low-income countries appear to outperform their slightly larger counterparts in terms of productivity measured by per acre yield (inverse relation between farm or plot size and productivity). However, such stylized fact is not applicable in developed countries, where productivity increases with farm scale. Foster and Rosenzweig argue that it is essential to adopt a global perspective, not only analysing the inverse relationship — the initial productivity decline as farm size increases from its lowest levels — but also examining the subsequent productivity gains as scale increases beyond a certain threshold.

The webinar delved into disentangling the reasons on why small farms appear to be more efficient in terms of their profitability/net income to slightly bigger farms, while medium-sized farms increase its efficiency with size. Using data from India, Foster and Rosenzweig explain that this is the case due to the cost-effectiveness of hiring additional labour, with relatively higher transactional cost for small farms. Furthermore, larger farms benefit from the efficient use of machinery with higher capacity and lower costs at larger scales.

One of the most striking takeaways from the webinar was the estimation that increasing farm size through consolidation in India, for example, could dramatically boost agricultural output. The authors calculated that the minimum farm size in India that would maximize land returns – using locally-available machinery – is 24.5 acres, which is over seven times the current average farm size in the country. Under this hypothetical scenario, the number of farms in India would decrease by 82 percent, total agricultural output would increase by 50 percent, and total output per worker would surge by 68 percent. The decrease in the number of farms would result in a 16 percent reduction in the labour force, leading to a surplus of workers seeking alternative sources of income outside of agriculture.

Considering these findings, Foster and Rosenzweig encourage a re-evaluation of agricultural strategies. They suggest that the promotion and research on the role of agricultural cooperatives – as a means of organizing and standardizing small farmers' production – is central for enabling them to harness machinery and technology, similar to large-scale farms.

The webinar has sparked a vibrant discussion among the participants, setting the stage for potential rethinking and reform in the agriculture sector to address productivity and income disparities among farmers. Specially because the apparent higher productivity of small farms has long motivated economists and policymakers working in developing countries to advocate for land reforms aimed at redistributing land from large farms to support smaller, more fragmented operations.