This article provides a comparison of levels and growth of agricultural total factor productivity between Australia, Canada, and the United States for the 1961-2006 period. A production account for agriculture that is consistent across the three countries is constructed to estimate output, input and total factor productivity, and a dynamic panel regression is used to link the productivity estimates to potential determinants. We show that investment in public research and development and

more »

... ucture plays an important role in explaining differences in productivity levels between countries. The findings provide useful insights into how public policy could be used to sustain agricultural productivity growth. GLOBAL AGRICULTURAL OUTPUT HAS more than tripled over the past half century, driven by new technologies and increased input use. This output growth has helped to satisfy increasing demand for food and fibre as population and income per capita have increased, and has thereby stabilized global food prices. Agricultural productivity growth has contributed significantly to these gains (Fuglie and Wang, 2012). One of the most important drivers of agricultural productivity growth is technological progress. This progress has followed two distinct paths in developed countries, depending on the initial endowment of resources. Those possessing relatively abundant capital and land, for example Australia, Canada and the United States, have been lead adopters of capital-intensive technologies such as reduced-till cropping, yield mapping and mechanised mustering, and thus have achieved high levels of output per worker. In contrast, land-scarce, labour-rich developed economies such as Japan, South Korea and Taiwan have adopted labour-intensive technologies such as green-housing and 1 Yu Sheng is a senior economist of Agricultural and Resource Economics and Sciences (ABARES) at the Australian Department of Agriculture and Water Resources; Eldon V.