Policy, Technology, and Efficiency of Brazilian Agriculture
by Nicholas Rada
and Constanza Valdes
Economic Research Report No. (ERR-137) 43 pp, July 2012
What Is the Issue?
Between 1985 and 2006, Brazilian agricultural production grew by
77 percent and Brazil emerged as a major international agricultural
exporter. The authors postulate that Brazil's agricultural
development was boosted by sustained public investments in science
and technology, leading to a stream of new technologies. These
investments, in an environment of improving economic liberalization
(initiated in the late 1980s) and stability (post-Real plan), may
have given farmers incentives to boost farm efficiency and
production. However, other policy drivers likely affected farm
efficiency, namely, public infrastructure and rural credit
investments. To test their hypothesis that science and technology
investments were the main impetus to Brazil's agricultural
productivity growth, the authors measure Brazil's national total
factor productivity (TFP-ratio of total output to total inputs
employed in production) and analyze the impact of each of the
policy drivers on that productivity growth.
What Did the Study Find?
We hypothesize that Brazil's agricultural development is a
result of sustained investments in science and technology that led
to a stream of new technologies. These policies, embedded in an
environment of macroeconomic stability and economic liberalization,
provided farmers with the incentives to boost farm efficiency and
production. The study finds:
• That Brazil's national average farm TFP growth increased at an
annual rate of 2.55 percent between 1985 and 2006.
• That TFP growth was driven by factor productivity of the most
efficient farms, which progressed at an average of 4.4 percent each
year. Total factor productivity of the most efficient farms (4.4
percent) is composed of impressive productivity growth in the
livestock (7.1 percent annually) and crops (2.9 percent)
• The most efficient producers achieved rapid TFP growth,
enabling these farms to produce 138 percent more in 2006 than in
1985, while maintaining the earlier input levels.
• The TFP growth of average farms was slower, with mean
technical efficiency levels declining from 93 percent in 1985 to 84
percent in 1995/6 and to 64 percent in 2006. These efficiency
levels imply that the average farm produced 93 percent of what the
most efficient farms produced in 1985, but only 64 percent of what
those most efficient farms produced in 2006.
• Despite an enlarging productivity gap between the most
efficient and average farms, the average farmer was able to produce
62 percent more in 2006 than in 1985, while maintaining the earlier
• Public agricultural research-provided through Embrapa, the
agricultural research agency linked to the Ministry of Agriculture
and Food Supply-appears to have had more influence on the most
efficient farms, widening the TFP gap between those farms and
average producers by 0.2 percent for each 1-percent increase in
Embrapa's research stock.
• Embrapa's national commodity research centers have been
especially important in boosting TFP growth, while its regional
resource research centers have not had the same measurable impact.
Boosting the national commodity research stock by 1 percent widened
the TFP gap by 0.23 percent, while a similar boost to the regional
resource stock did not change that TFP gap.
• Of the policies exerting a narrowing influence on the
productivity gap between the most efficient and the average
producers, rural credit and transportation infrastructure
investments have been factors, but primary school infrastructure
investments exerted the greatest impact.
How Was the Study Conducted?
To analyze Brazil's TFP growth, the authors use decennial
Brazilian farm census data obtained from the Brazilian Institute of
Geography and Statistics (IBGE). These data are combined with data
from Avila and Evenson (1995), Barros (1999), Brazil's agricultural
research agency, Embrapa, and Brazilian statistical yearbook
information. The authors use these data, controlling for a number
of factors, to evaluate Brazil's national agricultural productivity
growth and analyze policy impacts on productive efficiency. The
analysis entails estimating a stochastic distance frontier, which
makes it possible to distinguish technical change among Brazil's
principal commodity groups. The method further allows estimation of
the TFP growth of both the most efficient farms and the average
farms and of the TFP gap separating them. Technical effi ciency
statistics are estimated for each microregion and year and are used
in evaluating the effects of Government policy on Brazilian farm