In 2003, the member countries of the African Union launched a new initiative aimed at increasing food security and reducing poverty through the growth and development of the region’s agricultural sector. The Comprehensive Africa Agriculture Development Program , or CAADP, set a target of 6 percent annual average growth in the agricultural sector, as well as the allocation of 10 percent of total annual government expenditures to the sector. CAADP has been credited with significant achievements in improving the effectiveness and transparency of agricultural policy-making in the region; however, questions remain regarding the program’s impact on agricultural and economic growth and development outcomes.
A new IFPRI discussion paper aims to fill this knowledge gap using panel data from 46 African countries from 2001-2014. The author uses this data and a difference-in-difference treatment-effects model to estimate the impact of CAADP on agricultural expenditure, agricultural value-added, land and labor productivity, income (GDP per capita), and nutrition (prevalence of adult malnourishment).
Two definitions of treatment were used in the study: i) a variable signifying whether a country signed the CAADP compact and ii) a level signifying the stage of CAADP implementation reached (level 0 = pre-compact, level 1 = compact signed, level 2 = National Agricultural Investment Plan (NAIP) developed, level 3 = one external source of funding accessed, and level 4 = more than one external source of funding accessed). The paper notes that CAADP is a complex program that takes time to be implemented and whose effects would therefore take time to materialize. Thus, the treatment effects of the program should theoretically be larger and more positive for countries that signed the CAADP compact earlier than those that signed later and for those that have reached higher levels of implementation, according to the author.
The study finds that countries’ willingness to implement CAADP at higher levels appear to be positively influenced by a number of factors: the importance of agriculture in the national economy; the government’s political will, represented by its previous willingness to sign other AU charters; the share of neighbors currently implementing CAADP; and the government’s capacity to successfully implement the program. On the other hand, one factor has a negative impact on a country’s willingness to implement the program: the country’s negotiating position within the AU (measured by total expenditure per capita and share of a country’s economy in Africa’s overall economy). Richer countries appear less likely to implement the program, possibly because these countries have other non-agricultural sources with which to drive development.
In terms of the development outcomes studied, the results show that, in general, CAADP has positively impacted both agricultural value-added and land and labor productivity. The estimated impact (from the author’s fixed-effects regression estimation) of CAADP implementation on agricultural value-added for countries that have achieved level 4 implementation is 16.7 percent; this number is 10.8 percent for level 3 countries, 7.4 percent for level 2 countries, and 8.6 percent for level 1 countries. For land and labor productivity, CAADP has had a positive impact on countries at implementation levels 1, 3, and 4.
However, CAADP was found to have a generally negative impact on agricultural expenditure. This effect is particularly statistically significant for countries at level 4 implementation (between -23 percent and -20.1 percent). This suggests that a substitution effect exists between government spending on the agricultural sector and external sources of funding for the sector, which would make sense for level 4 countries, as this group has access to two or more sources of external agricultural funding.
In terms of nutrition, the estimated impacts are generally not significant. However, the results on nutrition did present a puzzling finding. The group of countries that signed the CAADP compact in 2007-2009 saw an increase in the prevalence of adult undernourishment, whereas countries who signed the compact in 2010-2012 saw a decrease. Again, this contradicts the assumption that countries that signed CAADP earlier would be better off. The author suggests that this finding on nutrition could be due to the quality and/or quantity of the actual investments made in nutrition-sensitive agricultural programs, compared to the stated budget allocations in a country’s NAIP. The author calls for further research on the actual amount invested in different nutrition interventions in CAADP countries.
Overall, the paper’s findings suggest that further development and monitoring of the CAADP process is needed, since some groups of countries see positive effects from the program and others see less positive or even negative effects. Efforts to sustain and raise the returns from the program should include transitioning countries from growth driven by expanded agricultural production to growth focused on improvements in agricultural technology and sustainability. In addition, further research is needed on how CAADP is developed and implemented in individual countries, as well as on how countries invest in productivity-enhancing and nutrition-security interventions.
By: Sara Gustafson