Climate Smart Agriculture: challenges, impacts, synergies and conflicts

Agriculture is responsible for 14% of global greenhouse gas emissions and will also be significantly impacted by climate change (IPCC 2014 , FAO, 2013). Furthermore, the FAO estimate that global agricultural production will need to increase by 60% by 2050 in order to meet the demands of population growth and dietary change (Alexandratos and Bruinsma, 2012). Faced with this 'perfect storm', the FAO and others argue there is a need to make agriculture more 'climate smart', i.e. to improve farmers' productivity, resilience to climate change, and reduce greenhouse gas emissions from agriculture. This work will explore what 'climate smart' agriculture could look like in practice and the effectiveness of initiatives to promote it in an African context. It will aim for the highest standards of academic rigour and policy relevance.
This research will address the following questions:
-An empirical assessment of how 'climate smart' is African agriculture. This stage would aim to estimate a 'baseline' in terms of emissions, livelihoods and climate preparedness in a selection of African agricultural settings.
-What has been the impact of policies and programmes to promote climate smart agriculture in Africa? For example what emissions reductions have been achieved; what has been the impact on livelihoods and adaptation?
-To what extent are there synergies and to what extent are there conflicts between the objectives of climate smart agriculture? For example do actions which help poor farmers adapt to climate change also contribute to improved livelihoods? Are there conflicts between mitigation and livelihood objectives and how could these be resolved?
Methods and challenges
The fundamental approach for this work will be quantitative and empirical. Access to data will be a key issue addressed in the next section. Appropriate networking and institutional ties with e.g. CGIAR institutes, FAO and in country statistical agencies could substantially help to develop the work.
The work could involve participation in primary surveys. However the extensive collection of household datasets is a resource intensive and technically difficult activity (Lamhauge et al 2011), and so it is likely that some form of collaboration with existing research(ers) will be needed should there be a need for additional data.
A particular methodological challenge for all evaluation work is the attribution of project outcomes to the particular intervention in question, and separating actual outcomes from a counterfactual 'business as usual' case (FAO, 2013, HMT 2011). Specific to the evaluation of climate change adaptation actions, there is the problem of disentangling the impact of local weather variability from long run trend effects of climate change (Lamhauge et al, 2011). The paucity of existing impact evaluations of climate smart policies will add to the challenge of this work.
Recent developments in econometrics provide powerful tools which can be used for the evaluation of public policies and interventions (Guido and Woolridge, 2009). One approach is the Difference-in Difference method which seeks to compare changes in a treatment group to changes in a control group who have not received the intervention (for example Udagawa et al 2014). A related method used in the agricultural adaptation literature is endogenous switching regression, which aims to control for the impact of non observable factors such as farmer skill on the effectiveness of adaptation (Di Falco et al, 2011, Gorst, 2015). This method is superior to Ordinary least Squares regression where there are systematic differences between those adopting for example climate smart practices and those not, i.e. the adoption decision is endogenous (Di Falco 2011).