Discussion Paper

No. 2013-12 | February 11, 2013
South African Food Security and Climate Change: Agriculture Futures
(Published in Special Issue Food Security and Climate Change)

Abstract

The projected changes in planted area, yield per area, net exports/imports and priced for five major agricultural crops in South Africa were simulated using the projections of four Global Circulation Models (GCMs) under three socio-economic scenarios. The GCM runs were those undertaken for the IPCC fourth assessment report. They show consistent strong warming over the subcontinent, but disagree with respect to future precipitation, from slight wetting (particularly on the eastern side) to overall slight drying. The future crop yields were simulated using the DSSAT crop model suite. The planted area, commodity prices and net exports were simulated using the IMPACT global food trade model. The results indicate slightly rising to stable yields per unit area up to 2050, despite climate change, largely due to the inbuilt assumption of ongoing agronomic and genetic improvements. The planted area remains fairly constant in both location and size. As a result of increasing food demand, net exports decline (i.e. imports increase) substantially, and so do prices due to simultaneous increases in global demand. The effects on food security in South Africa, measured as average calorific intake per person and malnutrition in children under the age of five, depends more on the assumptions regarding population and GDP growth than on climate change, since the study assumes that local shortages will be balanced by increased imports, if they are affordable. Thus the vulnerability to food insecurity at family and national level increases in the future under all but the most optimistic development scenarios, and is exacerbated by climate change, especially through global-scale, market-related mechanisms. Policies to increase local agricultural production, decrease its climate sensitivity and facilitate access to international markets are indicated, along with efforts to reduce agricultural greenhouse gas emissions.

Paper submitted to the special issue
Food Security and Climate Change

JEL Classification

Q55 Q18

Cite As

Sikhalazo Dube, Robert J. Scholes, Gerald C. Nelson, Daniel Mason-D’Croz, and Amanda Palazzo (2013). South African Food Security and Climate Change: Agriculture Futures. Economics Discussion Papers, No 2013-12, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2013-12

Assessment



Comments and Questions


Anonymous - Referee Report 1
February 25, 2013 - 11:49

see attached file


Gbadebo Odularu - Reader comment
February 28, 2013 - 08:47

see attached file


Anonymous - Referee Report 2
May 02, 2013 - 08:59

The study presents findings from detailed analysis of the effects of climate change on five key agricultural subsectors and food security in South Africa. As with most analyses of climate change and agriculture it is probably fair to say that even a 50-page report cannot fully capture the complexities of ...[more]

... this topic. However, the integrated model framework adopted in the study presents a useful indication of how various sectors might be affected under a range of future climate scenarios. I have no serious concerns about the methods or models used. However, below I list some issues related to the background analysis, model assumptions, and robustness of results which, if addressed, would in my view strengthen the paper.

Perhaps the main weakness of the paper is its report-like writing style, especially in the background section of the paper. Facts about the South African economy (or the agricultural sector in particular) and its population are presented in short paragraphs under numerous headings. A simplified structure would improve the flow. Furthermore, most paragraphs merely describe results in accompanying tables or figures rather than providing any interesting analysis. The result is that the paper lacks a binding narrative about South Africa’s agricultural sector, its links to poverty and food security, and how the sector might adapt to or mitigate the effects of climate change. I also feel there is a disconnect between the background work, on the one hand, and the model, its assumptions, and the results on the other. In particular, I would have liked a more in-depth discussion of the following:
* The notion that ;land expansion possibilities are limited given that large tracts of land are already "marginal". This implies that future food demand has to be either met by food imports or through yield increases. [And then in the modeling section: How do the authors account for these constraints in the model? What do the model results say about future land expansion into marginal lands?]
* The fact that the agricultural sector is largely characterized by large-scale intensive commercial farming, thus distinguishing the country from many other African countries where small-scale subsistence farming dominates. The implication is that poverty/food security is to a large extent "delinked" from the success of the agricultural sector, as most poor or vulnerable households are linked to nonfarm sectors or are unemployed. [And then in the modeling section: Is the disconnect between the poor and agricultural outcomes apparent in the model results, or are the consumption linkages strong enough so that poor households are affected via food prices?].

Another issue relates to the robustness of results. The results seem to suggest that climate change will not have a negative impact on crop yields; per capita caloric availability only declines in the pessimistic scenario; and the number of malnourished children declines in absolute terms despite population growth. Early on in the paper, however, the authors state that “projected changes in precipitation and increase in temperature will hamper food production”. The conclusion reiterates these sentiments: “climate change poses a challenge to food security in South Africa”. Am I missing something?
* First of all, the conclusion should clarify how climate change hampers food production or poses a challenge, because apparently it does not really, at least not under the climate scenarios considered here and the modeling assumptions accompanying them.
* Perhaps the conclusion should reiterate what is stated in the abstract, namely that yields increase despite climate change due to the “inbuilt assumption of ongoing agronomic and genetic improvements”. This assumption seems to be absolutely central to the results. I am not an expert on agricultural inputs/technology, but the assumption might be a fair one in countries where technology adoption is low and yield gaps are high. Does it also hold for commercial farmers already using modern technology as is mostly the case in South Africa?
* It is also noted that the analysis does not take into account weather variability. The authors claim that climate change will lead to “more extreme events that would have negative effects on agriculture” and that the negative effects of climate change are therefore underestimated. This raises two questions: (i) Can we really verify that climate change will lead to more extreme events? My understanding is that this issue is still under debate. (ii) Is it possible to speculate by how much these negative effects are underestimated (e.g., if they are large enough yields may actually be lower on average).

Finally, throughout the paper the authors note the need for “mitigation and adaptation”. The last section (on agricultural emissions) deals with greenhouse gas mitigation and the role of agriculture, but this section is almost out of place and appears to have been added as an afterthought rather than an integrated part of the paper. Besides, agricultural greenhouse gas emissions are so insignificant in the carbon-intensive South African context. A more interesting issue is adaptation, which seems to be largely ignored in the paper. It would greatly enhance the study if there were some discussion about what the agricultural sector can do to avoid an outcome such as the one in the pessimistic scenario, or to ensure that yields increase despite climate change (as the results suggest). This issue can be tied in with the issue of robustness of results discussed in the previous section.