The effect of drought and rising temperatures on total factor productivity in pasture-based dairy farming in the Eastern Cape

A region's climate is determined by its long-term rainfall, temperature, wind and evaporation profile, and weather is the short-term variations around these long-term expected values. This study uses weather to investigate the potential future effects of climate change without making any predictions about when and how much climate change will occur. The study is of the total factor productivity of pasture-based dairy production systems in the Eastern Cape. An unbalanced panel dataset of 206 observations collected from 62 farms over nine years was used to fit a stochastic frontier production function with technical inefficiency effects. Rainfall is part of the frontier because it is an essential input into dryland pasture production and temperatures are the main inefficiency effects. Since presented by Battese and Coelli (1995) this approach has become a mainstay in modern production economics and its application to climate change is just beginning to appear in the international literature (Qi et al., 2015). As far as could be established, this is the first study of its kind in South Africa. The stochastic frontier production function takes into account the number of cows in milk, the cost of concentrates fed to cows in milk, the cost of purchased roughage, the cost of fertiliser to grow pastures, and millimetres of rainfall per year. Capital, land and labour inputs were not available, but this was not insurmountable since all three of these factors are used in fixed proportions to herd size. A generalised likelihood ratio test confirmed the existence of significant inefficiency effects associated with daily minimum and maximum summer temperatures and an uneven rate of dissemination. The translog functional form was chosen over Cobb Douglas based on another generalised likelihood ratio test. Output elasticities indicate increasing returns to scale in production and identify cows as the main factor of production followed by the amount spent on concentrates and rainfall. Roughage and fertiliser expenditure are of minor importance in the production process. The mean level of efficiency was 94% and it varied between 82% and 99%. There is a small but statistically significant difference in efficiency between farms in the coastal belt and those in the interior which reflects the benefits of having access to a reliable source of irrigation water. Productivity is only marginally positively correlated with rainfall (r =0.100, p≤ 0.153). It is positively correlated with maximum summer temperatures, indicating better conditions for pasture growth, and strongly negatively correlated with minimum temperatures. This latter is difficult to explain and could be due to irrigation conditions rather than temperature. The importance of this study is that it demonstrates the viability of a method that could be used to assess the likely effects of the predicted climate change on a specific agricultural enterprise that could help that sector to mitigate those changes more effectively than it would have been able to do in the absence of such study.

Grantor

School of Economics

Year dc:date.issued

2021

Author dc:creator

• Renner, Andrea

Advisor dc:contributor.advisor

• Conradie, Beatrice

• Economics

Handle dc:identifier.uri

http://hdl.handle.net/11427/36137

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/36137