Can climate information improve the production efficiency of farmers in Ethiopia?

Agricultural productivity in Ethiopia has been inhibited by various factors, especially weather variability and climate change. Ethiopian agriculture is the highest at risk from climate change impacts, with Ethiopia among the top 15% of vulnerable countries with the least adaptive capacity worldwide. If the decrease in average annual rainfall for key agricultural zones follows the current trajectory, the country will lose more than 6% of its agricultural output yearly. Addressing the food security of a growing population of 107 million people, which is expected to reach 129 million by 2030 poses a formidable challenge for the sector. Yet, dominated by smallholder subsistence farmers, the production efficiency of the agriculture sector is 66%, indicating that 34% of production has been lost due to an inefficient use of resources. 

Climate information as a solution to improve production efficiency

In its climate-resilient green economy strategy, the Ethiopian government recently set out climate information as an instrument to minimise climate-induced shocks and build a climate-resilient agriculture sector. In this endeavour, climate information is expected to provide a double benefit. On the one hand, it would improve climate resilience by serving as a hedge against climate-related shocks. On the other hand, it would spur production efficiency as it assists farmers’ optimal use of agricultural inputs. Yet, appropriately utilising the potential of climate information to build climate change resilience requires answering two critical questions of policy relevance. 

1. What causes farmers’ reluctance in climate information utilisation for agricultural decision-making in Ethiopia? 
2. Would utilising climate information provide a positive payoff for smallholder farmers?

Our study aims to answer these questions. We use survey data from the Nile Basin of Ethiopia, which covers 34% of the country’s geographical area, collected in the 2015, 2016, and 2017 production seasons. One part of the survey was allocated to gather detailed information about farmers' access to temperature and rainfall forecast information, whether farmers use the information to decide on key agricultural activities (for those who had received the information), like choosing the crop type to be planted, the start of land preparation, the start of weeding and sowing dates; and the source of the information. For the data analysis, we jointly use frontier analysis methods with impact evaluation approaches, namely propensity score matching (PSM). 

For what purposes do farmers use climate information?

For those farmers who had access to climate information, we asked if they used the information to decide on various agricultural production activities. Most of the farmers in our sample use climate information to decide on harvesting times followed by threshing and planting times.  Pre- and post-harvest loss is a key challenge for smallholder farmers in Ethiopia. According to the recent CSA pilot survey on pre-and post-harvest crop losses, around 8.69% and 10% of maize and wheat annual production, respectively, were lost during the harvesting, stacking and threshing stages of production. Hence, climate information supports farmers to appropriately plan the harvesting and threshing time to reduce crop losses from these agricultural activities. 

Figure 1: Climate information use by types of production activities

What determines farmers’ climate information use for production decision-making? 

Farmers' decision to use climate information for agricultural activity decision-making is affected by different factors. Access to extension services and farmers' shock experience in agricultural production increases climate information use. This could be due to the benefits of climate information as a climate adaptation strategy and hedge against climate-related shocks. For example, a good understanding of the benefits of sustainable agricultural practices through extension services would be easily applied to identify and utilise related climate change adaptation measures, such as climate information. Similarly, a higher household size, being a male-headed household, and the perception of relying on a government during bad harvesting seasons enhance climate information utilisation. While reliance on the government in the absence of favourable weather conditions provides insurance for farmers, a higher household size could serve as a source of diversified agricultural knowledge and labour supply. This means that promoting climate information use combined with index-based climate insurance would widely facilitate climate information utilisation and open avenues towards reaping the benefits of climate information services. 

However, climate information utilisation is inhibited by temperature variability. One plausible reason for this would be the accuracy of the information prediction. If the predicted information fails to properly reveal the variability in actual temperature, farmers may not be willing to depend on the received climate information to decide on key agricultural activities during high-temperature variability. It could be also the case that farmers opt for other adaptation techniques, for instance engaging in off-farm income-generating activities, when they face a concurrent variability in temperature. 

Does climate information utilisation improve production efficiency?

Production efficiency, which we measure as farmers' capacity to produce a higher level of output using existing levels of agricultural inputs, is a key measure of farm performance. Our study findings show that using climate information for agricultural production decision-making increases the efficiency of smallholder farmers in Ethiopia. The technical efficiency level of climate information users is nearly 13 percentage points higher than non-users (Figure 2). 

Figure 2: The effect of climate information use on technical efficiency

Looking at the crop-specific effects, climate information use improves the technical efficiency of all four crops we considered in this study, namely, maize, wheat, teff, and barley. However, the highest payoff from climate information use is attributed to barley and wheat crops.

The policy implication of our study is that in subsistence farming where rain feed agriculture dominates the production system, climate information could serve as an early warning signaling tool to reduce climate-related risks such as crop failure. However, this requires developing appropriate planning for disaster and climate risk management and strengthening early warning systems for rapid-onset weather variability.