Nadja El Benni

Where is the risk? Price, yield and cost risk in Swiss crop production

Risk management strategies are of increasing importance in agriculture. An important question is what type of risk management strategies are required to reduce farmers’ income risks? Applying a variance decomposition approach using data from more than 3,000 Swiss farms over a five-year period, this paper quantifies the direct and indirect effects of yields, prices and costs on net revenue variability at the farm level. We find that costs play only a minor role in determining income variability, but price and yield risks are of outmost importance and very crop-specific. For instance, price risks dominate for conventional wheat and sugar beet producers; while corn and barley producers tend to suffer more from production risks. Group comparisons and logistic regressions results show that more intensively producing farms tend to suffer more from price risk, while yield risks are dominant for less intensive producers.

Alternative Specifications of Reference Income Levels in the Income Stabilization Tool

In this chapter, different approaches for the specification of reference income levels in the income stabilization tool (IST) are analyzed. The current proposal of the European Commission suggests a 3-year average or a 5-year Olympic average to specify the farm-level reference income that is used to identify if and to what extent a farmer is indemnified in a specific year. Using Monte Carlo simulations, we investigate the impact of income trends on indemnification if these average-based methods are used in the IST. In addition, we propose and investigate a regression-based approach that considers observed income trends to specify reference income levels. Furthermore, we apply these three different approaches to farm-level panel data from Swiss agriculture for the period 2003–2009. We find that average-based approaches cause lower than expected indemnification levels for farmers with increasing incomes, and higher indemnifications if farm incomes are decreasing over time. Small income trends are sufficient to cause substantial biases between expected (fair) and realized indemnification payments at the farm level. In the presence of income trends, average-based specifications of reference income levels will thus cause two major problems for the IST. First, differences between expected and realized indemnification levels can lead to significant mismatches between expected and real costs of the IST. Second, indemnity levels that do not reflect farm-level income losses do not allow achieving the actual purpose of the IST of securing farm incomes. Our analysis shows that a regression-based approach to specify reference income levels can contribute to bound potential biases in cases of decreasing or increasing income levels.