Identifying clusters of innovation barriers in farming enterprises: a K-modes clustering approach

Abstract

Despite the relevance of agriculture in terms of its contribution to food security and environmental sustainability, empirical evidence on the nature and effects of barriers to innovation, remains very limited. This problem has often led to inconsistent policy design, failing to meet the needs and expectations of farming enterprises. Innovation profiling and segmentation have emerged as important tools for understanding how businesses innovate. Using data from the South African Agricultural Business Innovation Survey (AgriBIS-2016-2018), this study applies the K-modes clustering algorithm, to group farming enterprises based on the innovation barriers they face. The analysis identified three distinct clusters with differing innovation profiles. The findings show that businesses in Cluster 2 recognised institutional barriers such as Lack of Government Support (65.4%) and Stringent Agricultural Policies (93.6%) as key barriers to their innovations, while Cluster 3 businesses highlighted environmental factors as critical impediments to their innovation. Cluster 1 businesses faced a diverse range of resource-related barriers. This study addresses a key knowledge gap and increases our understanding of the barriers to agricultural innovation. The findings have important implications for innovation policy instruments and strategies aimed at promoting agricultural innovation, sustainability, productivity and resilience in the face of barriers to innovation.