ABSTRACT
Chapter 5 describes and analyses the set of risk-based compliance strategies used to detect welfare non-compliance and fraud. It focuses on two specific risk profiling strategies, which were implemented in Australia in the mid-2010s: Data Mining Risk Profiling focused on detecting criminal fraud and Risk-based Compliance Reviews targeting welfare non-compliance and debt. The chapter also reflects on more recent developments in risk-based welfare surveillance tied to rapid technological advancements, including machine learning. Using these case studies, the chapter highlights and critiques the regressive effects of risk profiling in the realm of welfare compliance. It suggests that these techniques serve to increase the surveillance of the most economically and socially marginalised recipients, including poor single mothers. But, this chapter also demonstrates how risk logics are open to re-articulation alongside more progressive goals and agendas. This finding challenges strands of the risk literature, including scholarship inspired by the ‘new penology’ perspective, which assume that actuarial and algorithmic risk-based practices are necessarily regressive in their effects. Instead, inspired by scholars such as Pat O'Malley and Kelly Hannah-Moffat, this chapter illustrates the malleability of risk strategies and reinforces the importance of examining welfare governance in situ.
