The delivery of welfare services has become increasingly organised through quasi-markets, motivated by the idea that competition and choice can be harnessed to achieve services of higher-quality and lower-cost compared with monopolistic state provision. However, to incentivise the efficiency of service production and responsiveness of providers to clients, this requires quasi-markets to be frequently re-commissioned through periodic cycles of competitive tendering and public procurement. This leads to an inherent dilemma of transaction costs in quasi-markets: each episode of competitive procurement comes at considerable expense for governments and providers, and therefore increases the costs of quasi-marketisation. These costs of (re)commissioning quasi-markets rarely receive attention in public policy or research on quasi-markets. This study addresses this issue of the transaction costs associated with competitive tendering through an in-depth case study from Australia, which has had a welfare-to-work quasi-market market model for 30 years and undertaken multiple system redesigns and re-procurements. Drawing on interviews with executive leaders from agencies that bid for contracts during a wave of procurement and reform in the early 2020’s, as well as documentary analysis of hearings and reviews into the procurement of Australia’s welfare-to-work market, this study examines the heavy transaction costs that periodically re-procuring welfare-to-work markets create. In so doing, it draws attention not merely to the volume of these transaction cots but to their political role in skewing the terms of market participation towards particular kinds of organisations.