Abstract
The purpose of this article is to analyse the costs of introducing quasi-markets to the public sector in practice and to discuss if quasi-markets have been efficient when costs are added. Transaction costs, which will follow instead of hierarchy costs, are difficult to identify and therefore examine. In addition, costs will accrue from changing the mode of governance. The article concludes that creating quasi-markets produces some costs which reduce both technical and allocative efficiency, but what is most problematic is that those costs which are impossible to measure in practice are of greater significance.