New Public Management (NPM), particularly in its early stages in the 1980s, is often said to have concentrated on cost-cutting and efficiency, but few studies have examined how far NPM succeeded in cutting costs. Focusing on UK central government, often claimed to be a leading case of NPM, and analyzing three sets of data (for running costs, tax collection costs, and paybill), this study finds little evidence of real running cost reductions in the early NPM era. It concludes that, contrary to stereotype, 1980s NPM in UK central government does not offer a model for the scale of administrative cost-cutting currently planned, and even the cutbacks of the 1990s fell far short of those current plans. The conclusion for the academic study of NPM is that something more than a minor revision of the received view of NPM as a cost-cutting movement is needed to explain these observations.