Accountable Agents: Federal Performance Measurement and Third-Party Government



Ordinarily research articles on public sector third-party governance, or the state of agents, turn to the subject of accountability in their conclusions. Rather than leaving it to the conclusions, this article takes up the subject of public sector third-party governance as a problem of accountability. To consider accountability, we studied contemporaryperformance measurement practices in the federal government, particularly as they are applied in five agencies of the US Department of Health and Human Services. Our findings are presented using a six-part “promises of accountability” heuristic which captured the many and varied uses of accountability in contemporary policy and administration discourse. We found that the characteristics of third parties and the nature of their principal-agent relations are a key determinant of accountability; bureaucratic and hierarchical controls enhance agency and program accountability; building rules and program policies into grants andcontracts enhance accountability; agencies that practice lateral trust-based forms of “relational contracting” foster cultures of accountability; auditing and reporting requirements enhance program accountability;federal performance measurement regimes tend to be “top down” and“one size fits all;” federal performance measurement regimes are primarily executive branch forms of accountability; performancemeasurement often represents attempts to superimpose managerial logicand managerial process on inherently political processes embedded in the separation of powers; there is little evidence that performance-based accountability results in enhancing democratic outcomes or greater justice or equity.


Political Science

Publication Date


Journal Title

Journal of Public Administration Research and Theory


Oxford Journals

Digital Object Identifier (DOI)


Document Type



© The Author 2009. Published by Oxford University Press on behalf of the Journal of Public Administration Research and Theory, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org