Donald P. Moynihan

1. Experiences with performance management in the United StatesPresentation Towards a more result oriented Flemish public sector, January 10, 2014 Donald P. Moynihan

2. Part I: Overview

3. Overview • US Experience - background • Errors in understanding performance management • Expectations about implementation • The politics of performance management • Lessons: how do we encourage purposeful use?

4. Background

5. National government-wide changes • Government Performance and Results Act - GPRA (1993-2010) • Program Assessment Rating Tool (2002-2008) • GPRA Modernization Act (2010-) • State level variations on these models

6. Doctrinal logic for change

7. 20 Years of Learning? • Some lessons on how it went • Partly from study of topic • Reflected in some policy changes, especially GPRA Modernization Act

8. Expectations about implementation

9. Is the idea of performance management running out of steam? • OECD 2012 survey • Seems to be less use of performance data than in past • Performance targets not consequential • General sense of disappointment: we have systems in place, have not delivered desired results

10. Expectations problem • We define performance systems by the benefits we hope will occur (more rational budgeting, more efficient management) • The gap between our aspirations and the observed effects of these rules are usually large, resulting in disappointment • More grounded and accurate description: performance systems are a set of formal rules that seek to disrupt strongly embedded social routines

11. Confusion: Adoption vs. implementation • Speak of governments doing performance management • What do we mean? • Rules about measuring and disseminating data

12. Inattention to the use of data • Performance data by itself does not do much • Implementation of performance management means using the data • Why focus on performance information use? • Difficult to connect public actions to outcomes • Intermediate measure of effectiveness – performance information use • Without it, good things we want don’t happen • There are different types of use

13. The four types of use • Passive – minimal compliance with procedural requirements • Purposeful –improve key goals and efficiency • Political – advocate for programs • Perverse – behave in ways detrimental to goals (goal displacement and gaming)

14. Effect of performance reforms • Can observe if agencies comply with requirements (passive use), but not other types of use • Performance systems encourage passive use, not purposeful

15. The Politics of performance management

16. Apolitical performance reforms? • Performance data associated with neutrality • Offers objective account of the world, and will engender consensus • Reduces the role of politics by offering an alternative basis to make arguments • This is part of political appeal • Has implications for adoption and implementation

17. Politics of adoption • Elected officials motivated by symbolic values • Primary focus on adopting information reporting requirements, not broader change

18. Actual pattern of change

19. One basic reason for confusion • We fail to understand the nature of performance data • We assume data are • Comprehensive • Objective • Indicative of actual performance • Consistently understood • Prompts a consensus

20. The ambiguity of performance data • Examine sameprogramsbutdisagreeon data • Agreeon data butdisagreeonmeaning • Agreeonmeaning, butnotonnextactionsteps/resources • Notclearonhow data links tobudgetdecisions

21. The subjectivity of performance data • Actorswillselect and interpret performance informationconsistentwithinstitutionalvalues and purposes • Greater contesting of performance data and less potential for solution seeking in forums featuring actors with competing beliefs

22. Implications: Political use • Performance data • is socially constructed by individuals subject to personal biases, institutional beliefs, and partisan preferences • has qualities of ambiguity and subjectivity • These qualities make performance management likely to operate as part of political process, not as alternative to it

23. evidence of advocacy • “Spinning” (Hood 2006) • Claim credit when things go well, deny responsibility when things do not • Advocacy by agents seeks to avoid blame and respond to “negativity bias” • disproportionate citizen dissatisfaction with missed target (James 2011) • political officials pay more attention to high and low performers (Nielsen and Baekgaard 2013) • more bureaucratic explanations of failed performance (Charbonneau and Bellavance 2012)

24. Stakeholders • Political support for agency associated with performance information use (Moynihan and Pandey 2010) • May worry less about blame, freedom to experiment • Belief that stakeholders care about performance or performance measures associated with bureaucratic use (Moynihan and Pandey 2010) • More performance information use when: stakeholders perceived as more influential, more in conflict, and when there is more networking with stakeholders (Askim, Johnsen, and Christophersen2008; Moynihan and Hawes 2012)

25. Principal agent argument • Assumption: Use performance data to reduce information advantage that agencies have over center & elected officials • Reality: • Some evidence of partisan biases in implementation • As long as agencies play role in defining, collecting, and disseminating information, they retain information asymmetry

26. AN example: Program assessment rating tool (PART) • Bush-era questionnaire used by Office of Management and Budget to rank programs from ineffective to effective • Four sections: program purpose and design, strategic planning, program management, and program results/accountability • Burden of proof on agencies • Almost all federal programs evaluated

27. How might politics affect PART implementation? • Ostensibly neutral reforms may serve—or may be seen as serving—political ends: • Partisan reformers may implement reforms differently if programs/agencies are ideologically divergent • Managers of ideologically divergent programs may perceive bias (whether or not a reform effort is biased against their programs)

28. Was PART political? • Designed to be good government, politically neutral reform, and qualitative studies do not report overt partisanship, but… • More liberal agencies and programs get lower scores (Gallo and Lewis 2012; Gilmour and Lewis 2006) • PART scores only related to President’s budget proposals for liberal programs (Gilmour and Lewis 2006)

29. Did politics affect response to PART? • Liberal agencies, though smaller, had significantly higher PARTs completed • Two types of effort: • Observable: self-reported effort in completing PART – higher for managers in liberal agencies (Lavertu, Lewis and Moynihan 2013) • Discretionary: performance information use – lower for managers in liberal agencies (Lavertu and Moynihan 2012)

30. Why would PART impose a greater administrative burden on liberal agencies? • Liberal agencies likely concerned about making their programs look as good as possible, given preference divergence • Potentially greater scrutiny of liberal programs, requiring more costly agency data collection and reporting

31. Lessons: How do we encourage purposeful use

32. When does Perverse Use occur? • Goal displacement – e.g. cream-skimming • Data manipulation – including outright cheating • Becomes more likely when • Data is self-reported • Task is complex and hard to measure • High-powered incentives attached to measures • Especially in contracting • Job-training programs, tuition programs • Policymakers have imperfect knowledge of perversity, amend contracts after problems occur

33. Next generation performance system? GPRA Modernization Act of 2010 • Quarterly performance reviews • Goal leaders • Chief operating officers/performance improvement officers • High-priority goals • Cross-agency priority goals • For summary, see Moynihan 2013

34. Continuing challenge: how to make use of performance data • Create learning forums: routine discussions of performance data with supervisors/peers associated with use (Moynihan and Lavertu 2012) • GPRA Modernization Act: quarterly performance reviews • Not just routines, also learning culture • Tolerates error • Rewards innovation • Brings together multiple perspectives • Gives discretion to users • Tradeoff between learning and accountability • Accountability evokes defensive reactions and gaming

35. Look for actionable data • You might want to measure everything but you can’t manage everything • Problem with PART – equal attention to all goals • Modernization Act: focus on important targets, areas of opportunity (high priority goals, cross-agency priority goals)

36. Foster goal clarity • Clear goals increase performance information use (Moynihan and Pandey 2010); may not be easy if: • Service has many different aspects • Tension between: • Few enough measures to generate attention • Enough measures to avoid encouraging workers to ignore unmeasured aspects

37. Appeal to altruism • Appeal to altruistic motivations, not extrinsic reward (Moynihan, Wright and Pandey 2012) • Select goals that motivate • Clear line of sight between goals and actions • Celebrate achievement • Connect to beneficiaries

38. Integrate program evaluation and performance management • Performance data tells you if a measure moved up or down, evaluations tell you what affects performance • Discussion of evaluations should be incorporated into performance management • Assign evaluation funding for new policies • Example: Washington State Institute for Public Policy provides meta-analyses of research on different policies, and provides return-on-investment estimates to policymakers

39. Induce leadership commitment • Leadership commitment associated with use (Dull 2009; Moynihan and Lavertu 2012) • How do you create commitment? • Reputation: public commitments and responsibility (high priority goals) • Create leadership positions with oversight for performance (COOs, PIOs, goal leaders) • Select leaders based on ability to manage performance

40. Conclusion • Welcome your feedback and questions • Performance Information Project: • http://www.lafollette.wisc.edu/publicservice/performance/index.html • dmoynihan@lafollette.wisc.edu

41. References Askim, Jostein, ÅgeJohnsen, and Knut-Andreas Christophersen. 2008. Factors behind organizational learning from benchmarking: Experiences from Norwegian municipal benchmarking networks. Journal of Public Administration Research and Theory 18(2): 297–320. Charbonneau, Etienne, and François Bellavance. 2012. Blame Avoidance in Public Reporting. Public Performance & Management Review 35(3): 399-421 Gallo, Nick and David E. Lewis. 2012. The Consequences of Presidential Patronage for Federal Agency Performance Journal of Public Administration Research and Theory. 22(2): 195-217. Dull, Matthew. 2009. Results-model reform leadership: Questions of credible commitment. Journal of Public Administration Research & Theory 19(2): 255–84. Hood, Christopher. 2006. Gaming in targetworld: The targets approach to managing British public services. Public Administration Review 66(4): 515–21. James, Oliver. 2011. Managing Citizens’ Expectations of Public Service Performance: Evidence from Observation and Experimentation in Local Government Public Administration, 89 (4), 1419-35.   Gilmour, John B., and David E. Lewis. 2006a. Assessing performance budgeting at OMB: The influence of politics, performance, and program size. Journal of Public Administration Research and Theory 16:169-86. Lavertu, Stéphane and Donald P. Moynihan. 2013. Agency Political Ideology and Reform Implementation: Performance Management in the Bush Administration. Journal of Public Administration Research and Theory Moynihan, Donald P. and Daniel Hawes. 2012. “Responsiveness to Reform Values: The Influence of the Environment on Performance Information Use.” Public Administration Review 72(S1): 95-105.

42. Lavertu, Stephane, David Lewis and Donald Moynihan. 2013 Government Reform, Political Ideology, and Administrative Burden: The Case of Performance Management in the Bush Administration. Forthcoming in Public Administration Review Moynihan, Donald P. 2008. The Dynamics of Performance Management. Washington DC: Georgetown University Press. Moynihan, Donald P. 2013. The New Federal Performance System: Implementing the New GPRA Modernization Act. Washington D.C.: IBM Center for the Business of Government. Moynihan, Donald, and Sanjay Pandey. 2010. The big question for performance management: Why do managers use performance information? Journal of Public Administration Research and Theory 20(4): 849–66. Moynihan, D., Pandey, S., & Wright, B. (2012a). Prosocial values and performance management theory: The link between perceived social impact and performance information use. Governance, 25(3), 463–83. Moynihan, Donald P. and Daniel Hawes. 2012. Responsiveness to Reform Values: The Influence of Environment on Performance Information Use. Public Administration Review 72(S1): 95-105. Moynihan, Donald, and Patricia Ingraham. 2004. Integrative leadership in the public sector: A model of performance-information use. Administration & Society 36(4): 427–53 Moynihan, Donald P. and Stéphane Lavertu. 2012. “Does Involvement in Performance Reforms Encourage Performance Information Use? Evaluating GPRA and PART.” Public Administration Review 7(4): 592-602