Public Policy in an Uncertain World

Public Policy in an Uncertain World: Analysis and Decisions

Charles F. Manski
Copyright Date: 2013
Published by: Harvard University Press
Pages: 162
  • Cite this Item
  • Book Info
    Public Policy in an Uncertain World
    Book Description:

    Manski argues that public policy is based on untrustworthy analysis. Failing to account for uncertainty in an uncertain world, policy analysis routinely misleads policy makers with expressions of certitude. Manski critiques the status quo and offers an innovation to improve both how policy research is conducted and how it is used by policy makers.

    eISBN: 978-0-674-06754-7
    Subjects: Political Science, Economics

Table of Contents

  1. Front Matter
    (pp. i-vi)
  2. Table of Contents
    (pp. vii-xii)
  3. Preface
    (pp. xiii-xvi)
  4. Introduction
    (pp. 1-8)

    In recent times, Donald Rumsfeld has personified the difficulty of making public policy in an uncertain world. Rumsfeld served in many capacities in the federal government, culminating in a controversial tenure as secretary of defense from 2001 to 2006. He was often forthright about the limits to knowledge, but he may be remembered for a highly inaccurate prediction of a major policy outcome.

    Rumsfeld’s appreciation of the limits to knowledge is evident in the “Rumsfeld’s Rules” that circulated among public officials in Washington for many years. One was “Learn to say ‘I don’t know.’ If used when appropriate, it will...

  5. I Policy Analysis
    • 1 Policy Analysis with Incredible Certitude
      (pp. 11-46)

      To begin, I distinguish the logic and the credibility of policy analysis (Section 1.1) and cite arguments made for certitude (Section 1.2). I then develop a typology of practices that contribute to incredible certitude. I call these practices conventional certitudes (Section 1.3), dueling certitudes (Section 1.4), conflating science and advocacy (Section 1.5), wishful extrapolation (Section 1.6), illogical certitude (Section 1.7), and media overreach (Section 1.8). In each case, I provide illustrations.

      Policy analysis, like all empirical research, combines assumptions and data to draw conclusions about a population of interest. The logic of empirical inference is summarized by the relationship:


    • 2 Predicting Policy Outcomes
      (pp. 47-84)

      In an ideal world, persons who are not expert in research methodology would be able to trust the conclusions of policy analysis. They would be able to believe predictions of policy outcomes without concern about the process used to produce them.

      Unfortunately, the practices described in Chapter 1 indicate that consumers of policy analysis cannot safely trust the experts. Civil servants, journalists, and concerned citizens need to understand prediction methods well enough to be able to assess reported findings. They need to comprehend conceptually, if not in technical detail, how predictions depend on maintained assumptions and available data.

      With this...

    • 3 Predicting Behavior
      (pp. 85-112)

      This chapter continues to study prediction of policy outcomes, addressing problems more challenging than those examined in Chapter 2. Again the problem is to predict the outcomes that would occur under a policy mandating some treatment. The fresh challenge is that the data come from a study population in which no one received the treatment that would be mandated. For example, in the context of sentencing and recidivism, we may have data on outcomes under a policy where no juveniles were sentenced to confinement and we may want to predict the outcomes that would occur if all were confined. Thus,...

  6. II Policy Decisions
    • 4 Planning with Partial Knowledge
      (pp. 115-138)

      Part I explained the immense difficulty of predicting policy outcomes. I observed that the point predictions produced by analysts are achieved by imposing strong assumptions that rarely have foundation. Analysis with more credible assumptions typically yields interval rather than point predictions.

      Part II considers how policy making may reasonably cope with this difficulty. Chapter 4 and part of Chapter 5 use elementary ideas of decision theory to study policy choice by a planner—an actual or idealized solitary decision maker who acts on behalf of society. Part of Chapter 5 and Chapter 6 will discuss policy choice when a group...

    • 5 Diversified Treatment Choice
      (pp. 139-172)

      This chapter applies the framework for planning with partial knowledge introduced in Chapter 4 to the problem of allocating a population to two treatments. I also consider collective decision processes, where a group of policy makers jointly choose treatments. In both settings, I propose diversified treatment choice as a strategy to cope with uncertainty and reduce it over time. I originally developed this idea in Manski (2007a, 2009).

      Financial diversification has long been a familiar recommendation for portfolio allocation, where an investor allocates wealth across a set of investments. A portfolio is said to be diversified if the investor allocates...

    • 6 Policy Analysis and Decisions
      (pp. 173-176)

      An important objective of policy analysis should be to provide information useful in making policy decisions. Part I of this book described the practice of policy analysis and the inferential problems that researchers confront. I argued that credible analysis typically yields interval rather than point predictions of policy outcomes. Part II examined how a planner or a decision group might reasonably choose policy with partial knowledge. This short closing chapter presents some ideas that tie the two parts of the book together.

      Modern democratic societies have created an institutional separation between policy analysis and decisions, with professional analysts reporting findings...

  7. Appendix A: Derivations for Criteria to Treat X-Pox
    (pp. 177-178)
  8. Appendix B: The Minimax-Regret Allocation to a Status Quo Treatment and an Innovation
    (pp. 179-180)
  9. Appendix C: Treatment Choice with Partial Knowledge of Response to Both Treatments
    (pp. 181-182)
  10. References
    (pp. 183-192)
  11. Index
    (pp. 193-199)