The heart of policy analysis is trying to understand what happens when a certain policy is adopted. This is done by projecting how a policy will advance or detract certain policy criteria—hallmarks of good policy.
The problem is that good policy can be evaluated a number of ways. Does the policy grow the economy? Does it reduce poverty and inequality? Does it advance education and improve health? Does it make people happier? In order for a policy analysis to be manageable, a policy analyst needs to narrow this list by selecting criteria for the analysis.
Eugene Bardach details a few commonly used evaluative criteria in his book A Practical Guide for Policy Analysis: the Eightfold Path to More Effective Problem Solving. The first criteria Bardach deems “Hit the Target!” This includes goals like cutting water consumption by 5 percent for a quarter or de-leading all painted interior surfaces in a neighborhood by December 31st.
Bardarch then goes on to talk about “efficiency,” or the measure of the sum of economic welfare impacts of a policy. He also mentions “equality, equity, fairness, justice” as a criteria, which in short means how a policy will impact different groups of people. Bardach gets more abstract talking about “freedom, community, and other ideas”—criteria that help capture the type of community policymakers want to create and the types of lives that can be lived in those communities.
Lastly, Bardach mentions “process values,” a criteria that focuses on processes that keep democracy strong or give people opportunity to take part in the political process. This can be different than an efficiency or effectiveness analysis, but can nonetheless be helpful for policymakers who are interested in understanding if a policy will further participation and healthy democratic goals.
Bardach goes on to talk about how criteria relate to one another, making the claim that some evaluative criteria deserve more weight than others. The analyst can deal with this problem in a few ways. One way is for her to allow the political process to take care of it, with policymakers deciding how to weigh different criteria against one another. Another option is for the analyst to impose a solution, focusing on criteria that are overlooked in the political sphere so a new criteria and people served by that criteria are supported by good analysis. Another is to focus on the topic of “rights,” though Bardach is critical of this approach as it leaves little space for consensus to be achieved.
Beyond core criteria, practical criteria have a place in some policy analysis. One such criterion is legality, or whether a policy would conflict with a constitution or law superseding the level of the policy analysis. A second is political acceptability, or how much opposition or support a policy would attract. Another is administrative robustness: how easily will a policy be implemented with available or possible bureaucratic structure? Last, Bardach talks about political sustainability, or how long a policy is likely to survive in the long-term political climate.
Bardach also talks about criteria as logical constructs. This could mean minimizing or maximizing a criterion for policy analysis. For instance, if a policy analysis is built on the problem definition of “too many people are homeless in Cleveland, Ohio,” a criteria could be “minimization of homelessness,” with policies that minimize homelessness the most deemed fulfilling this criteria best. Another is what Bardach calls “linear programming,” or trying to maximize a criteria based on certain goals. For instance, “minimize homelessness spending no more than $2 million and weighting child homelessness at twice the level of adult homelessness.”
Analysts should also specify metrics, ideally using quantitative, specific metrics rather than qualitative, broad metrics. The goal here is to reduce confusion and philosophical argument and to shift the conversation toward empirics.
Bardach warns against confusing alternatives with criteria. Don’t let your criteria creep into your alternatives and then presuppose the effectiveness of a given alternative. Remember: alternatives are policy options someone can carry out, criteria are the values we use to assess them.