Public Compliance with and Deviance from Government Policies: The Final Frontier of Policy Implementation
R. Kent Weaver Georgetown University & the Brookings Institution The Starting Point: Most government policies are not “self-implementing” ̶ they require actions by the public, including:
• Purchase of more fuel- efficient vehicles to reduce greenhouse gas emissions • Use of community latrines to improve sanitation • Use of family planning to reduce population growth And sometimes by multiple targets • increase in condom use by commercial sex workers (CSWs) and their clients • Increased employment by single mothers requires cooperation from employers Compliance defined:
• Behavior that is consistent with government preferences that have been clearly articulated (though perhaps not effectively communicated) • Whether willing, grudging or coerced What is Compliance?
Level of Compliance
Low Medium High
High Illegally non- Fully compliant Pseudo- compliant compliant (Legally) abusing Government’s tax loopholes insistence on Acceptably Compliance deviant (Incentives & Modest violations sanctions + Ignoring voluntary of sped limit monitoring + tsunami evacuation order enforcement)
Smoking at home Inconsistent Partially consistent Fully consistent Feeding junk food to Low children The Big Question:
Why do program “targets” sometimes act the way that program designers intend and want (compliance) and sometimes fail to do so (deviance)?
…especially (but not exclusively) when compliance seems to be in their interests
For example…. • Why did many New Orleans residents not evacuate before Hurricane Katrina hit? • Why do people continue to smoke after exposure to graphic anti- smoking campaigns? • Why do parents sometimes reject vaccinations for children? • Why don’t parents send children to school even when it is free or inexpensive? Sometimes citizen compliance is very high
….as with the ban on smoking in Paris cafes ….and Irish pubs
G.T. Fong et al, “Reductions in tobacco smoke pollution and increases in support for smoke-free public places following the implementation of comprehensive smoke-free workplace legislation in the Republic of Ireland: findings from the ITC Ireland/UK Survey,” Tobacco Control, 2006 Sometimes compliance is fairly high…
Source: Ronald Clarke, The Distribution of Deviance and Exceeding the Speed Limit,” British Journal of Criminology, 1996, p. 172 Sometimes compliance is moderate and “normally distributed”….
Source: Ronald Clarke, The Distribution of Deviance and Exceeding the Speed Limit,” British Journal of Criminology, 1996, p. 175 …or fairly low and “normally distributed”
Source: Ronald Clarke, The Distribution of Deviance and Exceeding the Speed Limit,” British Journal of Criminology, 1996, p. 176 And sometimes it seems like compliance is the exception…
….as with parking rules in Rome
…or jaywalking in Wellington! Sometimes there is great variation in compliance: • Why do individuals sometimes engage in looting after a disaster or when social order breaks down, and sometimes refrain? • And why do Chileans pay their taxes when Argentines don’t? Sometimes initial compliance increases over time…
Edwards et al, “After the smoke has cleared: evaluation of the impact of a new national smoke-free law in New Zealand,” Tobacco Control, 2008 …as with the drop in complaints about violations of New Zealand’s smoking ban Sometimes initial compliance is high and then drops off…
PERCENTAGE OF NEW PPM ENTRANTS MAKING ACTIVE FUND CHOICE, 2000-2007
80 67.0 70 60 50 40 30 18.0 14.0 20 8.4 9.4 7.48.0 10 1.6 0 2000 2001 2002 2003 2004 2005 2006 2007 …as with active choice of pension funds by workers in the new Swedish pension system Sometimes compliance is so high… …that the initial policy objective is “overshot”…
Source: Sharon M. Lee, Gabriel Alvarez, J. John Palen, “Fertility Decline and Pronatalist Policy in Singapore,” International Family Planning Perspectives, Vol. 17, No. 2, (Jun., 1991), pp. 65-73 …and policies have to be reversed (but with little success) Some obvious contributors to non- compliance on the citizen-side:
“Habitual non-compliers” • Career criminals • Mentally ill • Addicted • Ideologically-driven • Teenagers
“Situational non- compliers” • Intoxicated Some obvious contributors to non- compliance on the government side:
• Government preferences may be: – not be clearly transmitted – Inconsistent across agencies • Monitoring and enforcement may be inconsistent The main arguments:
• Inadequate incentives, monitoring and enforcement aren’t the only barriers to compliance (although they are important) • Target heterogeneity is an important cause of non-compliance Bounded rationality and Choice architecture The “rational actor” perspective on target compliance: Program targets respond rationally to: • Information • incentives/sanctions • monitoring enforcement
A Normative Approach: Penalty = Damages / Probability of Detection
Gary Becker (1968), “Crime and Punishment: An Economic Approach,” Journal of Political Economy. Policy implications of the “rational actor” perspective on target compliance: • Get the incentives and sanctions right • Monitor compliance consistently • Sanction non-compliance reliably (but not necessarily severely) The “bounded rationality” perspective on target compliance argues that program targets: • have limited capacity to process information • make biased forecasts • are susceptible to variations in framing • are loss averse • heavily discount future benefits The “bounded rationality” perspective on target compliance:
Policy implications: • Structure options in ways that will skew choices toward socially desirable outcomes • Frame options in ways that will elicit desired response PERCENTAGE OF NEW PPM ENTRANTS MAKING ACTIVE FUND CHOICE, 2000-2007
80 67.0 • And don’t make 70 60 options too 50 complicated, or 40 30 18.0 14.0 people will 20 8.4 9.4 8.0 7.4 10 1.6 “choose” the 0 default option or 2000 2001 2002 2003 2004 2005 2006 2007 status quo NUMBER OF FUND OPTIONS IN PPM SYSTEM, 2000-2006
800 705 664 700 625 648 600 558 465 500 400 300 200 Number of funds of Number 100 0 2000 2001 2002 2003 2004 2005 Year Why Don’t People Comply?:
Eight Sources of “The Compliance Gap” A Model of Target Compliance and Non- Compliance
Stage 1 Stage 2 Stage 3 Stage 4
Determinants of Official compliance compliance regime (e.g, incentives Feedback loop regime (e.g., and sanctions) culture)
Determinants of Implementation of Individual Aggregate implementation compliance regime target levels of (e.g., bureaucratic (e.g., monitoring & compliance target capacity) enforcement behavior compliance
Determinants of Non-regime constraints on target capacity and individual target capacity willingness to and willingness to comply comply (e.g., income (e.g., resources, attitudes, distribution) peer effects Eight explanations of the “Compliance gap”: Incentive and sanction problems—positive and/or negative incentives are insufficient to ensure compliance Monitoring problems—target compliance may be difficult or costly to monitor Enforcement problems—sanctions for non- compliance are not implemented consistently Resource and access problems—targets lack resources and access needed to comply even if they want to Eight explanations of the “Compliance gap” (cont.): Autonomy problems—targets may want to comply but lack control over their own decisions Information and framing problems—targets lack information or interpret it in ways that makes compliance less likely Attitude and objectives problems—targets are hostile/mistrustful toward providers or programs Peer effect problems—non-compliance by those around targets makes them less likely to comply Incentive Problems may include: 1. Opportunity costs of compliance may be high—e.g., child labor 2. Perverse effects of poorly designed programs: Recipients of targeted wage subsidies stigmatized recipients Common responses to incentive and sanctions problems— 1. Employ more coercive strategies—moving up the hierarchy of: – Prohibiting and punishing non- compliant behavior – Imposing sanctions for non- compliance – Providing positive incentives for compliance – Manipulating choice architecture – Admonition – Providing information – Providing resources …from information… …to admonitions… …to incentives…
including conditional cash transfers to encourage school attendance, especially for girls …to prohibitions and requirements with penalties attached …or are used in combination 2. Strengthen existing incentives or sanctions-- e.g., increase taxes on gasoline and cigarettes to discourage their use But strengthening incentives and sanctions can create its own compliance problems, e.g., • Encourage costly avoidance behavior (e.g., smuggling to avoid cigarette taxes) • Be very costly in terms of resources • Pay a lot of people to do things that they would have done anyway (i.e., poor targeting) Monitoring problems—High rates of target compliance are unlikely where compliance is difficult or costly to monitor
• like polio vaccinations • or income of street vendors …or condom use or mandatory HIV testing by commercial sex workers and their clients—especially when the sex industry is decentralized and segmented
See: Wiwat Rojanapithayakorn “The 100% Condom Use Programme in Asia,” Reproductive Health Matters, vol 14, no, 27 (2004) Pages 41-52 and Amit Chattopadhyay, Rosemary G. McKaig, “Social Development of Commercial Sex Workers in India: An Essential Step in HIV/AIDS Prevention,” AIDS Patient Care and STDs, Vol. 18, No. 3 (March 2004), pp. 159-168 Common responses to monitoring problems—
• Find monitoring mechanisms that are cheaper, more reliable, less obtrusive and pose lower costs/risks on compliant targets But increased monitoring
• Is often very costly • And invasive of privacy • and may spark concerns about racial and ethnic profiling Enforcement problems— FLWs have may have incentives not to enforce because enforcement is: • dangerous (“plata o plomo” for Mexican drug enforcers) • or unpleasant (e.g., East German border guards who developed “poor aim”) • or front-line workers can gain from selective enforcement Common responses to enforcement problems—
• Ensure that sanctions or incentives are always applied by removing the “human factor”—e.g., traffic cameras at red lights, and tripwires at the Berlin Wall • Or increase the visibility and (perceived) certainty of enforcement Resource problems: High rates of compliance are unlikely where targets lack resources needed to comply even if they want to Resource shortages may include:
• Money income • Human capital • Financial and physical assets (e.g., car, home) • Physical and mental health • Social networks that provide: – access to employment opportunities, services, etc. – Role models of appropriate behavior • Poor physical access to or distance from jobs, public infrastructure (schools, public transportation), etc. Example: Many people didn’t leave New Orleans and coastal Mississippi before Hurricane Katrina’s landfall: • Many poor residents and institutionalized populations lacked private transportation (and Louisiana did not provide it) • At end of month, many more residents lacked resources to finance evacuation Multiple resource barriers increase the odds of “non-compliance”– e.g.: • Single mothers trying to make transition from welfare to work often burdened by more than one of: – low skills and low educational attainment – Lack of access to transportation and affordable child care – domestic violence – health and mental health problems
Sandra Danziger et al, “Barriers to the employment of welfare recipients,” in R.Cherry & W.M. Rodgers III (Eds.), Prosperity for all? The economic boom and African Americans, New York: Russell Sage Foundation, 2000. Common response to resource problems— provide resources needed to facilitate compliance… which is harder when: • Resource needs of clients are diverse • Compliance is costly and complicated for targets • Compliance stretches over time As in welfare-to-work transitions Autonomy problems: High rates of compliance are unlikely where targets’ actions are controlled by: • others with a stake in non- compliance, e.g. – Attendance at girls’ schools in Afghanistan (affected by Taliban) – Condom use by CSWs and their clients (affected by pimps and brothel owners) – Provision of information on crimes by individuals who fear retribution from criminals • Substance abuse (e.g., drug abuse, nicotine and alcoholism) Common responses to autonomy problems— • Weaken power of “secondary targets” over primary targets (e.g., weaken Taliban, witness protection programs) • Provide information to primary and secondary targets that compliance is in their interests • Increase incentives and sanctions for primary and secondary targets (e.g., rewards for information on crimes) Information problems—High rates of compliance are unlikely where • It is unclear to targets what constitutes compliance (e.g., vague goals for reducing energy usage) • targets lack information that would make compliance more likely (e.g., on HIV transmission mechanisms) Common response to information problems— general or targeted information (and sometimes admonition) campaigns…. …often with an emphasis on negative framing (potential losses from non- compliance) rather than positive framing (emphasizing advantages of compliance) • or asking residents to mark Social Security numbers in Hurricane Katrina evacuation
Although meta-analyses do not consistently show that negative framing is more effective than positive framing Problems with information strategies: • Information about risks may be uncertain, especially for rare events— and “crying wolf” increases resistance Attitude and beliefs problems: High rates of target compliance are unlikely where (1): • compliance contradicts deeply held cultural beliefs (e.g., strong desire for sons, religious objections to medical treatment, low cultural value on girls’ schooling) • targets are hostile toward or mistrust providers, programs or government (e.g., vaccination in northern Nigeria) Peer effect problems: High rates of target compliance are unlikely where:
• Injunctive norms—what people generally see as the right thing to do conflict with • Descriptive norms (what people usually do)
Example: “No one else pays taxes, why should I?”
Robert B. Cialdini “Crafting Normative Messages to Protect the Environment,” Current Directions in Psychological Science, Vol. 12, No. 4 (Aug., 2003), pp. 105-109 The “Broken windows” hypothesis: visible manifestations of non- compliance will lead to increased scope and severity of non-compliance
George Kelling and James Q. Wilson, “Broken Windows,” Atlantic Monthly, March 1982 Another peer effect: “Herd immunity” problems: if costs/risks of compliance are non-trivial, targets may fail to comply if they think they will be protected anyway Lessons for Policy Designers and Program Implementers
Are there “smarter” ways to obtain target compliance? 1. Don’t assume away the problem of target compliance • Many problems in target compliance can be anticipated in advance and incorporated into the design phase of policies, but • This is unlikely to happen unless it is done consciously and explicitly • Policymakers need to be convinced that policy implementation is not “someone else’s problem” 2. Invest “upfront” in understanding potential non- compliance issues • Use pilot projects when possible (but make sure the conditions are realistic) • Engage front-line workers and potential clients in understanding potential compliance issues • Analogize from other programs, but beware of how different conditions may lead to different outcomes • Don’t let a proposal’s momentum cause you to ignore what you have learned about potential compliance problems 3. Incentives, monitoring and enforcement are important determinants of levels of target compliance , but they are not the only determinants…. …so you need to think comprehensively about potential causes of non-compliance Incentive problems Monitoring problems Enforcement problems Resource problems Autonomy problems Information and framing problems Attitudes and beliefs problems Peer effect problems …and make sure you have the right diagnosis, which may be multi-causal… …as well as appropriate packages of “prescriptions”
Providing information Admonition Manipulating choice architecture Providing resources Providing positive incentives for compliance Imposing sanctions for non-compliance Prohibiting and punishing non-compliant behavior
that are targeted effectively 4. Think about ways to manipulate choice architecture to make compliance easier… For example Kiwisaver: • Has automatic enrollment unless worker opts out • Default fund are provided • Incentives and tax incentives provided …but choice architecture may be difficult to change once it is in place 5. Do not assume that your target clientele is homogeneous. When necessary, develop differentiated strategies for specific “sub- targets” who may need different “packages” of incentives, information, monitoring, etc. to comply, e.g., • welfare leavers • long-haul truck drivers who use services of commercial sex workers) So remember that getting compliance from the last 10% of targets • May be extremely expensive • May require a distinctive set of strategies rather than “more intensity of the same” strategies 6. Think creatively about “secondary targets” and how they may be useful in improving overall compliance levels 7. Try to anticipate target “situational imperatives” that will lower prospects for target compliance, e.g.: • Desperate family poverty in LDCs may lead to child labor rather than school attendance • Personal security concerns where drug violence is high will interfere with cooperation with police 8. Link your strategy for increasing compliance to the underlying reason for non-compliance, e.g.: Reason for Non-Compliance Strategic Response
• Advantages to targets of • Public information compliance are high and strategy; Change costs of compliance are low framing and targeting of information
• Non compliance with • Increase monitoring immunization or quarantine and sanctions for for diseases with high non-compliance mortality rates due to ‘herd immunity” beliefs 9. Analyze how much compliance is “good enough”—and by whom— for the program to be a success • Identify the most “mission critical” elements of target non-compliance and develop strategies, rules and informal practices to deal with them …and consider exempting groups for whom compliance is most problematic—e.g., “reasonable accommodation” for minority groups 10. Link your degree of insistence and strategy for increasing compliance to the severity of consequences of non- compliance, e.g.: • Strong case for compulsion in quarantining highly contagious and fatal diseases like SARS; H1N1 is less clear 11. Look for “leverage points” that: • reduce targets’ compliance costs • reduce monitoring costs or • simplify program delivery
for example
• Withholding income taxes
or Requiring that iodine be added to table salt Plus a massive public information campaign Providing free mosquito nets to mothers who get their children immunized …But also be conscious of potential challenges and risks in using leverage points • It may require developing partnerships with entities that do not see themselves as agents of government • Programmatic “silos” may need to be overcome • Implementation likely to require cooperation of front-line workers • Excessive leverage may put clients at risk 12.Have backup strategies
• When the primary strategy is problematic and the cost of non- compliance is high 13. Don’t forget that widespread non- compliance may indicate a problem with the policy rather than with the targets of policy
Number 27 September 2009
Target Compliance: The Final Frontier of Policy Implementation R. Kent Weaver
ABSTRACT urprisingly little Recent Issues in Governance Studies St heoretical attention has been “Blogs as Public Forums for Agency Policymaking” devoted to the (August 2009) final step of the public policy “The New Digital Press: How to implementation Create a Brighter Future for the News Industry” chain: (May 2009) understanding why the targets of “Center-Left America?” (April 2009) public policies do or do not “The Impact of Milwaukee “comply” — that is, © Reuters/Kevin Lamarque – Former U.S. President George W. Bush signing a bill. Charter Schools on Student behave in ways that are Achievement” (March 2009) consistent with the objectives of the policy. This paper focuses on why program “targets” frequently fail to act in the way that program designers intended and wanted, even when it appears to be in their self-interest to do so. This is a To view previous papers, visit: www.brookings.edu/governance/Issues- question that can be asked for a broad array of government policies. The paper in-Governance-Studies.aspx begins by briefly defining target compliance and the main approaches that governments may use to secure compliance. It then discusses the dominant theoretical approaches to understanding target compliance, and develops a general categorization of reasons for compliance or non-compliance and strategies that are generally used to cope with each of these causes. Finally, the author suggests some lessons about what program designers and implementers should do to address problems of policy non-compliance.
Target Compliance: The Final Frontier of Policy Implementation 1
Introduction
Voters elect governments to solve social problems. Governments design and implement a huge array of programs to ensure the public good. A sizable literature has developed suggesting that problems in program implementation are a major source of poor government performance, ranging from inadequate coordination between agencies and levels of government to front-line workers who disagree with the program and implement it with less than total enthusiasm. Surprisingly little theoretical attention, however, has been devoted to the final step of the implementation chain: explanations of why the targets of public policies do or do not “comply” with those policies – meaning that they behave in ways that are consistent with the objectives of the policy. Very few policies are “self-implementing” – that is, a declaration of policy requires no further actions to bring about the desired changes in policy outputs and societal outcomes. And most policies that are not self-implementing require actions by a broad array of ordinary citizens, corporations, or other actors if they are to achieve their objectives.
R. Kent Weaver is a senior Program “targets” frequently fail to act in the way that program designers fellow with Governance intended and wanted, even when it appears to be in their self-interest to do so. Studies at The Brookings For example, people continue to begin to smoke, even in countries where there Institution and a professor are well-organized anti-smoking programs, including ghastly pictures on at the Public Policy Institute at Georgetown cigarette packages prominently displaying the ill effects of smoking. Many University. residents of New Orleans and surrounding areas did not move to safer areas even after a mandatory evacuation order was issued before Hurricane Katrina struck in August 2005. Single mothers in the United States vary greatly in the degree to which they have responded to increased work incentives and work requirements by obtaining and retaining employment. Compliance with tax laws is much higher in Chile than in neighboring Argentina. Compliance with government policies also varies tremendously across policies. In some cases, it seems that compliance with government policies is hardly observed at all – think of parking regulations in Italy, for example – but in other cases, the compliance rate for government policies is surprisingly high, as with the imposition of a smoking ban in Irish bars and restaurants in March 2004 and in French restaurants and cafes in February 2008. Whether individuals or other actors outside government comply with government policy can all be said to depend on six broad factors. The importance of specific factors in influencing compliance or non-compliance will vary from policy sector to sector. In examining the roots of compliance and non- compliance, I will begin by briefly defining target compliance and the main approaches that governments may use to secure compliance. I then discuss the
Target Compliance: The Final Frontier of Policy Implementation 2
dominant theoretical approaches to understanding target compliance and develop a general categorization of reasons for compliance or non-compliance, and strategies that are generally used to cope with each of these causes. Finally, I suggest some lessons about what program designers and implementers should do to address problems of policy non-compliance.
Analyzing Barriers to Target Compliance
What is target compliance? The first contemporary definition of “compliance” offered by the Oxford English Dictionary is “acting in accordance with, or the yielding to a desire, request, condition, direction, etc.” Compliance may, but need not, involve willing agreement to behave in this way: grudging compliance is still compliance. Equally important, there are many gradations in how specific and insistent governments are in their degree of “direction” they give to the targets of policy, in the perceived stakes of compliance, and in the degree of compliance that is received. Sometimes governments simply admonish compliance with their policy objectives. Governments also use incentives to try to obtain compliance with government policies – requiring deposits on soft drink bottles and cans to encourage recycling, for example. Governments may also prohibit, regulate, or require specific behaviors, with penalties attached for non- compliance. These can take many forms, such as zoning laws and indoor smoking bans. Target compliance has both individual and aggregate elements. In both cases, the boundaries between “adequate” levels of compliance and inadequate levels of compliance that indicate or lead to policy failure are often unclear. In sectors where a specific standard for individual compliance needs to be set and it is unclear that there is a clear boundary between acceptable and unacceptable behavior, a modest level of non-compliance may be tolerated. In many jurisdictions, for example, driving a few miles per hour over the speed limit is unlikely to draw a speeding ticket. Formal rules are more likely to be adhered to, substantial resources are more likely to be devoted to monitoring and enforcement, and severe sanctions are more likely to be imposed in sectors where even very rare acts of non-compliance pose an unacceptable risk to policy objectives, such as the enforcement of airport security rules on bringing potential weapons or explosives onto airplanes. In short, there is no single, general answer to the question “What constitutes a compliance problem?” There are some groups (career criminals, for example), that are likely to be frequent non-compliers with government policies or almost any set of externally- imposed rules. But in what follows, the focus will be on people who are mostly compliers, as well as situations in which the “direction” being given (i.e.,
Target Compliance: The Final Frontier of Policy Implementation 3
information on what constitutes compliance) is reasonably clear and the stakes of compliance or non-compliance are perceived by those implementing the policy to be consequential.
Theoretical Perspectives on Compliance and Non-Compliance Policymakers and Two paradigms have dominated analyses of why targets comply or do not implementers comply with government policies. The first is the “rational actor” perspective. should structure This perspective sees program targets as responding rationally to incentives they face in a way that maximizes their self-perceived utility. The main policy options in ways implication of this approach is that it is critical to get the structure of incentives that will skew and sanctions right and to monitor and enforce compliance with the policy to choices toward ensure that compliance is appropriately rewarded and that non-compliance punished. The incentive structure should be tweaked occasionally to take socially desirable account for changes in behavior and changing public objectives. outcomes. The second common perspective on target compliance flows from the relatively new discipline of “behavioral economics,” and it can be seen in works like the recent book by Richard Thaler and Cass Sunstein, Nudge. Thaler and Sunstein argue that program targets have a limited capacity to process information, and that biases built into their projections of payoffs from various choices may lead them to make choices that are not optimal, either in terms of their own self-interest or that of society. The policy implications of the behavioral economics approach are several. First, policymakers and implementers should structure options in ways that will skew choices toward socially desirable outcomes. If it is desirable to increase voluntary retirement savings, for example, it would be wise to change the default by enrolling employees automatically in employer-administered voluntary retirement savings plans and requiring individuals to opt out, rather than making non-participation the default option. Secondly, instead of trying to maximize the number of options from which policy targets can choose, it is important to ensure sure that policymakers do not make options too numerous or complicated, or people will “choose” the default option, which is frequently the status quo. Third, individuals are susceptible to what a variety of social scientists call “framing effects” that emphasize the potential costs of non- compliance: in two communities threatened by Hurricane Katrina, for example, first responders found that asking holdouts from a mandatory evacuation order to write their Social Security numbers somewhere on their body with a permanent marker so that it would be easier to identify their corpses after the hurricane hit worked well in persuading them that the threat was severe and that they should leave the area.
Target Compliance: The Final Frontier of Policy Implementation 4
Sources of Compliance and Non-Compliance
Both the rational target and behavioral economics approaches have much to say that is useful. However, neither approach considers a sufficiently broad range of factors that affect policy compliance to serve as an adequate guide to policy designers and implementers. The seemingly disparate instances of the “target compliance gap” discussed above can, in fact, be understood as resulting from six broad underlying sets of factors: • Incentive and sanction problems where positive and/or negative incentives are insufficient to ensure compliance;
• Monitoring problems where target compliance may be difficult or costly to monitor;
• Resource problems where targets lack the resources to comply even if they want to; Policymakers who • Autonomy problems where targets do not have the power to make are considering decisions that comply with policy even if they want to; increasing • Information problems where targets lack information that would make incentives and compliance more likely; and enforcement need • Attitude and objectives problems where targets are hostile/mistrustful to consider the toward providers or programs. ways in which it I will look briefly at each of these in turn. may lead to need Incentives and Sanctions to new, and As suggested above, high rates of target compliance are unlikely where potentially more positive incentives and/or negative sanctions are not sufficient or certain enough harmful, forms of to ensure compliance. Low gasoline taxes are unlikely to cause people to stop non-compliance. buying sport utility vehicles, especially in places with long winters and lots of snow. The appropriate policy response to incentive and sanction problems is usually to strengthen those incentives or sanctions and ensure that they are enforced consistently rather than haphazardly. However, the assumption that “stronger incentives equals greater compliance” is not always correct: an obvious case is the U.S. experience with prohibition on sale of alcoholic beverages in the 1920s. At a minimum, policymakers who are considering increasing incentives and enforcement need to consider the ways in which it may lead to need to new, and potentially more harmful, forms of non-compliance.
Target Compliance: The Final Frontier of Policy Implementation 5
Monitoring High rates of target compliance may also be hard to achieve where compliance is difficult or costly to monitor. Monitoring is likely to be especially problematic where the activities involved are illegal, take place in private, or both. Both the privacy and illegality barriers to monitoring can be seen in government policies in many countries to encourage 100 percent condom use by commercial sex workers and their clients. The common responses to monitoring problems is to find monitoring mechanisms that are cheaper, more reliable, less obtrusive and pose lower costs/risks on compliant targets. Most monitoring mechanisms are not without risks, however, especially in relation to privacy concerns.
Resources Program targets may also lack the resources that they need to adapt to a policy, even if they want to comply and recognize the advantages of doing so. The kinds of resources that facilitate compliance with public policy may be diverse; these resources not only include cash assets, but also things like good health, human capital, strong social networks, and the ability to draw easily on existing public infrastructure. Many poor residents and institutionalized populations in New Orleans and nearby areas lacked private transportation to leave New Orleans before Hurricane Katrina struck, and the city of New Orleans and state of Louisiana did not provide it. Moreover, the hurricane struck at the end of August, and many individuals, especially those dependent on monthly government transfer payments, lacked the resources needed to support their families during an evacuation. Transitions from welfare to work have been particularly difficult for single mothers suffering from multiple resource barriers, such as low education, domestic abuse problems, and poor access to public transportation or a private vehicle.
Autonomy Problems Targets may also be non-compliant because they lack autonomy over their decisions and thus are unable to comply, even if they would prefer to do so. For example, children in developing countries are likely to lack autonomy in decisions over whether to attend school or engage in child labor. Victims of domestic violence may not seek medical attention or report the assaults they experience due to the threat they perceive at home. Addressing this type of non- compliance may involve (1) influencing the party with the decision-making power, or (2) empowering the target so they can gain autonomy over their actions.
Target Compliance: The Final Frontier of Policy Implementation 6
Information problems Another possible barrier to compliance is that targets of a policy lack information that, if they did possess it, would make them more likely to comply. It may be unclear to targets what constitutes compliance, as when governments have only vague goals for reducing energy usage, or what the stakes of compliance are. The most common response to information problems is to implement information campaigns. These campaigns can take many forms. Graphic warning labels on cigarette packages remind purchasers of the dangers of smoking. Large stickers on the windows of new cars that give their estimated gas mileage remind potential purchasers of the future environmental and economic consequences of their choices.
Attitudes and Beliefs Attitude and beliefs constitute a very broad set of influences on target compliance. They include beliefs concerning both the legitimacy of the policy itself and the government that imposes or enforces the policy. Thus compliance is likely to be higher when non-compliance is seen as socially unacceptable. As noted earlier, compliance with indoor smoking bans that have been enacted in many countries and sub-national jurisdictions in recent years has generally been very high, including in French cafés and Irish pubs where cultural stereotypes might suggest that this would not be the case. In jurisdictions with high rates of compliance, smokers appear to have accepted their “pariah” status, and accepted the argument that they should not inflict second-hand smoke on others.
Lessons for Policy Designers and Implementers
If policymakers and program implementers gain a better understanding of the causes of non-compliance, what lessons they can draw that may improve governmental performance? The first and most important lesson is that when policymakers are designing new public policies or revising old ones, they should not assume that they know how the targets of policy will react and that compliance will be automatic and universal. Compliance is instead an issue that must be explicitly addressed when designing policies and implementation strategies. An analysis of potential barriers to target compliance that draws on the set of factors examined here should be an important part of the initial policy design process, rather than going back to try to rectify problems later. But how can policymakers learn about programs that do not exist yet? One way is to look at experiences of other programs in the same political jurisdiction, or similar jurisdictions, that serve a
Target Compliance: The Final Frontier of Policy Implementation 7
similar clientele and have similar program characteristics. Another strategy for learning about potential implementation problems is to use pilot programs to get more detailed information on likely compliance rates. However, this option is expensive and is likely to conflict with politicians’ desires to get a new program implemented broadly as fast as possible so that they can claim credit for it in the next election. Focus groups of current front-line workers and potential clients Policymakers can also be useful in identifying potential compliance trouble spots. should not assume A second lesson is that it is important to think comprehensively about the causes of target non-compliance and how to address those causes. There is a that they know how tendency to think that if policymakers have addressed the most glaring causes of the targets of policy non-compliance – for example, getting incentive structures right – then that is all that needs to be done. Less obvious causes of non-compliance, such as lack of will react and that client resources or opportunity costs of compliance, may be ignored. It is compliance will be important for the government to make sure it has the right diagnosis, which may be multi-causal. automatic and A third lesson is that policymakers should think about ways to make universal. compliance with government policies easier. Here the principles of behavioral economics are particularly useful, notably in the ideas that choices should not be too complicated and that the default option should be one that is consistent with the government’s perception of socially-desirable outcomes. A fourth lesson is that the targets of policy sometimes face what James Q. Wilson refers to as “situational imperatives,” which are conditions that overwhelm any other considerations in a target group’s decision on whether to comply with policy. Sometimes these situational imperatives substantially lower the prospects for target compliance. For example, families in developing countries who are desperately poor are more likely to have their children engaged in child labor or in household labor, rather than attend school. Governments need to develop focused strategies to respond to situational imperatives that put target compliance, and policy success, at risk. Fifth, it is important to not assume that all targets of a policy are homogeneous and that strategies to secure compliance from the modal client will work for all. An effective analysis of target compliance should: (1) consider impacts of policy incentives, monitoring, resources, etc. on specific sub-groups of targets, (2) identify potential causes of target non-compliance and consider strategies to deal with them, and (3) analyze impacts on unintended targets and develop strategies to mitigate them, if necessary. Sixth, in cases where resources are limited, which is almost always, it is important to analyze how much compliance is “good enough,” and by whom, for the policy to be a success. It is vital to link the strategy for increasing compliance to the severity of consequences of non-compliance.
Target Compliance: The Final Frontier of Policy Implementation 8
A seventh lesson is that policymakers should be on the lookout for “leverage points” where they can reduce targets’ compliance costs, reduce monitoring costs, and/or simplify program delivery. Withholding of income taxes by employers and reporting and transfer of amounts withheld to government is a good example of a use of a leverage point that is so deeply institutionalized that we do not even think about it. An eighth overall lesson is that having backup strategies can be very important in securing compliance, especially when the primary strategy is problematic and the costs of non-compliance is high. A clear example is the multiple systems used to prevent additional terrorist hijackings of aircraft by even the most determined non-compliers. In addition to intensified airport screening of passengers, cockpit doors have been reinforced and air marshals put on some flights. A ninth lesson suggested by this discussion is that it is important to use framing to increase compliance with policy. The dire framing noted above to get residents in the path of Hurricane Katrina to comply with an evacuation order is a particularly striking example. In most cases, of course, framing compliance as a choice that is literally between life and death is not credible, and an overly dramatic framing may cause government authorities to lose credibility. However, sometimes life-or-death framing is appropriate, as in the case of cigarette warnings and HIV, and governments should not be afraid of using it. A final lesson, and this may be the hardest lesson for policymakers to learn, is that widespread failures of policy compliance may signal that there is something wrong with the policy, rather than that something is wrong with the targets who are being uncooperative by failing to comply with it. Policy experts tend to think that they know best, that they possess all the relative information, and the policy that they have come up with is the best option, given myriad constraints, including political constraints. It can be both frustrating and difficult for policymakers to comprehend why the targets of policy do not see things the same way they do. But history is replete with examples of situations in which policymakers in the past felt that they “knew better” about policies that we now regard as fundamentally wrong-headed. Policymakers must, therefore, listen to and learn from the “targets” of public policy – who are, after all, also citizens. It is critical to listen to both what they say and what they do. Failure to listen can keep governments from learning quickly what mistakes of omission or commission (or both) policymakers have made and correcting those mistakes. Failure to correct policy mistakes is likely to be a recipe for policy disaster, either in the near-term or later.
Target Compliance: The Final Frontier of Policy Implementation 9
Author
R. Kent Weaver is a senior fellow in the Governance Studies program at the Brookings Institution and Professor of Public Policy and Government Governance Studies The Brookings Institution at Georgetown University. His major fields of interest and expertise are 1775 Massachusetts Ave., NW American social policy, comparative public policy, and comparative Washington, DC 20036 Tel: 202.797.6090 political institutions. He would like to thank Julie Fitzgerald and Mahreen Fax: 202.797.6144 Hasan for their research assistance in preparing this Issues in Governance www.brookings.edu/governance.aspx Studies. Editor Christine Jacobs
Production & Layout John S Seo
Tell us what you think of this Issues in Governance Studies.
Email your comments to [email protected]
This paper from the Brookings Institution has not been through a formal review process and should be considered a draft. Please contact the author for permission if you are interested in citing this paper or any portion of it. This paper is distributed in the expectation that it may elicit useful comments and is subject to subsequent revision. The views expressed in this piece are those of the author and should not be attributed to the staff, officers or trustees of the Brookings Institution.
Target Compliance: The Final Frontier of Policy Implementation 10
References
Bergman, Marcelo S. (2003) “Tax Reforms and Tax Compliance: The Divergent Paths of Chile and Argentina,” Journal of Latin American Studies, vol. 35, no. 3, 593-624
Danziger, Sandra K., Mary Corcoran, Sheldon Danziger, Colleen Heflin, Ariel Khalil, Judith Levine, Daniel Rosen, Kristin Seefeldt, Kristine Siefert, and Richard Tolman (2000). Barriers to the employment of welfare recipients. In Robert Cherry & William M. Rodgers III (Eds.), Prosperity for all? The economic boom and African Americans. (pp. 239-272) New York: Russell Sage Foundation.
Edwards, Mark A. (2006). Law and the parameters of acceptable deviance. Journal of Criminal Law and Criminology, 97(1), 49-100.
Grindle, Merilee S. (2007). Good enough governance revisited. Development Policy Review, 25(5), 533–574.
Kahneman, Daniel and Amos Tversky. 1981. “The framing of decisions and the psychology of choice.” Science, 211(4481): 453-458.
May, Peter J. (2004). Compliance motivations: Affirmative and negative bases. Law & Society Review, 38 (1), 41-68.
Thaler, Richard H., and Cass Sunstein. 2008. Nudge:Improving Decisions About Health, Wealth and Happiness. New Haven: Yale University Press.
U.S. Senate Committee on Homeland Security and Governmental Affairs (2006). Hurricane Katrina: A Nation Still Unprepared (S. Rpt. 109-322). Washington, D.C.: U.S. Government Printing Office.
Wilson, James Q. (1989). Bureaucracy: What government agencies do and why they do it. New York: Basic Books.
Winter, Soren C., & Peter J. May (2001). Motivation for compliance with environmental regulations. Journal of Policy Analysis and Management, 20(4), 675-698.
Target Compliance: The Final Frontier of Policy Implementation 11