33362 World Bank Social Safety Nets Primer Notes 2003 No. 12 Public Attitudes Matter: Political Economy in the Design of Safety Nets Policies It has long been recognized that political econ- generational mobility do not appear to affect omy influences the design, implementation and views on income redistribution. outcome of safety nets, as well as other social programs, but there is no consensus about how When asked what their country needed most to to account for such concerns in policy deci- get ahead--productivity growth or more redistri- sions. This note draws on international public- bution--more than half of survey respondents opinion surveys and other research to highlight region-wide say productivity. Within countries, some of the implications for the design of safe- wealthier people are, on average, more likely to ty net policies. support productivity. However, across countries, mean levels of support for productivity are high- A Cross-Country Perspective er in poorer and more unequal countries (see Public opinion differs widely across countries figure). These apparent contradictions can be about government's role in the provision of explained in two ways. First, most poor coun- public services and about the extent to which tries in the region have relatively recently start- there is a collective responsibility for individu- ed serious market reforms and most citizens rec- als who are unable to provide for themselves. ognize the need to establish sustainable growth There are persistent differences in attitudes given the history of high debt, inflation and about how opportunities are distributed and stagnant economies. Second, these same coun- the government's role in equalizing both oppor- tries typically have weak state institutions and tunities and outcomes. Americans, Japanese, underdeveloped social welfare structures, con- and Australians, for example, spend much lower tributing to a skeptical view of the fairness and proportions of their gross national product on efficiency of redistributive policies. social welfare than do their European coun- terparts in the OECD. Americans also express What does the country need most to get ahead, more limited support for government responsi- productivity growth or redistribution bility for social welfare: lower percentages of 80 Americans than Europeans see jobs, health care, Honduras the elderly, and housing as essential government 75 responsibilities. Nicaragua 70 El Salvador (%) Based on evidence from the Latinobarometro 65 Bolivia multi-country surveys, there is a marked simi- 60 Costa Rica larity between Latin American and U.S. atti- supporters Guatemala Colombia tudes about the causes of poverty, about redis- 55 Peru Ecuador Mexico tribution, and about inter-generational mobility. 50 Paraguay For example, in one national survey 36 per cent Uruguay Productivity of Americans say that lack of effort is a major 45 Argentina Brazil cause of poverty, and 36 per cent of Latin Venezuela 40 R2 = 0.51 Chile Americans say that poverty is due to no effort Panama on the part of the poor. Similarly, a majority of 35 Americans and Latin Americans believe that 0 5,000 10,000 15,000 their children's future standard of living will be 1998 GDP per capita (PPP$) higher than their own. Attitudes about inter- Source: Graham, 2002. John Blomquist prepared this note based on Graham, Carol. 2002. "Public Attitudes Matter: A Conceptual Frame for Accounting for Political Economy in Safety Nets and Social Assistance Policies." Social Protection Discussion Paper No. 0233. World Bank. Washington, D.C. Who Deserves Assistance? political currents to implement policy change. To a larger extent, support for safety nets Three sets of issues directly influence the feasi- depends on who the public sees as deserving of bility and pace of implementation: assistance. Similarities and differences between · Institutional framework. The capacity of those needing assistance and the general public public-sector institutions and the relative also play roles. A number of generic considera- strength of particular interest groups often dic- tions are evident across countries: tate the choice of policies. In countries with · The poor versus the middle. The more simi- weak institutions and limited administrative lar a country's middle class is to its poor--in capacity, policies which rely on local commu- terms of income distribution and general nities and institutions are usually more appro- socioeconomic characteristics--the more likely priate. Where organized labor or other interest there will be broad political support for safety groups are strong, it will be difficult to dismiss nets and redistribution in favor of the poor. their concerns in any comprehensive reform This congruence of interests relates to beliefs initiative, suggesting a more decentralized about who is deserving of assistance, and may approach in some instances. be more important even than the perceived · Public attitudes and perceptions. If there are equality or fairness of government programs. It strong attitudes about the deserving versus the has been argued elsewhere that unless programs non-deserving poor, then work requirements are universal or at least seen to potentially ben- and other forms of self-help and private provi- efit the middle class as well as the poor, they will sion of services may help make programs polit- not be politically sustainable. Most Americans, ically feasible. Where there are strongly however, support assistance for the `deserving' entrenched views that certain services are pub- poor but not for the middle class. lic goods--such as in much of Europe and the · Regional balance. To the extent that poverty former Soviet states--reforms must be intro- is regionally concentrated and more remote or duced cautiously and with strong public rela- politically invisible, it is more difficult to mar- tions components, or on a limited basis, to shal political support for targeted assistance avoid polemical debates which can jeopardize policies. broader reform efforts. · Racial diversity. Large racial differences in the · Big bang versus reform by stealth. Sweeping population tend to create divisions on the change based on a centrally driven policy effort question of who is deserving. In the U.S., sur- is possible when there is a clear consensus on vey respondents' support for welfare can be the direction of change and central-level com- directly linked to their perceptions of the racial mitment and priority setting, and when these composition of poverty. In Latin America, there are accompanied by extensive efforts to publi- is significant racial homogeneity in countries cize and explain reforms. Examples include the with the most developed social-insurance and U.S. reform of welfare in 1996, Uruguay's edu- safety-net systems, in contrast to many of the cation reform and Columbia's health reform in Andean countries--e.g., Brazil and Mexico-- the early 1990s. Where entrenched public atti- where indigenous populations make up a high- tudes do not permit high-profile reform, er proportion of the total. changes can instead take place as pilot pro- · Beliefs about the causes of poverty. Support grams within a particular ministry or agency, for safety nets is likely to be weakest among implemented at a local level and scaled up if those who believe that poverty and vulnerabil- and when success is demonstrated. Examples of ity are due to a lack of effort or personal fail- this `stealth' approach include the CLAS ings. Research suggests that such aggregate (Comites Locales para la Administracion del country-level attitudes is at least as important Salud) in Peru and the Progresa program in as individual economic concerns in explaining Mexico, both of which relied on initial pilot differences in attitudes on redistribution. implementation--and formal evaluation of Progresa--to gain political support. Navigating the Political Context Once choices are made about deserving benefi- ciaries, policymakers must still navigate the The World Bank Social Safety Nets Primer series is intended to provide a practical resource for those engaged in the design and implementation of safety net programs around the world. Readers will find information on good practices for a variety of types of interventions, country con- texts, themes and target groups, as well as current thinking on the role of social safety nets in the broader development agenda. World Bank, Human Development Network Social Protection, Social Safety Nets http://www.worldbank.org/safetynets Printed on 100% post-consumer recycled paper