33370 World Bank Social Safety Nets Primer Notes 2005 No. 20 The Political Economy ofTargeted Safety Nets To be successful, Social Safety Net (SSN) programs require support from those "lucky"enough to be included in the three elements of policy design: technical correctness, program. administrative feasibility and political viability, yet the politically supportable aspect is often neglected. In this An example from Colombia illustrates the relationship paper, several features of political economy applicable between the available budget and the degree of target- to the choice, design and implementation of safety net ing. An existing food subsidy was transformed into a programs are discussed: modeling the electoral politics of targetedfoodstampprogramandthen,intheearly1990s, targeting; the roles of attitudes and perceptions; central- subsequently eliminated. The distinction between the ized versus localized control; internal and organizational technocratic and political economy approaches is illus- politics and finally, politics and the different social objec- trated by the World Bank's assessment of the Colombian tives of safety ropes and safety nets. policy shift:"Although the program seemed effective and welltargeted...itlackedpoliticalsupportandwasdiscon- Model of Electoral Politics ofTargeted Programs tinued." The analysis here suggests the possibility that Isittruethatwelltargetedprogramsdelivermoreresourc- the program's loss in political support may have occurred es to the poor? Or is it possible that the less well-targeted because, rather than in spite of, the effectiveness of its programs deliver greater benefits to the poor? Pritchett targeting. argues that if the size of the budget for redistribution is politically determined, the impact of targeting on the Interestingly, pro and anti-targeting supporters do not poor must account for this. Otherwise, even with a larger follow the traditional division of "left" and "right". Both share of spending going to the poor, the absolute magni- positions are represented in the entire political spectrum tude will fall sufficiently to make the poor worse off. dependingoneachone'sbeliefonthebudgetresponseto targeting and the related tax level needed to finance this Given a particular budget, the traditional model main- response (see table below). tains that resources would be used more efficiently and effectively if they were targeted. While targeting is op- Attitudes and Perceptions timal when a budget is fixed, theoretical models show The political viability of any SSN program is profoundly in- that targeting decisions influence the size of a budget. fluenced by corruption and the perceptions of horizontal For instance, if the incidence of benefits in a program is equity, process and administrative fairness, and effective- restricted through targeting, then political considerations ness. Corruption subverts all 3 perceptions and so is espe- (the desire for lower taxes, the narrowed constituency cially damaging to political support. for these transfers) would create pressure to decrease the size of the budget therefore negating any efficiency A fundamental principal of equity in many societies is gains. An alternative way to think about this issue is to that "like be treated alike". Introducing discretion into note that leakage of benefits to the non-poor is not nec- targeting criteria may give rise to choices in which per- essarily bad for the poor when it allows increased political sons similarly situated are treated differently and this may Pro-Poor Technocrat ("left" Fiscal Conservative ("right" ­ concerned about welfare of ­ concerned only about level poor) of taxation) Beliefs about Budget Response Flat"unresponsive" FavorTargeting OpposeTargeting toTargeting Steep"unresponsive" OpposeTargeting FavorTargeting ____________________________________________________________________________________________________ Azedine Ouerghi prepared this note based on Pritchett, Lant. 2005."The Political Economy ofTargeted Safety Nets."Social Protection Discussion Paper No. 0501.World Bank.Washington, D.C. make the program appear inequitable. Horizontal equity of the implementing organization there will be problems. is also threatened when a certain cause of poverty is not The Ministry of Public Works in Indonesia struggled with seen as requiring social spending (poverty that is ascribed their assigned task of designing and implementing labor- to laziness). intensive building programs during the 1998 East Asian crisis. Their institutional culture was centred on"building Not only does the targeting of benefits need to be care- things"and they balked at the inefficiency and poor stan- fullydone,theprocessbywhichitisdonemustbeseento dards of the unskilled labor they were required to hire. befair.Aspartofthis,programsmustincludemechanisms By contrast, it took only six months for subsidized rice to facilitatingappealstoanoutsideauthorityandalsomech- reach half of all Indonesian households during the same anisms for ensuring eligibility can be updated over time. crisis. The agency given the task was formerly responsible for the distribution of rice in Indonesia for purposes of Program design must also take into account the implica- price stabilization: the existing organizational structure, tions of managing expectations so as not to create the practices and identity ("we move rice") allowed the initia- impression that programs are not effective. The dilemma tive to be successful. here is that reaching out to certain segments of the popu- lation (for example, persons unemployed for more than Furthermore, politics across organizations can also affect one year) is unlikely to create successful outcomes (it is political support. The responsibility for a program can be harder for long-term unemployed persons to find work). grantedtoaparticularorganoftheStatenotonlytoavoid bureaucracy but also for image promotion purposes The Perceptions of program effectiveness can be enhanced location of Social Fund implementation agencies directly through scientific and independent evaluations that are inofficesofthePresidencyinseveralLatinAmericancoun- carried out in a transparent manner. trieshadbothapragmaticelementofcuttingthroughthe existing bureaucracy and a political element of allowing Centralized vs. Localized Control the President to take credit for the resulting benefits. Another example of the interplay between political econ- omy considerations and the targeting of social safety Politics and the Different Social Objectives of Safety nets revolves around the use of community discretion in Ropes and Safety Nets selecting beneficiaries. This can lead to better or worse Changes in the average poverty rate mask enormous targeting, depending on the interplay between national "churning" as households move in and out of poverty. politics and local politics and the pressures brought to This volatility creates the demand not just for transfer pro- bear at each level. Local leaders know better what is grams to those whose incomes are chronically low (safety socially and politically feasible. If local leaders choose to nets), but also for insurance-like programs that would pay award benefits to non-eligible households, this may be in- off not only when income was absolutely low, but also consistent with the program guidelines, but not necessar- when households experienced negative shocks (safety ilymiss-targeting. However,discretionwithoutconditions ropes). While safety "nets" seek to minimize income or for adequate local oversight can lead to abuses. Program expenditure poverty, the objective of safety "ropes" is to design can work either to limit community influences or mitigate risk. If the targeting of social programs is judged accommodate them by providing explicit local discretion. exclusively on poverty or benefit incidence based on a Centralized and localized targeting can be combined. The cross sectional snapshot, then risk mitigation programs national program could make regional allocations and benefiting households who have suffered large shocks criteria for targeting but stipulate local processes to be butwhoarenot"poor"mayappeartohavelarge"leakage" followed either in making the local decisions or in making when in fact they are simply serving an alternative social exceptions to the criteria. objective. While a "safety net" program might be more popular the more effectively it transfers from richer to Internal and Organizational Politics poorer households, a "safety rope" program might cause Organizational politics are an integral part of the political little net redistribution but be popular because it serves economy of targeted safety nets, overlooked by theory an important insurance function. but considered tremendously important by practitioners. There must be a good "fit" between the nature of a pro- gram and that of the implementing agency, in terms of ethos, image and internal culture. If either the substance or the very act of targeting is inconsistent with the culture TheWorld 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 contexts, 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