76055 THE WORLD BANK Inequality in Focus Community-Driven Development, Participation, and Inequality: What Does the Evidence Say? Ghazala Mansuri and Vijayendra Rao O ver the past two decades, donors and development aid agencies alike have invested heavily in partici- patory development. The World Bank alone has allocated almost $85 billion to participatory projects and decentralization efforts. Other development agencies—bilateral donors and regional development banks—have probably spent at least as much. Driving this massive injection of funding has been the underlying belief that involving communities in at least some aspects of project design and implementation cre- ates a closer connection between development aid and its intended beneficiaries. Indeed, local participation is proposed as a method to achieve a variety of goals, including sharpening poverty targeting, improving service delivery, expanding livelihood opportunities, and strengthening demand for good governance. The purpose of participatory programs is to enhance the © Simone D. McCourtie / World Bank involvement of the poor and the marginalized in community-level Residents of Aurangabad, India, attend a community decision-making bodies in order to give citizens greater say in de- meeting. Does such participation strengthen civil society? cisions that affect their lives. In practice, little is known about how effective participatory development approaches are in promoting Modalities of Local Participation equity. Do these programs result in choices that are better aligned The two major modalities for inducing local participation are with the preferences of the poor and the socially excluded? Do community development and decentralization of resources and they produce more resilient and inclusive local institutions? Do authority to local governments. Community development supports they reduce capture by the elites and corruption? Do participa- efforts to bring villages, urban neighborhoods, or other household tory projects face greater challenges in some types of communi- groupings into the process of managing development resources ties? In particular, is local inequality in wealth, status, or power a without relying on formally constituted local governments. Com- significant barrier to inclusion? These questions are at the heart munity development projects—variously labeled community- of the forthcoming World Bank Policy Research Report. This driven development, community-based development, community article briefly summarizes the evidence on these questions that are livelihood projects, and social funds—include efforts to expand central to the policy debate. community engagement in service delivery. Designs for this type of aid can range from community-based targeting, in which only This article is based on “Localizing Development: Does the selection of beneficiaries is decentralized, to projects in which Participation Work,� a World Bank Policy Research Report communities are involved to varying degrees in the design and to be published in November 2012, Ghazala Mansuri and management of resources. Vijayendra Rao, authors. Decentralization refers to efforts to strengthen village and mu- Poverty Reduction and Equity Department : : www.worldbank.org/poverty : : Volume 1, Number 3 : : October 2012 nicipal governments on both the demand and supply sides. On with the expansion in funding, has proceeded, in large part, with the demand side, decentralization strengthens citizens’ participa- little systematic effort to understand the particular challenges tion in local government by, for example, instituting regular elec- entailed in inducing participation or to learn from the failures tions, improving access to information, and fostering mechanisms of past programs. As a result, the process is, arguably, still driven for deliberative decision making. On the supply side, it enhances more by ideology and optimism than by systematic analysis, either the ability of local governments to provide services by increas- theoretical or empirical. ing their financial resources, strengthening the capacity of local officials, and streamlining and rationalizing their administrative Does Participatory Development functions. Benefit the Poor and the Socially Excluded? This note and the report on which it is based focus on the On balance, greater community involvement seems to modestly “demand-side� aspects of participatory development. Important improve resource sustainability and infrastructure quality. But “supply-side� aspects of governance (fiscal decentralization, the evidence suggests that people who benefit tend to be the most taxation policy, local government procedures, and bureaucratic literate, the least geographically isolated, and the most connected inefficiency) have been dealt with extensively elsewhere and are to wealthy and powerful groups. This picture may partly reflect beyond the scope of this work. the higher opportunity cost of participation for the poor. It also appears, however, that the poor often benefit less from participa- The History of Participatory Development tory processes than do the better off, because resource alloca- Participatory development and decentralization have common tion processes typically reflect the preferences of elite groups. intellectual origins. Deliberative decision making has been a cen- Participation thus appears to affect the distribution of benefits in tral feature of most religious and cultural traditions. In Athenian ways that suggest that “elite capture� is often not benevolent or democracy, for example, important decisions were made in public altruistic. deliberative settings in which all citizens (a group that excluded all Studies from a variety of countries show that communities in women, slaves, and children) were expected to participate. Mod- which inequality is high have worse outcomes, especially where ern notions of participation arguably derive from the eighteenth political, economic, and social power are concentrated in the and nineteenth centuries, notably from the work of Rousseau and hands of a few. “Capture� also tends to be greater in communities John Stuart Mill. that are remote from centers of power; have low literacy; are poor; In the early postcolonial period, the 1950s and 1960s, the U.S. or have significant disparities in caste, race, or gender. Agency for International Development (USAID) and other Project design and implementation rules play a critical role in donors helped drive the first wave of interest in participatory determining whether participatory programs are captured by development by funding and promoting cooperative institutions, elites and thus hinder opportunities for the poor. community-based development, and decentralization. By the For many years, willingness to contribute to programs and 1970s, however, interest in participatory development had waned projects has been seen as evidence of commitment and of the with the realization that cooperatives had largely failed and gov- sustainability of programs or of infrastructure. But this belief has ernment reform was difficult to implement or sustain. The focus little basis in evidence. What little is known suggests that co- of policy shifted to large-scale investments in agricultural and financing—the sine qua non of participatory projects—tends to industrial growth. By the mid-1980s, however, activists and schol- exclude the poorest, particularly when individuals or communities ars attacked this approach, seeing it as “top-down� and inher- self-select into a program. Evidence also suggests that cofinancing ently disempowering and biased against the interests of the poor. requirements for local governments can widen horizontal inequi- Economists such as Nobel laureates Amartya Sen and Elinor ties in targeted transfer programs, because poorer municipalities Ostrom made a vigorous case for a more bottom-up and delibera- or counties have an incentive to reduce the poverty threshold for tive vision of development that allows the “common sense� and transfer eligibility in order to reduce their own copayment burden. “social capital� of communities to play a central part in decisions Demand-driven, competitive application processes can also place that affect them. Their scholarship led to renewed interest in a greater burden on the weakest communities, which might self- community-based development, decentralization, and participa- select out of the program, exacerbating exclusion. tion by donors and governments. As the increased inequity and Policy design may also have unintended consequences on social costs of structural adjustment programs became evident by equity. A large injection of resources for a participatory develop- the early 1990s, donors began to actively fund such participatory ment project can, for example, attract the attention of the better approaches, with the aim of ensuring minimal levels of invest- off, making exclusion more likely. Participatory projects also ment in public services and infrastructure and in social programs often fail to build cohesive and resilient organizations. During the to protect the most vulnerable. course of a project, cash or other material payoffs induce people This renewed policy interest in participatory initiatives, along to participate and build networks—but these mechanisms tend to 2 : : October 2012 : : Inequality in Focus dissolve when the incentives are withdrawn. Only when projects addressed. One way of trying to do so is to mandate the inclu- explicitly link community-based organizations with markets, or sion of disadvantaged groups in the participatory process. There provide skills training, do they tend to improve group cohesive- is virtually no evidence from evaluations of community-driven ness and collective action beyond the life of the project. development projects on whether such mandates work. However, Spending decisions do seem to be better aligned with local needs a growing body of evidence from village democracies in India under democratic decentralization, and resources are reallocated shows impacts that are mixed but broadly positive. Quotas in in favor of the less advantaged. But much depends on the nature of village councils and presidencies for disadvantaged groups and electoral incentives and the capacity of higher levels of government women tend to change political incentives in favor of the interests to provide oversight and ensure downward accountability. of the group that is favored by the quota. Capacity also matters. The benefits of decentralization seem to Mandated inclusion also appears to provide an incubator for be weaker in more remote, more isolated, and less literate locali- new political leadership. Evidence indicates that women and ties. Such localities also tend to be more poorly served by mass other excluded groups are more likely to run for non-mandated media and other sources of information, and they are less likely to seats once they have had some experience occupying a mandated have adequate central oversight. seat. Quotas can also weaken prevailing stereotypes that as- sign low ability and poor performance to traditionally excluded Does Participation Strengthen Civil Society? groups. However, lasting change requires that the inclusion There is little evidence that induced participation builds long- mandates remain in place long enough to change perceptions and lasting cohesion, even at the community level. Group formation social norms. tends to be both parochial and unequal. Absent some kind of Democratic decentralization works because village and munici- affirmative action program, groups that form under the aegis of pal democracies incentivize local politicians to nurture their con- interventions tend to systematically exclude disadvantaged and stituencies. Because decentralized programs usually come with a minority groups and women. Moreover, because similar types constitutional mandate or other legal sanction from the center, of people tend to form groups with one another, projects rarely they are relatively permanent and can therefore change social and promote cross-group cohesion—and may actually reinforce exist- political dynamics over the long term. In contrast, community- ing divisions. based projects are usually ad hoc interventions that are unable to An important question in this context is the role of facilitators open political opportunities for real social change and inclusion. who work with communities. The evidence on this issue is scant, Participatory interventions have been used in post-conflict but the few studies that have tried to measure their effects find settings as a quick way of getting funds to the ground. The limited that facilitators strongly influence the stated preferences of com- evidence of their effectiveness suggests that such projects have munity members, who often tell facilitators what they think they made little headway in building social cohesion or rebuilding the want to hear. state. However, evidence from Africa seems to suggest that people Participation often tends to be driven by project-related incen- emerging from civic conflict have a strong desire to participate tives; people get together to derive benefits from project funds. in their communities and that well-designed and -implemented It is very difficult to know whether these effects will last beyond projects could draw on this need. the tenure of the project and the limited evidence indicates that it In sum, the evidence suggests that, although local actors may usually does not. There is some heartening evidence, though, that have an informational and locational advantage, they use it to participation may have intrinsic value. Communities tend to ex- the benefit of the disadvantaged only where institutions and press greater satisfaction with decisions in which they participate, mechanisms to ensure local accountability are robust. Local even when participation does not change the outcome or when oversight is most effective when other, higher-level institutions of outcomes are not consistent with their expressed preferences. accountability function well and communities have the capacity The ballot box, though far from perfect, appears to provide to effectively monitor service providers and others in charge of a clearer mechanism for sanctioning unpopular policy choices public resources. Local participation appears to increase, rather or excessive rent seeking by traditional or political elites than than diminish, the need for functional and strong institutions at more informal forums for deliberation. In decentralized settings, the center. It also implies that implementing agencies for donor- credible and open elections help align the decisions of politicians funded projects need to have the capacity to exercise adequate with the demands of their constituents. When participatory and oversight. There is little evidence that they can substitute for deliberative councils exist in such settings, they can foster a sig- a nonfunctional state as a higher-level accountability agent, nificant degree of civic engagement. It is less clear how citizens can however. Reforms that enhance judicial oversight, allow for collectively sanction negligent or corrupt officials or local leaders independent audit agencies, and protect and promote the right to where such venues for the exercise of voice are not available. information and a free media appear to be necessary for effective Repairing civic failures requires that social inequalities be local oversight. Inequality in Focus : : October 2012 : : 3 Policy Implications for the Evidence that function well usually do so because they have Three main lessons emerge from distilling the evidence and strong built-in systems of learning and great sensitivity thinking about the broader challenges in inducing broad- and adaptability to variations in context. based participation that promotes equity. 3. Effective civic engagement does not develop within a predict- able trajectory. Instead, it is likely to proceed along a 1. Induced participatory interventions work best when they are “punctuated equilibrium,� in which long periods of supported by a responsive state. The state does not neces- seeming quietude are followed by intense, and often sarily have to be democratic—though being demo- turbulent, change. Donor-driven participatory proj- cratic helps a great deal. But in the sphere in which the ects often assume a far less contentious trajectory. intervention is being conducted—at the level of the Conditioned by bureaucratic imperatives, they often community or the neighborhood—the state has to be declare that clear, measurable, and usually wildly responsive to community demands. optimistic outcomes will be delivered within a speci- Parachuting funds into communities without any fied time frame. There is a danger that such projects set monitoring by a supportive state can result in the cap- themselves up for failure that derives not from what ture of decision making by elites who control the local they achieve on the ground but from their unrealistic cooperative infrastructure, leading to a high risk of cor- expectations. ruption. In the absence of a supportive state, participa- tory engagement might still be able to make a difference, One important reason for this overly ambitious approach is but projects implemented in such environments face that many donors’ institutional structure continues to derive much greater challenges and higher risks of increasing from a focus on capital-intensive development and recon- inequities. struction. Building dams, bridges, and roads, or even schools 2. Context, both local and national, is extremely important. and clinics, is a much more predictable activity than chang- Outcomes from interventions are highly variable across ing social and political systems. Repairing civil society and communities; local inequality, history, geography, the political failure requires a shift in the social equilibrium that nature of social interactions, networks, and political derives from a change in the nature of social interactions and systems all have a strong influence. Local inequality is from modifying norms and local cultures. These much more particularly important in determining the likelihood of difficult tasks require a fundamentally different approach to capture. The variability of these contexts is sometimes development—one that is flexible, long term, self-critical, and so large, and their effect so unpredictable, that projects strongly infused with the spirit of learning by doing. Enhancing the Redistributive Role of Fiscal Policy in Developing Economies Francesca Bastagli, David Coady, and Sanjeev Gupta and-transfer) income inequality in 150 advanced and developing P ast contributions to Inequality in Focus have high- countries. These data confirm that inequality of disposable lighted how income inequality has increased in income has indeed increased in most advanced and many devel- many countries worldwide over recent decades. oping countries over recent decades. They also show that there Advanced countries clearly have been able to is greater inequality in developing countries than in advanced tackle high inequality better than developing economies. countries. A key factor in explaining this difference is the Inequality trends between 1990 and 2005 are especially greater redistributive impact of fiscal policy in advanced enlightening because data are available for a large sample of countries. advanced and developing countries. Our analysis uses the In a recent paper, we reviewed evidence on the redistribu- Gini coefficient, a commonly used inequality measure, which tive impact of fiscal policy in both advanced and developing runs from 0 (where everyone in the economy has the same countries (Bastagli, Coady, and Gupta, 2012). We assembled a income) to 1 (where one person has all the income). In most comprehensive database on trends in disposable (i.e., post-tax- countries, the Gini is between 0.25 and 0.6. The data, shown in 4 : : October 2012 : : Inequality in Focus Table 1 Changes in Disposable Income Inequality Across Regions, 1990–2005 (Percentage-point change in Gini coefficient) Small Decrease Large Increase Medium Increase Small Increase Medium Decrease Large Decrease Change (-3 < Change < (Change ≥ 5) (3 ≤ Change < 5) (0 < Change < 3) (-5 < Change ≤ -3) (Change ≤ -5) 0) Argentina, Latin America Colombia, Honduras, Bolivia, Costa Rica, El Salvador, Brazil, Chile, Ecuador, 1990-2005 Dominican Republic, Belize, Mexico and Caribbean Paraguay, Venezuela Uruguay Panama Nicaragua, Peru Guatemala, Jamaica Burkina Faso, Central African Republic, Ethiopia, Guinea, Sub-Saharan Côte d'Ivoire, Ghana, Niger, Mozambique, Burundi, Madagascar, Cameroon, 1990-2005 Gambia Guinea-Bissau, Kenya, Africa Rwanda, South Africa Tanzania Zambia Nigeria, Uganda Lesotho, Malawi, Mali, Namibia, Senegal, Swaziland China, Indonesia, Rep. of Bangladesh, Asia and Paci�c 1990-2005 Korea, Lao PDR, Nepal, Sri India, Mongolia, Thailand Malaysia Cambodia, Taiwan Lanka Philippines, Vietnam Middle East Kyrgyz Republic, Tajikistan, Egypt, Mauritania, 1990-2005 Djibouti Pakistan Iran, Jordan and North Africa Turkmenistan, Uzbekistan Morocco, Tunisia Belarus, Bosnia and Herzegovina, Bulgaria, Azerbaijan, Hungary, Croatia, Czech Republic, Albania, Georgia, Emerging Europe 1990-2005 Serbia, Slovenia, Armenia, Estonia Kazakhstan, Latvia, Russian Federation Turkey Lithuania, Macedonia, Moldova, Poland, Ukraine Austria, Belgium, New Zealand, Norway, Canada, Finland, Denmark, Australia, Italy, Japan, 1980-2005 Portugal, United Kingdom, Germany, France, Greece, Switzerland Netherlands, Spain United States Luxembourg, Ireland Sweden Advanced Countries Austria, Japan, Australia, Luxembourg, New Denmark, Belgium, Canada, Norway, Portugal, United Zealand, Spain, France, Greece, 1990-2005 Finland, Germany, Switzerland States Sweden, United Ireland, Italy Kingdom Netherlands F. Bastagli, et al., “Income Inequality and Fiscal Policy.� Staff Discussion Note, International Monetary Fund, 2012 Table 1, confirm that large increases in income inequality have • Latin America and the Caribbean: Inequality in this region been observed in all regions: has been substantially higher than in most other regions. • Advanced Economies: Inequality increased in 15 of 22 ad- Over this period, inequality increased in 11 of 20 coun- vanced countries. The Gini coefficient increased by more tries, and the Gini increased by more than 3 points in than 3 points in eight countries. This followed increases seven countries. However, nearly all countries in the re- in equality in most countries over the previous decade. gion have experienced a decrease in inequality since 2000. • Emerging Europe: Inequality increased in virtually all • Sub-Saharan Africa: This was the only region where aver- (20 of 22) Emerging European countries. Most of this age inequality decreased over the whole period. Still, the increase happened between 1990 and 1995, during the Gini coefficient increased in 10 of the 26 countries, rising transition to a market economy. The Gini coefficient in- by more than 3 points in seven countries. creased by more than 3 points in 15 countries and by more • Asia and the Pacific: Inequality increased in 13 of 15 than 5 points in 12 countries. countries in the region. The Gini coefficient increased by Inequality in Focus : : October 2012 : : 5 Figure 1 Benefit Incidence of Education and Health Public Spending (Percent of Public Spending Going to Poorest 40 Percent of Households) Education Health 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Namibia 2003 Lesotho 2002 South Africa 2000 Argentina 2009 Brazil 2009 Peru 2009 Mexico 2008 Albania 2002 Bosnia&Herz Kenya 2006 Costa Rica 2001 Cambodia 2002 Turkey 2001 Azerbaijan 2001 Liberia 2008 Kosovo 2000 Thailand 2008 Bolivia 2007 Uzbekistan 2000 Côte d'Ivoire 2008 Benin 2003 Mozambique Egypt 2005 Uganda 2006 Bangladesh 2000 Zambia 2009 Nepal 2004 Mozambique 1997 Argentina 2009 South Africa 2000 Brazil 2009 Bolivia 2007 Egypt 2005 Belarus 2002 Ghana 1998 India 1996 Ecuador 1998 Guatemala 2006 Mexico 2008 Mongolia 1995 Bangladesh 2000 Zambia 2009 Turkey 2003 Bulgaria 1995 Thailand 2008 Romania 1997 Honduras 2004 F. Bastagli, et al., “Income Inequality and Fiscal Policy.� Staff Discussion Note, International Monetary Fund, 2012 more than 5 points in six countries (including China) and most redistribution in OECD countries is achieved through the by more than 3 points in nine countries. expenditure side of the budget, especially via non-means-tested • Middle East and North Africa: Inequality increased in transfers (including public pensions and universal child ben- nine of 12 countries in the region. The Gini coefficient efits). On average, the redistribution achieved by public cash increased by more than 5 points in four countries and by transfers is twice as large as that achieved through taxes. On the more than 3 points in 5 countries. tax side, personal income taxes achieve the greatest amount of redistribution. However, possibly the most striking message is the enormous The redistributive impact of fiscal policy is even higher if difference in inequality between advanced and developing in-kind transfers, such as public education and health spend- countries, which is substantially larger than changes in average ing, are included. The Gini coefficient for disposable income regional inequality over time. Between 1990 and 2005, average decreases by as much as another 6 points when these are taken inequality in each region changed by less than 4.5 percentage into account. Indirect taxes, on the other hand, are typically points. In contrast, while average inequality in the two most highly regressive. The effective indirect tax rate, calculated as unequal regions (Sub-Saharan Africa and Latin America) the share of consumption taxes in total household income, is exceeded a Gini of 0.45 every year, average inequality in the two on average, three times higher for low-income families than it is most equal regions (Emerging Europe and Advanced Econo- for those in the top decile of the income distribution. mies) was less than 0.34, a difference of 11 percentage points. Differences in the redistributive impact of fiscal policy The Limited Redistributive Impact of Fiscal Policy explain the bulk of the differences in inequality between these in Developing Economies regions. The different role of fiscal policy in advanced coun- Fiscal policy plays a much more limited role in reducing tries, compared to fiscal policy in developing countries, pro- inequality in developing economies. Their higher income vides a basis for identifying inequality-reducing fiscal reforms inequality is often explained by lower levels of taxation and in the latter. public spending, as well as a greater reliance on less progressive tax and spending instruments. The Redistributive Impact of Fiscal Policy Average tax ratios for advanced economies exceed 30 percent in Advanced Countries of GDP. Ratios in developing countries generally fall in a range Taxes and public transfers have played a significant role in from 15 to 20 percent of GDP. Consequently, spending is also offsetting the increase in inequality in advanced countries. substantially lower in developing countries. This is especially Over the past two decades, fiscal policy decreased inequality by true in Asia and the Pacific and in Sub-Saharan Africa, with about one-third in OECD countries. low transfer spending explaining most of the difference. Although income taxes are important in many economies, Three-quarters of the difference in disposable income 6 : : October 2012 : : Inequality in Focus inequality between Latin America, the region with the most can be progressive. This regressivity reflects lack of access by unequal distribution of income in the world, and advanced low-income households to key public services such as educa- countries can be explained by fiscal policy. Six Latin American tion and health. Aggregate education and health spending is economies (Argentina, Brazil, Chile, Colombia, Mexico, and regressive in many developing economies, especially in low- Peru), have fiscal policies that reduce income inequality by only income countries, as shown in figure 1.The progressivity of about 2 percentage points, from 0.52 to 0.50. This compares to a primary health care spending is dominated by the regressivity decrease of about 20 percentage points in 15 European econo- of higher-level health spending. The progressivity of primary mies, from 0.46 to 0.27. education spending is dominated by the regressivity of second- Low tax and spending levels are compounded by a heavy reli- ary and tertiary education spending. However, future in-kind ance on regressive tax instruments as well as by the low cover- spending increases to finance the expansion of basic education age and benefit levels of transfer programs. and health services in developing countries are likely to be • Indirect Taxes: Greater reliance on indirect taxes and much more progressively distributed. narrower consumption tax bases limit the redistributive The recent expansion of “conditional cash transfer� pro- potential of taxes in developing countries. These taxes grams provides a promising approach for enhancing the have only a small progressive impact on income inequal- distributive power of public spending in developing countries. ity in developing countries. Taxes on imports, which These programs target income transfers at poor households continue to be important in low-income economies, of- and continued receipt of the transfer occurs only if households ten appear to be among the most regressive, while excise invest in the education and health of family members. These taxes—such as those on fuel, alcohol, and tobacco—tend programs have been adopted in many developing countries, to be progressive. Although the distributive impact of including some low-income African countries, albeit on a smaller scale. Roughly 17 economies in Latin America are cur- value-added taxes is mixed, there is strong evidence that rently operating conditional cash transfer programs. Typically the exemption of small businesses (including agriculture program expenditures fall below 1 percent of GDP. The largest and the informal sector) can lead to more progressive programs, in Brazil and Mexico, have reduced the Gini for incidence. disposable income by 2.7 points. These programs alone have • Direct taxes: Personal income and property taxes in accounted for about a fifth of the recent decreases in inequality. developing countries are generally progressive. However, However, these programs are most cost-effective when targeted high levels of tax noncompliance combined with narrow at the poorest households, which tend to be most disadvan- income tax bases can contribute to low income tax ratios taged in terms of human capital. Thus, expansions need to and low income tax progressivity. Often this results due be carefully designed in order to continue to generate human to widespread exemptions and the preferential treatment capital impacts and avoid labor supply disincentives. of capital and other income. Resource taxation can be progressive as well as efficient, though it is applied mostly Enhancing the Redistributive Role of Fiscal Policy to foreign incomes. in Developing Economies? • Expenditures: Low spending and poor targeting limit the The challenge in developing countries is to enhance the redistributive capacity of transfer programs. A large redistributive role of fiscal policy, while also promoting growth informal sector further complicates the development of and maintaining fiscal sustainability. This requires strengthen- such programs. In most developing economies, participa- ing the resource mobilization capacity of governments. But tion in social insurance schemes is restricted to high- it equally requires the development of more comprehensive income workers in the formal sector and to public sector social protection systems, including better-targeted safety net employees. In the early 2000s, the share of the population programs. above the legal retirement age in receipt of a pension in Tax policy should focus more on broadening tax bases rather developing countries was, on average, about 40 percent. than increasing tax rates. Expanding corporate and personal This compares to 90 percent in European countries. income tax bases by reducing tax exemptions, closing loop- In addition, expenditure on social assistance programs holes, and improving tax compliance can raise revenues to is often low and poorly targeted. Moreover, the fiscal finance progressive transfers. Expanding the consumption tax space for expanding more distributive social transfers is base (e.g., through broader adoption of the value-added tax) constrained by large expenditures on regressive universal can increase tax revenues. These consumption taxes can be de- price subsidies, especially energy price subsidies. signed to mitigate adverse distributional impacts (e.g., through In-kind public spending has been found to be regressive in appropriate treatment of small businesses and the application many developing countries, although individual components of excise taxes to luxury goods). In many countries, eliminating Inequality in Focus : : October 2012 : : 7 fiscally costly universal price subsidies, which are typically both a deterrent to reform in many countries. Increasing progres- inefficient and inequitable, can generate substantial resources sive public expenditures can help foster political support for in the short term. Especially important are energy subsidies, reforms. Expanding targeted safety net programs can help including tax subsidies (i.e., forgone tax revenues), which can reduce poverty, while expanding education, health and physical escalate to very large percentages of GDP when international infrastructure programs can also benefit middle- and upper- energy prices rise sharply. income groups, promote growth, and thus broaden political However, continued tight revenue constraints and the large support. The recent success of conditional cash transfer pro- demands on these resources to finance broader development grams in many economies suggests that these programs should objectives means that greater emphasis will need to be placed play a greater role in the social protection strategies in devel- on improving the progressivity of public spending. This can oping countries. Broadening the coverage of public pension be achieved through greater reliance on better-targeted social systems would also play an important role in reducing inequal- expenditures aimed at protecting households from poverty ity. Where their expansion is constrained over the short term and improving education and health outcomes among disad- by administrative capacity and fiscal constraints, greater use of vantaged households. But it is also important that these social targeted social pensions may be warranted. programs are monitored and implemented carefully to ensure that resources reach their intended beneficiaries. Reference Development of effective social protection systems can help Bastagli, F., D. Coady, and S. Gupta. 2012. “Income promote tax reform and, more generally, structural reforms. Inequality and Fiscal Policy,� Staff Discussion Note, SDN/12/08 Such reforms, even when well designed, can have an adverse (Revised), International Monetary Fund. impact on poverty and income distribution, and this has been http://www.imf.org/external/pubs/cat/longres.aspx?sk=40024 The Inequality in Focus series aims at informing the public debate on equity, inequality of opportunity, and socioeconomic mobility. It features articles written by World Bank staff, as well as researchers and policy makers from the broad development community. The views and interpretations in the articles are those of the authors and do not necessarily represent the views of the World Bank, its Executive Directors, or the countries they represent. The Inequality in Focus series is not copyrighted and may be reproduced with appropriate source attribution. Editorial Committee: Pedro Olinto (managing editor), Jaime Saavedra, Francisco Ferreira, Luis-Felipe Lopez-Calva, John Newman, Gabriel Demombynes, and Anna Reva Editor: Mary Anne Mulligan THE WORLD BANK