"Targeting errors exclude the poor; universal transfers achieve better coverage despite fiscal cost"

"2.11, panel 1). Cash transfer programs, moreover, may improve human capital accumulation and help households to smooth income shocks, reducing future inequality. This is especially true when benefits are conditional on requirements such as children’s school attendance or regular health checkups (Parker and Vogl 2018; Barrera-Osorio, Linden, and Saavedra 2020). In contrast, fee waivers have little effect on poverty, because these programs are not usually well targeted. Spending is not always allocated to the programs with the largest effect on poverty (Figure 2.11, panel 2), suggesting that governments have significant room to increase the alloc- ative efficiency of social assistance spending. More progressive taxation, along with higher reve- nue mobilization (especially in countries with lower tax capacity) that finances social spending, has significant potential to reduce inequality, especially in countries where taxation and its progressivity are relatively low. Since the 1980s both the average market income inequality and the capital share of income at the top of distribution have risen (Saez and Zucman 2016). Tax policy has meanwhile become less progressive, with significant declines in top marginal tax rates for both labor and capital incomes (Figure 2.12).12 Various other more complex measures also point to a declining trend in tax progressivity—the degree to which the average tax rate rises with income (October 2017 Fiscal Monitor; Gerber and others 2020). Several countries may readily increase top mar- ginal income tax rates (October 2013 Fiscal Monitor; Kindermann and Krueger, forthcoming), although bal- ance needs to be struck against labor supply and invest- ment distortions, as well as potential tax avoidance and evasion from higher taxes. Tax deductions that predom- inantly benefit higher incomes can also be reformed, such as some universal deductions proportional to tax- payers’ incomes or mortgage interest deductions. Coun- tries with flat tax rates could grant (in-work) tax credits for low-income households to heighten progressivity. Should they wish to increase progressivity also at the top of the distribution, they could consider raising tax rates on higher incomes. Addressing loopholes in the 12In addition to the decline in statutory rates, tax expenditures can often further weaken the progressivity of the benchmark system. For example, about 75 percent of the benefit of the preferential rates on long-term capital gains and qualified dividends in the United States is estimated to accrue to the top 1 percent of households by income (Toder and Baneman 2012). Cash transfers Social pensions Fee waivers In-kind (food) Other social assistance programs Poverty headcount reduction (percent) 18 16 14 12 10 8 6 2 4 0 Spending (percent of GDP) 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.4 0.2 0.6 0 1. Poverty Reduction 2. Spending Sources: World Bank ASPIRE database; and IMF staff calculations. Note: Each box shows the 25th and 75th percentiles of the variable of interest. The midline/× corresponds to the median/mean. Poverty reduction is defined as the percentage change in poverty headcount. Data cover 110 emerging market and developing economies for the most recent available year during 2010 to 2018. Examples of in-kind (food) include food stamps, vouchers, nutrition programs, school feeding, school supplies, and in-kind/nonfood emergency support. Examples of fee waivers include reduced medical"

"(Eissa and Liebman 1996). Tax and Transfer Policies (Redistribution) Direct taxes and transfers have, in the long term, reduced income inequality by more than one-third in advanced economies. This redistribution accounts for 85 percent of the difference in disposable income inequality between advanced economies and emerg- ing market and developing economies (October 2017 Fiscal Monitor). Three-quarters of fiscal redistribu- tion in OECD countries is achieved through direct transfers and the remainder through taxes (Causa and Hermansen 2018); the former helps reduce inequality mostly at the bottom, and the latter at the top.10 Coverage and adequacy determine the effective- ness of social assistance programs in reducing poverty and inequality.11 These programs are particularly important in developing economies, where high labor market informality limits social insurance. Countries where both coverage and adequacy are high are more effective in fighting poverty and tend to reduce poverty more for a given amount of social assistance spending (Figure 2.10). Low coverage is a weakness exposed by COVID-19, preventing many countries from providing 10Beyond direct transfers, the distribution of indirect taxes and in-kind transfers also matters. 11Coverage is defined as the share of low-income households that benefit from social assistance. Adequacy is defined as the ratio of social assistance benefits relative to an individual’s income before the transfer. High coverage, high adequacy Other countries Sources: World Bank ASPIRE database; and IMF staff calculations. Note: Poverty reduction is defined as the difference between poverty headcount after and before transfers divided by poverty headcount before transfers. Data are taken from the most recent available year, ranging from 2008 to 2018. High coverage/adequacy is defined as the level above the median. Social assistance spending (percent of GDP) 0 0.5 1 1.5 2 2.5 3 3.5 4 Poverty reduction (percent) 25 20 15 10 5 0 Figure 2.10. Effectiveness of Social Assistance Spending (Percent) Poverty reduction is higher where both coverage and adequacy are high. 36 International Monetary Fund | April 2021 FISCAL MONITOR: A Fair Shot timely lifelines to vulnerable households. To improve coverage, governments need comprehensive social registries, including those that cover the informal sector. A reliable citizen identification system, such as a biometric ID, integrated with socioeconomic databases, is essential (Prady 2020). Identification systems need to be complemented by effective payment mechanisms, such as e-payments (Una and others 2020). Where access to bank accounts is limited, governments can use mobile money transfers (Davidovic and others 2020). Some social assistance programs better reduce poverty than others and could encourage human capital accumulation. Cash transfer programs tend to have the largest effect of all social assistance programs in reducing poverty (Figure 2.11, panel 1). Cash transfer programs, moreover, may improve human capital accumulation and help households to smooth income shocks, reducing future inequality. This is especially true when benefits are conditional on requirements such as children’s school attendance or regular health checkups (Parker and Vogl 2018; Barrera-Osorio, Linden, and Saavedra 2020). In contrast, fee waivers have little effect on poverty, because these programs are not usually well targeted. Spending is not always allocated to the programs with the largest effect on poverty (Figure 2.11, panel 2), suggesting that governments have significant room to increase the alloc- ative efficiency"