Tax Revenue Reforms And Income Distribution In Developing .
Tax Revenue Reforms andIncome Distribution inDeveloping CountriesSanjeev Gupta and João Tovar JallesAbstractCenter for GlobalDevelopment2055 L Street NWFifth FloorWashington DC 20036We explore the impact of major revenue mobilization episodes on income distributiondynamics using a new “narrative” database of major policy changes in tax and revenueadministration systems, covering 45 emerging and low-income countries from 2000 to 2015.Our main finding is that after a tax reform (particularly those affecting the personal incomeor the operation of the revenue administration), the Gini index falls and the bottom incomeshare rises. This result does not hold for sub-Saharan Africa, calling into question the designof tax reforms implemented in the region (mostly fragile states in the sample). In general, toreduce more rapidly income inequality (and improve the income prospects of the pooreststrata of the population), it would be more effective to implement tax reforms when theeconomy is growing relatively slowly. Finally, the smaller the government and the smallerthe tax system, the larger the beneficial impact of tax reforms on income distribution. Ourresults are robust to a battery of sensitivity and robustness tests.202-416-4000www.cgdev.orgThis work is made availableunder the terms of the CreativeCommons AttributionNonCommercial 4.0 license.www.cgdev.orgCGD Policy Paper 175June 2020
Tax Revenue Reforms and Income Distributionin Developing CountriesSanjeev GuptaCenter for Global DevelopmentJoão Tovar JallesISEG, University of LisbonThe authors are grateful to Benedict Clements and Mark Plant for veryhelpful suggestions on the earlier draft. Any remaining errors are theauthors’ sole responsibility. The usual disclaimer applies.The Center for Global Development is grateful for contributions from theBill & Melinda Gates Foundation in support of this work.Sanjeev Gupta and João Tovar Jalles. 2020. “Tax Revenue Reforms and IncomeDistribution in Developing Countries.” CGD Policy Paper 175. Washington, DC: Centerfor Global Development. https://www.cgdev.org/publication/ nter for Global Development2055 L Street NWWashington, DC 20036202.416.4000(f) 202.416.4050www.cgdev.orgThe Center for Global Development works to reduce global povertyand improve lives through innovative economic research that drivesbetter policy and practice by the world’s top decision makers. Use anddissemination of this Policy Paper is encouraged; however, reproducedcopies may not be used for commercial purposes. Further usage ispermitted under the terms of the Creative Commons License.The views expressed in CGD Policy Papers are those of the authors andshould not be attributed to the board of directors, funders of the Centerfor Global Development, or the authors’ respective organizations.
Contents1. Introduction . 22. Literature Review . 53. Econometric Methodology. 84. Data and Stylized Facts . 94.1. Income Distribution . 94.2. Tax Reforms.114.3. Other Data.145. Empirical Results.165.1. Baseline .165.2. Sensitivity and Robustness .205.3. The Role of Business Cycle conditions.215.4. Does the Size of the Tax System or the Size of the Government matter? .246. Conclusion and Policy Implications.26References.28Appendix 1. List of Countries by Region .35Appendix 2. Sensitivity and Robustness Checks .37
1. IntroductionThere is considerable emphasis on both emerging and low-income countries collecting more taxes fromdomestic sources to help achieve the Sustainable Development Goals (SDGs). This is because the resourcesrequired to finance the SDGs are huge and the ability of advanced countries to transfer resources through aidare heavily constrained by their debt-to-GDP ratio, which now exceeds 100 percent of GDP on average andare likely to increase even further following the COVID-19 crisis. The Addis Ababa Agenda for financingdevelopment pays special attention to domestic resource mobilization in emerging and low-income countriesand SDG 17.1 tracks country level domestic resource mobilization efforts. However, domestic resourcemobilization should not come at the cost of impoverishment of the poor or widening of income disparities.Developing countries are on pace to collect public revenues of about 4,444 per person annually by the endof 2020, which corresponds to over 1 per person per day. This compares with 16,200 per person in richercountries (2015 data). If developing countries were to improve revenue-to-GDP by an extra 2 percentagepoints by the end of 2020, their annual public revenues would increase collectively by 144bn—whichcorresponds to more than the total amount of development aid recorded in 2016 (Oxfam, 2019).1Generally speaking, inequality refers to the degree to which distribution of economic welfare generated in aneconomy differs from that of equal shares among its inhabitants (KNBS and SID, 2013). Inequality is amultifaced concept that can be reflected in terms of access to basic services, opportunities, income, amongothers. Our focus in this paper is on income inequality. Even though some (positive) degree of incomeinequality is unavoidable, and it may even stimulate investment and innovation, there is ample evidenceshowing that elevated levels of inequality can cause financial, political and social instability and undermine thepace and sustainability of economic growth (Benabou, 1996; Berg and Ostry, 2017; Cingano, 2014, Ostry etal., 2014; Agnello et al., 2017).Due to high levels of income inequality around the world, many governments have made redistribution ofincome a major policy goal.2 In Figure 1 we plot the median Gini coefficient on disposable income from theStandardized World Income Inequality Database (SWIID) between 1970 and 2015 in selected geographicalregions. We can observe that, at the end of the time span, sub-Saharan Africa is the region with the highestinequality.3 While countries in Asia and the Middle East have been relatively stable over time, there has been arising trend in inequality in both European and North American regions.This Oxfam calculation is based on 2000–2015 trends in population growth and domestic revenue mobilization. Based onconservative GDP growth projections, Oxfam calculated two scenarios for revenue increases. With those trends, they project that theaverage revenue-to-GDP ratio would be 24 percent and collective GDP would be around US 7.225 trillion. This would generatearound 1.738 trillion in domestic revenues collectively by developing countries in 2020. With projected population of 3.913 billionpeople in developing countries in 2020, the 1.738 trillion in revenues would be equivalent to 1.22 per person daily revenues.2 Piketty (2014) presents a new approach to understanding the evolution of income and wealth distribution by examining historicalevidence. According to Krugman (2014), unequal compensation and high incomes of a few individuals have led to accumulation ofwealth.3 Gini indices are measured on a [0–100] scale with larger values denoting higher inequality.12
.25.3.35.4.45.5Figure 1. Gini (disposable income) by region over time, 1970–2015197019801990year20002010median lacmedian menapNote: blue line refers to Latin America; green line refers to Sub Saharan Africa; dark red line refers to Middle Eastmedian ssamedian eurand North Africa; grey line refers to Asia; light red line refers to North America; orange line refers to Europe.median asiamedian namrcSource: SWIID.The list of underlying cyclical and structural forces affecting the level and dynamic evolution of inequality islong. These range from demographic trends to persistent cross-country differences in macroeconomicpolicies or different institutional settings (see e.g. Jalles and Mello, 2019). One that has been receivinggrowing attention is the role attributed to fiscal policy which shapes a country s distribution of income andpoverty (Clements et. al 2015; DeFina and Thanawala, 2004). This has made the role of government inincome redistribution—in particular, the tax-benefit system—all the more relevant (Ortiz and Cummins,2011; Rosen and Gayer, 2014).4 Governments need to strike a balance between efficiency and redistributionwhen designing a tax system (Clements et. al, 2015; Diamond and Mirrlees, 1971).5Studies have shown that in many countries a substantial proportion of the poor are made poorer (or nonpoor made poor) by the tax and transfer system. Higgins and Lustig (2016) find that in ten of twenty-fivecountries they studied at least one-quarter of the poor paid more in taxes than they received in transfers. Thisis because tax and transfer systems in low and middle-income countries are typically far less effective thanthose in OECD countries at reducing poverty and inequality (Bastagli et al., 2015). In the OECD countries,the tax and transfer systems have lowered average market income Gini by one-third (Coady et al., 2015). Therelative ineffectiveness of tax and transfer systems in emerging and low-income countries is attributable tolow level of revenues and poor targeting of spending programs.As we discuss in more detail in section 2, many authors have looked at the role attributed to different taxes inaffecting both economic growth and income inequality in an effort to understand taxes that make the overallTaxation provides resources to the government to perform critical roles such as economic stabilization, allocation and redistribution(Musgrave, 1959).5 The theory of optimal taxation looks at the tax design that seeks to maximize social welfare, an analysis that is beyond the scope ofthis paper.43
revenue system pro-poor and more inclusive. The majority of these studies focus on advanced economies andmany have not reached a consensus on the nature of the tax-inequality relationship (Atkinson and Leigh,2013). Tax and revenue administration reforms are typically not randomly assigned in time and space and varywidely across developing countries, and their effects in improving income distribution, if any, are not clearlyestablished. This paper contributes to filling this gap by analyzing the relationship between inequality and taxpolicy by using a new dataset that does not suffer from the limitations of cross-sectional or time units, as hasbeen the case with other studies (e.g. Piketty and Saez, 2003). From a policy point of view, as Bird (2005)suggested, the issues related to income distribution not only are relevant to tax policy, but they also affect theminds of policy makers with regard to vertical and horizontal justice. A good understanding of thedistributive impact of ordinary taxes should help in moving towards justice-oriented tax systems withoutsacrificing efficiency (Askari, 2011).We use a “narrative” database of major increases in government tax revenue, stemming from bothimprovements in revenue administration and specific tax policy measures, put together by Akitoby et al.(2019) for 45 developing economies (23 emerging and 22 low-income) during 2000 to 2015. Using thisdatabase, we estimate the dynamic short to medium-term response of income distribution proxies to differenttax policy measures. An important novelty and strength of this database is the precise timing and nature ofkey legislative tax actions (or tax reform “shocks”) over the 15-year period. We rely on the local projectionmethod (Jordà, 2005), which has been used to study the dynamic impact of macroeconomic shocks such asfinancial crises (Romer and Romer, 2017) or fiscal shocks (Jordà and Taylor, 2016). Because the short-termeffects of tax reforms may differ depending on the phase of the business cycle prevailing at the time of thereform and initial conditions (such as the size of the existing tax system or general government), we alsoexplore the role of these non-linearities in shaping the dynamic response of income distribution proxies to taxreforms6, using the smooth transition autoregressive model developed by Granger and Teräsvirta (1993).While we argue that the dataset used in the paper provides an exogenous source for tax reforms, endogeneitycan still be a potentially significant concern in our framework since i) revenue mobilization efforts may notnecessarily be exogenous events; and ii) several authors have found it difficult to control for the two-waycausality between income and inequality in most cross-country studies. We try to address these majormethodological challenges by controlling for expected economic growth at the time of revenue reforms andother possible short-term drivers of income inequality. In order to mitigate potential endogeneity concerns,we also estimate our specifications using an Instrumental Variable (IV) approach. On the one hand, toinstrument revenue mobilization reforms we draw instruments from the political economy literature for thedrivers of structural reforms more broadly (and not necessarily only fiscal or economic reforms in general).On the other hand, we use the growth rate of real GDP of main trading partners to instrument outputgrowth, following Cevik and Correa-Caro (2020).The main findings can be summarized as follows. After a general tax reform, the Gini index slowly falls, andthe bottom income share slowly rises. This result is driven largely by countries outside sub-Sahara Africa.Particularly effective in improving the income distribution seems to be personal income tax (PIT) reformsand also reforms in the operation of the tax revenue administration. The results further show that the designof tax reforms has been ineffective in reducing disposable income inequality in sub-Saharan Africa; unlikeSee Cacciatore et al. (2016) and Duval and Furceri (2018) for examples on the nonlinear effects of reforms on macroeconomicvariables depending on the phase of the business cycle. See IMF (2019) for the role of the fiscal stance in affecting reforms coveringmacroeconomic variables in developing countries.64
other country groups, PIT reforms worsened the Gini coefficient. Our results are robust to a battery of testsincluding the exclusion of country fixed effects, the addition of time effects and time trends, controlling forother short-term drivers of inequality (including expectations about future growth). Conducting endogeneityrobust estimations, we found stronger negative (positive) effects of tax reforms on inequality (bottom incomeshare). We also found that in order to reduce more rapidly income inequality and improve the incomeprospects of the poorest strata of the population, it is more effective to implement tax reforms when theeconomy is growing relatively slowly. Finally, the smaller the government and the smaller the tax system, thelarger the beneficial impact of tax reforms on income distribution.The remainder of the paper is structured as follows. As background, context and motivation for our empiricalanalysis, Section 2 provides an overview of related literature. Section 3 presents the empirical strategyfollowed to study the dynamic response of income distribution proxies to revenue mobilization reforms.Section 4 presents the data and key stylized facts. Section 5 discusses the baseline empirical results, sensitivityand robustness checks. Section 6 concludes and elaborates on the policy implications.2. Literature ReviewAn extensive literature is available on the determinants of income distribution.7First and foremost, most authors tend to agree that a high degree of income inequality affects economicgrowth negatively (Alesina and Rodrik, 1994; Perotti, 1996; Cingano, 2014; Bourguignon, 2004; KayizziMugerwa, 2001).8 There are several reasons for this. High inequality is typically associated with unequal accessto basic facilities and opportunities, political instability and social problems (such as crime and other conflictsand the use of illegal drugs which further worsens social inequalities) (Wilkinson and Pickett, 2010; Ortiz andCummins, 2011). Globalization, technological change, changes in demography, unemployment and disparitiesin the distribution of wages and salaries are seen as the major causes of income inequality (Krugman, 2007;Kayizzi-Mugerwa, 2001; Stiglitz, 2012).At the same time, higher taxes can generate negative consequences for growth by affecting consumption andinvestment decisions (Feldstein, 2012). Earlier theoretical studies on taxation show how higher taxes tend todiscourage investment rates (Auerbach and Hasset, 1991) as well as labor supply of individuals (Hausman,1985) and productivity growth, the latter adversely affecting research and development.9 Subsequently, someauthors used endogenous growth models to simulate the effects of tax reforms on economic growth andfound that a decrease in the distorting effects of the current tax structure may lead to a permanent increase ineconomic growth (Gale, 1996).Much of the empirical work examining the effect of income inequality on growth argues that inequalityaffects growth through its effect on taxes and redistribution (Barro, 2000; Milanovic, 2002; Perotti, 1992;Persson and Tabellini, 1994). The conventional wisdom is that taxes (particularly individual and corporate)The inequality literature can be divided into two strands: the first group studies income inequality (Golding and Margo, 1992;Feenberg and Poterba, 1993, 2000; Frank, 2009; Atkinson et al., 2011); the other focuses on wealth inequality (Saez and Zucman,2016; Johannesen and Zucman, 2014; Alstadsaeter et al., 2019).8 According to Kaldor (1995), income inequality can also be good for growth by enabling more savings and, hence, capitalaccumulation.9 The main channel is that corporate and personal income taxes reduce incentives to raise supply through capital accumulation orproductivity enhancements (Schwellnus and Arnold, 2008; Vartia, 2008; Galindo and Pombo, 2011).75
were largely motivated politically by “concerns a
Tax Revenue Reforms and Income Distribution in Developing Countries www.cgdev.org Center for Global Development 2055 L Street NW Fifth Floor . 1 This Oxfam calculation is based on 2000–2015 trends in population growth and domestic revenue mobilization. Based on conservative GDP growth projections, Oxfam calculated two scenarios for revenue .
Income Tax Act 2007 2007 No 97 BC 6 Income tax liability of filing taxpayer 106 BC 7 Income tax liability of person with schedular income 106 BC 8 Satisfaction of income tax liability 108 Subpart BD—Income, deductions, and timing BD 1 Income, exempt income, excluded income, non- 108 residents' foreign-sourced income, and assessable income
reforms in terms of tax revenue mobilization. The rest of this paper is organized as follows. Section 2 provides a literature review on the effect of reforms on tax revenue. Section 3 provides a descriptive analysis and an econometric framework for investigating the effect of reforms on tax revenue in Senegal.
between tax reforms and domestic revenue mobilization. Findings on tax reforms indicated that more emphasis was put on indirect tax reform but direct tax heads was given negligible attention. Majority of tax payers do not get the sufficient tax education and this leads to non compliance on the part of tax payers. Domestic
401(k) 457 Roth IRA Traditional IRA Lower tax bill now! Tax-free growth! Tax deferred growth! Tax deferred Tax deferred After-tax deposits May be tax-deductible Pay income tax Pay income tax Tax-free Pay income tax when withdrawn when withdrawn withdrawals when withdrawn Deposits Payroll-deduction (if allowed by employer) Rollovers
the customs enforcement reforms leading to more tax collected per customer should be given more examination due to its high standard deviation. Key Words: Tax Reforms, Revenue Collection I. Introduction Tax reform is aimed at raising adequate revenues to finance public expenditures on social goods and services.
impose revenue constraints also examine the effect on tax revenue.4 1 Hatta (1977a, 1977b) are among the pioneering studies in the field of trade and domestic tax/subsidy reform policies. While the first paper considered reforms of trade policies, the second one examined reforms of domestic tax policies.
tax reforms in developing countries, both emerging and LIDCs. To do that, we used a new “narrative” database of major tax reforms implemented in 45 countries during 2000-2015 as compiled by the IMF. This database considers large tax revenue changes in the aggregate but also identifies tax
additif alimentaire ainsi que d’une nouvelle utilisation pour un additif alimentaire déjà permis. Les dispositions réglementaires pour les additifs alimentaires figurent à la partie B du titre 16 du RAD. L’article B.16.001 énumère les exigences relatives à l’étiquetage des additifs alimentaires. En particulier, l’article B.16.002 énumère la liste des critères qui doivent .
