This special feature explores the potential for Asian countries to increase tax revenue in the wake of the COVID-19 pandemic while discussing the constraints and challenges they face in doing so. It examines the use of tax expenditures in the region and identifies opportunities to generate additional revenues through value-added tax, income taxes and international co-operation on taxation.
Revenue Statistics in Asia and the Pacific 2022
Strengthening Tax Revenues in Developing Asia
Revenue Statistics in Asia and the Pacific
2. Strengthening tax revenues in developing Asia
Abstract
Tackling developing Asia’s1 varied challenges will require substantial public investments. To meet these challenges and to restore the health of public finances after COVID-19 passes, the region’s economies urgently need to mobilise additional fiscal resources, especially tax revenues. Drawing on the Asian Development Outlook 2022 theme chapter (ADB, 2022[1]), this special feature2 examines the challenges and opportunities for developing Asia to increase tax revenues. It first presents estimates of tax potential and outlines some of the key constraints to tax revenue mobilisation in the region. Next, it examines the challenge of optimising tax expenditures, which can cause significant revenue losses. Finally, it analyses options for increasing revenue from the key sources of value added, personal income and corporate income taxes.
Developing Asia has the potential to increase tax revenues
The scope to increase tax revenue in developing Asia can be broadly assessed by estimating tax capacity and comparing this to current tax collection. Tax capacity is the theoretical maximum tax revenues an economy can mobilise given its structural characteristics. General underdevelopment, low education levels and large agriculture sectors have been found to reduce tax capacity (Mawejje and Sebudde, 2019[2]). Tax effort is the ratio of actual tax revenues to tax capacity. High tax effort implies actual tax collection is close to tax capacity and the scope to increase revenues may be more limited. In contrast, low effort implies tax collection falls far short of tax capacity, possibly reflecting low tax rates, narrow tax bases or poor compliance.
Figure 2.1. Tax potential in Asia: tax capacity and tax revenue, latest available year
Note: CAM = Cambodia, FIJ = Fiji, FSM = Federated States of Micronesia, HKG = Hong Kong, China, IND = India, INO = Indonesia, KOR = Korea, Republic of, LAO = Lao People's Democratic Republic, MAL = Malaysia, MYA = Myanmar, PAK = Pakistan, PHI = Philippines, PNG = Papua New Guinea, PRC = People's Republic of China, SAM = Samoa, SIN = Singapore, SOL = Solomon Islands, THA = Thailand, TIM = Timor-Leste, TON = Tonga, VAN = Vanuatu, VIE = Viet Nam.
Source: Gupta, and Jalles (2022[3]).
New indicative tax capacity estimates using stochastic frontier analysis techniques suggest that developing Asia can increase tax revenues by 3.6% of GDP on average, which is significant given current tax levels (Gupta and Jalles, 2022[3]). These estimates benchmark tax revenues across countries controlling for key characteristics including GDP per capita, the size of the agricultural sector, education spending and trade openness. There is substantial sub-regional variation in tax capacity and tax effort (Figure 2.1). Tax effort is lower and tax potential correspondingly greater in South and Southeast Asia, particularly in Malaysia, Pakistan and Thailand. By contrast, tax effort and actual revenue are relatively high in the People’s Republic of China (PRC), Korea and some Pacific Island economies. In a few countries, notably Lao People's Democratic Republic (Lao PDR), Papua New Guinea and Timor Leste, tax effort is high despite actual taxes being low. This regional variation underscores the need for country-specific approaches to increasing revenues while avoiding excessive tax burdens that stifle economic growth.
Tax policy challenges and priorities
Raising tax revenues across developing Asia requires governments to make the most of their key sources of revenue, consistent with local priorities and capacity. These include value-added tax (VAT), personal and corporate income taxes, and property taxes. Some countries have very narrow tax bases and base broadening is therefore an important objective for efficiently increasing revenues. However, particularly in small island economies, narrow tax bases reflect undiversified economies, which can limit their options. In many developing countries, enforcement capacity can be weak and hamstrung by limited third party information on taxpayers from sources such as large employers and financial intermediaries, which plays a key role in developed countries (Kleven, H. et al., 2011[4]; Kleven, Kreiner and Saez, 2016[5]; Pomeranz and Vila-Belda, 2019[6]). The acceleration of the digital economy during the COVID-19 pandemic is creating new economic opportunities but also significant tax challenges, especially on cross-border transactions.
While growth-friendly tax systems are essential, strengthening the equity of tax systems may also be important in countries with high inequality and weak transfer systems. In developing Asia, market income inequality (before government taxes and social benefits) is lower than in Latin American and high-income countries, as indicated by the lower pre-tax and benefit Gini coefficient (Table 2.1). However, taxes and social benefits achieve relatively modest redistribution due to the limited progressivity of taxes and the modest size and poor targeting of social benefits (Clements, Gupta and Jalles, 2022[7]). Fiscal policies can help promote economic growth without exacerbating income inequality. Empirically, such inclusive growth episodes are more likely to occur in developing Asia when the population is better educated, spending on health and social benefits is higher, and tax-benefit systems are more redistributive (Clements, Gupta and Jalles, 2022[7]).
Table 2.1. Redistributive effects of fiscal policy, latest available data
Note: Calculations based on latest available data for each country over the years 2010-2020. Developing Asia comprises 38 countries, Latin America and the Caribbean comprises 26 countries and Advanced Economies comprises 30 countries. The redistributive effect of fiscal policy is the difference between the market income Gini coefficient and the net Gini coefficient.
Source: Clements, Gupta and Jalles (2022[7]).
Therefore, in weighing options for enhancing equity, tax and spending policies need to be considered together. Often the primary contribution of tax revenues to inclusive development is providing the resources for equity-promoting public spending. Indeed, even in OECD countries, which rely heavily on progressive personal income taxes, transfers account for around three-quarters of the combined reduction in income inequality achieved by taxes and transfers (Causa and Hermansen, 2017[8]).
A further consideration is fiscal decentralisation, with sub-national governments (SNG) generating their own revenues and accounting for sizeable spending in many developing Asian economies (Figure 2.2). Well-designed fiscal decentralisation, and prudent SNG fiscal policy, can contribute to overall fiscal stability and efficient government. Notably, SNG proximity to residents can help improve the design and implementation of public spending.
Figure 2.2. Sub-national governments’ share of total government revenues and spending
Note: (1) For Malaysia and the Philippines, subnational figures represent the difference between general and central government revenues. (2) Figure shows the most recent year with non-missing entries for both revenue and expenditure. Economies with 2015 data – Republic of Korea; with 2017 data – Afghanistan and Cambodia; with 2018 data – Kiribati, Philippines, Mongolia, Uzbekistan, PRC, and USA. All other economies are presented with 2019 data.
Source: ADB estimates based on IMF GFS Revenue and Expenditure by Function of Government, https://data.imf.org (accessed 31 January 2022); and the Bureau of Local Government Finance for Philippine SNG expenditure data, https://blgf.gov.ph (accessed 16 February 2022).
However, the effectiveness of fiscal decentralisation is constrained by imbalances between SNG expenditure obligations and their own fiscal resources. Except for in the PRC, SNGs’ share of tax collection in total tax revenues remains far lower in developing Asia than in advanced economies. Such imbalances leave SNGs vulnerable to inadequate financial support from central governments, which would constrain their ability to provide adequate public goods and services. Such risks can be mitigated by strengthening SNG own revenues. Certain taxes, notably on property as discussed below, are especially attractive for SNGs. However, there is no one-size-fits-all approach to strengthening SNG tax revenue mobilisation.
Tax expenditures need to be used strategically and transparently
Optimising taxes and ensuring tax system integrity in developing Asia require carefully managing special exemptions. Tax expenditures are preferential tax treatments granted to specific sectors, activities or groups, and they imply a loss of revenues. They include exemptions, deductions, credits, deferrals and lower tax rates, and they are intended to enhance social welfare, promote development, and support other policy goals (Haldenwang, Redonda and Aliu, 2021[9]). Tax incentives, a type of tax expenditure, are often designed to encourage investment in specific sectors or activities, including foreign direct investment (FDI) (IMF et al., 2015[10]; Celani, Dressler and Wermelinger, 2022[11]).
Unlike direct expenditures, tax expenditures are typically not reported in a reliable, comparable and open manner (CBO, 2012[12]; CRS, 2019[13]). Moreover, the cost and benefits of tax expenditures are infrequently assessed and often unknown. In a survey of 43 G20 and OECD countries, only a small minority, including Australia and Korea, published regular, comprehensive and rigorous tax expenditure reports (Redonda and Neubig, 2018[14]). Thus, tax expenditures are characterised by a lack of transparency and accountability and could be conducive to “spending” outside the budget (Burman and Phaup, 2011[15]; Celani, Dressler and Wermelinger, 2022[11]).
Aside from reducing revenue, tax expenditures reduce the efficiency and, depending on their incidence, the equity of the tax system by narrowing the tax base and distorting the rate structure (Bird, 2008[16]). Concessions favouring certain taxpayers require governments to offset revenue losses by imposing a higher tax burden elsewhere or reducing expenditure. Tax incentives that focus on certain businesses can create an unlevel playing field, undermine competition, and reduce tax burdens on investments that would have occurred without the incentive. While it is often claimed that tax incentives create new investment that ultimately boosts revenues, tax incentives are empirically associated with lower overall corporate tax revenues (Kronfol and Steenbergen, 2020[17]). Finally, tax expenditures are likely to increase enforcement costs, give rise to fraud, and encourage rent-seeking behaviour that leads to further tax base erosion.
Figure 2.3. Foregone revenue due to tax expenditures, selected DMCs
Note: ARM = Armenia; BHU = Bhutan; IND = India; INO = Indonesia; KOR = Korea, Rep. of; MON = Mongolia; PHI = Philippines; PNG = Papua New Guinea; SRI = Sri Lanka; TON = Tonga. The figure reflects the latest reference year with available data: 2019 (ARM, KOR, PHI, INO, SRI);
2018 (MON, IND, BHU, PNG); 2014 (TON).
Source: Global Tax Expenditure Database. https://gted.net (accessed 1 Dec 2021); Asian Development Bank staff estimates.
Available data suggest tax expenditures are widely used in developing Asia and cause significant revenue losses. Surveys of regional tax authorities indicate tax incentives are used in almost every country in the region, with tax holidays and tax rate reductions particularly prevalent (ADB, 2022[18]). Available data from the Global Tax Expenditure Database indicate tax expenditures in the region typically include exemptions, deductions, and reduced rates granted for taxes on income and consumption. While lower than the global average (possibly due to lower average tax rates), forgone revenue in a sample of ADB developing member countries (DMCs) was nevertheless substantial, averaging around 2% of GDP, or 14% of total tax revenue, and was highest in Armenia, Tonga and the Philippines (Figure 2.3). It should be noted that comparing the value of tax expenditures across countries poses challenges due to differences in approaches to reporting, and these figures represent a lower bound since costs are unavailable for some tax expenditures (Haldenwang, Redonda and Aliu, 2021[9]).
As with direct expenditures, tax expenditures should have clear policy objectives and justifications, including meeting goals in an efficient, cost-effective manner superior to policy alternatives. Well-designed tax expenditures can advance social policy objectives; for example, a long-standing earned income tax credit scheme in the United States effectively supports poorer households (Hoynes and Patel, 2018[19]). Tax expenditures can also be used to promote gender inclusion. However, as discussed below, exemptions and other special arrangements for income and consumption taxes are often poorly targeted and harm equity.
While a lower overall corporate income tax burden can help attract FDI and increase investment overall, evidence casts doubt on the effectiveness of tax incentives, particularly in attracting greenfield investment (Kinda, 2014[20]; Klemm and Van Parys, 2012[21]; Appiah-Kubi et al., 2021[22]). Investor surveys consistently report that tax is only one of many factors that influence investment location and is less important than factors such as political stability and the regulatory environment (World Bank, 2018[23]). Surveys also indicate that many investments would proceed even in the absence of incentives, meaning that existing incentives result in windfall gains for investors (James, 2013[24]). Incentives for FDI motivated by access to large markets or resources are likely to be particularly ineffective and wasteful (Andersen, Kett and Von Uexkull, 2018[25]). Hence broader, less costly reforms to improve the business environment may be more effective for attracting FDI than tax incentives.
Improved reporting, including cost estimates, and stronger governance are essential for ensuring effective tax expenditures. In developing Asia, few countries provide regular, detailed tax expenditure assessments, although headway is being made. A tax incentives revenue impact statement is published with the Indian annual budget3 and some Southeast Asian countries publish sporadic tax expenditure estimates (ADB, 2021[26]). In many countries, numerous government agencies, notably investment boards and sector ministries, can grant tax incentives. This can create a proliferation of poorly designed and costly incentives (James, 2016[27]).Codifying tax incentives in law, with only the finance minister authorised to grant discretionary incentives, might reduce this tendency.
VAT is a revenue mainstay but can be further optimised
VAT revenue will likely remain a mainstay of tax systems in developing Asia and must therefore be optimised. The self-enforcement property of VAT, which is a relatively efficient tax, is valuable. The tax is paid by firms only on the value-added portion of their sales, with a tax deduction for inputs. To receive the deduction, firms require their suppliers to provide a receipt, creating an auditable paper trail and an incentive for firms to correctly report their activities to tax authorities (Pomeranz, 2015[28]). This self-enforcement can be further strengthened. For example, a new digital invoice system in the PRC reduced the scope for misreporting of input costs and substantially boosted VAT revenues (Fan et al., 2021[29]).
VAT is often considered regressive since poorer households spend more of their income. However, this depends on consumption patterns and tax design. Exemptions and lower rates often apply to food and other necessities, and evidence on equity is mixed (Alavuotunki, Haapanen and Pirttila, 2019[30]; IMF, 2019[31]). Further, many poor households grow their own food and purchase goods and services from small vendors, leaving much of their consumption beyond the VAT net. After accounting for consumption patterns, VAT may even be progressive in developing countries (Bachas, Gadenne and Jensen, 2021[32]).
VAT revenue performance indicators indicate scope to increase VAT collection in developing Asia. C‑efficiency is the ratio of actual to potential revenues assuming a single VAT rate without exemptions across all consumption with perfect compliance. High C-efficiency values therefore imply strong revenue performance. C-efficiency has gradually increased in developing Asia and is comparable to OECD countries (Figure 2.4). However, C-efficiency is much lower in some DMCs.
Figure 2.4. Value-added tax C-efficiency for select DMCs (latest available year) and regional averages (2000-2018)
Note: ARM = Armenia, AZE = Azerbaijan, CAM = Cambodia, DMC = developing member country, FIJ = Fiji, GEO = Georgia, INO = Indonesia, KGZ = Kyrgyz Republic, MON = Mongolia, NEP = Nepal, OECD = Organisation for Economic Co-operation and Development, PHI = Philippines, PRC = People's Republic of China, SRI = Sri Lanka, THA = Thailand, UZB = Uzbekistan, VAN = Vanuatu, VAT = value-added tax. Figures for individual DMC represent the latest available data: 2017 (MON, NEP, PRC, VAN); 2018 (ARM, AZE, CAM, FIJ, GEO, INO, KGZ, PHI, SRI, THA, UZB). Figures for each region represent simple annual average for countries with available data and exclude observations where estimated VAT C-efficiency>1.
Source: Gupta and Jalles (2022[3]).
To increase VAT revenues, exemptions and reduced rates should be reviewed and tightened to broaden the tax base, which is more conducive to economic growth than increasing tax rates (Acosta-Ormaechea and Morozumi, 2021[33]). VAT exemptions are generally an inefficient way to improve equity; while they may benefit the poor, they often benefit the wealthy – who consume more – by a greater amount. Adverse impacts on the poor from removing exemptions can be offset by higher pro-poor spending financed by higher revenues. Simplifications including removing exemptions and adopting a streamlined rate structure can also improve compliance. Complexity was a factor motivating Malaysia to abolish its GST in 2018 (Nutman, Isa and Yussof, 2021[34]).
Across developing Asia, VAT tax rates average 12.7%, lower than in Latin America (14.9%) and high-income OECD countries (19.8%). At 10% or less, VAT rates are especially low in Fiji, Sri Lanka, Thailand and Viet Nam. While less efficient than base broadening, there is scope to increase VAT rates to lift revenues, particularly in DMCs where existing rates are very low (Gunter et al., 2021[35]). Indeed, as part of major tax reforms to increase revenues, Indonesia hiked its VAT rate from 10% to 11% in 2022 and plans a further increase to 12% in 2025.
The appropriate VAT registration threshold for firms needs to balance revenue with compliance costs. Lower thresholds support revenue but increase compliance burdens for very small businesses and tax authorities, and they may encourage firms to underreport activity to stay small (Liu et al., 2021[36]). Such behaviour can have adverse consequences. For example, Muthitacharoen et al. (2021[37]) find that unregistered firms in Thailand have significantly lower growth rates than registered firms, especially near registration thresholds.
Finally, rising digital commerce in the region needs to be managed carefully. Normally, VAT on imported goods can be collected at the border, but this is not the case for imported digital products, which are provided directly to the consumer. Foreign suppliers of those services are often not required to register for and pay VAT. Imports may also be tax exempt due to VAT exemptions for low value imports. The sharing economy, where transactions are mostly made via a digital platform, poses another challenge to VAT.
Applying VAT on digital imports bolsters revenues and ensures a level playing field for domestic suppliers. VAT on imported digital goods is gradually being introduced across developing Asia. Around half of the countries with a VAT in the region have rules that apply VAT to the digital economy (Mullins, 2022[38]). However, at this stage, few countries have VAT rules for goods and services supplied via domestic digital platforms. India does have rules requiring suppliers and platform operators to register for the GST, and the operator to collect tax from suppliers.
The ADB, in partnership with the OECD and World Bank Group, recently released the VAT Digital Toolkit for Asia-Pacific to support reforms of VAT regimes needed to address the VAT challenges of e-commerce growth in the region. This Toolkit presents detailed guidance for the design, administration and operation of the internationally agreed standards and guidance for the collection of VAT on digital trade developed by the OECD, notably building on the successful implementation of these reforms by countries in the region and in other parts of the world.
Personal income taxes can raise more revenue and improve progressivity
Personal income taxes with rising marginal tax rates (imposing proportionately higher tax liabilities on higher earners) are central to a progressive tax system. They are a key source of revenue in OECD countries but much less prominent in developing countries (ADB, 2022[18]). For example, in developing Asia, personal income tax revenue amounts to around 2.0% of GDP compared with around 8.6% in high-income OECD countries. Moreover, their redistributive capacity is relatively weak (Vellutini and Benitez, 2021[39]).
Revenue productivity for personal income tax (the amount of revenue generated by an incremental increase in the tax rate) across developing Asia is similar to other regions but varies enormously within the region (Gupta and Jalles, 2022[3]). In some DMCs, including Bangladesh, it is very low, suggesting significant policy design and enforcement challenges. Therefore, there may be scope for personal income taxes to generate more revenue and become more progressive.
Developing economies characteristically have a large informal sector, which presents several development challenges, notably widespread tax avoidance and weaker fiscal capacity (ADB, 2022[18]). A key challenge to increasing personal income tax revenue in developing countries is prevalent self-employment, which reduces access to third-party information on taxable income, hinders enforcement and shrinks the tax base (Jensen, 2022[40]). Indeed, personal income taxes are often only paid by high-income earners. However, a gradual transition from self-employment to paid employee jobs allows governments to expand the personal income tax base.
Typically, self-employment recedes with development as countries transition from agriculture to manufacturing and services (Gindling and Newhouse, 2014[41]). Developing Asia is no exception, with self-employment on average falling from 56% of total employment in 2000 to around 46% by 2019, still significantly higher than the OECD average but converging towards levels in Latin America (Figure 2.5). However, within developing Asia there is substantial variation, ranging from very high levels in Afghanistan, India and Lao PDR, to less than one quarter of total employment in Kazakhstan and Korea (Figure 2.6).
Figure 2.5. Self-employment in selected regions, 2000 to 2019
Source: World Bank World Development Indicators. https://databank.worldbank.org/source/world-development-indicators (accessed 1 Dec. 2021); Asian Development Bank staff estimates.
Figure 2.6. Self-employment in selected DMCs, 2000 and 2019
Note: AFG = Afghanistan, ARM = Armenia, AZE = Azerbaijan, BAN = Bangladesh, BHU = Bhutan, BRU = Brunei, CAM = Cambodia, FIJ = Fiji, GEO = Georgia, HKG = Hong Kong, China, IND = India, INO = Indonesia, KAZ = Kazakhstan, KGZ = Kyrgyz Republic, KOR = Korea, Republic of, LAO = Lao People's Democratic Republic, MAL = Malaysia, MLD = Maldives, MON = Mongolia, MYA = Myanmar, NEP = Nepal, PAK = Pakistan, PHI = Philippines, PNG = Papua New Guinea, PRC = People's Republic of China, SAM = Samoa, SIN = Singapore, SOL = Solomon Islands, SRI = Sri Lanka, TAJ = Tajikistan, THA = Thailand, TIM = Timor-Leste, TKM = Turkmenistan, TON = Tonga, UZB = Uzbekistan, VAN = Vanuatu, VIE = Viet Nam.
Source: World Bank World Development Indicators. https://databank.worldbank.org/source/world-development-indicators (accessed 1 Dec. 2021); Asian Development Bank staff estimates.
While potentially contributing to more equitable tax systems and substantial revenues, personal income tax design requires carefully balancing the objectives of equity, efficiency and administrative simplicity. Personal income taxes can be economically costly, especially if marginal tax rates rise steeply (Heathcote, Storesletten and Violante, 2014[42]; Blundell, 2016[43]; Keane and Wasi, 2016[44]; Wheaton, 2022[45]). Higher tax rates reduce work incentives and can dampen labour supply, especially for highly skilled and internationally mobile workers (Akcigit, Baslandze and Stantcheva, 2016[46]; Kleven et al., 2020[47]). Personal income taxes levied on household incomes can discourage female workers, exacerbating gender inequality. Married females are more likely than males to be second earners and evidence suggests that joint taxation arrangements disincentivise married female labour market participation (Bick and Fuchs-Schündeln, 2018[48]). By reducing life-time earnings, progressive income taxes may weaken incentives to invest in human capital, compounding efficiency and output losses (Guvenen, Kuruscu and Ozkan, 2014[49]).
The top personal income tax rate on labour income has generally declined across developing Asia (Figure 2.7). Although the average rate of around 27% is much lower than the 40% in OECD countries, there is enormous variation within the region. In countries including the PRC, India and Papua New Guinea, the rate is comparable to the OECD average or even higher. Meanwhile, many Central Asian economies apply a flat tax rate of around 10% to 13%.
Figure 2.7. Personal income tax top marginal rates in Developing Asia
Note: Figure comprises 22 DMCs.
Source: Gupta and Jalles (2022[3]).
Except where flat taxes apply, the top marginal rate also often applies at relatively high income levels. On average in developing Asia, the top marginal rate applies to income of more than 12 times the level of GDP per capita, much higher than 6.6 times GDP per capita in Latin America and 4.1 times GDP per capita in OECD countries (Figure 2.8). In DMCs, where this threshold is high relative to average incomes, it could be lowered to boost personal income tax revenue.
Figure 2.8. Personal income tax maximum and minimum rate thresholds in select regions
Note: The maximum threshold is the income threshold above which the top marginal personal income tax rate applies. Comprise 23 DMCs, 18 Latin America and the Caribbean and 26 OECD countries. For each box plot, X marks the mean and the middle line the median; its height is from 75th percentile (top) to 25th percentile (bottom); and the extended whiskers cover the maximum and the minimum in the sample, excluding the following outliers: Developing Asia (Philippines = 46.3); High-income OECD (Austria = 22.2).
The minimum threshold is the income level at which personal income taxes start being levied. Comprise 24 DMCs, 16 Latin America and the Caribbean and 27 OECD countries. For each box plot, X marks the mean and the middle line the median; its height is from 75th percentile (top) to 25th percentile (bottom); and the extended whiskers cover the maximum and the minimum in the sample, excluding the following outliers: Latin America and the Caribbean (Guatemala = 8.3, Trinidad and Tobago = 9.2); High-income OECD (Slovak Republic = 3.0).
Source: Gupta and Jalles (2022[3]).
Many developing Asian economies apply a tax-free threshold, or zero tax bracket. By exempting the lowest earners, this threshold promotes progressivity and can reduce potentially high compliance costs. On average in DMCs, the threshold applies above the level of GDP per capita, lower than in Latin America but much higher than in OECD countries (Figure 2.8). To expand the tax base, the zero-tax bracket could be lowered, particularly where it is comparatively high and where enforcement capacity and access to third-party information on earnings is stronger. Where it is not currently applied, including in countries with flat taxes, a zero-tax bracket could be added to strengthen progressivity. In some DMCs, there is also potential to expand personal income tax collected through withholding arrangements (ADB, 2021[26])
Taxing capital income can promote progressivity since wealthy individuals earn a disproportionate share of capital income. It can also help ensure self-employed entrepreneurs, who can shift their income from labour to capital, are appropriately taxed. For this reason, the gap between personal, capital and labour income taxation should be minimised. However, in developing countries, poor access to third-party information and the highly mobile nature of capital complicate the taxation of capital income. The wealthy often locate their assets offshore, especially in tax havens, hindering tax enforcement. For example, a voluntary disclosure scheme in Colombia revealed hidden wealth worth around 1.7% of GDP, of which 87% had been concealed offshore (Londoño-Vélez and Ávila-Mahecha, 2021[50]). Initiatives to improve the exchange of information between tax authorities, notably the Exchange of Information on Request and the Automatic Exchange of Information, in which many developing Asian economies participate, can help.
While many countries apply preferential tax rates to certain types of capital income, this practice can create investment distortions and reduce progressivity. For example, tax breaks to encourage retirement savings may encourage a shift into lower tax accounts, resulting in revenue losses that worsen inequality (Chetty et al., 2014[51]).
Multilateral initiatives can help address corporate income tax challenges
Corporate income taxes represent an important revenue source in developing Asia but are under pressure. Motivated by a desire to attract internationally mobile capital and maintain competitiveness, corporate income tax rates have steadily declined across the region over the past few decades, following a global trend. They now average 20% across DMCs, down from 30% in 2000, a little lower than the Latin American and OECD country average (Figure 2.9). As noted above, tax incentives for business are widespread in developing Asia, reducing the effective tax rate.
Figure 2.9. Statutory corporate income tax rates in developing Asia
Note: Comprise 36 DMCs.
Source: Gupta and Jalles (2022[3]).
Corporate tax revenues are also reduced by tax avoidance. Weaknesses in the international tax framework and differences in tax policies – especially tax rates and corporate tax incentives – across countries can be exploited by Multinational Enterprises (MNEs) to reduce their tax liabilities. For example, firms can engage in transfer pricing – mispricing transactions between firm subsidiaries – and can also route income through countries with the most advantageous tax treaties. This allows incomes and profits to be shifted to lower tax jurisdictions, reducing tax liabilities. Global revenue losses from tax avoidance have been estimated at around 4% to 10% of corporate income tax revenue but losses are larger for developing countries (IMF, 2014[52]; OECD, 2015[53]; Crivelli, De Mooij and Keen, 2016[54]). Within Asia, Cobham and Jansky (2018[55]) estimate revenue losses for South Asia at 1.7% to 1.9% of GDP and for East Asia and the Pacific 0.6% to 0.7% of GDP. MNE tax avoidance distorts the location of FDI and mobile assets, and it gives MNEs an unfair competitive advantage.
The rise of the digital economy in developing Asia exacerbates challenges with respect to international taxation (Mullins, 2022[38]). While the international tax system is based on a concept of physical presence, digitalisation means that it is possible to do business in a country without being physically present there. For example, software might be sold from a platform in one country but downloaded by a user in another country. Further, intangible assets such as licenses, trademarks and data, which are easy to shift to lower tax jurisdictions, are prevalent in the digital economy, meaning digitally-intensive firms may be able to shift assets and profits more easily than other firms. The difficulty of determining arm’s length prices for digital intangibles increases firms’ ability to exploit transfer pricing. It is likely that MNEs in digital sectors benefit significantly from tax planning and enjoy lower effective tax rates (OECD, 2018[56]; Mullins, 2022[38]).
Widespread attention to the issue of tax avoidance, especially by MNEs, has fueled concerns that firms do not always pay their fair share of tax. The need to address international tax challenges related to digitalisation has led a growing number of governments, including in developing Asia, to introduce discrete digital service taxes. However, these have generated little revenue and provoked retaliatory trade measures (ADB, 2022[18]; Mullins, 2022[38]). Attempts to address the tax challenges related to digitalisation have also spurred efforts to secure multilateral solutions, notably the G20-OECD project on base erosion and profit shifting (BEPS) and the subsequent work of the Inclusive Framework (IF), which includes 20 DMCs.
In 2021, the IF endorsed a new international tax framework featuring two pillars. Under Pillar 1, profits and taxing rights are reallocated to market jurisdictions, while Pillar 2 proposes a global minimum corporate income tax rate of 15%. Together, Pillars 1 and 2 aim to increase fairness and provide limitations to the ‘race to the bottom’ competition on corporate income tax rates and provide greater certainty for taxpayers and tax administrations.
Estimates suggest that Asian countries may benefit more from Pillar 2 than from Pillar 1. For developing Asia, the revenue impact from the reallocation of taxable profits under Pillar 1 is estimated to be small, with gains or losses of no more than 0.01% of GDP (IMF, 2021[57]). Resource-rich countries which host MNEs, such as Papua New Guinea, and larger economies such as the PRC and Korea gain the most as a percentage of GDP while Singapore and Hong Kong, China lose the most due to their role as investment hubs. Additional global tax revenues from Pillar 2 are estimated at around USD 150 billion per year (OECD, 2021[58]).
References
[33] Acosta-Ormaechea, S. and A. Morozumi (2021), The Value-Added Tax and Growth: Design Matters, International Tax and Public Finance 28.
[18] ADB (2022), A Comparative Analysis of Tax Administration in Asia and the Pacific (5th edition), Asian Development Bank.
[1] ADB (2022), Asian Development Outlook 2022: Mobilizing Taxes for Development, Asian Development Bank, Manila.
[26] ADB (2021), A Comprehensive Assessment of Tax Capacity in Southeast Asia, Asian Development Bank.
[46] Akcigit, U., S. Baslandze and S. Stantcheva (2016), Taxation and the International Mobility of Inventors, American Economic Review 106(10).
[30] Alavuotunki, K., M. Haapanen and J. Pirttila (2019), The Effects of the Value Added Tax on Revenue and Inequality, The Journal of Development Studies 55(4).
[25] Andersen, M., B. Kett and E. Von Uexkull (2018), Corporate Tax Incentives and FDI in Developing Countries, Global Investment Competitiveness Report 2017/2018, World Bank.
[22] Appiah-Kubi et al. (2021), Impact of Tax Incentives on Foreign Direct Investment: Evidence from Africa, Sustainability 13(8661).
[32] Bachas, P., L. Gadenne and A. Jensen (2021), Informality, Consumption Taxes and Redistribution, Faculty Research Working Paper RWP21-026, Harvard Kennedy School.
[48] Bick, A. and N. Fuchs-Schündeln (2018), Taxation and labour supply of married couples across countries: A macroeconomic analysis, The Review of Economic Studies, 85(3), 1543-1576.
[16] Bird, R. (2008), Tax Challenges Facing Developing Countries, Inaugural Lecture of the Annual Public Lecture Series of the National Institute of Public Finance and Policy.
[43] Blundell, R. (2016), Labor Supply and Taxation, Oxford University Press.
[15] Burman, L. and M. Phaup (2011), Tax Expenditures: The Big Government Behind the Curtain, VoxEU.org.
[8] Causa, O. and M. Hermansen (2017), Income Redistribution through Taxes and Transfers across OCED Countries, OECD Economics Department Working Papers No. 1453, OECD Publishing, Paris, https://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=ECO/WKP(2017)85&docLanguage=En.
[12] CBO (2012), Tax Expenditures Have a Major Impact on the Federal Budget, Web Blog Post, Congressional Budget Office.
[11] Celani, A., L. Dressler and M. Wermelinger (2022), Building an Investment Tax Incentives database: Methodology and initial findings for 36 developing countries, OECD Working Papers on International Investment, No. 2022/01, OECD Publishing, Paris, https://doi.org/10.1787/62e075a9-en.
[51] Chetty et al. (2014), Active vs. Passive Decisions and Crowd-Out in Retirement Savings Accounts: Evidence from Denmark, The Quarterly Journal of Economics 129.
[7] Clements, B., S. Gupta and J. Jalles (2022), Fiscal Policy for Inclusive Growth in Asia, Asian Development Bank.
[55] Cobham, A. and P. Janský (2018), Global distribution of revenue loss from corporate tax avoidance: re-estimation and country results, Journal of International Development, https://doi.org/10.1002/jid.3348.
[54] Crivelli, E., R. De Mooij and M. Keen (2016), Base Erosion, Profit Shifting and Developing Countries, FinanzArchiv 72.
[13] CRS (2019), Spending and Tax Expenditures: Distinctions and Major Programs, Congressional Research Service.
[29] Fan et al. (2021), The Dynamic Effects of Computerized Invoices on Chinese Manufacturing Firms, NBER Working Paper No. 24414, National Bureau of Economic Research.
[41] Gindling, T. and D. Newhouse (2014), Self‑Employment in the Developing World, World Development 56.
[35] Gunter et al. (2021), Non-Linear Effects of Tax Changes on Output: The Role of the Initial Level of Taxation, Journal of International Economics 131.
[3] Gupta, S. and J. Jalles (2022), Priorities for Strengthening Key Revenue Sources in Asia, Asian Development Bank.
[49] Guvenen, F., B. Kuruscu and S. Ozkan (2014), Taxation of Human Capital and Wage Inequality: A Cross‑Country Analysis, Review of Economic Studies 81.
[9] Haldenwang, C., A. Redonda and F. Aliu (2021), Shedding Light on Worldwide Tax Expenditures: GTED Flagship Report 2021, German Development Institute.
[42] Heathcote, J., K. Storesletten and G. Violante (2014), Optimal Tax Progressivity: An Analytical Framework, NBER Working Paper No. 19899, National Bureau of Economic Research.
[19] Hoynes, H. and A. Patel (2018), Effective Policy for Reducing Poverty and Inequality? The Earned Income Tax Credit and Distribution of Income, Journal of Human Resources 53(4).
[57] IMF (2021), Fiscal Monitor: A Fair Shot, International Monetary Fund.
[31] IMF (2019), Macroeconomic Developments and Prospects in Low-Income Developing Countries 2019, IMF Policy Paper, International Monetary Fund.
[52] IMF (2014), Spillovers in International Corporate Taxation, IMF Policy Paper Volume 2014: Issue 071, International Monetary Fund.
[10] IMF et al. (2015), Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment, https://www.oecd.org/tax/tax-global/options-for-low-income-countries-effective-and-efficient-use-of-tax-incentives-for-investment.pdf.
[27] James, S. (2016), Tax Incentives around the World, Columbia University Press.
[24] James, S. (2013), Tax and Non-tax Incentives and Investments: Evidence and Policy Implications, World Bank.
[40] Jensen, A. (2022), Employment Structure and the Rise of Modern Tax System, CID Faculty Working Paper 371, Center for International Development at Harvard University.
[44] Keane, M. and N. Wasi (2016), Labour Supply: The Roles of Human Capital and the Extensive Margin, Economic Journal 126.
[20] Kinda, T. (2014), The Question for Non-Resource‑Based FDI: Do Taxes Matter?, IMF Working Paper WP/14/15, International Monetary Fund.
[21] Klemm, A. and S. Van Parys (2012), Empirical Evidence on the Effects of Tax Incentives, International Tax and Public Finance 19.
[47] Kleven et al. (2020), Taxation and Migration: Evidence and Policy Implications, Journal of Economic Perspectives 34.
[4] Kleven, H. et al. (2011), Unwilling or Unable to Cheat? Evidence from a Tax Audit Experiment in Denmark, Econometrica 79(3).
[5] Kleven, H., C. Kreiner and E. Saez (2016), Why Can Modern Governments Tax So Much? An Agency Model of Firms as Fiscal Intermediaries, Economica 83(1).
[17] Kronfol, H. and V. Steenbergen (2020), Evaluating the Costs and Benefits of Corporate Tax Incentives: Methodological Approaches and Policy Considerations, World Bank.
[36] Liu et al. (2021), VAT Notches, Voluntary Registration, and Bunching: Theory and UK Evidence, Review of Economics and Statistics 103(1).
[50] Londoño-Vélez, J. and J. Ávila-Mahecha (2021), Enforcing Wealth Taxes in the Developing World: Quasi-Experimental Evidence from Colombia, American Economic Review: Insights 3.
[2] Mawejje, J. and R. Sebudde (2019), Tax Revenue Potential and Effort: Worldwide Estimates Using a New Dataset, Economic Analysis and Policy 63.
[38] Mullins, P. (2022), Taxing Developing Asia’s Digital Economy, Asian Development Bank.
[37] Muthitacharoen, A., W. Wanichthaworn and T. Burong (2021), VAT Threshold and Small Business Behavior: Evidence from Thai Tax Returns, International Tax and Public Finance 28.
[34] Nutman, N., K. Isa and S. Yussof (2021), GST Complexities in Malaysia: Views from Tax Experts, International Journal of Law and Management 64(2).
[58] OECD (2021), Two-Pillar Solution to Address the Tax Challenges Arising from the Digitalisation of the Economy, OECD Publishing, Paris, https://www.oecd.org/tax/beps/statement-on-a-two-pillar-solution-to-address-the-tax-challenges-arising-from-the-digitalisation-of-the-economy-october-2021.htm.
[56] OECD (2018), The Role and Design of Net Wealth Taxes in the OECD, Tax Policy Studies No. 26, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264290303-en.
[53] OECD (2015), Examples of Successful DRM Reforms and the Role of International Co-operation, OECD Discussion Paper, OECD Publishing, Paris, https://www.oecd.org/ctp/tax-global/examples-of-successful-DRM-reforms-and-the-role-of-international-co-operation.pdf.
[28] Pomeranz, D. (2015), No Taxation without Information: Deterrence and Self-Enforcement in the Value‑Added Tax, American Economic Review 105(8).
[6] Pomeranz, D. and J. Vila-Belda (2019), Taking State‑Capacity Research to the Field: Insights from Collaborations with Tax Authorities, Annual Review of Economics 11.
[14] Redonda, A. and T. Neubig (2018), Assessing Tax Expenditure Reporting in G20 and OECD Economies, Discussion Note 2018/3, Council on Economic Policies.
[39] Vellutini, C. and J. Benitez (2021), Measuring the Redistributive Capacity of Tax Policies, IMF Working Paper WP/21/252, International Monetary Fund.
[45] Wheaton, B. (2022), The Macroeconomic Effects of Flat Taxation: Evidence from a Panel of Transition Economies.
[23] World Bank (2018), Global Development Report.
Notes
← 1. Developing Asia is broadly defined as the developing members of the Asian Development Bank. This includes: Developing members: Afghanistan, Armenia, Azerbaijan, Bangladesh, Bhutan, Brunei Darussalam, Cambodia, Cook Islands, People’s Republic of China, Georgia, India, Indonesia, Fiji, Hong Kong, China, Kazakhstan, Kiribati, the Republic of Korea, Kyrgyz Republic, Lao People’s Democratic Republic, Malaysia, Maldives, Marshall Islands, Federated States of Micronesia, Mongolia, Myanmar, Nauru, Nepal, Niue, Pakistan, Palau, Papua New Guinea, Philippines, Samoa, Singapore, Solomon Islands, Sri Lanka, Taipei,China, Tajikistan, Thailand, Timor-Leste, Tonga, Turkmenistan, Tuvalu, Uzbekistan, Vanuatu, Viet Nam.
← 2. This special feature was written by Sam Hill (World Bank), Yothin Jinjarak (Asian Development Bank), Donghyun Park (Asian Development Bank), and Shu Tian (Asian Development Bank). Sam Hill undertook this work while he was a staff of the Asian Development Bank.