ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY NOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY LOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY NOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY Indonesia Rong Qian, Grzegorz Poniatowski Prosperity Insight Series ECONOMIC POLICY Corporate Income Tax (CIT) Gaps in Estimating Value Added Tax (VAT) and ECONO | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY ECONOMIC POLICY | Estimating Value Added Tax (VAT) and Corporate Income Tax (CIT) Gaps in Indonesia | ECONOMIC POLICY Rong Qian, Grzegorz Poniatowski A verified Reproducible package for this paper is available at http://reproducibility.worldbank.org CONOMIC POLICY Click here for direct access © 2025 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Some rights reserved This work is a product of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, links/footnotes and other information shown in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The citation of works authored by others does not mean the World Bank endorses the views expressed by those authors or the content of their works. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2025. Estimating Value Added Tax (VAT) and Corporate Income Tax (CIT) Gaps in Indonesia. © World Bank.” Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e mail: pubrights@worldbank.org. Cover image: © Vector Point Studio/ shutterstock.com. Used with the permission of Victor Point Studio. Further permission required for reuse. Cover design: Arsianti/arsianti04@gmail.com TABLE OF CONTENTS Executive Summary 1 1. Introduction 3 2. Analytical framework and methodology 5 3. Results 10 3.1 VAT analysis 11 3.1 CIT analysis 16 4. Summary and Conclusion 20 Appendix One. Methodology and data used for VAT gaps’ estimation 24 Appendix Two. Parameter estimation 27 Appendix Three. Methodology and data used for CIT gaps’ estimation 28 Bibliography 31 ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA v Prosperity Insight Figures Figure 1.1: VAT and corporate taxes on income and capital gains as percent of GDP (2002-22) 4 Figure 1.2: VAT and corporate taxes on income and capital gains as percent of GDP, cross-country 4 comparison Figure 2.1: Sources of forgone revenue 6 Figure 2.2: Components of the NIR 7 Figure 2.3: Core methodologies used for tax gap estimation 8 Figure 3.1: VAT C-efficiency in Indonesia (2016-21) 11 Figure 3.2: C-efficiency and VAT revenue-to-GDP ratios (2019) 11 Figure 3.3: VAT compliance gap (2016-21) 12 Figure 3.4: VAT compliance gap and VAT revenue in Indonesia (trillions of IDR) (2016-21) 12 Figure 3.5: VAT compliance gap in Indonesia and selected other middle-income countries (2019 or most 12 recent available data) (% of the VTTL) Figure 3.6: VAT/GST compliance gap in selected high-income countries (2021 or most recent available data) 13 (% of the VTTL) Figure 3.7: Total VAT compliance gap in the EU (2000-21) (% of the EU-wide VTTL) 13 Figure 3.8: Composition of the VTTL (%, average) (2016-21) 14 Figure 3.9: Changes in VAT compliance gap and real GDP growth (2017-21) 14 Figure 3.10: VAT policy gap in Indonesia (2016-21) 15 Figure 3.11: Core components of the VAT policy gap (2021) 15 Figure 3.12: CIT collection efficiency (2016-21) 17 Figure 3.13: CIT revenue and revenue potential (trillions of IDR) (2016-21) 17 Figure 3.14: CIT compliance gap and CIT revenue in Indonesia (trillions of IDR) (2016-21) 18 Figure 3.15: CIT compliance gap (2016-21) 18 Figure 3.16: CIT policy gap in Indonesia (2016-21) 19 Figure 3.17: CIT policy gap and its components (% of GDP) (2021) 19 Figure 4.1: Contributions to changes in VAT revenue (year-over-year, 2017-21) 22 Figure 4.2: Contributions to changes in CIT revenue (year-over-year, 2017-21) 22 Tables Table 4.1: Summary results (average 2016-21) 21 Boxes Box 3.1: The VAT compliance gap in advanced economies 13 ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA vi Prosperity Insight Abbreviations AFT Alternative Final Tax BPS Badan Pusat Statistik C-NTB Current-Year Net Tax Base C-TB Curremt-Year Tax Base CIT Corporate Income Tax FAP Financial Accounting Profit GFCF Gross Fixed Capital Formation GOS Gross Operating Surplus HHC Household Consumption IC Intermediate Consumption MoF Ministry of Finance NIR Notional Ideal Revenue NPISH Non-Profit Institutions Serving Households SFA Stochastic Frontier Analysis SUT Supply and Use Tables TB Tax Base THL Tax Harmonization Law TTL Total Tax Liability VAT Value Added Tax VTTL VAT Total Tax Liability WAR Weighted Average Rate ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA vii Prosperity Insight Acknowledgements This paper is one of several background notes prepared in support of a World Bank flagship report on corporate taxation and Value Added Taxes in Indonesia. The paper was prepared by a World Bank team consisting of Rong Qian (Senior Economist) and Grzegorz Poniatowski (consultant), with contributions from Ratih Dwi Rahmadanti (Economist), Anthony Obeyesekere (Economist), Ahya Ihsan (Senior Economist), and Dwi Endah Abriningrum (Research Analyst). The team would like to thank valuable comments from Rajiv Kumar (Senior Economist), Ralph Van Doorn (Senior Economist), Oyebola Okunogbe (Economist), and guidance from Habib Rab (Lead Economist) and Lars Moller (Practice Manager). The authors would like to thank staff of the Ministry of Finance’s Directorate General of Taxes, BKF, and Macroeconomic Policy Center for their valuable feedback. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA viii Prosperity Insight Abstract Despite their subdued collection efficiency, VAT and CIT are the main sources of government revenue in Indonesia. The sources of the forgone revenue in VAT and CIT remained largely unknown. This paper aims to close this knowledge gap. It finds that the gaps in VAT and CIT accounted for 6.4 percent of GDP on average between 2016 and 2021. Overall, non-compliance had a larger impact on the tax gap—contributing to 58 percent of total forgone notional revenue from these tax instruments. The compliance gaps are very high compared to international peers. They have increased considerably in 2020, likely caused by the economic consequences of the COVID-19 pandemic, resulting in higher incentives to evade and defer tax payments. Despite some decline observed in 2021, the compliance gaps remained elevated compared to the pre-pandemic period. The policy gaps are relatively smaller but also contributed to significant revenue loss. The policy gaps have declined because of the deflation of nominal thresholds, reducing the scope of concessions in the VAT and CIT systems for small and medium-size companies. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 1 Prosperity Insight ES. EXECUTIVE SUMMARY This note presents the results of the analysis of Value Added Tax (VAT) and Corporate Income Tax (CIT) compliance and policy gap estimates in Indonesia between 2016 and 2021. The objective of the study is to quantify the magnitude of the tax gaps and to identify drivers behind the low efficiency of collection of the two main sources of tax revenue. Top-down approaches based on national accounts were used to estimate the tax gaps. To derive the VAT gap using the top-down consumption-side approach, the VAT Total Tax Liability (VTTL) was estimated as the sum of the liability from five main components: final consumption by households, government, and non- profit institutions serving households, intermediate consumption, and gross fixed capital formation using primarily national accounts’ use table. The CIT gap was estimated using the International Monetary Fund’s (IMF) Revenue Administration Gap Analysis Program framework. This top-down approach uses data from national accounts’ gross operating surplus (GOS) and makes appropriate adjustments to reflect conceptual differences between the GOS and CIT base. The estimated gaps in VAT and CIT accounted, on average, for 6.4 percent of GDP or IDR 944 trillion between 2016 and 2021. The estimates of the policy gap are consistent with the government estimates of tax expenditures. Nevertheless, they are somewhat higher indicating that tax expenditures (that is, the actionable policy gap) account for only a fraction of the overall policy gap. The estimates of the compliance gaps are higher than the available estimates of the share of the underground economy. This indicates that informality is a predominant factor but not the only driver of tax non-compliance. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 2 Prosperity Insight Compliance gaps have followed a sideways trend policy gaps needs to be interpreted with caution. during the studied period whereas the policy gaps Small companies and many service providers are have declined. The compliance gaps increased currently taxed at lower effective rates. As non- considerably in 2020, which was likely caused compliance among these taxpayers is likely very by the economic consequences of the COVID-19 high, to reduce markedly this component of the pandemic, resulting in higher incentives to evade policy gap, the compliance gap would need to be and defer tax payments. In 2021, the compliance addressed in parallel. Moreover, the sectors and gaps have declined and the fast revenue growth in types of companies subject to lower CIT and VAT 2022 and 2023 suggests there has been further tax burden are largely subject to alternative regimes improvement in taxpayer compliance in recent (for example, Alternative Final Tax (AFT) or local years. The policy gaps have declined because tax on hospitality services). If these taxpayers were of the deflation of nominal thresholds, thereby brought under the umbrella of CIT and VAT, the reducing the scope of concessions in the VAT and alternative regimes would need to be discontinued CIT systems for small and medium-size companies. or their scope would need to be narrowed to avoid double taxation. In consequence, the impact of Non-compliance had a larger impact on theoretical such a policy change on total tax revenue would be VAT and CIT revenue than the policy decisions. lower than the value of the estimated CIT and VAT CIT and VAT compliance gaps contributed to policy gaps. 58 percent of total forgone notional revenue. Non-compliance ratios are very high compared Relatively high thresholds for VAT and CIT (of to Indonesia’s regional peer countries and in a IDR 4.8 billion) contributed to the large size of broader international comparison. The policy both compliance and policy gaps. The scale of gaps are found to be relatively modest. The major the overall estimates confirms other evidence source of forgone revenue due to policy decision of the threshold’s substantial impact on tax in VAT was the non-taxability of unprocessed revenue. Firstly, a relatively large proportion of food products, education, health services, and taxpayers outside the VAT and CIT regime below hospitality services. The main components of the threshold contributes to the relevant policy forgone revenue in CIT from policy decisions were gaps. Secondly, companies below the threshold are the exemption for companies with turnover below subject to simplified reporting and less scrutiny, IDR 4.8 billion and the rate discount for companies which likely increases taxpayer non-compliance. with turnover below IDR 50 billion. Moreover, this large cliff in the system incentivizes and facilitates bunching. Lowering the threshold as Due to large compliance gaps and alternative well introducing legal prohibition of bunching could tax regimes to VAT and CIT, the value of the narrow the VAT and CIT gaps. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 3 Prosperity Insight 1. INTRODUCTION Indonesia significantly underperforms in tax Value Added Tax (VAT) and Corporate Income revenue collection. Its tax revenue-to-GDP ratio is Tax (CIT), the main sources of tax revenue, among the lowest in the world at just 9.1 percent perform below their potential. In 2021, CIT and in 2021. This level is markedly lower than that of VAT accounted for about 66 percent of total tax other regional middle-income countries such as collection, equivalent to about 6 percent of GDP Cambodia (18.0 percent), Malaysia (11.9 percent), (Figure 1.1). Although VAT and CIT were more Philippines (15.2 percent), Thailand (15.7 percent), productive2 than other tax instruments, the revenue and Viet Nam (14.7 percent).1 Furthermore, generated from VAT and CIT remains relatively low Indonesia has experienced a concerning negative compared to Indonesia’s structural and regional trend in tax revenue-to-GDP ratio over the past peers (see Figure 1.2).3 This can be attributed to a decade. In comparison to the ratio observed combination of factors, including low compliance, ten years ago, the 2021 figure reflects a decline a relatively low effective tax rate, and a narrow tax of approximately 2.1 percentage points. The base. COVID-19 crisis compounded this issue, resulting in a sharp drop to 8.3 percent of GDP in 2020. 1. IMF, Article IV country reports. 2. Understood as revenue-to-GDP ratio. 3. The group of structural and regional peers used in this report includes Cambodia, Malaysia, Philippines, Thailand, and Viet Nam. Two countries in this group (Malaysia and Thailand) are classified as upper middle income, whereas three of them (Cambodia, Viet Nam, and the Philippines) are classified as lower middle income. Despite structural and geographical similarities, CIT and VAT systems and tax collection efficiency in those countries are somewhat differentiated which allows for informative comparisons. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 4 Prosperity Insight Figure 1.1: VAT and corporate taxes on income and capital Figure 1.2: VAT and corporate taxes on income and capital gains as percent of GDP (2002-22) gains as percent of GDP, cross-country comparison 10% 9% 9% 8% 8% 7% 7% 6% 6% 5% 5% 4% 4% 3% 3% 2% 2% 1% 1% 0% 0% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 IDN (2019) MYS (2018) THA (2019) VNM (2019) PHL (2019) Corporate taxes on income and capital gains VAT Corporate taxes on income and capital gains VAT Source: Own elaboration based on data from OECD, BKF Source: UNU-WIDER. (2023), and BPS. Note: IDN stands for Indonesia, MYS for Malaysia, THA for Thailand, VNM for Viet Nam, and PHL for Philippines. Existing estimates suggest that Indonesia’s tax 0.7 percent to 1.2 percent of GDP per annum collection efficiency4 is below that of peers and from 2022 to 2025 (World Bank 2021). Despite the tax gap (forgone revenue) is rising. Stochastic these efforts, tax collection in Indonesia still faces Frontier Analysis (SFA), which links a set of country challenges and needs further improvement. characteristics tax-to-GDP, was used to estimate To further improve tax collection, it is important to Indonesia’s overall tax gaps. The SFA shows that understand the extent and nature of the forgone Indonesia’s tax collection efficiency is below that revenue. This note aims to quantify the magnitude of peers and the tax gap is rising. Indonesia’s of the tax gap and to identify drivers of tax gaps by efficiency score, averaged since 2015, is among examining the compliance and policy gaps in VAT the lowest in the selected sample of middle- and CIT between 2016 and 2021. It provides a income countries and has been declining over time. comprehensive analysis of forgone revenue in VAT This has translated into a growing distance to the and CIT, considering all forms of non-compliance frontier, estimated at about 6 percentage points and policy gaps. This information can be used of GDP for 2018, however, this analysis provides to develop and tailor measures to improve tax only an approximation of Indonesia’s tax gap using collection and reduce revenue loss. a cross-country approach. It does not provide analysis by type of tax or by type of gap (compliance The note is structured into four sections. Following or policy gap) (World Bank 2021). this introduction, Section Two discusses a summary of the analytical framework and methodology The government has taken steps to enhance tax used in the analysis. Section Three discusses the revenue collection through the passing of a Tax findings of overall collection efficiency, compliance Harmonization Law (THL) in October 2021. The and policy gaps analysis for VAT and CIT. Section government increased the VAT standard rate and Four provides a summary and concludes the report. eliminated certain exemptions. In addition, the Detailed methodological considerations and previously regulated CIT rate reduction has also approach are presented in the appendixes. been cancelled. The implementation of THL is expected to boost tax revenues by an estimated 4. Ratio of actual to notional revenue—that is, the revenue that would be collected if the entire theoretical base was taxed and all taxpayers were fully compliant. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 5 Prosperity Insight 2. ANALYTICAL FRAMEWORK AND METHODOLOGY The sources of forgone tax revenue could be linked parts of tax base. This rate is determined by the to the three factors defining revenue in all taxes standard rate and tax/contributions expenditures in and contributions—that is, tax/contributions’ the form of exemptions, rate reductions, increased base, effective rate, and compliance rate (1): allowances, and tax holidays, among others. The latter component is also considered as a part of Tax revenue = tax base x effective rate x compliance ratio (1) the policy gap. Finally, the revenue depends on the The tax base—the total amount of income, compliance ratio and the compliance gap. property, assets, consumption, transactions, or The compliance gaps refer to all sources of other economic activity subject to taxation by a noncompliance, including underreporting, tax authority—is one of the factors behind the tax evasion, fraud, insolvencies, bankruptcies, revenue. The departure of the actual base from administrative errors, and legal tax optimization. the notional one covering the entire base could, This broad approach ensures that all sources of therefore, be an element of the forgone revenue (see forgone revenue are considered. Similarly, the Figure 2.1). It also constitutes one of the elements policy gaps are a comprehensive measure of of the so-called policy gap. Another factor in the forgone revenue from the departure of the tax base revenue equation (1) is the effective rate—that is, from the notional one and the difference between the true mean rate applicable on the entire or certain the effective and standard statutory rate. The gaps ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 6 Prosperity Insight Figure 2.1: Sources of forgone revenue Forgone revenue Departure from Effective rate lower Low standard tax/ Non-compliance: notional tax/ than standard rate contributions' rate 1) Intentional contributions’ base due to breaks in the 2) Non-intentional systems Source: European Commission (2022). are expressed in nominal terms and in relation entire base. The difference between the NIR and the to the tax potential. The compliance gaps are TTL is the tax policy gap. The difference between reported in reference to the total tax liability, which the two concepts is equivalent to revenue losses means the revenue that would have been collected from applying reduced rates and exemptions, if everyone had fully complied with tax laws. On the while referring to potential revenue rather than to other hand, the policy gaps are referred to as the actual collections. In other words, the policy gap Notional Ideal Revenue (NIR) which assumes full is an indicator of the additional revenue that could compliance and the entire notional tax base taxed be collected under the assumption of perfect tax at the standard rate. compliance and maximally broad tax base. The policy gap can be expressed in absolute (4) or in The tax compliance gap is the difference between relative terms (5): the tax revenue that would be collected in the case of full compliance (assuming an unchanged Tax policy gap = notional ideal revenue – TTL (4) tax base)—referred to as the Total Tax Liability (notional ideal revenue – TTL) Tax policy gap (%) = (5) (TTL)—and the actual revenue. Most often, the (notional ideal revenue compliance gap is expressed in absolute terms (2) or in relation to the benchmark—that is, in relation Importantly, tax policy gap measures may differ to the TTL (3): from tax expenditures. Firstly, the benchmark or taxable income for estimating tax expenditure is Tax compliance gap = TTL — tax revenue (2) often narrower than the NIR. Secondly, the policy TTL — tax revenue Tax compliance gap (%) = (3) gap is a simpler measure that: (i) does not account TTL for revenue compensation from other taxes (as, for instance, Regional Tax that substitutes VAT To assess the relative impact of reduced rates, on certain services and the Alternative Final Tax exemptions, and other tax breaks on revenue (AFT) for firms below the threshold); (ii) includes losses, the liability according to the tax law negative expenditures (these variations increasing needs to be compared with the potential revenue the revenue compared to the standard statutory that could be collected in a tax system with a treatment); and (iii) does not account for the uniform rate and the broadest possible base. This economic effects of counterfactual changes in tax benchmark—the NIR—is the revenue that would be burden (that is, the tax base may change if some tax collected if the standard rate was applied to the incentives are removed). It is also important to note ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 7 Prosperity Insight that the policy gap is a proxy of forgone revenue disincentives for taxpayers and complicate the while assuming perfect taxpayer compliance. In system—thereby, deterring enforcement of tax other words, the revenue gains from decreasing the compliance. policy gaps could only materialize if the compliance There is a relationship between the gaps and gap was addressed in parallel. Moreover, the the reference values used for their calculation. policy gap is calculated under the assumption The difference between the NIR (100 percent of no behavioural changes, as a decrease in the compliance and reference policy structures) and overall volume of consumption and adjustments the receipts is the sum of the policy and compliance of consumption baskets would be observed if the gaps, which accounts for all revenue losses in a effective tax rates were shifted upwards. given tax system (see Figure 2.2). The tax expenditures underlying the policy gap A measure that could be used as a proxy of both the shall not be treated as undesirable per se as VAT compliance and the policy gap is C-efficiency. they could allow meeting other fiscal objectives C-efficiency is an indicator of the departure of of the tax system. Nonetheless, the cost of these the VAT from a perfectly enforced tax levied at a expenditures could outweigh the benefit so, uniform rate on all consumption. It is expressed as: in addition to quantifying forgone revenue, the governance of tax expenditures should also include CE= VR (6) measures to assess their efficiency and equity.5 tC Otherwise, the proliferation of tax expenditures where, VR stands for VAT revenue, t for statutory could have negative consequences, hindering standard rate, and C for final consumption (net the mobilization of resources necessary for of VAT). The values of the measure could range governments to provide public goods and services. from zero to one, however, values larger than 65 Moreover, unequal treatment resulting from the percent are rarely observed (see Keen 2013). Even proliferation of exemptions could create various in a utopian situation of full compliance and a flat Figure 2.2: Components of the NIR ACGD = Current ACHE = Reference potential revenue potential revenue 100 percent G H compliance C Compliance gap Current F Policy gap compliance level B Level of compliance A D E Current policy Reference policy structure structure Policy structure Source: IMF (2017). 5. See, for example, World Bank (2011) that analyzes the distibutional impact of the reduction of the policy gap in Indonesia envisaged under the reforms carried out in 2021-22. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 8 Prosperity Insight rate system, C-efficiency should be considerably and CIT gaps involve top-down approaches. These lower than one as domestic final consumption in methods rely on national accounts which cover the denominator of C-efficiency is broader than the the full tax base and all productive activities. On actionable VAT base. Domestic final consumption, the contrary, bottom-up approaches rely on data which is the core of the notional VAT base, includes gathered by tax administrations—including audits, components that cannot be taxed. The list of surveys, and enquiry programs. These data allow elements that are difficult or impossible to tax estimating non-compliance in VAT or CIT for specific includes government and consumption by non- taxpayer groups and types of non-compliance (see profit institutions serving households (NPSIH)—like Figure 2.3). imputed consumption of public education services Both the VAT gap and CIT gap analysis use top- which are largely untaxed. down up approaches based on national accounts. In contrast to the C-efficiency that is a broadly The bottom-up methodologies that could serve as accepted measure for VAT, there is no harmonized a complement or a substitute to top-down methods approach to measuring efficiency of CIT collection. could not be implemented due to unavailability More specifically, there is no commonly accepted of sufficient quality audit results and granular benchmark for CIT base. For this purpose, we information from tax returns. At the same time, define a broad reference base, which covers all the implementation of the top-down methods sources of income of all incorporated businesses, was possible thanks to the availability of national while assuming the current depreciation schedule accounts data as a source of information on the tax and the rules of loss carry forward in the reference base and the availability of detailed administrative tax base. and third-party data that are used for estimating parameters of the model. The World Bank team investigated the potential application of different well-established In addition to relatively low data requirements, methodologies for estimating the tax gap. The top-down approaches based on national accounts most common methods used to estimate the VAT are considered to be accurate and consistent in Figure 2.3: Core methodologies used for tax gap estimation Core tax gap estimation methods Top-down method Bottom-up methods Based on national Based on random accounts enquiries Based on risk-based Indicator methods audits Survey-based Source: World Bank staff elaboration. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 9 Prosperity Insight time (European Commission 2022). Since national Aside from different methodologies used, the accounts are a complete source of information on estimates of tax gap could also be differentiated all economic transactions and they are compiled by the treatment of tax collected through audit using standardized methodologies that do not activities and assessed but not collected on time. change from year to year, potential imprecisions to The estimates presented herein show a net gap— the estimates of the scale of the gap rarely impact meaning that they account for all revenue including the estimated trends. In contrast, the accuracy late payments and VAT collected because of an of bottom-up methodologies may change over audit. The estimates of a gross gap, containing only time due to, as an example, changes in the audit the liabilities paid on time, would be larger. selection process or the effectiveness of auditors. The precision of tax gap estimates based on top- At the same time, bottom-up methodologies allow down approaches primarily hinges on the accuracy for the scrutiny of the sources and components of parameter estimates, reliability of data on the of the compliance gaps, while the top-down tax base (national accounts), and the alignment approaches could often derive only the aggregate of tax revenue figures with information on the tax scale of the gap. base. Of these elements, potential inaccuracies in The method applied to the VAT gap analysis, parameter estimates and imprecisions in national the top-down consumption-side approach, is accounts data could lead to somewhat inaccurate often used as an initial method to estimate VAT estimates of the scale, but usually have a minor gaps thanks to relatively low data requirements impact on changes in the gaps over time. The (European Commission 2022). The alternative latter element, the misalignment of tax revenue method, the production-side approach, could figures and tax liability estimates, typically affects not be employed as it requires national accounts’ the trend, while rarely impacting the scale of the statistics such as supply and international trade gap over longer time frames. Such misalignment tables, which are not available. Moreover, a sectoral might be caused by late payments, varying pace breakdown of VAT revenue statistics that would of granting refunds and changing amounts of tax allow for a breakdown of the compliance gap by the recovered by the audit function. The inaccuracy of sector of economic activity is not available. the tax base data is often related to the difficulties in accounting for the underground economy, The CIT gap was estimated using the IMF’s especially if tax records serve as a core source of Revenue Administration Gap Analysis Program information for compiling national accounts. (RA-GAP) framework. This top-down approach uses data from national accounts’ gross operating surplus (GOS) and makes appropriate adjustments to reflect conceptual differences between the GOS and CIT base (Ueda 2018). ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 10 Prosperity Insight 3. RESULTS 3.1. VAT analysis Over the core period of the analysis, the main Following the destination principle, exports are parameters of the VAT system in Indonesia ‘zero-rated’ (that is, exempt with no right to remained stable. Before the introduction of the deduct). The VAT registration threshold since 2014 2021 THL (effective in April 2022) the standard has been set at a relatively high value of IDR 4.8 VAT rate was set at 10 percent. While no reduced billion (US$320,000 in 2021). rates were in place, Indonesia has excluded VAT C-efficiency relatively broad categories of goods and services from the VAT base or exempted them entirely. The Indonesia’s estimated ratio of actual VAT revenue VAT treatment of both ‘non-taxable’ and exempt to notional potential VAT revenue was 52.8 categories remained the same, with the main percent in 2021 and also averaged 52.8 percent difference between the two lying in the reporting between 2016 and 2021. This suggests that if and invoicing obligations. The list of exclusions and addressing the entire compliance and policy gap exemptions covered mining and drilling (except was possible and desired, the country could nearly coal mining); agricultural products/staple foods double its current VAT receipts.6 Over the last (for example, rice, grain, eggs, salt); electricity (for decade, C-efficiency followed a downward trend connections below 6,600 watts); domestic public (see Figure 3.1). From 2016 to 2018, C-efficiency7 transport; health care; education; employment showed slight improvement, however, it continued services; and margin-based financial services. 6. Assuming no impact of the increased VAT burden on the value of the tax base. 7. C-efficiency ratio is the most commonly used indicator for evaluating the revenue performance and overall efficiency of the VAT system. It is simply the ratio of actual revenues to theoretical revenues from a perfectly enforced tax levied at a uniform rate on all consumption (Ueda 2017). ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 11 Prosperity Insight to decline after 2018, reaching a low of 44.5 statutory rate. This suggests that Indonesia is percent in 2020. The decrease in C-efficiency in experiencing relatively significant amounts of 2020 was primarily caused by the introduction forgone revenue due to non-compliance. of deferrals and late payments resulting from the VAT compliance gap COVID-19 crisis.8 Between 2016 and 2021, the average VAT Indonesia’s C-efficiency was consistently below compliance gap accounted for 43.9 percent of the average of its structural and regional peers. the VTTL or 2.6 percent of GDP (Figure 3.3). In In 2018, when C-efficiency reached a short-term nominal terms, the compliance gap amounted to peak at 58 percent, it was still 1 percentage point an average of approximately IDR 386 trillion for the below the average of developing Asian countries same period (Figure 3.4). In 2020, forgone revenue (Gupta and Jalles 2022). In 2019, however, when due to non-compliance was relatively the largest comparable data was available, Indonesia’s VAT reaching 50.7 percent of the VTTL or IDR 463 C-efficiency ratio was about 17 percentage points trillion, equivalent to over 102 percent of actual lower than the average of its regional peers. This VAT receipts.9 The estimated VAT compliance gap is reflected in VAT revenue accounting for only 3.4 fluctuated over the analyzed period ranging from percent of GDP—which was nearly one percentage 39.5 percent in 2018 to 50.7 percent of the VTTL in point below the average of its regional and income 2020. Between 2016 and 2018, the compliance gap peers (Figure 3.2). For instance, Thailand, which as a percentage of VTTL declined by 5.6 percentage operates a VAT system with similar scope of points, however, in 2019, the trend reversed where exemptions, and policy gap size, has a C-efficiency the compliance gap increased to 43.6 percent. In of 76.7 percent and VAT revenue-to-GDP ratio of 2020, the estimated compliance gap increased 4.7 percent, despite a substantially lower standard further by 7.1 percentage points—reaching 50.7 Figure 3.1: VAT C-efficiency in Indonesia (2016-2021) Figure 3.2: C-efficiency and VAT revenue to GDP ratios (2019) 70% 64.7% 90% 6% 61.7% 57.9% 58.0% 80% 60% 56.3% 5% 51.9% 53.0% 52.8% 70% 50% 44.5% 60% 4% 50% 40% 3% 40% 30% 30% 2% 20% 20% 1% 10% 10% 0% 0% PHL IDN THA VNM 0% 2013 2014 2015 2016 2017 2018 2019 2020 2021 C-efficiency (left axis) VAT revenue to GDP (right axis) Source: World Bank staff calculation. Source: World Bank staff estimates. Note: IDN stands for Indonesia, THA for Thailand, VNM for Viet Nam, and PHL for Philippines. 8. The estimates use cash-based rather than accrual-based revenue. 9. The relatively high level of compliance gap in 2020 is likely contributed by the unavailability of information to control for deferrals and late payments. As a part of the stimulus measures, Indonesia allowed for six months of import VAT payments to be deferred in 19 manufacturing sectors. Tax debt payments, including VAT debt, were delayed up to six months for micro loan credit for businesses affected by COVID-19 (see https://kpmg.com/xx/en/home/insights/2020/04/indonesia-government-and- institution-measures-in-response-to-covid.html). Later payments of VAT liability could also result from the liquidity problems of the businesses affected by the pandemic but not covered by the stimulus package. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 12 Prosperity Insight percent. This increase can be attributed to the Figure 3.3: VAT compliance gap (2016-21) economic consequences of the pandemic, as 60% 4.0% well as the challenge to control for deferred VAT 50.71% 3.5% 50% 45.12% 43.56% 43.12% payments in this calculation. When the economic 41.39% 39.54% 3.0% situation improved in 2021, the compliance gap fell 40% 3.00% 2.5% 2.73% by 7.6 percentage points—down to 43.1 percent of 30% 2.50% 2.37% 2.59% 2.47% 2.0% the VTTL. 1.5% 20% 1.0% A steep increase in VAT revenue in 2022 suggests 10% 0.5% that the VAT compliance gap has further declined. 0% 0.0% In 2022, VAT revenue went up by approximately 2016 2017 2018 2019 2020 2021 25 percent year-to-year and the accumulated two- VAT compliance gap (% of the VTTL) VAT compliance gap (% of GDP) year growth between 2020 and 2022 amounted to Source: WB staff calculation. approximately 53 percent. At the same time, the VAT base proxied by household final consumption Figure 3.4: VAT compliance gap and VAT revenue in Indonesia increased at a much slower pace (about. 10 percent (trillions of IDR) (2016-21) compared to 2021 and about 14 percent compared 1,200 to 2020). Despite the upward shift of the statutory 1,000 VAT rate from 10 to 11 percent as of April 2022, relatively fast growth or revenue should be partially 800 418 351 410 attributed to the increase in taxpayer compliance. 340 463 600 339 It could be expected that the compliance gap fell in 2022 below the pre-pandemic level. 400 481 537 532 552 200 412 450 The VAT compliance gap in Indonesia was relatively high compared to its structural peers. While only 0 2016 2017 2018 2019 2020 2021 limited middle-income countries and Indonesia’s VAT revenue VAT compliance gap peers in the Asia-Pacific region undertake estimations of the VAT compliance gap and publish Source: WB staff elaboration. their estimates, the existing evidence suggests that Figure 3.5: VAT compliance gap in Indonesia and selected other Indonesia’s performance is relatively weak. In fact, middle-income countries (2019 or most recent available data) (% the estimated compliance gap for Indonesia (as a of the VTTL) percentage of the VTTL), is significantly higher than the median in the region. More specifically, it falls 50% 44% 45% among the 25 percent of the highest gaps observed 38% 40% in Asia and the Pacific as documented by the IMF’s 35% 31% RA-GAP program, which covers data up to 2022 30% (Gupta and Jalles 2022). Moreover, Indonesia’s VAT 25% 20% 20% compliance gap surpasses that of the Philippines 15% 10% where the IMF estimated gap was 38 percent in 10% 5% 2015. It is also higher than the compliance gap 5% 0% observed in those other middle-income countries Republic of Bulgaria Turkey Costa Rica Philippines Indonesia where reliable estimates of the compliance gap are South Africa (2019) (2012) (2014) (2016) (2015) (2019) publicly available (Figure 3.5). Source: World Bank staff calculation based on Canıkalp et al. (2016), IMF (2015), IMF (2018), OECD (2018), and European Commission (2022). ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 13 Prosperity Insight Box 3.1: The VAT compliance gap in advanced economies The available estimates of the VAT compliance gap in high-income countries point to non-negligible losses of VAT revenue in these jurisdictions. The VAT compliance gaps in advanced economies typically range from 2 to 20 percent (Figure 3.6). This indicates that advanced economies also face taxpayer compliance challenges and complete sealing of the compliance gap is difficult, if not impossible. At the same time, there is broad evidence of success stories, where jurisdictions have much lower gaps than other comparators. Overall, more than one-half of the European Union (EU) member states recorded a compliance gap of below 5 percent of the VTTL in 2021. The evolution of the compliance gap in recent years in selected high-income countries also indicates that the compliance gaps are actionable. In 2009, the EU-wide compliance gap reached its peak of over 18 percent before reducing to 5.3 percent in 2021 (Figure 3.7). The increase in taxpayer compliance was accompanied by numerous reforms and efforts of tax administrations. For example, actions such as the introduction of domestic reverse charge, later replaced with the split-payment mechanism, the introduction of e-reporting, and increased penalties for fraud have contributed to a decrease in the VAT compliance gap in Poland from 24 percent in 2015 to under 10 percent in 2021. Figure 3.6: VAT/GST compliance gap in selected high- Figure 3.7: Total VAT compliance gap in the EU (2000-21) income countries (2021 or most recent available data) (% (% of the EU-wide VTTL) of the VTTL) 20% 20% 18% 18% 16% 14% 16% 12% 14% 10% 8% 12% 6% 10% 4% 2% 8% 0% 6% 4% 2% 0% Source: European Commission (2023), Canada Revenue Source: Own elaboration based on European Commission Agency (2022), Australian Taxation Office. (https://www. (2023). ato.gov.au), HM Revenue & Customs (2023). Non-deductible investment and intermediate VAT deduction. This includes purchases made by consumption (where there is a large share of government entities, companies operating below taxpayers without the right to deduct) contributed the VAT registration threshold, the mining sector, significantly to the VTTL (Figure 3.8).10 On average, food production, and agricultural industries. Gross Fixed Capital Formation (GFCF)—comprising The primary contributor to the VTTL, however, household, public, and exempt sectors—accounted was household final consumption—representing for 22 percent of the VTTL. This points to relatively about 50 percent of the VTTL. On the other hand, frequent breaks in VAT chains that bring the government and NPISH final consumption, mainly VAT system closer to the turnover tax regime. through in-kind social transfers, accounted for Approximately 24 percent of the VTTL resulted a comparatively modest 4 percent of the VTTL in from hidden VAT on intermediate consumption of total. goods and services that do not qualify for input 10. See Appendix One for the description of the VTTL components. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 14 Prosperity Insight Figure 3.8: Composition of the VTTL (%, average) (2016-21) Figure 3.9: Changes in VAT compliance gap and real GDP growth (2017-21) Intermediate Household final 8.0 7.2 consumption consumption 6.0 5.1 5.2 5.0 liability liability 4.0 24% 50% 3.7 4.0 2.0 0.0 -2.0 -1.9 (2.1) -4.0 -3.7 GFCF -6.0 liability -8.0 22% -7.6 -10.0 Government final 2017 2018 2019 2020 2021 NPISH final consumption liability consumption liability 3% Change in VAT compliance gap (pp) Real GDP growth 1% Source: World Bank staff calculation. Source: World Bank staff calculations. Tax administration measures, alongside robust the observed decline in VAT compliance during that economic growth pre-pandemic, contributed to year. the decline in VAT non-compliance. The reduction VAT policy gap of the VAT compliance gap in 2017 and 2018 can be attributed to the introduction of mandatory The VAT policy gap was relatively low, averaging e-Invoicing which became applicable to all VAT- 13.5 percent of NIR or 0.9 percent of GDP between registered taxpayers from July 2016 onwards. 2016 and 2021 (Figure 3.10). This estimated value Similar experiences in other countries indicate that of the policy gap is in line with the estimates of improved access of tax administration to taxpayer tax expenditures (BKF 2023) at about 0.8 percent information through digital reporting requirements of GDP in 2021. The slightly higher value of the increases voluntary compliance and strengthens policy gap presented in this note (1 percent of GDP enforcement measures.11 At the same time, the in 2021) is due to the broader notional tax base decline of the compliance gap during this period than the VAT benchmark set by BKF.12 At the same coincided with favourable economic conditions, time, the tax expenditure study does not calculate characterized by annual real GDP growth exceeding the impact of the “variations” that increase the 5 percent (Figure 3.9). Robust economic growth revenue.13 plays an important role in bridging tax compliance gaps by mitigating business bankruptcies and The stability of the tax rates’ system prior to the improving firms’ financial liquidity, thereby introduction of the THL resulted in a stable VAT incentivizing for compliance. In 2020, however, policy gap over time. Changes in the economic the economic headwinds likely had an adverse structure–primarily driven by changes in household impact and were among the primary reasons for final consumption baskets and the share of companies falling below the statutory thresholds in 11. See European Commission (2022) on the analysis of the impact of digital reporting requirements in the EU, https://op.europa. eu/s/yKu6 12. See paragraph 11 for the relevant discussion. 13. Certain exemptions in VAT may increase VAT revenue as shown in Figure 3.9 and discussed below. 14. Changes in household final consumption baskets impact the share of exempt goods and services in the VAT base, which is one of the core drivers of the policy gap. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 15 Prosperity Insight the system–led to fluctuations in the policy gap.14 products accounted for 1.4 percent of the NIR, The slight increase in the weighted average rate equivalent to IDR 15 trillion. Similar revenue loss (WAR)15 observed up to 2020 can be attributed could be attributed to non-taxability of companies to changes in consumption patterns. Despite an below the VAT registration threshold. Despite the overall decrease in household final consumption, relatively large share of companies below the the values of purchases of non-taxable staple threshold, forgone revenue was relatively low due goods showed an upward trend in 2020 and 2021. to the prevalence of small companies in branches In 2021 another interesting pattern might be that predominantly provided non-taxable goods observed. The policy gap denoted as a percentage and services such as agriculture and hospitality of the NIR increases, whereas the relation of the services. Furthermore, the value of forgone revenue gap to GDP increases. This is due to the components was partially mitigated by the inclusion of hidden, of GDP growth, mainly driven by ‘zero-rated’ (that non-deductible VAT on inputs of small companies. is, exempt with the right to deduct) exports (36 In other words, since companies below the VAT percent increase year to year). registration threshold have no right to deduct, they contribute to the VTTL and VAT revenue by The major sources of forgone revenue were non- purchasing taxed goods and services. taxability of financial and insurance services (1.9 percent of the NIR), private education and The non-taxability of mining and quarrying health services (1.7 percent of the NIR), and activities contributed to a decrease in the value hospitality services (1.7 percent of the NIR) in of the policy gap by 3.1 percent of the NIR 2020 (Figure 3.11).16 Forgone revenue due to non- or 0.2 percent of GDP in 2021. This value has taxability or final taxation of food and agricultural substantially declined due to the inclusion of coal Figure 3.10: VAT policy gap in Indonesia (2016-2021) Figure 3.11: Core components of the VAT policy gap (2021) 16% 14.9% 15.2% 1.2% 20% 14% 13.0% 12.9% 15% 12.4% 12.7% 1.0% 1.0% 1.0% 10% 12% 15.2% 1.4% 0.9% 0.9% 0.8% 1.7% 0.9% 0.9% 5% 10% 1.7% 1.9% 0% 1.4% 8% 0.6% -3.1% -5% 6% 0.4% VAT policy gap Main components of the policy gap 4% Policy gap VAT policy gap (% of the NIR, lhs) Non-taxability of hospitality services 0.2% 2% VAT policy gap (% of GDP, rhs) Non-taxability of health and education services (excludes public services) Non-taxability of financials 0% 0.0% Non-taxability of foodstuffs and agricultural products 2016 2017 2018 2019 2020 2021 Non-taxability of mining and drilling Source: World Bank staff calculation. Source: World Bank staff elaboration. Note: Due to the interdependence between non-deductible VAT and rate applied on intermediate use, components of the VAT policy gap are not cumulative. 15. Tax rate attributed to the entire household final consumption consisting of goods and services taxed at different rates. 16. In contrast to the estimates of tax expenditures, the calculation of the VAT policy gaps does not account for revenue from other taxes such as, for instance, Regional Tax. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 16 Prosperity Insight mining under the standard VAT treatment as of the In contrast, these categories accounted for only third quarter of 2020. The positive impact of non- 7 percentage points in Indonesia. The size of the taxability of mining and quarrying results from the policy gap in Indonesia is also influenced by the fact that the output of this industry is used mostly type of relief implemented—namely exemptions as intermediate inputs. In such cases, when the without the right to deduct input tax, non-taxability, final products and services are still subject to or flat-rate taxation without the right to deduct taxation but the input VAT cannot be deducted VAT input tax. Middle-income countries that rely cascades, hence mining and quarrying producers heavily on reduced rates, such as Turkey (with a overpay the VAT—resulting in a decrease in the VAT policy gap above 40 percent) and South Africa policy gap. (with a VAT policy gap of about 30 percent), tend to have higher gaps. The reduced rate applicable Compared to other middle-income and to goods sold for final use reduce liability on the aspirational peers, Indonesia’s VAT policy gap entire chain (proportionally to the difference is relatively low, primarily due to relatively low between the reduced and standard statutory rate). expenditure on health, education, and financial The exemptions without the right to deduct often services which typically remain untaxed in most have more moderate impact on the liability. This VAT systems. For instance, in the EU , where the is because they create hidden VAT, which reduces average policy gap is approximately 46 percent of the fiscal impact of exemptions while creating an the NIR, health, education, and financial services undesirable cascading effect. contribute about 21 percentage points to the gap. 3.2. CIT analysis Between 2016 and 2019, the CIT rate in Indonesia same time, firms’ eligibility was now subject to applicable to business income after adjusting for sunset clauses. allowable deductions such as costs of materials, CIT collection efficiency employment, depreciation, and losses carried over from previous years, was set at 25 percent. The estimated revenue potential for CIT was more In 2020, the CIT rate was reduced to 22 percent. than double the actual CIT revenue.17 On average, Stockmarket-listed companies ware entitled to the ratio of CIT revenue to the CIT potential, was a three-percentage point discount, whereas the about 42 percent between 2016 and 2021. In other businesses with annual turnover below IDR 50 words, if all incorporated businesses were fully billion were eligible to a 50 percent tax discount compliant, CIT revenue would see a remarkable of the standard rate on taxable income on the increase of over 130 percent.18 The estimated part of gross turnover up to IDR 4.8 billion. During revenue potential and collection efficiency the period covered by this analysis incorporated exhibited volatility, primarily due to substantial businesses, standard businesses, and individual year-over-year shifts in the components of financial businesses could opt for the presumptive AFT profit accounting which are strongly influenced by regime. Importantly, from 2018 the rate applicable economic conditions such as financial profits. in this alternative regime was reduced, and, at the 17. In line with the calculation of the CIT collection efficiency, the theoretical base for CIT assumes that all interpolated businesses are subject to the standard rate with the right to deduct the cost of all business expenses (under the current depreciation scheme) and could take advantage of the loss carry-forward. 18. CIT revenue is defined as CIT payments and withholding taxes credited against CIT payments collected during the year (PPh Pasal 22 and PPh Pasal 23). The calculation does not account for the impact of the burden on the value of tax base. See more in Appendix Three. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 17 Prosperity Insight Figure 3.12: CIT collection efficiency (2016-21) Figure 3.13: CIT revenue and revenue potential (trillions of Rp) (2016-21) 60% 1000 54.0% 857 50.0% 900 783 50% 744 800 671 688 700 604 46.8% 40% 35.5% 600 32.5% 500 37.3% 366 372 30% 400 302 305 250 242 300 20% 200 100 10% 0 2016 2017 2018 2019 2020 2021 0% 2016 2017 2018 2019 2020 2021 CIT notional ideal revenue CIT revenue Source: World Bank staff elaboration. Source: World Bank staff calculation. Note: Estimates for 2021 may be subject to larger imprecision due to data unavailability. The collection efficiency increased sharply the average yearly forgone revenue due to non- between 2016 and 2019 from 37.3 percent compliance in CIT was about IDR 160 trillion (Figure to 54 percent but declined rapidly in 2020 to 3.14). These estimates are broadly in line with the 32.5 percent (see Figure 3.12). Despite all main estimates of Anjariadi (2011) and the compliance components of the financial accounting profit and ratio among audited companies (Al Rikabi 2018). CIT base increasing in 2020, CIT revenue declined Moreover, these estimates substantially exceed by about 35 percent (see Figure 3.13). This the estimates of the underground economy which decline was partially related to the government’s is about 20 percent according to Medina and decision to allow a six-months postponement of Schneider (2018). CIT payments and the challenge of accurately The estimated CIT compliance gap fluctuated controlling for late payments in this calculation.19 significantly between 14.2 percent (2017) and Nevertheless, despite the economic backwinds 51.5 percent (2020). During the pre-pandemic in 2021, it is expected that collection efficiency period it followed a sideways trend but increased improved only marginally.20 significantly in 2020 and remained elevated also in CIT compliance gap 2021 (Figure 3.15). This hike was 30 percentage points due to the economic consequences of the Between 2016 and 2021, the average CIT pandemic as well as the challenge in controlling compliance gap accounted for about 33 percent of deferred CIT payments. The considerable the CIT Total Tax Liability (CTTL), equivalent to 1.1 fluctuations of the estimated CIT gap in other percent of GDP (Figure 3.15). In nominal terms, years could also result from misalignment in time 19. OECD (2020). The analysis of measures implemented by tax administrations in different countries shows that the majority of them provided additional time for taxpayers for dealing with tax affairs. These deferrals ranged from several weeks to several months and were both automatic and on request. 20. Estimates for 2021 may be subject to greater imprecision due to data unavailability. The value of GOS for 2021 was not available, thus it is projected using 2020 values and growth rates of GDP and intermediate consumption between 2020 and 2021. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 18 Prosperity Insight Figure 3.14: CIT compliance gap and CIT revenue in Indonesia Figure 3.15: CIT compliance gap (2016-21) (trillions of IDR) (2016-21) 700 60% 2.5% 51.5% 600 50% 2.0% 40% 35.5% 500 30.8% 1.5% 163 269 1.7% 400 89 30% 257 19.4% 1.0% 138 50 20% 14.2% 300 1.1% 1.1% 10% 0.5% 200 366 372 0.6% 302 305 0.4% 250 242 0% 0.0% 100 2016 2017 2018 2019 2020 2021 0 CIT compliance gap (% of the CTTL, LHS axis) 2016 2017 2018 2019 2020 2021 CIT compliance gap (% of the GDP, RHS axis) Linear (CIT compliance gap (% of the CTTL, LHS axis)) CIT revenue CIT compliance Gap Linear (CIT compliance gap (% of the GDP, RHS axis)) Source: World Bank staff calculation. Source: World Bank staff elaboration. Note: Estimates for 2021 may be subject to greater Note: Estimates for 2021 may be subject to greater imprecision due to data unavailability. imprecision due to data unavailability. Dotted lines indicate simple linear trend lines. between tax revenue figures and liability estimates. percent (2017). Poland, on the other hand, had More specifically, variations in CIT revenue an estimated CIT compliance gap of about 30 recovered by the audit function and tax amnesty in percent in 2019 and 2020 (Sawulski et al. 2022). 2016 and 2017 have likely led to a non-negligible In the Slovak Republic, the estimated gap ranged difference between cash and not available accrual- from approximately 41 percent in 2012 (which was based revenue. The fluctuations in the estimated similar to average value estimated for Indonesia gap may indicate measurement error21 but also in the time period covered), to 8 percent in 2016 the underlying issue–a significant contribution (Fiscalis 2018). In the UK, Her Majesty’s Revenue from the largest taxpayers who may change their and Customs (2019) estimated the compliance gap tax optimisation practices over time.22 Due to the for CIT in 2017-18 at approximately 8.7 percent of existence of imprecisions that inflate the volatility the CTTL. of the estimated CIT compliance gap, the point CIT policy gap estimates should be interpreted together with the trend lines observed over the time period. On average, the CIT policy gap accounted for about 35.9 percent of its potential or 1.8 percent The high estimate of the CIT compliance gap for of GDP between 2016 and 2021 (Figure 3.16). The Indonesia is in line with publicly available estimates estimates of the CIT policy gap are approximately reported for other countries.23 For example, in three times higher than the estimates of tax 2015, the estimated CIT compliance gap in Costa expenditures in CIT by BKF (2023) which point to Rica was about 60 percent (IMF 2018). In the case three big-ticket sources of forgone CIT revenue: of another middle-income country, South Africa, (i) effective rate reduction for certain enterprises Jansen et al. (2020) found that the CIT compliance with gross turnover of not more than IDR 4.8 gap fluctuated between 8 percent (2015) and 12 21. Due to inability to control accurately for tax optimization and unavailability of granular information on GOS, imprecisions related to estimating CIT gaps are considered to be higher than in the case of VAT gaps. 22. Tax optimization is the use of processes to reduce a company’s or individual’s tax charges to a minimum by using the law’s advantages. It entails recognizing tax deductions, credits, exemptions, and other tax-saving options while adhering to tax law. 23. The estimates of the CIT compliance gap for regional peers were not available. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 19 Prosperity Insight billion; (ii) tax rate discount for small companies remain unchanged (with the prevailing compliance with yearly turnover of IDR 50 billion; and (iii) tax rate) if tax expenditures were withdrawn. rate reduction for public companies. The disparity Between 2016 and 2018, the CIT policy gap in estimates shows that tax expenditures—the declined sharply from 42.2 percent to 32.3 percent, actionable policy gap—account for only a fraction of whereas between 2018 and 2021 it remained the overall policy gap. It also stems from different stable (Figure 3.16). The long-term downward assumptions. The policy gap estimates assume trend in the CIT policy gap is closely linked to the full compliance, whereas the estimates of tax structure of the policy gap and its main elements, expenditures are derived based on actual revenue namely the discount rate for companies with figures and compliance rates. The policy gap and turnover below IDR 50 billion and the exemption the estimated fiscal cost of tax expenditures must, for companies in the final regime (Figure 3.17). Due therefore, be interpreted differently. The policy to price inflation and (real) economic growth, the gap accounts for the theoretical revenue loss that share of companies covered by these preferential would be obtained if all taxpayers were compliant. regimes has gradually declined—reducing forgone The fiscal cost of tax expenditures reported by BKF CIT revenue. reflects the actionable growth in revenue under the assumption that behaviour of taxpayer would Figure 3.16 CIT policy gap in Indonesia (2016-21) Figure 3.17: CIT policy gap and its components (% of GDP) (2021) 42.2% 41.8% 33.1% 45% 3.0% 35% 40% 30% 2.3% 2.5% 32.3% 33.0% 32.9% 33.1% 25.8% 35% 1.9% 25% 30% 1.7% 1.7% 2.0% 1.6% 25% 1.4% 20% 1.5% 20% 15% 15% 1.0% 10% 10% 0.5% 5% 2.5% 5% 0% 0.0% 0% 2016 2017 2018 2019 2020 2021 CIT policy gap Exemption to Exemption to companies in the companies and with CIT policy gap (% of CIT potential, LHS axis) final regimes turnover<50 billions CIT policy gap (% of GDP, RHS axis) of IDR Source: World Bank staff calculation. Source: World Bank staff elaboration. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 20 Prosperity Insight 4. SUMMARY AND CONCLUSION On average, the estimated gaps in VAT and CIT currently taxed at lower effective rates. As non- accounted for 6.4 percent of GDP on average, or compliance among these taxpayers is likely very IDR 944 trillion, between 2016 and 2021 (see high, to reduce markedly this component of the Table 4.1). Overall, non-compliance had a larger policy gap, the compliance gap would need to be impact on VAT revenue than the policy decisions. addressed in parallel. Moreover, the sectors and In contrast, the CIT policy gap was larger on types of companies subject to a lower CIT and average than the compliance gap. CIT and VAT VAT tax burden are largely subject to alternative compliance gap contributed to 58 percent of total regimes (for example, AFT or local tax on hospitality forgone notional revenue for both tax instruments. services). If these taxpayers were brought under Non-compliance ratios are very high compared the umbrella of CIT and VAT, the alternative regimes to Indonesia’s peer countries and in a broader would need to be discontinued or their scope would international comparison. Meanwhile, the policy need to be narrowed to avoid double taxation. In gaps are relatively modest. consequence, the impact of such a policy change on total tax revenue would be lower than the value Due to large compliance gaps and alternative of the estimated CIT and VAT policy gaps. tax regimes to VAT and CIT, the value of the policy gaps needs to be interpreted with caution. Changes in taxpayer compliance between 2016 Small companies and many service providers are and 2021 were the primary factor behind the ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 21 Prosperity Insight Table 4.1: Summary results (average 2016-21) Actual collection Potential Compliance gap (4) Policy gap (5) (1) Current policy (2) Benchmark (3) VAT IDR trillion 494 881 1,019 387 138 % GDP 3.3 5.9 6.9 2.6 0.9 CIT IDR trillion 306 467 725 161 258 % GDP 2.1 3.1 4.9 1.1 1.7 Total IDR trillion 800 1,348 1,744 548 396 % GDP 5.4 9.1 11.8 3.7 2.7 Source: World Bank staff elaboration Note: numbers may not add up to rounding. Values denotes in % refer to yearly averages than total values over the period. shifts in revenue. The value of the actual VAT and that CIT and VAT compliance have further improved CIT revenue can be decomposed into components, in recent years. which is helpful in understanding the underlying A relatively high threshold for VAT and CIT (of IDR sources of its evolution. Between 2016 and 2018, 4.8 billion) had a sizable contribution to the large the increase in tax base for VAT contributed size of both policy and compliance gaps. Although about 17 percent of revenue growth, whereas the top-down approaches do not allow a breakdown the increased compliance contributed to about of the compliance gaps into components, the scale 10 percent growth (Figure 4.1). Since 2019, the of the overall estimates confirm other evidence of trend reversed, changes in non-compliance were the threshold’s substantial impact on tax revenue. the main factor hampering revenue growth. At the Firstly, a relatively large proportion of taxpayers same time, statutory VAT rates were stable, thus outside the VAT and CIT regime below the threshold only minor shifts in the VAT effective rate resulting contributes to the relevant policy gaps. Secondly, mostly from changes in household consumption companies below the threshold are subject to structure. CIT compliance rate and tax bases have simplified reporting and less scrutiny, which likely been volatile, suggesting potential measurement increases taxpayer non-compliance. Moreover, this errors in estimating the base primarily using large cliff in the system incentivizes and facilitates national accounts data (Figure 4.2). Nonetheless, bunching. As shown by Saputro (2020) using the figures for 2020 and 2021 indicate that for administrative data, an excess mass of the firms both VAT and CIT follow a similar pattern. In 2020, in Indonesia were bunched in the turnover interval the revenue decline outpaced the liability, primarily of IDR 4.4-4.8 billion, which may result in non- due to decreased compliance and some deferred compliant behaviour like underreporting or artificial tax payments. In 2021, in the case of both VAT and separation of business. Lowering the threshold as CIT tax base and compliance have increased. Fast well as introducing legal prohibition of bunching growth of VAT and CIT revenue in 2022 and 2023, could decrease the VAT and CIT gaps. following the period covered by this study, suggests ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 22 Prosperity Insight Figure 4.1: Contributions to changes in VAT revenue (year- Figure 4.2: Contributions to changes in CIT revenue (year- over-year, 2017-21) over-year, 2017-21) 30% 60% 25% 40% 20% 15% 20% 10% 5% 0% 0% -5% -20% -10% -40% -15% -20% -60% 2017 2018 2019 2020 2021 2017 2018 2019 2020 2021 VAT compliance VAT base* CIT compliance CIT base VAT effective rate VAT revenue CIT effective rate CIT revenue Source: World Bank staff calculation. Source: World Bank staff calculation. Note: See Analytical framework and methodology for the Note: See Analytical framework and methodology for the description of VAT revenue components. description of VAT revenue components. Indonesia’s sizeable underground economy the Ministry of Finance (MoF), the revenue forgone contribute to tax compliance gaps.24 A relatively from tax expenditures accounted for about 1.5 low VAT and CIT revenue-to-GDP ratio compared percent of GDP in 2019. Additionally, due to the to Indonesia’s peer countries operating at implementation of fiscal stimulus measures from similar rates indicates a lack of efficiency in tax COVID-19 pandemic, this figure increased by collection. One important factor contributing to about 0.4 percentage point in 2020. Most of the this inefficiency is the prevalence of informality, revenue loss, approximately 1 percent of GDP, which results in gaps in tax compliance. According was attributed to VAT and sales tax exemptions on to a study by Medina and Schneider (2018), luxury goods. The remaining 0.5 percent of GDP in the underground economy in Indonesia was revenue loss was a result of tax incentives offered estimated to account for 21.8 percent of GDP in in income taxes. 2015. Similarly, Marhamah and Zulaikha (2020) The CIT policy gaps have declined since 2016, estimated it to be about 17.6 percent on average whereas the VAT policy gap has increased. The between 2016 and 2019. Unfortunately, recent shifts in the CIT policy gap were largely driven by estimates of the compliance gaps in taxes paid by the real deflation of the thresholds in the CIT system corporations are not available. Previous evidence that define the scope of preferential treatment for suggests, however, that the amount of tax revenue small- and medium-size companies. The change in forgone due to noncompliance could exceed the the VAT policy gap was largely driven by the change estimates of the underground economy.25 in the consumption structure of households, as Tax expenditures are estimated to have a moderate the share of expenditure on non-taxable financial, impact on the efficiency of tax collections and education, health care, and hospitality services has policy gaps. According to estimates provided by increased. 24. Underground economy refers to unregistered activities by tax and statistical authorities. 25. See Anjariadi (2011). Note that underground economy often presents in the sectors taxed at lower rates (for example, small companies that would be in the AFT regime if their activity was recorded). Furthermore, the underground economy and tax gap is not the same. The compliance gap includes errors, omissions, and bankruptcies that are not a part of the underground economy. At the same time, companies may evade taxes on recorded transactions (for example, overstate their input VAT). ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 23 Prosperity Insight The 2021 Tax Harmonization Law (2021 THL) government’s priority. Expanding access to timely implemented many changes in the system that information on taxpayers’ operations by utilizing have likely affected the size of the gaps. The third-party data and effectively utilizing available introduced reforms include, among others, the information today is a prerequisite for enhancing extension of a flat rate scheme to a number of compliance enforcement. Without improvements in businesses that were outside the VAT base (as taxpayer compliance, potential revenue gains from financial services and agricultural companies). THL removing tax expenditures—such as reducing the has also increased the scope of VAT obligations policy gap—will be constrained. Simultaneously, by reclassifying non-tangible goods and services taxpayers who remain outside of the system to exempt. To verify the impact of these and other (small taxpayers and non-taxable sectors) should changes, the gaps in CIT and VAT shall be monitored be brought under the umbrella of VAT and CIT further. reporting to fully integrate them into the system in the future. To increase revenue and fairness in VAT and CIT, improving taxpayer compliance remains the ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 24 Prosperity Insight APPENDIX ONE: METHODOLOGY AND DATA USED FOR VAT GAPS’ ESTIMATION Methodology The estimates of the VTTL necessary to estimate both the compliance and policy gap, were derived using a top-down “consumption-side” approach. This approach hinges on national accounts’ use tables as a source of information on tax base. The parameters are calibrated or derived based on third-party or fiscal data. These figures are used to estimate the VAT liability generated by different sub-aggregates of the total economy: household, government, and final consumption of NPISH; intermediate consumption (IC); gross fixed capital formation (GFCF); and other specific adjustments. The VTTL could be denoted using the following formula (6): ∑ (HHC rate x HHC Value ) N VTTL = i i i=1 + ∑ ( GOV rate + GOV Value )+∑ (NPISH rate + NPISH Value ) N N i i i i i=1 i=1 + ∑ ∑ (IC rate + prorata + IC Value ) N M i j i,j i=1 j=1 ∑∑ N M + (GFCF ratei + prorataj + GFCF Value i,j ) i=1 j=1 where: i denotes groups of products and services, j denotes sectors of economic activities, (HHC, GOV, NPISH, IC, GFCF) Rate are the effective rates for respective sub-aggregates of economy and groups of products and services, (HHC, GOV, NPISH) Value are respective components of the final use (denoted in net terms), IC Value is the value of IC (denoted in net terms), GFCF Value is the value of GFCF (denoted in net terms), and Prorata is the percentage of output in a given sector that is exempt from VAT or non-taxable (without the right to deduct). ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 25 Prosperity Insight Use tables are recorded in purchasers’ prices, thus final use values need to be net out before applying the formula (6). At the same time, the values of intermediate use in the use tables contain non-deductible VAT (but exclude deductible VAT). Net values of intermediate consumption liability for a group of products and services i and for consuming sector j are thus calculated as: tax ratei *pro-rataj IC Valueij = IC Gross Valueij * +1 tax ratei *pro-rataj Household consumption liability The core component of VTTL is the household consumption liability,26 which is a product of the effective VAT rates and household consumption values of each of the groups of products and activities. Household consumption values, similar to other components of the use tables, are recorded in purchasers’ prices, thus they require correction for the included VAT component. Moreover, the calculation requires adjustment for non- taxable consumption, in that self-supply and imputed rents. Government and NPISH consumption The government and NPISH consumption liabilities are estimated as a product of respective VAT rates and government and NPISH consumption values. Contrary to household consumption, most government and NPISH transactions do not constitute a taxable event. The exception is transfers in kind, which are one of the components of individual government consumption. Intermediate consumption The liability from intermediate consumption is computed for each industry as a product of intermediate use of each of the inputs times the average VAT rate for these groups of inputs times industry average proportion of non- deductible VAT in the intermediate consumption. The latter, propex or non-deductibility pro-rata coefficient, is estimated using the breakdown of sectoral production to narrow categories of goods and services. Importantly, as intermediate consumption is reported in purchasers’ prices, it includes non-deductible VAT that needs to be excluded from the use tables to reflect the net tax base. GFCF liability Similar to intermediate consumption liability, non-deductible investment is estimated as a product of tax rate, propex, and the base—that is, industry’s GFCF. The core components of this liability component include investment in dwelling (household investment) and public investment. Net adjustments In addition to the core components of the base, the estimation method involves corrections that are accounted for aside the main formula of the VAT gap model. More specifically, the main adjustments are: (1) the correction for small businesses under the VAT threshold, (2) exemption for public transport, (3) exemption for electricity, except for houses with power above 6600 VA. 26. See, for example, Barbone et al. (2013) for comparison of the VTTL components in European Union member states ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 26 Prosperity Insight Data sources A complete description of data, their sources and treatment are summarized in Table A1.1. Table A1.1: Data sources summary for VTTL calculation DESCRIPTION PURPOSE SOURCE COMMENT 1 Household final Estimation of Statistics Indonesia For 2016, Available for 185 groups expenditure by groups of Household (Badan Pusat Statistik: of products and services. For other products and services. Consumption (HHC) BPS): Use tables and years, values for 185 categories were liability. aggregate national forecast using up-to-date values accounts. broken down by main product and service categories. 2 Government final Estimation of BPS: Direct request expenditure by groups of government liability. of the value of social products and services. transfers in kind. 3 NPISH final expenditure Estimation of NPISH by groups of products and liability. services. 4 Intermediate consumption Estimation of BPS: Use tables and Available for 2016 for 185 groups broken down by groups of Intermediate aggregate national of products and services and 185 products and services. Consumption (IC) accounts. sectors (185x185). For other years, liability of industries values for 185 categories were for which VAT on inputs forecast using up-to-date values cannot be deducted. broken down by main product and service categories. 5 Investment (GFCF). Estimation of BPS: GFCF by For other years than 2018, rescaled VAT liability from Industrial Origin and using more aggregate national investment of exempt 24 Types of Assets accounts. sectors (housing (Billions of IDR, 2018). investment inclusive). 6 WARs for HHC and IC. Estimation of HHC and Own calculation. See following section. Rates derived IC components. for 2016-20. 7 WAR for government and Estimation of Own calculation: See following section. NPISH. government and NPISH Economy-wide WAR. liability. 8 Pro-rata coefficients for Estimation of IC liability Own calculation. See following section. estimating IC liability. of industries for which VAT on inputs cannot be deducted and companies below the VAT threshold. 9 Pro-rata coefficients for Estimation of GFCF Own calculation base See following section. estimating GFCF liability. liability of industries on the sectoral split for which VAT on inputs in GFCF by Industrial cannot be deducted. Origin and 24 Types of Assets. 10 Percentage of output of Net adjustments. World Bank Enterprise See following section. companies below the VAT Survey triangulated registration threshold. with administrative data. 11 Other net adjustments. Net adjustments. Tax expenditure - report. 12 VAT revenue. VAT revenue. Fiscal Policy Agency, Due to the unavailability of accrual MoF. revenue, cash collections had to be used. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 27 Prosperity Insight APPENDIX TWO: PARAMETER ESTIMATION Effective rates and 24 Types of Assets. It was assumed that the following sectors cannot deduct the VAT accrued The effective rates, or the WARs, were estimated on their investment: consistently for all products and services categories (185 groups) using the statutory list of exemptions. • A Agriculture, Forestry and Fishing For all categories except air transport, a zero or • B Mining and Quarrying standard rate (10 percent in the entire analyzed period) was assigned. Such a simple categorisation • K Financial and Insurance Activities was possible thanks to high granularity of products breakdown and homogeneity of the distinguished • O Public Administration and Defence; categories. Compulsory Social Security For air transport, the weighted rate was estimated • P Education at about 6.3 percent using the statistics on the number of passengers in domestic and international • Q Human Health and Social Work Activities flights provided by BPS. Percent of output of companies below the VAT Pro-rata coefficients registration threshold Similarly to the WARs, the estimation of pro-rata For manufacturing (International Standard coefficients for intermediate consumption, was Industrial Classification (ISIC) sectors 10 to 33), based on a categorisation (exempt vs. standard- the calculation of the liability for each sector and rated category). It was assumed that all fully groups of products uses the fraction of value added exempt categories have no right to deduct input created by micro and small companies published VAT (pro-rata = 100 percent). The values of the by BPS. For the primary sectors and services, due coefficients’ also account for the fact that the to unavailability of exact information, a 20 percent fraction of companies below the VAT registration proxy is used. The calculation assumes that the threshold cannot deduct their input VAT. ratio of intermediate use (that generates non- deductible VAT) to output is two times smaller than The pro-rata coefficient for estimating GFCF liability for medium and large companies. was calculated using GFCF by Industrial Origin ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 28 Prosperity Insight APPENDIX THREE: METHODOLOGY AND DATA USED FOR CIT GAPS’ ESTIMATION Methodology any other expenses that are not recognised as intermediate inputs (or calculated differently Similarly to the estimation of the VTTL, the in national accounts and companies’ books). estimation of the CTTL uses a top-down approach. It also requires an adjustment for the increase More specifically, the approach is based on RA-GAP in taxable income due to not qualified/denied methodology (Ueda 2018). Under this approach the expenses (Figure A3.1). tax base is calculated from existing macroeconomic data carefully adjusted to match with the statutory • Current-year tax base (C-TB): An aggregated tax base. result of profit-making corporations in a current year; this is before deducting carried- The starting point for estimating the tax base is over losses. GOS. GOS denotes the magnitude of value added that is produced by resident corporations and not • Tax base (TB): An aggregate result of profit- allocated to employees through compensation or making corporations in a current year minus to governments through taxes on production and deductions for carried-over losses; this is the imports. base for calculating aggregate CIT liability in a year (Ueda 2018). To match the actual tax base GOS needs to be adjusted, first to derive current-year net tax base, On top of estimating the tax base, in systems then current-year tax base, and finally tax base: where more than one statutory CIT rate is in place, the CTTL requires an estimate of the effective or, • Current-year net tax base (C-NTB): An in other words, weighted average CIT rate. This aggregated result reflecting both profit-making calculation, to the extent possible, should use corporations and loss-making corporations in third-party information, to account for varying a current year; this is generally smaller than non-compliance across types of companies. Data current-year tax base because losses made by sources and assumptions taken in this calculation loss-making corporations are netted out from for Indonesia are presented in the following sub- aggregate profits in C-NTB. The derivation of sections. C-NTB requires adjusting GOS by net capital gains, payments of interest and rents, and ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 29 Prosperity Insight Figure A3.1: Steps in calculating CIT base OUTPUT Intermediate Consumption Gross Domestic Product (GDP) Taxes on Gross Operating + Receipts of interests/dividends/rents Compensation of Mixed production and imports (D1.1) + Foreign source of income (taxable) employees income less Surplus (GOS) subsidies + Capital gains/Holding gains [D1] - GOS households/government/NPISH + Increases in taxable income Financial Accounting } } } - Payment of interests/rents (D2.1) (D1.2) calculation (denied costs/added [GOS] + [D1.1] - [D1.2] = [FAP] - Depreciation (accounting) Profits (FAP) revenues) - Capital losses/holding losses - Other expenses not recognised as intermediate consumption [D2] - Decreases in taxable income calculation Current-year Net + Losses of corporations (D2.2) (D3.1) (deduction/allowances for tax purposes) Tax Base (C-NTB) (current year) [FAP] + [D2.1] - [D2.2] = [C-NTB] Current-year Tax Base (C-TB) [D3] - Deductions for [C-NTB] + [D3.1] - [D3.2] = [TB] carried-over Losses (D3.2) Tax Base (TB) (previous years) Source: Ueda, J. (2018). Estimation of tax base The estimation algorithm is presented step-by-step in Table A3.2. Table A3.2: Adjustments to derive CIT tax base SOURCE OF STEP NAME OF ADJUSTMENT PURPOSE COMMENT INFORMATION From GOS to Financial Accounting Profit (FAP) 1 Adjustment of GOS Estimation of FAP, BPS: Integrated Added to GOS. for received interest, correction for elements Institution Balance dividends, and rents. that are included in Sheets for S.11 and statutory tax base but S.12. 2 Adjustment of GOS Subtracted from GOS. not in GOS. for received interest, dividends, and rents. 3 Adjustment of GOS for net Added to GOS. capital gains. Adjustment of FAP for other taxes 4 Adjustment of FAP for Adjustment of FAP, BKF (tax revenue Income taxes that are not a income taxes other than correction for elements figures). component of CIT base are CIT. that are included in deducted. statutory tax base but 5 Correction for the share of BKF and World Bank not in GOS/FAP. tax base in the final regime. Enterprise Survey. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 30 Prosperity Insight From FAP to C-NTB 6 Accounting depreciation. Correction for the Own calculations Accounting depreciation is not proxy of accounting based on BPS. included in GOS and must be depreciation. subtracted. Due to unavailability of data, accounting depreciation was proxied by the consumption of fixed capital adjusted by the share of GFCF purchased by corporations and shifts in prices of capital. From C-NTB to C-TB 7 Correction for losses. Adjustment of C-NTB Individual level data Own calculations of loss as since it is a sum of from EMIS, 833 percentage of gross profit. positive and negative largest taxpayers. values. From C-TB to TB 8 Correction for loss carry Adjustment of C-TB Individual level data Own calculations of loss carry forward. since some companies from EMIS, 833 forwards as percentage of net profit, are carrying forward largest taxpayers. maximum 5-year loss carry forward their losses, which assumed. decreased their liability. Estimation of the effective rate The effective rates were estimated at: Table A3.3: Effective CIT rate 2016 2017 2018 2019 2020 2021 23.67% 23.67% 23.67% 23.67% 21.09% 21.09% The estimates are based on the estimated share of public companies, companies with turnover below the IDR 4.8 billion threshold and between IDR 2.4 and IDR 50 billion thresholds subject to exemption and rate reduction, respectively. Other adjustments On top of estimating tax base and the effective rate, the liability from CIT was adjusted by exemptions, allowances, and other expenditures that could not be deducted from tax base. These adjustments are based on the tax expenditure reports. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 31 Prosperity Insight Bibliography Al Rikabi. 2018. Anjariadi, R. 2011. “Estimating Indonesia’s Corporate Income Tax Potential 2006-2010: An Analysis Of Input Output Table Home.” Jurnal BPPK, Vol 3. Badan Kebijakan Fiskal (BKF). 2023. “Tax Expenditure Report 2022.” Jakarta: Fiscal Policy Agency, Ministry of Finance. Barbone L., M. Belkindas, L. Bettendorf, R.M. Bird, M. Bonch-Osmolovskiy, and M. Smart. 2013. “Study to Quantify and Analyse the VAT Gap in the EU-27 Member States.” CASE Network Reports No. 116/2013. Canada Revenue Agency. 2022. “Overall Federal Tax Gap Report: Estimates and key findings for non-compliance, tax years 2014-2018.” Ottawa: Canada Revenue Agency. Canıkalp, E., I. Ünlükaplan, and M. Çelik. 2016. “Estimating Value Added Tax Gap in Turkey.” International Journal of Innovation and Economic Development 2 (3): 18-25. European Commission, Center for Social and Economic Research (CASE). Poniatowski, G., M. Bonch-Osmolovskiy, A. Śmietanka, and A. Pechcińska. 2022. “VAT gap in the EU – Report 2022.” Luxembourg: Publications Office of the European Union. https://data.europa.eu/doi/10.2778/109823 ———. 2023. Poniatowski, G., M. Bonch-Osmolovskiy, A. Śmietanka, and A. Sojka. “VAT gap in the EU – Report 2023.” Luxembourg” Publications Office of the European Union. https://data.europa.eu/doi/10.2778/911698. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 32 Prosperity Insight European Commission Fiscalis 2020 Tax Gap Project Group. 2018. “THE CONCEPT OF TAX GAPS Report II: Corporate Income Tax Gap Estimation Methodologies.” Brussels: European Commission. Gupta, S., and J.T. Jalles. 2022. “Priorities for Strengthening Key Revenue Sources in Asia.” Center for Global Development Working Paper 612. Washington DC: Center for Global Development . HM Revenue & Customs. 2023. “Measuring tax gaps 2023 edition: tax gap estimates for 2021 to 2022.” https://www.gov.uk/government/statistics/ measuring-tax-gaps International Monetary Fund (IMF). 2015. “South Africa Technical Assistance Report—Revenue Administration Gap Analysis Program—The Value-Added Tax Gap.” IMF Country Report No. 15/180. Washington DC: IMF. ———. 2017. “The Revenue Administration–Gap Analysis Program: Model and Methodology for Value-Added Tax Gap Estimation.” Technical Notes and Manuals No. 2017/004. Washington DC: IMF. ———. 2018. “Costa Rica Technical Assistance Report—Revenue Administration Gap Analysis Program—Tax Gap Analysis for General Sales Tax and Corporate Income Tax.” IMF Country Report No. 18/124. Washington DC: IMF. Jansen, A., W. Ngobeni, A. Sithole, and W. Steyn. 2020. “The corporate income tax gap in South Africa: A top-down approach.” WIDER Working Paper No. 2020/40. Helsinki: The United Nations University World Institute for Development Economics Research (UNU-WIDER). Keen, M. 2013. “The Anatomy of the VAT.” IMF Working Paper WP/13/111. Marhamah and Zulaikha. 2020. “How are the Underground Economy Levels in Indonesia?” Advances in Economics, Business and Management Research, Vol. 169. Medina, L., and F. Schneider. 2018. “Shadow Economies Around the World: What Did We Learn Over the Last 20 Years?” IMF Working Paper No. 2018/017. Nurfauzi, E.A., C. Nuryakin, and B.C. Putra. 2019. “Firms Bunching Response to Indonesian Income Tax Threshold.” Jejak: Jurnal Ekonomi dan Kebijakan (Journal of Economics and Policy) 12(1): 153–167. https://doi.org/10.15294/jejak.v12i1.18678 Organisation for Economic Co-operation and Development (OECD). 2018. “Revenue Statistics in Asian and Pacific Economies.” Paris: OECD Publishing. https://doi.org/10.1787/9789264308091-en ———. 2020. “Tax Administration Responses to COVID-19: Measures Taken to Support Taxpayers.” Paris: OECD. ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA 33 Prosperity Insight Saputro, H. 2020. “How Do Small Taxpayers Respond to Tax Simplification: Evidence from the Indonesian Turnover Tax Reform?” In 113th Annual Conference on Taxation. https://nta.confex.com/nta/2020/mediafile/ExtendedAbstract/Paper3934/ SME_Himawan_111720.pdf Sawulski, J., M. Klucznik, M. Reczek, and J. Rybacki. 2022. „Luka CIT w Polsce.” Working Paper, nr 2, Polski Instytut Ekonomiczny, Warszawa. Ueda, J. 2017. “The Evolution of Potential VAT Revenues and C-Efficiency in Advanced Economies.” Working Paper No. 2017/158. IMF. ———. 2018. “Estimating the Corporate Income Tax Gap: The RA-GAP Methodology.” Technical Notes and Manuals No. 2018/002. IMF, Fiscal Affairs Department. World Bank. 2011. “Analysis of Tax harmonization law impact on VAT and PIT.” ———. 2021. “Indonesia Economic Prospects — A GREEN HORIZON: Toward a High Growth and Low Carbon Economy.” Jakarta: The World Bank. Prosperity Insight ESTIMATING VALUE ADDED TAX (VAT) AND CORPORATE INCOME TAX (CIT) GAPS IN INDONESIA MIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY 34 | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY MIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY | ECONOMIC POLICY