Revenue System in Vietnam
Revenue System in Vietnam
Revenue System in Vietnam
66317
Tax Reform in Vietnam:
Toward a More Efficient and
Equitable System
Gangadha Prasad Shukla
Duc Minh Pham
Michael Engelschalk
Tuan Minh Le
(Editors)
3
10. Summary of Key Findings and Recommendations.................................................................................97
Chapter 4: Forecasting Tax Revenue: Methods And Tools by Gangadhar Prasad Shukla,
Daniel Alvarez, Tuan Minh Le, and Duc Minh Pham............................................................................101
1. Summary..........................................................................................................................................................101
1.1. Organizational Set Up and Forecasting Process..............................................................................101
1.2. Data Collection and Data Integrity....................................................................................................101
1.3. Present Status of Revenue Forecasting in GDT...............................................................................102
1.4. Revenue Forecasting Techniques and Modeling.............................................................................102
1.5. Strategy for Revenue Forecasting in Vietnam..................................................................................103
2. Introduction.....................................................................................................................................................104
2.1. Why Revenue Forecasting?..................................................................................................................105
2.2. Present Status of Revenue Forecasting in Vietnam.........................................................................107
2.3. Organizational Structure......................................................................................................................107
2.4. Legal Framework....................................................................................................................................108
2.5. Reporting Procedure.............................................................................................................................108
2.6. Coordination within MoF....................................................................................................................108
2.7. Linkage of Revenue Projection with the Budget Process..............................................................109
2.8. Data Collection, Data Maintenance, and Data Integrity................................................................109
2.9. Present Status of Database and Revenue Forecasting in GDT.....................................................109
2.10. Data from other Sources.....................................................................................................................111
3. Conclusion.......................................................................................................................................................111
3.1. International Practice in Revenue Forecasting.................................................................................111
3.2. Revenue Forecasting Techniques and Modeling.............................................................................112
3.3. Strategy for Revenue Forecasting in Vietnam..................................................................................114
3.4. Revenue Forecasting Model and Results for Vietnam....................................................................115
3.5. Summary of Model Results..................................................................................................................127
3.6. Suggestions for Improving Revenue Forecasting............................................................................128
Chapter 5: Tax Reform and Revenue Assignments in Vietnam by Jorge Martinez-Vazquez and
Minh Van Nguyen..............................................................................................................................................129
1. Introduction.....................................................................................................................................................129
2. Review of the Existing Revenue Assignments..........................................................................................130
3. What is to be Accomplished by Revenue Assignments?........................................................................140
4. Desirable and Less Desirable Forms of Tax Autonomy..........................................................................143
5. Significant Policy Issues with Revenue Assignments in Vietnam.........................................................145
6. Guidelines for Reforming Tax Assignments in Vietnam........................................................................147
6.1. What kinds of Tax Instruments are best suited for Subnational Governments?......................147
6.2. Selecting Tax Instruments for Assignment at the subnational level...........................................149
4
7. Recommendations for Reform....................................................................................................................155
Chapter 6: Integration and Government Revenues: Assessing the Implications of Vietnams
Accession to the WTO by Robert Warner.................................................................................................159
1. Summary..........................................................................................................................................................159
2. Introduction.....................................................................................................................................................160
3. Vietnams system for Taxing and Managing Trade...................................................................................161
4. Analyzing the Revenue Impact of Tariff Bindings...................................................................................176
5. Industry and Tariff Protection in Vietnam................................................................................................182
Chapter 7: Value-Added Tax by Gangadhar Prasad Shukla, Tuan Minh Le, and
Duc Minh Pham................................................................................................................................................191
1. Introduction.....................................................................................................................................................191
2. Exemptions......................................................................................................................................................198
3. Taxation of Special Sectors of the Economy.............................................................................................204
4. Zero-rating and VAT Refunds......................................................................................................................206
5. Policy and Administrative Recommendations.........................................................................................211
Chapter 8: Taxation In Vietnam: Who Pays What? by Jonathan Haughton..................................215
1. Introduction.....................................................................................................................................................215
2. Current Tax Structure....................................................................................................................................215
3. The Structure of Household Income and Expenditure...........................................................................219
4. Tax Incidence...................................................................................................................................................224
5. Property Tax....................................................................................................................................................236
6. Checking for Robustness..............................................................................................................................238
7. An Assessment of Incidence and Its Implications....................................................................................239
Chapter 9: Property and Land Taxes by Bruno Moser and Nicolaus Tideman............................243
1. Property and Land Taxes: Legal Background and Revenue..................................................................243
1.1. Summary of laws related to land taxes...............................................................................................243
1.2. The House and Land Tax......................................................................................................................243
1.3. Agriculture Land-Use Tax....................................................................................................................244
1.4. Land Use Right Transfer Tax...............................................................................................................245
1.5. Land and Water Surface Rental Charges...........................................................................................245
1.6. Land Use Levies......................................................................................................................................245
1.7. Registration Fee......................................................................................................................................246
2. The Economics of Property Taxes...............................................................................................................247
3. Land Tax...........................................................................................................................................................248
4. Equity................................................................................................................................................................251
5. Tax on Improvements....................................................................................................................................252
6. Tax on Real Estate Transfers.........................................................................................................................253
7. Policy Implications.........................................................................................................................................254
8. Land Recovery................................................................................................................................................254
5
9. Settling Land Disputes...................................................................................................................................255
10. Land Fragmentation and Consolidation.................................................................................................256
11. Internalizing Externalities...........................................................................................................................256
12. Land Taxes and the Poor.............................................................................................................................257
13. Application and Exemptions......................................................................................................................258
14. Equitization of State Owned Enterprises (SOEs).................................................................................258
15. Taxes on Rental Income from Real Estate...............................................................................................259
16. Protecting against Sudden Increases in the Rental Value of Land......................................................259
17. Pilot Projects.................................................................................................................................................259
18. Practical Application of Taxing Scarce Space: Londons Congestion Charge.................................260
19. Administrative Issues of Land Value Taxation.......................................................................................260
20. The Optimal Duration of Land Use Rights.............................................................................................262
21. Revenue Potential of Land Rent...............................................................................................................263
22. Sharing the Revenue from Rent Collection among Levels of Government.....................................264
23. Should Land be Rented or Sold?...............................................................................................................266
24. Dealing with Land that Has Been Sold by the State..............................................................................266
25. Securing the Natural Value of Land by Bonds........................................................................................268
26. Implementing Land Value Taxation.........................................................................................................269
Annex....................................................................................................................................................................273
References...........................................................................................................................................................295
Tables
Table 1.1: Average Buoyancy of Tax System, 19972009............................................................................40
Table 2.1: South Africa: Tax and Accounting Costs as Percentage of Turnover......................................63
Table 2.2: Revenue-Depleting Impact of Corruption...................................................................................65
Table 2.3: The Bigger the Shadow Economy, the Greater the Tax Leakage..............................................69
Table 2.4: Estimation of Operating Surplus by Economic Sector, 2005...................................................71
Table 2.5: Estimation of CIT Tax Gap, 2005...................................................................................................72
Table 2.6: Tax Filing Compliance by Major Taxes, 2010..............................................................................73
Table 3.1: Paying Taxes in Vietnam...................................................................................................................79
Table 3.2: Registered Taxpayer Population: An International Comparison.............................................81
Table 3.3: A New GDT Results Monitoring Framework..............................................................................92
Table 3.4: Which Are the Most Corrupt Government Agencies in Vietnam?.........................................93
Table 3.5: Key Issues in Anticorruption Strategy...........................................................................................94
Table 4.1: Budget Cycle.....................................................................................................................................105
Table 4.2: VAT Policy Impact Scenarios, 2007.............................................................................................118
Table 4.3: Average Tax Rates for Different Levels of Income Tax Ordinance 2004 and Current Law
6
Enacted in 2009...................................................................................................................................................123
Table 4.4: Assessing Progressivity of Old (2004 and New (2009) PIT System....................................124
Table 4.5: 2007 Revenue Forecasting Summary..........................................................................................127
Table 5.1: Tax Sharing Rates for the Provinces 200406, 200710, and 201115 Stability Periods.....131
Table 5.2: Provincial Revenues by Revenue Sources and Summary Statistics, 19972010................134
Table 5.3: Provincial Revenue per 1,000 Population and Summary Statistics, by Revenue Sources,
19972010...........................................................................................................................................................137
Table 5.4: Outcomes of State Budget Revenue Decentralization in Vietnam........................................139
Table 6.1: Composition of Revenues..............................................................................................................161
Table 6.2: Summary of Vietnams Preferential or MFN Tariff Structure, 200509...............................162
Table 6.3: Summary of Vietnams CEPT tariff schedule.............................................................................163
Table 6.4: Imports by Region of Origin: ASEAN and the Rest of the World, 19952005.................163
Table 6.5: Application of VAT to Goods since 2009...................................................................................165
Table 6.6: Special Consumption Tax on Goods (as at August 2007)......................................................167
Table 6.7: Excise Tariff since April 2009.......................................................................................................167
Table 6. 8: Vietnams Tariff Binding Transition.............................................................................................172
Table 6. 9: Tariff Bindings for Countries Acceding to the WTO..............................................................173
Table 6.10: Value of Imports by Customs Processing Treatment.............................................................178
Table 6.11: Real Imports, GDP, Growth, and Elasticities..........................................................................179
Table 6.12: Trajectory of Collection Efficiency............................................................................................181
Table 6.13: Vietnams Tariffs and Tariff Bindings by Economic Activity................................................183
Table 6.14: Regional Distribution of Manufacturing Activities and Average MFN and Bound Tariffs....186
Table 6.15: Summary Measures of Changes in ERPs..................................................................................188
Table 7.1: VAT Rates and Revenues by Regions..........................................................................................192
Table 7.2: VAT in Selected OECD Countries...............................................................................................193
Table 7.3: VAT in Asian Countries..................................................................................................................194
Table 7.4: Production and Consumption Efficiency Ratios for VAT by Region...................................194
Table 7.5: Productivity and Consumption Efficiency in Vietnam and Neighboring Countries........216
Table 8.1: Revenue from Taxes and Fees, 19972008................................................................................218
Table 8.2: Summary of Main Taxes in Vietnam............................................................................................221
Table 8.3: Sources of Household Income for Vietnam Households, by Income Per Capita Decile, 2006...222
Table 8.4: Breakdown of Household Expenditure by Vietnamese Households,...................................223
Table 8.5: Breakdown of Household Expenditure and Income, by Region and Urban/Rural, 2006....224
Table 8.6: Spending Comparison (in VND 000 per Household per Year, 2006).................................225
Table 8.7: Tax Paid by Vietnamese Households in 2006, by Expenditure Per Capita Deciles...........226
Table 8.8: Tax Paid by Vietnamese Households in 2006, by Income Per Capita Deciles....................227
Table 8.9: Tax Paid by Vietnamese Households in 2006, by Region and Urban/Rural Residence...229
Table 8.10: Measures of Tax Progressivity.....................................................................................................230
7
Table 8.11: Reconciling Survey with Budget Data.......................................................................................231
Table 8.12: Structure of Current and Previous Personal Income Taxes..................................................232
Table 8.13: The Distributional Effects of Revising the Personal Income Tax Law................................233
Table 8.14: Income and Estimated Tax for High-Income Households, 2005........................................235
Table 8.15: Estimating the Net Effect on Tax Revenue of Personal Income Tax Changes, 2006.......236
Table 8.16: The Distributional Effects of Introducing a Tax on Nonagricultural Real Estate.............237
Table 8.17: Tax Paid by Vietnamese Households, by Region and Urban/Rural Residence...............238
Table 9.1: Revenue from Land-Related Taxes (million VND)..................................................................246
Figures
Figure 01: Change in Tax Revenue Structure.................................................................................................18
Figure 02: Changes in Revenue Structure.......................................................................................................19
Figure 03: A Progressive Tax Burden................................................................................................................24
Figure 04: Gap in Operating Process................................................................................................................26
Figure 05: Tax Administration Modernization Objective............................................................................27
Figure 2.1: Determinants of Noncompliance and Revenue Leakage.........................................................61
Figure 2.2: Higher Corruption Associated with Greater Revenue and Tax Loss....................................66
Figure 2.3: Negative Correlation between Size of Shadow Economy and Tax-to-GDP Ratio.............68
Figure 3.1: Staffing Proportion in Hearquarters.............................................................................................77
Figure 3.2: Operational Gap............................................................................................................................... 78
Figure 3.3: Taxpayer Segmentations: Population vs. Revenue Contribution..........................................82
Figure 3.4: Tax Administration Modernization Objective and Indicators................................................88
Figure 4.1a: Typical Taxpayer Model Structure...........................................................................................121
Figure 4.1b: Aggregate Model Structure........................................................................................................121
Figure 4.2: PIT Progressivity (wage income)...............................................................................................123
Figure 4.3: PIT Progressivity (all sources of income).................................................................................125
Figure 5.1: Total Provincial Revenue, by Revenue Sources.......................................................................136
Figure 5.2: Coefficient of Variation of Provincial Revenue, by Revenue Sources.................................136
Figure 5.3: Average of Provincial Revenue per 1,000 Population, by Revenue Sources......................138
Figure 5.4: Coefficient of Variation of Provincial Revenue per 1,000 Population,................................138
Figure 5.5: Share of Local Budget Revenue in Vietnam..............................................................................139
Figure 5.6: Share of Local Budget Revenue in Vietnam..............................................................................140
Figure 6.1: Evolution of Vietnams Tariff Schedule, 1995 to 2005, and Bound Rates to 2019...........175
Figure 6.2: Distribution of Actual and Bound Tariff Rates, 2005 to 2019..............................................175
Figure 6.3: Potential and Realized Rates of Import Duty Collection......................................................177
Figure 6.4: Nonpetroleum Import Revenue Scenarios...............................................................................180
8
Figure 6.5: Petroleum Revenue Scenarios.....................................................................................................181
Figure 6.6: Distribution of Manufacturing Output and Employment by Average Actual and
Bound MFN Tariffs, 2005 and 2019...............................................................................................................185
Figure 6.7: Changes in Effective Rates of Protection for Import Competing Production Associated
with Implementation of Final Bound MFN Tariff Rates............................................................................187
Figure 6.8: Changes in Average Effective Rates of Protection Associated with Implementation of
Final Bound MFN Tariff Rates.........................................................................................................................188
Figure 6.9: Distribution of Industrial, Mining, and Agricultural Value Added in I/O Industries,
by ERPa, 2005 and 2019...................................................................................................................................189
Figure 8.1: Incidence of Taxes by Expenditure Per Capita Deciles in Vietnam, 2006..........................227
Figure 8.2: Incidence of Taxes by Expenditure Per Capita Deciles in Vietnam, 2006..........................228
Figure 8.3: Checking the Robustness of the Incidence Results................................................................239
Boxes
Box 01: Recommendations Based on Framework of a Good Tax System for Vietnam......................21
Box 02: Land Administration Reform and Property Tax.............................................................................22
Box 3.1: How Costly Is It to Comply with Tax Regulations?.......................................................................80
Box 3.2: Developing a Compliance Management Strategy..........................................................................82
Box 3.3: Examples of Tax-Related Simplification Proposals as Part of Project 30..................................86
Box 3.4: Computerization of Tax Administration Operations....................................................................89
Box 3.5: Key Elements of a Modern Taxpayer-Tax Administration Relationship...................................96
Box 3.6: Service Pledge Inland Revenue Authority of Singapore............................................................97
Box 6.1: Exemptions from Import and Export Duties...............................................................................164
Box 6.2: Other Issues in the Duty Payment System....................................................................................170
Annex
Annex 1: Vietnam: Key Economic Indicators...............................................................................................273
Annex 2: State Budget Revenues (share of GDP)........................................................................................274
Annex 3: Annual Revenue Buoyancy.............................................................................................................275
Annex 4: Features of a Good Tax System.......................................................................................................276
Annex 5: Documents Examined for Analyzing Major Tax Regimes in Vietnam...................................278
Annex 6: Implications of Joining the WTO for Various Sectors of the Economy.................................280
Annex 7: Tax-Royalty Regime versus Production Sharing Arrangements..............................................282
Annex 8: Suggestions about Decentralization of Charges and Fees.........................................................284
9
Annex 9: Tenets of a Good Tax Administration...........................................................................................285
Annex 10: Summary of Statistics of Variables Used in Empirical Analysis, 19922002......................287
Annex 11: Corruption and Revenue Leakage: Cross-country Estimates...............................................288
Annex 12: Summary of Three Methodologies to Measure Shadow Economy......................................291
Annex 13: Shadow Economy, Tax Collection, and Leakage......................................................................292
Foreword
Vietnams taxation system has undergone a fundamental reform in recent years. Tax policy
has been modernized taking into account regional and global trends and the need to facilitate
tax compliance of the rapidly growing private sector. The structure and transparency of
the tax policy framework has improved in line with the principles of the socialist-oriented
market economy. Cornerstones of this reform process were the creation of the necessary
legal conditions for accession to the World Trade Organization, the overhaul of the Personal
Income Tax system, moving from a taxation of high-income earners to the introduction of
a universal income tax, and the adoption of a Natural Resource Tax Law to promote their
efficient use and to ensure social equality in exploiting and using the natural resources of the
country.
At the same time, steps have also been taken to strengthen the administration of the tax
system in parallel with tax policy reform. A fundamental reform to organize the structure of
tax administration has been launched, key operational functions have been modernized, and
taxpayer services and assistance have strengthened voluntary compliance with the tax system
and reduced compliance costs. A backbone of the new phase of tax administration reform is
the development and deployment of an Integrated Tax Administration Information System,
based on an automation of reengineered business processes. It requires substantial investments
in terms of training of tax officials and strengthening the integrity and accountability by
undertaking organizational and procedural reforms.
Results achieved so far have been encouraging. The performance of the revenue collection
system has improved considerably, and net collections almost doubled between 2006 and
2010. It is expected that by 2015 a tax-to-GDP ratio of 23 to 24 percent will be achieved. The
share of domestic nonoil revenues in total revenue is expected to increase further, reducing
the dependency of the budget on revenues from natural resources.
Tax policy reform remains a priority for Vietnam, and there are major reform challenges still
to be addressed. The Value-Added Tax (VAT) system should further study the international
best practice. There is no VAT threshold in place and two different VAT rates continue
to apply. The direct tax system requires further strengthening to address newly emerging
challenges, such as cross-border tax evasion, transfer pricing, and e-commerce transactions.
The use of the tax system for ensuring better protection of the environment should be
developed further.
Reviewing the design and impact of the tax policy system in Vietnam to inform the future
reform process thus is a critical necessity for the Government. The studies in this publication
provide an important contribution to this review process. They were initially prepared for a
workshop on tax policy reform organized by the Ministry of Finance in cooperation with the
11
World Bank in Hanoi in May 2008. The studies have subsequently been revised and updated.
I am very pleased to introduce this important publication of the World Bank to a broader
audience.
12
Acknowledgments
This report was prepared by a team led by Duc Minh Pham and comprised of Gangadha
Prasad Shukla, Michael Engelschalk, and Tuan Minh Le under the guidance of Martin Rama
and Deepak Mishra. Viet Dinh Tuan of the World Bank was manager of the project.
The individual chapters and the overall outline and content of the report were discussed with
government representatives of Vietnam during a workshop held in Hanoi on May 2930,
2008. All chapters were subsequently updated by the respective authors. Ms.Nguyen Thi
Thanh Hoai of the Academy of Finance contributed significantly to the review and updating.
The editors would like to thank experts from a large number of government and
nongovernment institutions in Vietnam who provided invaluable comments and suggestions
during the preparation of the report. Special thanks are due to Mr. Do Hoang Anh Tuan, Vice
Minister of Finance; Mr. Vu Van Truong, Deputy Director General, General Department of
Taxation; Mme. Le Hong Hai, Deputy Director General, General Department of Taxation;
Mme. Nguyen Thi Cuc, Chairwoman of the Tax Consultancy Association; Mr. Nguyen
Van Phung, Deputy Director, Tax Policy Department, Ministry of Finance; Mr. Le Van
Ai, Academy of Finance; Mr. Pham Dinh Thi, Ministry of Finance; Mr. Quach Duc Phap,
Institute of Finance, Ministry of Finance; Mr. Dang Hung Vo, Former Vice Minister of
Ministry of Natural Resources and Environment; Mr. Vo Tri Thanh, Central Institute of
Economic Management; Ms. Mai Thi Thu Van, Ministry of Finance; Mr. Bui Ba Cuong,
General Department of Statistics; Mr. Nguyen Dinh Cung, Central Institute of Economic
Management; and Mr. Hoang Huu Phe, Director General, VINACONEX R&D. The
excellent editorial work of Diane Stramm is duly acknowledged.
The project benefited from secretarial support provided by Ms. Linh Anh Thi Vu and Ms. Chi
Kim Tran.
Special acknowledgments go to Mr. Vikram Nehru and Ms. Victoria Kwakwa of the
World Bank, who provided valuable support and guidance during the preparation process.
The project also benefited from comments and suggestions received from the staff of the
International Monetary Fund and from numerous World Bank colleagues. Their support is
gratefully acknowledged.
13
Abbreviations
AFTA ASEAN Free Trade Area
BOT build-operate-transfer
BT build-transfer
BTA Bilateral Trade Agreement
CIT Corporate Income Tax
CPAs Certified Public Accountants
EU European Union
FDI foreign direct investment
GDP gross domestic product
GDT General Department of Taxation
GSO General Statistics Office
IMF International Monetary Fund
IT information technology
ITAIS Integrated Tax Administration Information System
MFN most favored nation
MIS Management Information System
MoF Ministry of Finance
MPI Ministry of Planning and Investment
MTEF Medium Term Expenditure Framework
OECD Organisation for Economic Co-operation and Development
OOG Office of Government
PIT Personal Income Tax
PREM Poverty Reduction and Economic Management
SCT Special Consumption Tax
SEDP Socio-economic Development Plan
SMEs small and medium enterprises
SOE state-owned enterprise
TAMP Tax Administration Modernization Project
TIN Taxpayer Identification Number
U.K. United Kingdom
VAT Value-Added Tax
VIES VAT information exchange system
WCO World Customs Organization
WTO World Trade Organization
15
Executive Summary
OVERVIEW OF THE TAX SYSTEM IN VIETNAM
By Martin Rama, Deepak Mishra, and Duc Minh Pham
In 2010, after two decades of rapid economic growth, Vietnam passed the threshold to
become a lower-middle-income economy. Sustained market-oriented reforms combined
with intensive efforts to integrate into the world economy are among the key drivers of
this achievement. The reform of tax policy and administration has been a vital part of this
transition. This is leading to a fundamental change in the composition of taxpayers, from
large state-owned enterprises (SOEs) and foreign-invested companies to a myriad of small
and medium private enterprises. Economic transition is also leading to an equally important
change in the sources of government revenue, away from cross-border trade-related taxes and
revenue collection from crude oil toward a greater share of domestic tax revenue, in particular
taxation of business profits, labor income, and capital gains on land. However, completing
the transition to a market economy will require changes going beyond tax collection and
administration procedures, and will involve changes to the tax instruments themselves. At
the end of this process, Vietnam should have a set of taxes that is simple and transparent,
secures a stable flow of revenues for the government, encourages an efficient allocation of
resources, and does not risk constituting a source of inequality or unfairness.
An ambitious strategic target has been set in the Socio-Economic Development Plan 2011
2015 (SEDP). During the next five-year period, the state budget revenue is aimed at securing
macroeconomic stability and investments in infrastructure. Average total revenue collection
is expected to account for approximately 25.1 to 25.4 percent of gross domestic product
(GDP), out of which the revenue from taxes and fees will account for 23 to 24 percent of
GDP. Total revenue is expected to reach approximately 3,880 trillion dong, or one and a half
times the 200610 level. This would represent an average annual growth rate of state budget
revenues of approximate 15.6 percent per year.
The purpose of the series of studies in this volume is to shed light on the issues Vietnam
will be facing in the process of reforming its tax policy and administration. The studies are
also expected to lead to concrete policy recommendations contributing to the preparation of
key policies and legislative documents to ensure the achievement of the state budget revenue
target and other tax administration reform targets in the SEDP 20112015. It is expected that
the individual studies in this series will become useful inputs into the debate surrounding the
issuance of new laws and regulations. It is also hoped that the volume will support the reform
momentum in the tax policy area, leading to increased efficiency, transparency, and equity.
A Tax System in Transition
The first two major tax policy reforms undertaken in Vietnam in 1990 and 1998 were to adjust
to the transition to a market economy. A comprehensive legal framework was developed to
restructure the dated tax system in the centrally controlled economy. Taxes imposed on the
state, nonstate, and agricultural sectors were integrated into major, more standard instruments,
notably turnover and profit taxes. Subsequently, these were transformed into a Value-Added
17
T ax R eform in V ietnam
Tax (VAT) and a Corporate Income Tax (CIT), respectively. The tax structure has evolved
further to include natural resource taxes; Personal Income Tax (PIT); excises; customs
duties; and a number of minor taxes, fees, and charges. Since then, the structural change in
the Vietnamese tax system is characterized by three main features: (i) reduced dependence
on oil revenue and tariff revenue, (ii) a reduced share of the state sector and an increased share
of the nonstate sector in total revenue, and (iii) an increased importance of the VAT.
Revenue from tariffs is diminishing in importance given a rapid global integration. Vietnam signed
a bilateral trade agreement with the United States in 2001, participated in the ASEAN Free Trade
Area in 2006, and obtained World Trade Organization (WTO) membership in 2007. During this
process, tariff rates have been reduced and nontariff barriers and trading right restrictions have
been phasing out. More recently, this decline in tax revenue has been more than compensated by
the solid performance of other tax instruments, in particular by a highly productive VAT (figure
01). Import and export duties still accounted for 19 percent of total tax revenue in 2001, when
the reform process accelerated. They fell to 11.4 percent in 2006 but recovered to 15.6 percent
in 2010 owing to the significant increase in volume of international trade in recent years (figure
02). In addition, increasing global integration is leading to a higher ratio of imports to GDP, so
the reduction in tariff rates has been partly offset by the increase in the tax base.
Figure 01: Change in Tax Revenue Structure
Agricultural tax
PIT
2000: 2.2%; 2010:
2000: 2.4%; 2010:
0.01%
5.1%
Excise
2000: 6.7%; 2010: CIT
7.8% 2000: 37.2%; 2010:
32.4%
Natural
Resources Tax
2000: 9.6%;
2010:5.7%
2000
Export &
Import Tax
2000: 17.2%;
2010: 15.6%
2010
VAT
2000: 21.9%; 2010:
32.4%
The Vietnamese tax system in the current transition has the capacity to guarantee a
sustained flow of revenues for financing the government-increased recurrent expenditure
and investments. Despite the decline in trade-related proceeds, total average tax revenue
was 22.2 percent of GDP for 200109, a favourable comparison with regional neighbours.
This record has been achieved despite a consistent underestimation of tax revenue due
partly to poor forecasting capacity. Buoyant revenues are partly the result of high prices of
oil in international markets. However, they also stem from the operation of an extremely
efficient VAT. Tax productivity, defined as percentage points of GDP in revenue divided by
percentage points of the basic tax rate, is a standard performance indicator, in particular for
18
E xecutive S ummary
the VAT. Vietnams VAT productivity is close to 0.6, which exceeds the average for industrial
countries. Some cascading-inducing features of the Vietnamese VAT may contribute to this
exceptional performance. But there is more than cascading involved, which explains why the
VAT has become one of the workhorses of the Vietnamese tax system. The other workhorse
of the system is the CIT. Together with the VAT, this has been a dominant source of revenue,
with each of the two instruments contributing slightly less than 6 percent of GDP.
Oil revenue still constitutes the major source of revenue, despite the recent reduction of its share
in total revenue. Oil revenue as a share of total revenue decreased from 29.2 percent in 2005 (a
peak year in the last decade) to 13.4 percent in 2010 (figure 02, panel A). Meanwhile, nonoil
revenues increased steadily from 50.9 percent of total revenue in 2000 to 55.1 percent in 2008.
The structure of nonoil revenues has changed over time, as well. The importance of SOEs in
the economy has been decreasing steadily as Vietnam moves to a market economy, and the
trend is bound to continue. In contrast, the non-state sector has increased in importance.
The emergence of a vibrant private sector is changing the potential number and composition
of taxpayers. At present, some 500,000 enterprises and organizations, and about 1.7 million
business households and 9.0 million individuals, are registered to pay. These numbers are
expected to increase rapidly. The expectation is that the number of enterprises will increase by
close to 50 percent within five years. The number of small taxpayers could reach at least 2.3
million by 2012. Although tax revenues from all economic sectors have increased in absolute
terms over the past decade, as a result of their declining importance in the economy the share
in total budget revenues from the state-owned sector has decreased steadily while revenues
from the private sector and the foreign-invested sector increased (figure 02, panel B).
Figure 02: Changes in Revenue Structure
Panel A Panel B
A collection priority shifting from the and focusing more on
border to the domestic market, and emerging SMEs
reducing dependence on oil revenue
35.0 25.00
30.0 State-Owned
20.00
25.0 Oil
20.0 15.00
15.0 Foreign
10.00
10.0
Tariff Nonstate
5.00
5.0
0.0 0.00
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009/e2010/e 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009/e
Oil Revenue Tariff Revenue State-Owned Sector Foreign Invested Sector Nonstate Sector
19
T ax R eform in V ietnam
20
E xecutive S ummary
complicates administration and encourages tax evasion while doing little to promote equity.
Importantly, these brackets are not indexed to inflation, which may result in an unplanned
increase in the already high number of potential taxpayers. Given that business households
will be migrating from the CIT to the PIT, the highest tax rate for the latter should be similar
to the CIT tax rate. At 35 percent, the current VAT is much higher.
Despite its overall strong performance, the Vietnamese tax system is still unbalanced.
Revenue from some tax instruments is almost negligible, at less than 1 percent of total
revenue. Effort made could help include the large set of minor taxes, charges, and fees but
would result in unnecessary complexity. Tax instruments with a strong potential to influence
the allocation of resources and reduce waste are missing. And the main instrument being
introduced at present, the universal PIT, will not yield high revenue for some time but could
distort incentives if it is not phased in carefully. Current efforts to modernize the tax system
are focused on the introduction of a broader PIT. But the challenge is that PIT compliance is
typically low in developing countries, and an attempt to expand the tax base too rapidly could
penalize the formalization of employment, thus hindering modernization and job creation.
See box 01 for a discussion of what constitutes a good tax system.
21
T ax R eform in V ietnam
In addition, the number of potential taxpayers is growing much faster than current
administrative capacity can handle. The mushroom growth of small and medium enterprises
(SMEs) suggests that the missing middle of the size distribution pyramid could be populated
quite rapidly. Furthermore, taxing a much larger number of domestic private businesses as
they grow into sizable enterprises is bound to be challenging. Efforts have been made to
shift the responsibility for collection of these taxes to local governments, but decentralization
creates additional accounting and supervision problems. Taxpayers complain about
ambiguities and contradictory provisions in the tax system, which increase the compliance
burden and the discretionary powers of tax inspectors. Revenue forecasting capacity remains
weak. Projections for the following year are primarily based on the growth rate of revenues
during the current year, adjusted after consultations with line ministries and the General
Department of Taxation (GDT) field offices. No econometric or micro-simulation modeling
tools are used, and the quality of the data is poor. To address these concerns, a third wave of
reforms to tax policy and tax administration is underway.
At present, natural resources may be placed in four categories for the purpose of taxation.
Taxing natural resources has the potential to increase economic efficiency and reduce
environmental damage. Minerals are only subject to royalties. Natural forest products are
also subject to royalties, with rather high rates up to 35 percent applying. While high rates
aim to protect forest resources, they may also encourage the exploitation of high-value natural
wealth and create an incentive for evasion, thus negating their purpose. Natural aquatic
resources and natural water are subject to a combination of CIT and a low rate of royalty.
Most items in this category are taxed at 1 to 5 percent, except for pearls and sea slugs, which
are taxed at 6 to 10 percent. A similar combination applies to oil and natural gas, although
in this case CIT rates vary (from 6 to 40 percent) depending on the amount and depth of
extraction.
The combination of CIT and royalty is a sensible model, and it should be generalized to all
types of natural resources. The introduction of CIT in the case of minerals, and especially
of natural forest products, should allow reducing royalty rates. While the aim of such high
rates is to protect natural resources such as forests, in practice they tend to promote the
exploitation of high-value natural wealth and foster tax evasion. Royalty rates should be
reduced in number and should not be kept at an abnormally high level. This would also help
reduce administrative and compliance costs. However, the CIT rate should not be reduced
below 20 or 25 percent - not even in the case of the oil and natural gas sectors.
Property tax reform is another challenge which currently is being addressed. Box 02
summarizes key issues in the land administration reform in relation to property tax reform.
22
E xecutive S ummary
from property, this instrument can also lead to significant improvements in the quality of land
use. Taxing property can also help contain land speculation, one reason for the real estate bubble,
and thus contribute to improved land administration and equity.
Property taxation may result in a situation where poor people are unable to pay. Exempting
subsistence farms from property taxes and introducing a minimum imposable threshold are
obvious solutions to poor people owning land of little value. The problem of poor people with
land of significant value (particularly in cities) is that they may be reluctant to move to cheaper
land for understandable social reasons, especially in the case of the elderly. A possible way to
address this concern is to allow any poor person to defer his or her taxes (plus interest) until
death, and then recover the debt from the value of the property.
There is another concern that taxing property may be complex and costly compared to the revenues
it may potentially generate. A fundamental challenge is property revaluation on a regular basis.
Systematic property assessments conducted by area can bring the cost down. The self-assessment
of property prices is another practical solution. Avoidance risks can be mitigated through the
provision that the self-assessed value is what will be paid as compensation by government if the
property is reclaimed for the purpose of infrastructure development.
However, a more serious difficulty in Vietnam is the incomplete issuance of land-use right
certificates, especially in urban and upland areas. Household survey data from 2006 suggest that
only 76 percent of agricultural land parcels, 68 percent of urban land parcels, and 34 percent of
forestland parcels have been granted land-use right certificates. In practice, this means that about
two-thirds of the total land parcels of Vietnam still lack a certificate. Despite the promulgation
of the Land Law in 2003, the main reason for this problem is the shortage of basic infrastructure
for an effective operation of land administration, including cadastral mapping, transaction
registrations, and record management to the provision of land administration services. The
system itself is cumbersome, not transparent, and inefficient, and does not provide the quality
services that end users can rely on. As a result, land administration is generally recognized as one
of the most severe constraints to business development and transparent governance in Vietnam.
Combined with limited local capacity, this calls for a step-by-step approach in introducing this
new tax instrument.
23
T ax R eform in V ietnam
mostly everybody is confronted with the VAT, whereas a meager 0.1 percent of households
pay the PIT. A study using household survey data from 2006 shows that the Vietnamese tax
system also does well in terms of equity. Richer groups, whether measured by income per
person or by expenditure per person, face a proportionately higher tax burden (figure 03).
Compared to results of a similar study for 1998, the tax burden is now heavier, and it is more
progressive. The 1998 study found that taxes represented about 8 percent of expenditures for
all but the top quintile, where they stood at just over 10 percent. The figures for 2006 show
a tax burden of about 8 percent only for the bottom two population deciles, rising steadily to
reach a plateau of about 14 percent in the four top deciles.
This higher, more progressive, tax structure is due largely to the introduction of the VAT in
1999, when it replaced a complex system of turnover taxes. The VAT is not progressive in
every country. But in Vietnam poor people provide for many of their own needs rather than
purchasing them, which in turn reduces the VAT they must pay. Over time this will change, as
home production becomes less important, making the tax system gradually less progressive.
Figure 03: A Progressive Tax Burden
16
14
12
Tax burden
10
8
6
4
2
0
1 2 3 4 5 6 7 8 9 10
Population decile (from poor to rich)
Percent of income Percent of expenditure
24
E xecutive S ummary
The tax reform program adopted by the government confronts GDT with a series of new
challenges. Shifting collection priorities from the border to the domestic market, reducing
dependence on oil revenues, focusing more on private sector enterprises and less on SOEs
and foreign-invested companies, adjusting to self-assessments as opposed to inspections,
introducing modern taxpayer service processes, and adapting to the electronic formats
increasingly used by taxpayers to keep their records are among the major issues to address.
The Tax Administration Law that became effective in July 2007 was an important step toward
addressing these challenges. The law provides a foundation for ensuring transparency and
integrity in all tax administration operations. In addition, it eliminates gray areas potentially
fostering noncompliance by taxpayers and abuse by tax administration officials. Among
several important innovations, the law introduces a unique Taxpayer Identification Number
and stipulates that third-party information can be acquired from various agencies for effective
audit and enforcement, while safeguarding its confidentiality. But there are some weaknesses,
too. While the law stipulates the general rights and responsibilities of the tax administration
agency and other related sub-national organizations, it does not include provisions on the
division of responsibilities between various levels of tax administration, or on the division of
responsibilities between the tax administration agency and Peoples Councils and Peoples
Committees at various sub-national levels.
An assessment of GDTs operational processes was conducted by reviewing 15 functional
aspects of the tax administration performance (figure 04). Overall performance is below
average in all main and support functions scored in accordance with an international best
practices research scale, with supporting data for the framework gathered from 30 developing
and developed countries tax administrations. Among them, the debt collection and
enforcement function turned out to be the area with the most advanced processes in place.
Multiple payment methods (cash and bank transfer) and the use of technology to access
taxpayer account data combined with semi-integrated collection systems to monitor taxpayer
delinquency are among key factors contributing to this advanced function. In addition,
provisions for the counseling and advising of GDT staff on legal issues exist, so that at least
basic support is available to help tax administration staff deal with more complex legal matters.
However, the legal support is not yet comprehensive and additional processes to perform
critical legal tasks, such as creating the capacity to issue private letter rulings to taxpayers, still
have to be developed.
Existing processes are less satisfactory in most other areas, with the tax audit function being an
area of particular concern. Procedures and methodologies for a risk-based audit selection have
not yet been developed, and auditor discretion in the selection of cases for a desk or field audit
is substantial. In addition, existing processes for the conduct of an audit provide substantial
discretion to the audit team, and the overall audit planning and supervision process does not
yet include a systematic ongoing monitoring of audit efficiency. Anti-evasion appears to be
the weakest area, where no strategic plan for addressing evasion cases exists.
25
T ax R eform in V ietnam
Pre-Filing Taxpayer
Services
Support Functions
(Financian and 5 Filing - Return
Human Resources Processing, and Tax
4 Accounting Function
Management, and IT
3
Operations 2
Mangement-Strategic 1 Audit Function
Planning, Perf Mgmt,
QA, Communications 0
A taxpayer baseline survey conducted in 2007 and 2008 provides some information on
how GDT is doing so far. On the positive side, 61.3 percent of the taxpayers interviewed
considered the services to the public provided by tax offices to be above average. However,
only 3.9 percent of them rate these services as very good, suggesting that there is still scope
for improvement. The taxpayer baseline survey also reveals that taxpayers struggle to
understand major components of the CIT legislation, such as preferential regimes or the
determination of deductible expenses. Less than 40 percent of respondents found the VAT
refund system easy to understand and implement. And more than 60 percent reported
finding out about changes in the laws and procedures less than a month prior to their
introduction and enforcement. This highlights the need to improve taxpayer information
and communication.
A 10-year tax system reform strategy has been drafted to set a target that taxes and fees
(excluding revenue from crude oil and export revenue) should constitute 70 to 75 percent of
the total state budget revenue by 2015, and 75 to 80 percent by 2020, comparing favorably
with the current level of 62.5 percent in 2010. An ambitious program supported by the
World Bank to modernize the tax administration is underway to improve governance in tax
administration and raise the level of volunteer compliance with the tax system. Through this
program, government resources, combined with financial and technical assistance from the
donor community, will be mobilized to implement a comprehensive reform plan to improve
the business climate, sustain revenue collection, and reduce corruption.
The Tax Administration Modernization Project was prepared in 2007. It is considered
to be the main driver to achieve the target that, by 2015, Vietnams tax agency will rank in
the top five in the East Asia region. The project aims at assisting GDT to strengthen its
governance in tax administration and to increase the level of voluntary compliance with
the tax system by improving the effectiveness, efficiency, transparency, and accountability
of the tax administration. These reforms are expected to have a positive impact on the
26
E xecutive S ummary
business climate by developing predictable and enforceable tax administration processes and
procedures. The sustainability of revenue collection and greater equity will also be supported
through implementation of this project. The projects development objectives, therefore, are
complementary to two higher-level objectives of the countrys four reform pillars, namely
business-led development and strengthened governance systems. The development and
deployment of an Integrated Tax Administration Information System based on reengineered
business processes reflecting international best practice is assumed to be the backbone of the
five-year plan and 10-year strategy (figure 05).
Figure 05: Tax Administration Modernization Objective
Business-led Strengthened
Development State Governance
Source: Based on the World Bank-funded Tax Administration Modernization Project Document.
This comprehensive modernization program is not risk free. A reform program that
intends to reduce the opportunity for part of the tax officers to charge discretionary rents
might be seriously resisted. Implementing an ambitious multifaceted reform agenda, with
interdependence between a complex computerization component and other required
activities might lead to delays in meeting the targets of the reform agenda, considering in
particular the weak implementation capacity of the administration.
Five factors, therefore, are considered critical for the sustainability of the reforms:
Strong government ownership of, and a high level of commitment to, its reform agenda,
including the Tax System Strategy for 2011-20
Mobilization of private sector support and partnership
Sound and quantifiable performance indicators and client standards introduced, so
that any slippage in performance will be easily identifiable and tax officials can be held
accountable
Development of a sustainable, in-house capacity for continuous improvement following
full implementation of the Reform Plan
27
T ax R eform in V ietnam
28
E xecutive S ummary
By identifying the key private and public stakeholders and the role they play in defining the
strategy and supporting the implementation of reforms, the chapter outlines the elements of
a comprehensive strategy to improve voluntary tax compliance that reduces compliance costs
while ensuring integrity and transparency in tax administration. In assessing the efficacy
of these reforms to meet these objectives, the chapter highlights the challenges faced and
concludes that organizational changes in the direction of reengineering business processes
and strengthening of the tax administration headquarters and the establishment of specialized
administrative structures for the internal audit, the appeals, and the compliance monitoring
functions need to be supported by enhanced information technology support. Reforms in
human resource management with regard to training and capacity building to enable tax
administration staff to perform highly specialized tasks in a function-based organizational
structure, as well as workforce planning and workload analysis, and the design of appropriate
incentives to reward high performance, are also recommended.
As with the preparation of government budgets, effective tax administration requires accurate
forecasting of tax revenues. As with most developing countries, both statistical data and
institutional capacity to make these forecasts need improvement. This is one task of the tax
administration reform and the theme of Chapter 4. The chapter provides an overview of the
forecasting tools and methods currently in use and lays out the options available, including
specialized modeling tools like input-output-based modeling for the VAT. Some of these
methods and tools are applied to illustrate forecast of key taxes such as the VAT, the CIT,
and the PIT. Findings from these exercises are used to show how they can be used to assist
policy makers in making appropriate decisions. These findings are, in the sequence for
implementation:
The need to strengthen the tax database that GDT can access
Training of personnel in relevant economic theory and quantitative techniques
Model building
Proper sequencing according to an appropriate time frame of up to 24 months.
Once tax revenues are received, how they are allocated becomes important. This importance
is enhanced in the context of tax reform. Lack of coordination between tax reform and
assignment can lead to processes working at cross purposes. The most important objectives of
tax reformsimplicity, low economic distortions or efficiency, and equity in the distribution
of the tax burdenare less vital in decentralization reform, where revenue adequacy and
local government accountability take center stage. Given Vietnams drive toward fiscal
decentralization, budget allocation norms have been established for recurrent and capital
expenditures. These norms have been translated into rules for tax revenue assignment. In
Chapter 5, an overview of tax assignment to and revenues derived from sub-national levels
reveals arrangements that are ad hoc. These arrangements are then assessed in terms of
their contribution to the accountability of local officials and their compatibility with the
tax administration reform strategy. This chapter concludes with proposals to strengthen
arrangements for tax revenue sharing. These relate to strengthening revenue adequacy at the
sub-national level, improvement on the apportionment of tax revenues between the central
and the provincial governments, and the introduction of own revenue sources for local
29
T ax R eform in V ietnam
30
E xecutive S ummary
with higher revenue and the other lower. Finally, the chapter looks at the likely impact of a
property tax on nonagricultural real estate, which Vietnam does not have, and comes to the
conclusion that it can lead to substantial yields and be very progressive.
Chapter 9, the final chapter, provides observations on property and land taxes. Despite a
modest contribution to tax revenue, land taxes help broaden the tax base, potentially increase
economic efficiency, and improve equity. The property tax becomes an increasingly useful
instrument to provide local governments with resources to invest in local infrastructure.
Taxing property may be complex and costly compared to the revenue it might generate. The
chapter suggests a distinction between the land and the capital improvement components
of property taxes, given the different elasticity of their supply. A fundamental challenge is
that property should be reevaluated regularly. Ineffective operation of land administration
is one of the most severe constraints to business development and transparent governance in
Vietnam.
31
The Legal Framework for
Tax Administration I
By Duc Minh Pham, Tuan Minh Le, and Gangadhar Prasad Shukla
1. Summary
Vietnam is currently following a course of economic integration with the world economy. To
facilitate this process, the government has undertaken a series of trade and tax policy reforms
and has improved its fiscal management. As a result, the country has enjoyed spectacular
growth during the past two decades, and most economic indicators have shown a very robust
trend. Inflation has been generally under control, the trade deficit mild, and the exchange rate
fairly stable. The budget performance has also been encouraging. The growth in revenues has
been impressive, while expenditures have been contained, thus keeping the budget deficits
within manageable limits.
The country has also been taking steps to reform its tax laws and modernize its tax
administration. The tax system has undergone three phases of reform in the past two
decades, and the third and current phase of reform has been particularly motivated by the
increasing integration of Vietnam into the world economy and its effort to join the World
Trade Organization (WTO). The existing tax regime in Vietnam is analyzed below under
the framework of a Good Tax System, that is, equity, economic efficiency, stability, and
technical efficiency.
33
T ax R eform in V ietnam
iv. The credit method should be adopted across-the-board, and the subtraction method
applied to small traders must be abolished when the exemption threshold is established.
v. The VAT refund system needs to be rationalized and simplified to lower collection
(administration and compliance) costs and to alleviate cash flow problems for businesses.
Excises or Special Consumption Tax (SCT)
The tax base consists primarily of goods whose consumption is regarded as socially and
environmentally undesirable, and luxury goods. The tax rates on goods are highly dispersed,
varying between 10 percent (gasoline; air conditioners of 90,000 BTU or less; and electrically
operated passenger cars of between 16 seats and under 24 seats) to 70 percent (votive gilt
papers and votive objects) and on services between 15 percent (lottery business) to 40 percent
(dance halls). The tax on gasoline, however, appears to be low compared to neighboring
countries. The excise regime should be reformed as follows:
i. Tax rates are quite dispersed. Some top excise rates should be lowered while some low
rates may be increased.
ii. In the short run, the government may retain the low excise rate on gasoline, due to the
increase in the international price and high inflation. However, over the medium to long
term, the rate may be increased to the level comparable with neighboring countries.
iii. A cost-benefit analysis of levying an excise tax on some of the entertainment items may
be conducted to compare the cost of administration and compliance with the revenues
collected.
iv. The provision of tax reductions and exemptions for establishments facing financial
difficulty should be eliminated.
v. Expansion of the tax base under excises may be considered by bringing into the tax net
some additional luxury items primarily consumed by higher-income groups.
Trade Taxes
The tariff rates could be put into four categories: (i) preferential rates applied to goods
originating from countries in the most favored nation (MFN) category; (ii) rates applied
to countries without any trade agreements (50 percent higher than MFN countries); (iii)
special preferential tariff rates for countries with special trade agreements with Vietnam;
and (iv) antidumping and antidiscrimination laws. Thus, Vietnam has a multiplicity of tariff
rates. The tariff rates are first assigned based on country of origin (five rates) and then on the
classification of goods (four categories) resulting in 20 tariff rates. The following suggestions
should be considered to reform the import taxes:
i. There should be a basic set of tariffs covering preferential, MFN, or ordinary tariffs,
supplemented by a set of special preferential tariffs for countries with treaties. For
countries not accorded favored treatment, the MFN rate plus a certain percentage should
be used. Preferably, the number of categories based on origin should not exceed three.
ii. Within the base category, three rates based on classification of goods are suggested: (a)
the lowest rate (for example, 5 percent), applicable to inputs; (b) the next-higher rate (say
15 percent), applicable to intermediate goods; and (c) the highest rate (say 25 percent),
applicable to pure consumption goods.
34
T he L egal F ramework for T ax A dministration
iii. For goods whose imports are to be discouraged, a high tariff rate may be specified.
Alternatively, imports of these goods may be licensed to keep the number of rates at three.
iv. The government should implement formal WTO agreements, in consultation with the
World Customs Organization (WCO), trade facilitation measures such as harmonization
of tariffs, advance rulings, specialized appellate authorities, etc.
Tax Administration
A Tax Administration Law became effective on July 1, 2007, and provides the legal basis to
ensure transparency and integrity in tax administration. In general, the law is comprehensive
36
T he L egal F ramework for T ax A dministration
and high quality, and it stipulates clear provisions for governing taxpayer services, tax appeals,
and dispute settlement. Some commendable features of the Tax Administration Law are as
follows:
i. The law clearly specifies the rights of taxpayers during the process of tax assessment,
examination, audit, investigation, and appeals. The law makes it mandatory for tax
administration to compensate for damages caused to the taxpayer.
ii. The law emphasizes the granting and use of Taxpayer Identification Numbers (TINs),
which is the first step toward a modern tax administration.
iii. There is symmetry of treatment of tax payment and tax refunds. If taxpayers delay paying
taxes, they are liable to pay a penalty. In the case of overpayment of taxes, a refund must
be made within a prescribed time.
iv. The law stipulates that third-party information can be acquired from various agencies,
while confidentiality of the information for tax purpose must be safeguarded.
The law has also exposed the following major shortcomings that should be resolved:
i. While the law stipulates the general rights and responsibilities of the tax administration
agency and other related sub-national organizations, it does not include specific provisions
on the division of responsibilities between various levels of tax administration (that is,
the General Department of Taxation [GDT], provincial offices, and district offices).
ii. There is a lack of a detailed legal basis in the following administrative operations: (a)
the law does not specify cooperation between the tax and customs agencies; (b) the law
does not include provisions on accounting and bookkeeping; (c) there is no provision
governing electronic filing and the recognition of electronic signature; and (d) while the
law is clear about the rights of appeal, it does not yet specify levels of appeal.
2. Introduction
Vietnam is currently following a course of economic integration with the world economy.
To facilitate this process, the government has undertaken a series of trade and tax policy
reforms and has improved its fiscal management. As a result, the country has enjoyed
spectacular growth during the past two decades, and most economic indicators have shown
a very robust trend. Inflation has been generally under control, the trade deficit mild, and the
exchange rate fairly stable. The budget performance has also been encouraging. The growth
in revenues has been impressive, while expenditures have been contained, thus keeping the
budget deficits within manageable limits. Selected economic indicators for 200110 are
shown in Annex 1.
The country has also been taking steps to reform its tax laws and modernize its tax
administration. The features of the existing tax regimes of direct and indirect taxes are
analyzed in this document within the framework of a Good Tax System and in the context
of international best practices. Suggestions for improving the tax regimes are presented and
their linkage with improved tax administration and compliance are explained. The existing
Tax Administration Law is also scrutinized within the framework of Tenets of a Good Tax
Administration, and directions for further reform are presented.
37
T ax R eform in V ietnam
Trade Liberalization
During the past two decades, the trade policy regime has undergone significant changes
mainly in three directions: (i) restrictions on trading rights (that is, the right to import and
export) have been relaxed; (ii) tariff rates have been reduced; and (iii) nontariff measures
have been slackened. This has also been accompanied by changes in tax policy.
Over the years, licensing requirements have been significantly reduced for exporting and
importing. Currently, all enterprises with business licenses may engage in foreign trade in
the goods specified in their business license without an import-export license except for a few
restricted categories. The tariff schedule was rationalized following the countrys accession to
the ASEAN Free Trade Area (AFTA) in 1992 and in preparation for the countrys accession
to the WTO in 2007. Finally, in the past few years, Vietnam has significantly reduced the use
of nontariff barriers. Another significant step was taken in 2000, when Vietnam signed the
Bilateral Trade Agreement (BTA) with the United States, which envisages phasing out all
nontariff barriers over a period of three to seven years.
and the tax is efficient from an economic perspective. The tax is the largest source of state and local
taxes in the United States and elsewhere in the developed world. However, the tax is still largely
marginal in developing countries, and Vietnam is not expected to collect large revenues from
this tax over the short term. Reasons for this may be attributable to meager performance in the
developing world: a tax culture and tradition combined to form political and social resistance to
the tax; institutional constraints (in particular, the capacity of revenue administration agencies is
typically weak, while the valuation and collection of the tax are sometimes completely segregated);
and valuation of real property (both land and improvements on land) are inherently complex.
38
T he L egal F ramework for T ax A dministration
and keeping the system simple and transparent. This should also generally take care of the
transfer pricing problem.
Revenue Performance
Tax revenues in 2010 constituted around 86 percent of total government revenues, and about
24.6 percent of GDP. Total revenue collection is currently about 28.6 percent of GDP, which
compares favorably with Vietnams neighboring countries. The fact that the tax revenues have
been consistently high and total revenues have been more than 20 percent of GDP over the past
10 years is a remarkable achievement for a country like Vietnam. The revenues from different
taxes and their buoyancy over the past decade are shown in Annex 2. The annex reveals that
the CIT and the VAT are the two major sources of revenue, each contributing slightly less than
6 percent of GDP. The trade tax remains another major tax, accounting for approximately 3
percent of GDP. Excise and natural resource taxes have shown an upward trend, and were 2.5
and 1.3 percent of GDP, respectively. The contribution of the Personal Income Tax, although
on an upward trend, is still quite low at 1.3 percent of GDP, according to 2010 statistics. Other
taxes are almost negligible at less than 1 percent of total tax revenues. This shows that the tax
system is currently somewhat skewed and needs to be suitably reformed.
39
T ax R eform in V ietnam
Ultimately, the tax system should become elastic and not just buoyant; that is, growth in tax
revenues should be at least proportionate to growth in GDP without the need for making
frequent discretionary changes in the base or rates. The recent dramatic improvement in
buoyancy may cause some concerns that a higher tax burden may choke off investment and
innovation. From the revenue efficiency perspective, however, higher buoyancy relates to more
revenue stability and better opportunity for revamping the tax regimes to achieve the seemingly
opposite aspects of a good tax system: enhanced economic efficiency and targeted equity. 2
Analysis of Specific Tax Regimes under the Framework of a Good Tax System
The following major direct and indirect taxes are examined in order to find the key issues
that need attention during the current reform phase: (i) the Value-Added Tax (VAT), (ii)
the Excise Tax, (iii) the Trade Taxes, (iv)the Personal Income Tax (PIT), (v) the Corporate
Income Tax (CIT), (vi) Natural Resource Taxation, and (vii) fees and charges.
The analysis of these tax regimes is conducted under the framework of a Good Tax System,
targeting, in particular, the aspects of efficiency, equity, transparency, revenue adequacy, and
stability. The features of a good tax system are briefly described in Annex 4. An overall picture
of the tax system in Vietnam is presented, followed by an analysis of specific taxes.
Economic Efficiency
At present, there are several non-neutralities with respect to both tax policies and their
implementation. A lack of harmonization among taxes, a multiplicity of rates, and a large
number of exemptions and tax incentives give rise to a great deal of inefficiency in resource
allocation, because taxpayers have incentives to base their business decisions on minimizing
their tax burden instead of optimizing output and production.
For a detailed discussion of the basic concepts of tax buoyancy, tax elasticity, taxable capacity, and
2
tax efforts, see, for example, Le et al. (2008); Musgrave and Musgrave (1989); and Shome (1988).
40
T he L egal F ramework for T ax A dministration
Equity
Under the current system, there are numerous violations of horizontal equity. For instance,
taxpayers under the PIT are taxed at different rates depending on the type of income earned.
The numerous exemptions also lead to a violation of vertical equity. 3 The same thing is true
for indirect taxes with multiple VAT and excise rates and many exemptions.
Transparency
Transparency primarily refers to tax laws that are not overly complex and discriminatory
and tax regulations and administrative requirements that are easy to comply with. From this
perspective, Vietnam has made significant strides in achieving transparency, although a great
deal of reform is yet to be implemented. For example, in its present form, the VAT has zero
plus two other rates and a long list of exemptions. Excises have many different tax rates and
exemptions. The CIT has several tax rates and a series of incentives related to the nature of
industries and their location that leave effective tax rates different from their statutory rates.
3. Value-Added Tax
Main Features of the Current VAT Regime
The VAT has proved to be the main workhorse for raising tax revenues in Vietnam. There are
three VAT rates: zero percent, mostly for exports; 5 percent tax, for 15 goods and services;
and 10 percent (the standard rate), for the remaining items. Twenty six goods and services
are VAT exempt. At present, there is no threshold based on turnover for exempting small
3
Horizontal equity refers to a tax system that induces individuals of similar income level, regardless
of their sources of income, to pay the same amount of taxes as a share of their gross income. Vertical
equity refers to a tax system that induces individuals of higher income levels to pay more in taxes as
a share of their gross income.
41
T ax R eform in V ietnam
producers and traders from the VAT net. The credit method is used in parallel with the
subtraction method, which is applied exclusively to small traders who fail to fully observe
regulations on accounting, invoices, and documents, and to gold, silver, and gem-trading
activities. The present VAT refund rules in Vietnam are complicated and vary with the nature
and activity of the business.
42
T he L egal F ramework for T ax A dministration
tradeoff between the cost of administration and expected revenue generated. Instituting
a turnover threshold for small traders will help the administration focus on big taxpayers
rather than wasting its resources on a large number of taxpayers who contribute very little
to tax revenues. A simplified tax regime can be applied to all traders below the threshold.
iv. The credit method should be adopted across-the-board and the subtraction method
currently applied to small traders should be abolished when an appropriate exemption
threshold is established. This will improve tax compliance since it leaves a trail of invoices
in all cases that may be audited and verified.
v. The current VAT refund system needs to be rationalized and simplified to lower collection
(administration and compliance) costs and to alleviate cash flow problems for businesses
claiming genuine VAT refunds.
43
T ax R eform in V ietnam
the United States it is around 10 U.S. cents per liter, while in France, Germany, and the U.K. it
is 60 to 80 U.S. cents). The tax rates on tobacco and alcohol also vary significantly. Denmark
and the U.K. tax them heavily, while Italy, Portugal, and Spain tax them lightly. Most countries
use specific tax rates to target the alcohol in drinks and discriminate more against dangerous
forms of smoking.
When it comes to automobiles and motoring fuel, at least four types of taxes are imposed in
OECD countries. They include (i) taxes on motor vehicles, (ii) taxes on motor fuels, (iii)
license fees, and (iv) users fees and charges (driving license fees, road tolls, parking fees).
Taxes on betting and gambling are targeted less at revenue generation and more at regulation.
Excise taxes in OECD countries also address environmental issues.
There is a wide variety of excise taxes in Asian countries. Both specific and ad-valorem rates
are used, and they vary widely across items and countries. Taiwan taxes rubber tires, cement,
beverages, gasoline as well as diesel and kerosene, fuel oils, electric appliances, and motor
vehicles. China taxes tobacco, alcoholic beverages, cosmetics, jewelry, and motor vehicles. In
Malaysia, mainly motor vehicles, beverages, and tobacco are taxed. The Philippines has a very
long and differentiated excise tax schedule on beverages, tobacco products, fuel, automobiles,
and cosmetics. Singapore applies excise taxes on alcohol products, tobacco, petroleum
products, and motor vehicles. In Korea, excise taxes apply to a wide range of goods, including
alcoholic beverages, luxury items, petroleum products, casinos, and nightclubs. Thailand
levies excise taxes mainly on four product groups: petroleum products, beverages, tobacco,
and motor vehicles. In Indonesia, excises are applied on 37 groups of goods including luxury
items, electrical appliances, alcoholic beverages, and furniture.
44
T he L egal F ramework for T ax A dministration
5. Trade Taxes
Trade taxes have been of special significance in Vietnam, because they used to bring about
one-third of total tax revenues. This share has been falling mainly due to tariff reduction
following Vietnams increased international commitments, and it decreased from about 21
percent in 1997 to 11.2 percent in 2010. With Vietnam joining the WTO, the contribution of
trade taxes is likely to decline further. Tariff rates fall into the following three broad categories:
i. Preferential rates applied to goods originating from countries in the most favored nation
(MFN) category.
ii. Common tariff rates applied to goods from other countries without any trade agreements.
For those countries, tariff rates in this case were 50 percent higher than for MFN countries.
iii. Special preferential tariff rates on goods from countries with special trade agreements
with Vietnam.
There is a fourth type of import duty under which several supplementary tax laws have been
enacted that are in the nature of antidumping and antidiscrimination laws.
45
T ax R eform in V ietnam
There is a special provision of exemptions from customs duty on first-time imports into
the country. Most of these goods are common consumer items subject to a VAT regime.
Therefore, the government should specify the mode of subsequent taxation of these goods
when they are disposed by the first-time importer.
Another provision in the law prohibits the import of certain goods, particularly used consumer
goods including garments and home appliances. Thus, the duty-exempt goods are new goods,
and when the duty-free importers dispose of the used goods in the domestic market, they are
conferred a rent. The rationale for doing this is unclear.
Provisions for the remission of duties and refunds have been specified in detail. These
provisions are modern and in conformity with the laws in many developing countries.
46
T he L egal F ramework for T ax A dministration
The main thrust of these tax policy reforms is to further rationalize tariff rates. This is bound
to simplify customs administration and help enhance compliance.
6. Direct Taxes
Revenues from the PIT and the CIT combined were about 8.7 percent of GDP in 2010. CIT
revenue accounts for about 90 percent of total income tax revenue.
4
The exemption threshold is set at 5 million dong per month for Vietnamese citizens and 8 million
dong for foreign residents, or about than US$314 and US$503 at the current exchange rate of
15,900VND/US$, respectively.
47
T ax R eform in V ietnam
between 2 to 24 percent of the income of the average production worker. The threshold
has been abolished according the 2009 PIT law.
v. The Individual as a Taxable Unit: The taxable unit is the individual income earner and this is
in conformity with the international PIT trend. The Ordinance did not include provision
for personal allowances for spouse, children, medical expenses, etc., but rather allowed for
a flat high exemption. This practice facilitated the administration but narrows the tax base
considerably. The law promulgated in 2009 has been revised to include this provision.
vi. Inequity in the Tax System: The tax system is highly inequitable in Vietnam, and the
principle of horizontal equity is violated in many ways for a variety of reasons, including
the exemption of different kinds of allowances and savings incomes.
vii. Tax Rates and Tax Brackets: Tax rates are highly differentiated, and this needs to be
rationalized. According to the 2009 PIT law, there are seven tax rates between 0 percent
and 35 percent. The highest tax bracket of 35 percent is too high compared to international
best practices and developing countries, where the highest tax rates are converging to 30
percent. This should be seen in the context of tax competition and integration of PIT
and CIT rates. With the highest tax rate in a developed country at 40 to 45 percent,
it becomes attractive for workers to move to a country where the highest tax bracket is
around 30 percent. At present, there is substantial disparity between the CIT rate of
25 percent and the highest PIT rate of 35 percent. Thus, this rate should be lowered to
around 30 percent.
48
T he L egal F ramework for T ax A dministration
threshold applies on a case-by-case basis rather than on the total annual income from
these sources.
vi. Social and medical insurance: These contributions are explicitly deductible from business
income or from salaries and wages (Article 21), but their payment is not clearly mentioned
as income. These contributions might escape taxation at both ends unless the word
pension, taxable under the current PIT law, is meant to cover these items.
vii. Family deduction: A deduction for the taxpayer and dependents has been introduced. For
the taxpayer, the monthly amount is 4 million dong, and for each qualified dependent it
is 1.6 million dong. Thus, the threshold has gone down from 5 million dong to 4 million
dong, but the family deduction effectively increases the threshold. This may further
decrease the number of taxpayers and lower tax revenues.
viii. Charitable contributions: A deduction for charitable contributions has been introduced,
which is in accordance with international practice.
ix. Tax rates: The law has seven tax brackets: 5, 10, 15, 20, 25, 30, and 35 percent. The highest
tax bracket has been reduced from 40 percent to 35 percent, which is a welcome step,
but this has been offset by the increase in the number of tax brackets. Presumably, this
is to promote vertical equity, but international experience suggests that having too many
brackets complicates administration and encourages tax evasion while doing little to
promote equity. The number of tax brackets should preferably be limited to three or four.
A scheduler tax structure has been adopted for incomes from capital investment (interest
and dividends), royalties and franchising, winning prizes, inheritance or donations, and
the sale of immovable property.5
x. Threshold for tax liability: A threshold of 10 million dong is applied for income from
royalties, franchising, winning prizes, inheritance, etc. Thus, these incomes also receive a
special tax break under the new law. There is little justification for this tax break.
xi. Narrow tax base: The problem of a narrow tax base persists. However, this may be
acceptable in the short run in the interest of formalization of the labor market and the
economy.
xii. Nonresidents: There are scheduler rates instead of a flat rate. A tax rate of 20 percent
is applicable for income from salaries or wages. Tax rates applicable to income from
business are 1, 2, and 5 percent.
Is the new PIT law more progressive or does it generate more revenue?
A microsimulation model for PIT revenue forecasting indicates interesting results about
relative productivity and progressivity of the two PIT regimes, current and previous (see
Chapter 8: Taxation in Vietnam: Who Pay What). If the Ordinance and the 2009 PIT law
were applied exclusively to income from wages and salaries, the new law would yield lower
5
Taxing income from capital investment is a complex issue, and a tradeoff between collection and
economic efficiency costs need to be made. In a classical regime, corporate income is taxed at
different levels; for example, corporate income is taxed at the corporate level, and then distributed
dividends are taxed at the individual level. Such a regime is simple for administration, but is likely
to generate unintended economic distortion. Hence, some countries attempt to introduce in their
income tax system some form of partial integration (for example, in the United States, dividends
are taxed at a substantially lower rate than the highest marginal rate of personal income tax).
49
T ax R eform in V ietnam
revenues and look less progressive. This is not surprising in view of the newly introduced
personal and family deductions. If, however, other forms of income as provided in the new
law are incorporated into the tax base, the system becomes more progressive and yields higher
tax revenues.
50
T he L egal F ramework for T ax A dministration
vi. The distinction between regular and irregular incomes still exists, and royalties and lottery
prizes are taxed at lower rates and with a threshold of 10 million dong for each transaction.
This income should be treated in the same way as regular income. Alternatively, a
withholding tax of around 15 percent may be used, and it should become the final tax.
vii. While schedular tax rates may be used in some cases (transfer of immovable property,
interest and dividends, inheritance), the law applies them to too many items.
viii. To account for inflation, the threshold and tax brackets should be adjusted periodically
for bracket creep.
51
T ax R eform in V ietnam
negative, indicating that the excessive capital gains would need to be taxed. Assets under
accelerated depreciation should be kept in a separate pool.
d. Depreciation rules may still be used as instruments of tax incentive for particular types of
investment or investments in notified geographic areas.
e. Accelerated depreciation up to twice the normal rate is applicable for high income
efficiency establishments.
(iii) Disallowed deductions and limits on deductions:
Some payments by companies are not allowed under the law, although they are legitimate
business expenses under internationally acceptable practices. Also there is a cap on the
following forms of expenditures:
a. Deductions for benefits in kind and payment of bonuses; such payment of bonuses would,
of course, be taxable under the PIT paid by the employee.
b. There is a cap on expenditures for advertising and other forms of sales promotion,
including entertainment, and such costs cannot exceed 10 percent of the total deductible
expenditure. For a new enterprise, such expenses cannot exceed 15 percent of deductible
expenses for the first three years from the date of establishment. This kind of expenditure
is generally legitimate and should not be capped. There has to be clear rules about
allowing entertainment as an expense.
(iv)Tax rates
a. The present tax regime sets the standard tax rate at 25 percent. A higher rate - between
32 and 50 percent - is stipulated for oil, gas, and other precious and rare natural resources.
Reduced tax rates of 20 percent and 10 percent as part of the CIT incentive scheme are
applied to investment in preferred economic sectors or geographic areas. Multiple rates
make the administration complicated.
b. The existing regulatory framework does not differentiate tax rates on foreign company
branches and domestic enterprises. A higher income tax rate on branches of foreign
companies is advocated because branches are taxed only once, and there is no occasion
to apply a withholding tax.
(v) Tax incentives
a. The CIT provides incentives for investments made in specific sectors, specific locations,
and if they meet some social criteria (female or handicapped people).
b. At present, incentives are very complex, and they need to be greatly rationalized. There
are three major types of incentives: (i) preferential or reduced rates (20 percent and 10
percent), (ii) outright tax exemption or a tax holiday for a prescribed period of time,
and (iii) accelerated depreciation. There are detailed criteria for each category that are
sometimes unclear.
52
T he L egal F ramework for T ax A dministration
CIT rate, but a few exempt capital gains if reinvested in business. Subject to restrictions in
some countries to prevent tax avoidance, all expenses incurred wholly and exclusively in
earning the income or maintaining the assets are tax deductible.
All countries provide a depreciation allowance, but the rules vary. Some standardized
method, such as the straight line or declining balance method, is followed. Tax incentives
are generally offered in the form of accelerated depreciation at a higher rate than normal. All
OECD countries allow loss carry forward ranging from five years to indefinitely, while some
allow loss carry backward from one to three years. Tax rates in developed countries also vary
considerably (from 19 percent in Canada to 45 percent in Germany). Top tax rates have been
declining over the years under the increasing pressure of tax competition.
Selected Asian countries apply various tax laws, and all offer different forms of fiscal incentives
for certain types of investments or investments in designated geographic locations. The
standard rates vary widely. For example, Indonesia has three rates - 10, 15, and 30 percent -
while Malaysia has 28 percent for non-petroleum companies and 38 percent for petroleum
companies. Korea applies two rates - 15 and 25 percent depending upon income - whereas
the Philippines, Singapore, and Thailand have a single rate, which are 32 percent, 26 percent,
and 30 percent, respectively.
53
T ax R eform in V ietnam
54
T he L egal F ramework for T ax A dministration
iv. Just as for the oil and natural gas sectors, a higher CIT should be imposed on minerals and
precious metals. Even when the normal CIT rate is reduced to 20 to 23 percent, the rate
for this sector may be kept at a higher level.
v. The concessions and exemptions granted to certain categories of investors need to be
closely examined and reduced in scope since such incentives are not effective and only
result in revenue losses. This also simplifies the administration.
vi. The government should consider adopting a production sharing arrangement in the
oil and gas sectors. A royalty and tax regime can also be maintained along with the
production sharing arrangement for the sake of revenue maximization. These are well
tried instruments and would in the long run help in administration and in generating
higher revenues.
55
T ax R eform in V ietnam
iv. Need for decentralization: Many user charges and fees are centralized, but only a few
are decentralized to provincial authorities. The international trend is toward more
decentralization of both service delivery and revenue collection. A major exercise should
be undertaken to determine which services must be retained by the central government
and which are to be delegated to the provincial and local levels. Some degree of
decentralization is suggested in Annex 8.
56
T he L egal F ramework for T ax A dministration
v. By their very nature, rates for most charges and fees are fixed in dong and not as a
percentage of the price. With a given level of inflation, the government or the agency
providing the service will begin to lose revenue unless there is a well established system
of revising the rates on a regular basis, in keeping with the inflation rate.
9. Tax Administration
The adoption of the new Tax Administration Law, which was enacted on November 29, 2006,
and which became effective on July 1, 2007, was a milestone in the ongoing tax administration
modernization. It provides a legal basis for GDTs concerted effort to ensure transparency
and integrity in all tax administration operations. In addition, it is expected to help eliminate
gray areas and uncertainties serving as breeding grounds for noncompliance by taxpayers and
abuse by tax administration officials.
In general, the law is comprehensive and of high quality. It consists of 14 chapters, covering all
key tax administration business functions (tax registration, filing and accounting, assessment,
audit and investigation, collection and debt management). The law has clear provisions for
governing taxpayer service, tax appeals, and tax law violation settlement.
57
T ax R eform in V ietnam
iv. Third-party information for effective audit and enforcement. The law stipulates that third-
party information can be acquired from various agencies, but the confidentiality of the
information for tax purpose must be safeguarded.
However, the new law also has some shortcomings (see below) that should be resolved in the
next round of revisions.
58
Tax Compliance and Sources of
Revenue Leakage: Conceptual II
Framework and Assessment
By Tuan Minh Le, Duc Minh Pham, G. P. Shukla, and Gregory Kisunko
1. Summary
With respect to taxes, Vietnam faces the same challenge as other developing countries: how
to effectively plug revenue leakage and thereby enhance compliance. A growing body of
literature indicates that low compliance has the following detrimental impacts on economic
growth and development: countries have fewer resources for development, the tax system
tends to be more distorted and less equitable, and tax policies are more vulnerable to frequent
ad-hoc changes. The typical paradox of the missing middle reflects the reality in Vietnam,
a country still in transition; that is, a myriad of household or small businesses coexist with
large state-owned and foreign-invested companies, while medium-size enterprises are still
emerging. This chapter provides a conceptual framework for assessing tax compliance and
presents practical approaches to measuring compliance for Vietnam.
While tax compliance or noncompliance is a straightforward concept, its measurement
and especially, any measurement of tax potentialis technically difficult. In principle, tax
potential can be quantified in three distinct ways: (i) revenue forecasting, using micro
data and applied by tax type; (ii) analyzing tax administration data to gauge compliance
in key functions, specifically tax filing, accounting, and payment; and (iii) relying on
taxpayers self-reporting for compliance. However, taxpayers have little incentive to reveal
their true estimates, and thus relying on their estimates is not feasible. Given the limited
data available for Vietnam, we apply the first two approaches to estimate compliance of
selected taxes and of selective tax functions. The results indicate that the key challenge
for enhancing tax compliance for Vietnam is twofold: how to sustain the relatively high
compliance rate by the formal, large businesses, and how to improve compliance by small
or household businesses. Such a challenge is not unique. As Vietnam embarks upon a new
wave of tax policy and administration reforms, it can draw on the following lessons from
international experience:
Analytical capacity within the General Department of Taxation (GDT) must be improved as
an integral component of successful tax administration modernization. Without a systemic
approach to develop a quality database and train sufficient staff for revenue forecasting,
GDT will face an uphill battle to gather intelligent information on the perceived risks of
noncompliance.
Effective reform of tax administration requires successful tax policy reforms. Vietnam is
moving in this strategic direction; it is in the process of revamping all its tax policies while
concurrently modernizing its tax administration.
59
T ax R eform in V ietnam
Tax reforms need to be implemented in parallel with public sector reforms, in general, and public
financial management reform, in particular. Tax administration cannot become an island
of integrity in an overall public sector facing a high incidence of corruption, and at the
same time, corruption or wasteful spending of public expenditures undermines peoples
willingness to comply with their tax obligations.
Any focus on a large group of taxpayers as the main stable source of revenues needs to
be complemented with improved compliance by small and medium enterprises (SMEs).
Voluntary compliance by the hard-to-tax SMEs can only be achieved with lower
compliance costs through a simplified tax regime, greater reliance on automatic filing,
and a different approach to compliance management that would emphasize service rather
than discretionary punishment.
2. Introduction
Higher tax compliance with a broader base creates an opportunity for a country to reduce tax
rates across-the-board and make the tax regime more stable, not affected as much by sectoral
fluctuations in the economy, and less reliant on ad-hoc changes in the tax structure to meet
revenue targets. Many developing countries with weak governance and accountability, an
expansive shadow economy, and low tax administration capacity face similar challenges of
how to plug the revenue leakage and thereby enhance compliance. This challenge is also
true for Vietnam, where a myriad of household enterprises and small businesses coexist with
large state-owned enterprises and foreign-invested companies while the middle part of the
distribution of enterprises is only emerging. Compliance by domestic private micro and
small businesses is estimated to be very low, with enforcement costs often higher than what is
being collected, whereas compliance is relatively high for large, established enterprises.
The analysis of determinants of revenue leakage would facilitate the understanding of
the specific nature and systematic patterns in noncompliance, which is necessary for
developing a new risk-based compliance management philosophy in tax administration.
This chapter provides a conceptual framework for assessment of compliance and presents
practical approaches to measuring compliance and reasons for noncompliance in the case
of Vietnam.
The structure of the chapter is as follows. Section 2 establishes a conceptual framework for
understanding compliance and noncompliance. Sources of noncompliance and its detrimental
impact on the economy are explored. Section 3 presents measurements of compliance and
noncompliance. Two different approachesrevenue forecasting modeling and estimation
based solely on tax administration dataare applied to measuring tax revenue potential and
level of compliance in Vietnam. Section 4 offers a conclusion.
60
T ax C ompliance and S ources of R evenue L eakage
of an economy, and researchers still lack agreement on what exactly is being measured.
For example, the Organisation for Economic Co-operation and Development (OECD)
Compliance Measurement Practice Note (2001) poses the following question: How
is compliance definedis it compliance according to the tax authoritys or the taxpayers
interpretation of the law and its application to the facts? Is it from another more neutral
perspective? (p. 3). Tax administrations tend to consider the different interpretation of the
laws by taxpayers as evidence of noncompliance without regard to the fact that tax lawsas
observed in many developing countriesmay not be consistent and transparent, or they may
have grey areas open to widely different interpretations by taxpayers and tax administrations.
While the means to define compliance or, related to it, noncompliance, are still subject to
much dispute, the sources of noncompliance or channels for revenue leakage are conceptually
straightforward.
Appeals and
Compliance
Review
Cost
System
Conceptually, tax evasion and avoidance are different, but both lead to revenue leakage,
less equitable tax systems, and more distortions to the competitive market framework.
Taxpayers are said to engage in tax evasion practice when they use illegal means to evade
paying their full tax liability. Tax evaders choose not to report or to underreport their
earned incomes, sales revenues, or wealth while they attempt to overreport the amount
of deductible expenses or even participate in outright smuggling or money laundering
schemes. In tax avoidance, taxpayers abuse loopholes in the existing tax regimes in order
to lower their tax liability.
61
T ax R eform in V ietnam
62
T ax C ompliance and S ources of R evenue L eakage
Table 2.1: South Africa: Tax and Accounting Costs as Percentage of Turnover
(Based on Highest Turnover in Each Turnover Bracket)
Source: FIAS.
Apparently, some evidence shows that reasons for noncompliance may be different in South
Africa and Vietnam. According to the results of an Enterprise Survey sponsored by the
Enterprise Analysis Unit of the World Bank, South African businesses are expected to give
a gift to a tax official in 3.13 percent of cases, while in Vietnam it is a much more common
practice (33.68 percent).8 However, a number of factors of the gift-giving practice and,
related to it, the compliance pattern, (for example, social culture), may be vastly different
between the two countries and are not accounted for in such surveys. The practice is not so
different if one compares Vietnam with other countries in the region. For example, according
to the same survey, almost 39 percent of businesses in China report giving gifts to tax officials
(survey 2003), and the figure was around 22 percent for the Philippines in 2009.
63
T ax R eform in V ietnam
9
Applying the ordinary least squares in a fuller empirical test, Bird et al. (2004) also find that
institutional qualities, such as corruption or political stability, as measured by the governance
quality index, have a statistically significant impact on a countrys tax intake. They use different
instrumental variables (including ethnic fractionalization, language, and latitude) and the
Hausman Chi-square test to examine but fail to detect the presence of simultaneity of the tax or
revenue intake and institutional variables.
64
T ax C ompliance and S ources of R evenue L eakage
Annex 10 describes the data and summarizes the statistics of the variables. To ensure
consistency, the data used are taken from the same sources: data on revenue or tax-to-GDP
ratios and real GDP per capita are from World Bank World Development Indicators 2005
(WDI 2005), and data on the corruption index are from the International Country Risk
Guide (ICRG). The original ICRG corruption index ranges from 0 (most corrupt) to 6 (least
corrupt).10 To make it consistent and easy to compare with the results in Tanzi and Davoodi
(1997), we have rescaled the ICRG index by multiplying it by -10/6 so that the index ranges
from -10 (least corrupt) to 0 (most corrupt).
Our empirical results lend support to Tanzi and Davoodis hypothesis that corruption is
associated with lower government revenue (table 2.2). The results indicate that controlling
for the level of income, an increase in corruption by one standard deviation (2.1) reduces the
revenue and tax collection as a share of GDP by 1.64 and 1.39 percentage points, or by 7 and
9 percent, respectively (see annex 10 and table 2.2).
Table 2.2: Revenue-Depleting Impact of Corruption
Using the results of the regression analysis (table 2.2), we estimate the specific losses in revenue
and tax collection due to corruption for those countries that have higher corruption indexes
than the mean (the revised corruption index of -5.6, see Annex 11). The corruption impact is
negative and significant. Annex 11 provides a detailed account of corruption-induced leakage
10
The ICRG provides alternatives for gauging the quality of the institutional setting of a country,
particularly the corruption index and bureaucratic quality. The corruption index ranges from 1
to 6, where a higher number means lower corruption. For detailed presentation of the ICRG
methodology, see http://www.prsgroup.com/ICRG_Methodology.aspx.
65
T ax R eform in V ietnam
by country illustrated in figure 2.2.11 The revenue and tax leakage is presented for three
groups of countries with low, medium, and high corruption. The low and high corrupt groups
include the top and bottom 20 percent of the whole country sample, respectively (Annex 11).
The tax and revenue loss is small for countries with low corruption (in the magnitude of 0.2
and 0.3 percent of GDP, respectively), but it is significant for countries with high corruption
(approximately 2 and 2.2 percent of GDP, respectively).
Figure 2.2: Higher Corruption Associated with Greater Revenue and Tax Loss
Tanzi and Davoodi (2000) further hypothesize that statistically, corruption has a detrimental
impact on the equity of a tax system. They find corruption has a larger negative impact on
direct taxes than on indirect taxes. The results of their empirical work indicate that corruption
exacerbates the imbalances in the tax take from direct and indirect taxes; it reduces the tax
collection from a certain group of taxpayers (and normally those well-to-do businesses that
have the ability to give bribes to tax officials), and increases relatively the collection from the
poorer groups of taxpayers. However, independent of the corruption, by its very nature, the
collection of direct taxes is technically more challenging than the collection of indirect taxes,
and this is even more so in developing countries.
The revenue and tax losses for a particular country in Annex 11 are estimated as a product of the
11
corresponding corruption coefficients (-0.78 and -0.66, respectively), and the difference between
the rescaled corruption index of this country and the mean rescaled corruption index (-5.6, Annex
10). For example, the revenue loss for Congo would be: [-0.78] * [-1.02 (-5.6)] = -3.3 (Annex 11).
66
T ax C ompliance and S ources of R evenue L eakage
to tax authorities.12 Choi and Thum (2005) examine the connection between corruption
and the shadow economy and their implications for the official economy. Surprisingly, they
demonstrate that the shadow economy and official economy may be compliments rather than
substitutes. They explain that an open option for an entrepreneur to flee the official economy
constrains the ability of corrupt officials to artificially introduce distortions for businesses
operating formally in order to extract private gains. The shadow economy thus mitigates
government-induced distortions and leads to more efficiency of the official economy. However,
the more common belief is that the shadow economy is a substitute for the formal economy.
For example, Johnson et al. (1997) in their study of the pattern of the unofficial economy
in the Former Soviet Union and Eastern Europe show that bad public policiesincluding
burdensome combinations of taxes and excessive bureaucratic regulations, and corruption
may lead to the expansion of the unofficial sector at the expense of the formal sector and as a
consequence result in the leakage of the governments tax collections, continued deterioration
of good public service provision, and poor overall economic performance. Shneider and Enste
(2000) and Johnson, Kaufmann, and Zoido-Lobaton (1998) also conclude that corruption
and the shadow economy are complimentary; a country with more corruption tends to have a
larger shadow economy; this implies that there should be some negative correlation between
the official and shadow economies.
In any case, since we are dealing with ghost activities, we recognize that estimating the size of a
shadow economy would be technically challenging. Schneider and Enste (2000) and Schneider
and Klinglmair (2010) offer an extensive review of the three most widely used methodologies
for estimating the size of the shadow economy: (i) the direct or micro approaches, (ii) the
indirect or indicator approaches, and (ii) the model approach (see annex 12 for a summary).
67
T ax R eform in V ietnam
shadow economy dataset by Schneider, Buehn and Montenegro (2010) with the data on
tax-to-GDP ratios from World Development Indicators (WDI), we find that the correlation
between the size of the shadow economy and tax collection is negative and substantial (-0.23)
(figure 2.3 and annex 13).
Figure 2.3: Negative Correlation between Size of Shadow Economy
and Tax-to-GDP Ratio
30
25 Vietnam
20
15
10
5
0
0 10 20 30 40 50 60 70
Size of Shadow Economy (% of GDP)
68
T ax C ompliance and S ources of R evenue L eakage
Table 2.3: The Bigger the Shadow Economy, the Greater the Tax Leakage
Sources: Schneider and Klinglmair 2010; WDI 2007; and authors estimate.
69
T ax R eform in V ietnam
For designing compliance strategies, advanced tax administrations normally apply both
approaches as compliments. Given the limitation of data, we attempt to apply the two
approaches to estimate compliance and noncompliance of selected taxes (the VAT and CIT)
and of selective tax functions.
Modeling of Tax Potential, Tax Compliance, and Noncompliance for the VAT
and CIT
B v = Biv iv + K j (1)
i j
Where:
Biv = VAT base for commodity i
iv = VAT taxable proportion for commodity i
K j = Business inputs purchased by tax-exempt sector j.
The potential VAT revenues for the economy (Rv) is equal to the summation of all the adjusted
tax bases across commodities and sectors multiplied by the VAT rate, at the full compliance
rate.
R v = B v v v (2)
Where:
q v = VAT compliance rate (100 percent for estimation of VAT revenue potential)
rv = VAT rate.
Finally, the tax gap (Gv) is estimated as the difference between the potential VAT revenue
(Rv), and actual VAT collection (TTRv).
G v = R v TTR v . (3)
In the VAT model under consideration, assuming the full compliance rate, the VAT has
a revenue potential of approximately 48,980 billion dong for FY06, whereas the actual
collection was 36,469 billion dong. This implies that the VAT compliance rate in Vietnam
is 74 percent and that the tax gap is estimated to be 12,511 billion dong, equivalent to 1.3
percent of GDP.
Jenkins, Kuo, and Shukla (2000) present in detail the I-O modeling methodology. For a discussion
14
of various methodologies for estimating the VAT base and an application for the case of Romania,
see Le (2007). United Nations (1999) provides a framework for compilation and analysis of I-O
tables.
70
T ax C ompliance and S ources of R evenue L eakage
Table 2.5 presents the estimate of the CIT compliance and collection gap, with the CIT
potential estimated on the basis of the standard CIT rate of 28 percent. The estimation
excludes the foreign investment sector due to a complex set of rules effectively applicable
to this sector, for example, various statutory rates and special treatments for foreign direct
investment in the natural resource sectormostly in the oil sectoror in different types of
business, such as build-operate-transfer (BOT) or build-transfer (BT).
With a comprehensive database of the CIT returns, which is part of the objectives of the ongoing
15
tax administration reform process, the GDT will be able to estimate the base more directly using a
microsimulation approach, rather than relying on the OS concept.
71
T ax R eform in V ietnam
Table 2.5: Estimation of the CIT Tax Gap, 2005 (Billions of dong, current prices)
The domestic nonstate sector, consisting mostly of small and medium private businesses,
stands out as the most problematic in compliance. Statistics show that the sector accounts
for more than 90 percent of total CIT taxpayers but contributes less than 9 percent of total
collection.16 Table 2.5 reaffirms this; compliance by this sector is just over 26 percent, while
compliance by the state-owned sector is more than 71 percent. The VDR 2008 also reports
that the SOEs remain the most important taxpayers, accounting for approximately 54 percent
of total CIT. A sensitivity analysis indicates that if the compliance rate for the domestic
nonstate sector is improved to the approximate level by the state sector (70 percent), CIT
collection can be improved by almost 3 percentage points of GDP.
The magnitude of the foregone revenues is large, and some other policy-related factors like
16
differences in thresholds, definition of small business taxable income, and exemptions may be
contributing to this.
72
T ax C ompliance and S ources of R evenue L eakage
At present, the tax administration does not have sufficient information for estimation of
reporting or payment compliance.17 However, GDT has managed to maintain the tax filing
database, which allows for analysis of filing compliance. Table 2.6 provides a summary of tax
filing compliance by major taxes (business income tax, individual income tax, VAT, excise,
and the natural resource tax). The estimates indicate that the composite filing compliance
varies substantially. In common with other developing countries, individual income tax has a
very low level of compliance (28 percent). Compliance management of the VAT, the business
income tax, and the excise tax is fairly good; in particular, compliance reaches 84 percent for
the VAT, which does not have any threshold.
Table 2.6: Tax Filing Compliance by Major Taxes, 2010
Filing Compliance of Compliance of Composite Filing
Tax Type
Registered Taxpayers Filing on Time Compliance
Business income tax
71 68 58
(preliminary figure)
Individual income tax 44 36 16
VAT 91 79 72
Excise duty 47 40 19
Sources: GDT and authors estimates.
3. Conclusion
Noncompliance in the payment of taxes depletes the tax base and leads to revenue leakage.
Countries with low compliance levels are likely to require frequent tinkering with their tax
regimes to maintain revenues at a certain percentage of GDP. Frequent and ad-hoc changes
in tax regimes, in turn, risk making the system less stable and equitable. Enhancing voluntary
compliance has become a key institutional and operational objective of worldwide tax reforms.
Due to the limited availability of quality microdata on business operations and economic sectors,
and the lack of a comprehensive tax administration database, only some preliminary estimates of
tax compliance can be made. While the results indicate that Vietnam has attained commendable
achievements in compliance with major taxes, a number of issues remain. The challenge is
twofold: to sustain the relatively high compliance rate by the formal, large businesses, and to
improve the compliance of SMEs. The government is currently embarking on a comprehensive
tax administration modernization reform program, with a focus on improved integrity, enhanced
taxpayer service, and risk management to ensure high tax compliance and better equity.
The following lessons can be learned from other revenue reforms:18
17
As part of the ongoing Tax Administration Modernization Project (TAMP), the GDT is to
reengineer tax procedures and automate the administration, which will facilitate the development
of a reliable database for complete compliance analysis.
18
These lessons are also featured in the 2007 World Bank Development Committee report on Fiscal
Policy for Growth and Development. For the report, see:
http://siteresources.worldbank.org/DEVCOMMINT/Documentation/21289619/DC2007-
0004(E)-FiscalPolicy.pdf. Also, for country experience in revenue reforms and lessons learned,
see Casanegra de Jantscher and Bird (1992).
73
T ax R eform in V ietnam
1. Analytical capacity within GDT must be improved to support revenue administration reforms.
The new philosophy in revenue administration based on voluntary compliance requires
effective compliance measures and improved taxpayer service. At present, GDT does not have
sufficient in-house analytical capacity to estimate compliance, by tax, or by major economic
sector. Without a systematic approach to develop a quality database and to train sufficient
staff in revenue forecasting, GDT will face an uphill battle to gather intelligent information on
the perceived risks of noncompliance.
2. Effective reform of tax administration requires successful tax policy reforms.
Tax policy and tax administration are linked. In countries with poor tax administration
capacityposed as a binding institutional constraint to tax reformstax policy reform
should take into account the administrative capacity and contribute to reducing opportunities
for corruption. Frequent adjustments of tax regimes may help keep the system buoyant, but
there are no credible short-term solutions for enhancing compliance that deal with tax policy
or tax administration reforms in isolation. Vietnam is currently revamping its entire direct
and indirect tax policies while modernizing its tax administration.
3. Tax reforms need to be implemented in parallel with public sector reforms, in general, and public
financial management reform, in particular.
Corruption in the public sector in general, and in tax administration in particular, is one of
the key determinants of low compliance. The recent trend in tax administration reforms
shows that projects increasingly focus on good governance, with anticorruption being at the
core. However, there is no island of integrity, and at the same time, corruption or wasteful
spending of public expenditures undermines peoples willingness to comply with their tax
obligations. Thus, tax reforms should be conducted in parallel with public administration
reforms and reforms in budgeting and public expenditures.
4. Any focus on a large group of taxpayers as the main stable source of revenues needs to be
complemented with improved compliance by SMEs.
There are strong arguments for widespread inclusion of SMEs in the tax net as a direct
approach to shrink the shadow economy and enhance the equity of the tax system (ITD
2007). Although the benefits for SMEs of being in the tax net are substantial (for example,
easier access to finance and the possibility to have formal contracts with the formal sector),
their compliance costs are disproportionably high. While good analytical capacity to project
revenues, and reforms to streamline tax laws and reduce opportunities for corruption are
necessary, they are not sufficient to change compliance behavior (especially in the SME
sector), if taxpayers needs and views are not taken into consideration.
Voluntary compliance by this segment of taxpayers can only be achieved with lower
compliance costs through a simplified tax regime, greater reliance on automatic filing,
and a different approach to compliance management that would emphasize service rather
than discretionary punishment. As part of the overall reform program, GDT is to focus its
resources on establishing a working system of feedback mechanisms and improving capacity
to act on the results of feedback assessment.
74
Tax Administration
By Duc Minh Pham and Michael Engelschalk III
1. Role and Challenges of Tax Administration
in Vietnam
With almost 44,000 employees, the Tax Administration of Vietnam (General Department of
Taxation, GDT) is one of the largest public sector institutions in the country. It is organized
at three levels, with tax administration headquarters in Hanoi, 63 tax offices at the provincial
level, and 694 district tax offices. GDT thus maintains a physical presence in all provinces and
districts in the country. GDT is charged with collecting all major domestic taxes, in particular
the Value-Added Tax (VAT), the Corporate Income Tax (CIT), excise taxes, and natural
resource taxes, which account for 76.4 percent of domestic revenue collection.19 GDT is also
responsible for the collection of a relatively large number of minor taxes, such as the tax on the
transfer of land use rights, the land and housing tax, and the small business tax. Domestic tax
revenues, including tax revenues from crude oil, collected by GDT, amounted to 25.1 percent
of GDP in 2010. GDT is operating in a rapidly and fundamentally changing environment,
which will substantially increase the challenge of ensuring effective tax revenue collection.
The greater integration of Vietnam into the international trade system has led to a substantial
decrease in the collection of revenues from international trade taxes. A milestone in the
trade integration process has been the accession of Vietnam to the World Trade Organization
(WTO) in January 2007. Vietnams international trade integration has been accompanied
by tariff reductions, and the share of total tax revenues from import and export taxes has
decreased from 28.9 percent in 1996 to 13.2 percent in 2010. Shifting collection priorities
from the border to the domestic market, however, puts a much greater burden on the
compliance management system and increases the risk of tax evasion.
Taxes from state-owned enterprises (SOEs) still constitutes the major source of tax revenues,
although with the move to a market economy, the importance of SOEs in the economy and
for revenue collection is decreasing rapidly. According to 2010 estimates, 35.1 percent of
CIT revenues and 36.4 percent of VAT revenues from domestic production are paid by
SOEs. The increasing role of the private sector in the economy so far has not been reflected
in the revenue collection results, and the contribution of the nonstate sector to total revenue
collection remains small at 17.4 percent in 2009.
The decrease in trade tax revenues has been compensated primarily by the operation of
an extremely efficient VAT. The efficiency ratio of the VAT in Vietnam is close to 0.6 and
thus exceeds the average VAT efficiency in Organisation for Economic Co-operation and
Development (OECD) countries. While some cascading-inducing features of the Vietnamese
VAT system have contributed to the exceptional revenue performance at the expense of
economic efficiency, the short-to-medium-term challenge is that the planned revision of the
19
Including domestic tax revenues from crude oil.
75
T ax R eform in V ietnam
VAT, including the rationalization of the exemption and enhancement of the VAT refund,
may result in a revenue loss.
76
T ax A dministration
strategic planning process must be introduced and stakeholder involvement in the strategic
planning process must be solicited to ensure wide acceptance of the plan. Operational and
annual business plans will need to be prepared to ensure consistency in operations and
implementation of the strategic plan.
Capacity development will also require increasing the staffing level at GDT. While many
modern tax administrations assign around 10 percent of their staff to headquarters functions,
only 1.1 percent are assigned in Vietnam. This is not sufficient to efficiently perform all
relevant planning, monitoring, and analysis tasks.
An assessment of GDTs operational processes was conducted by reviewing 15 functional aspects
of tax administration performance (figure 3.1). Overall performance is below average in all main
and support functions scored in accordance with an international best practices research scale,
with supporting data for the framework gathered from 30 developing and developed country tax
administrations. Among them, the debt collection and enforcement function turned out to be
the area with the most advanced processes in place. Multiple payment methods (cash and bank
transfer), the use of technology to access taxpayer account data combined with semi-integrated
collection systems to monitor taxpayer delinquency, are among the key factors contributing to
this advanced function. In the legal area, provisions for the counseling and advising of GDT staff
on legal issues exist, so that at least basic support is available to help tax administration staff deal
with more complex legal matters. However, the legal support is not yet comprehensive, and
additional processes to perform critical legal tasks, such as creating the capacity to issue private
letter rulings to taxpayers, still have to be developed.
Figure 3.1: Staffing Proportion in Headquarters
50,000
40,000
30,000
20,000
10,000
0
Australia Malaysia Sweden Indonesia Vietnam
Total tax administration staffing Headquarters stafing
Existing processes are less satisfactory in most other areas, with the tax audit function
being of particular concern. Procedures and methodologies for a risk-based audit selection
have not yet been developed, and auditor discretion in the selection of cases for a desk or
field audit is substantial. In addition, existing processes for the conduct of an audit provide
substantial discretion to the audit team, and the overall audit planning and supervision
process does not yet include a systematic and ongoing monitoring of audit efficiency. Anti-
77
T ax R eform in V ietnam
evasion appears to be the weakest area, where no strategic plan for addressing evasion cases
exists (figure 3.2).
Figure 3.2: Operational Gap
Pre-Filing Taxpayer
Services
Support Functions
(Financian and 5 Filing - Return
Human Resources Processing, and Tax
4 Accounting Function
Management, and IT
3
Operations 2
Mangement-Strategic 1 Audit Function
Planning, Perf Mgmt,
QA, Communications 0
78
T ax A dministration
Therefore, GDT must establish a long-term ICT strategy to effectively support the future
business model.
New small and medium enterprises (SMEs) entering the tax net require information and
services tailored to their needs in order to reduce tax compliance costs and facilitate business
formalization. Compliance costs tend to be largely regressive and, therefore, put a special
burden on SMEs. GDT needs to develop its taxpayer service function to be able to respond
to the service needs of SMEs.
The World Banks Doing Business report ranked Vietnam 124 out of the 183 nations surveyed
in 2011 in terms paying taxes, a ten-grade improvement over the 2010 ranking. The costs of
paying taxes in Vietnam include the following:
Payments: This indicator measures the total number of taxes and contributions paid per year.
In Vietnam, this figure is more than twice the average number of payments in OECD countries
and significantly higher than the average in the East Asia & Pacific region (table 3.1).
Time: This indicator reflects the time it takes to prepare, file, and pay (or withhold) the CIT,
the VAT, and social security contributions (in hours per year). The time for complying with
taxes was 1,055 hours per year, which was unchanged during 200610. It has been slightly
reduced to 941 hours in 2011, remaining significantly higher than the average of the East Asia
& Pacific region and OECD countries (table 3.1).
Profit tax: The profit tax indicator measures the amount of taxes on profits paid by the
business as a percentage of commercial profits, while the labor tax and contributions
indicator reflects the amount of taxes and mandatory contributions on labor paid by the
business as a percentage of commercial profits. Vietnam tends to tax profits at a lower rate
than the average of OECD and East Asia & Pacific economies, while labor taxes are almost
at the same level as OECD countries and are twice as high as in East Asia & Pacific region
countries (table 3.1).
Table 3.1: Paying Taxes in Vietnam
See box 3.1 for a discussion of the costs of complying with tax regulations in Vietnam.
79
T ax R eform in V ietnam
80
T ax A dministration
This demonstrates that the level of private entrepreneurial activity in the country is still at
a rather moderate level. The registration of private businesses, mainly SMEs, is on the rise,
however, and the registered enterprises in 2010 - 430,000 - is five times higher than it was in
2006.
From a compliance management standpoint, the transition to a market economy and the
rapid development of a private sector creates two major challenges. First, ensuring large
81
T ax R eform in V ietnam
taxpayer compliance will increasingly require dealing with private business operations instead
of SOEs. Simultaneously, the international dimension of business operations is of growing
importance. This is not only reflected in the growing level of foreign direct investment in
Vietnam, but also in cross-border business transactions of Vietnamese enterprises. With
access to high-quality tax advice now available in the country (all major international tax
consulting firms operate branch offices in Vietnam), GDT will be confronted with the use
of sophisticated tax avoidance schemes, including international transfer pricing operations.
Second, an emerging active medium and small segment of the taxpayer population requires
the development of new compliance management strategies. It also poses capacity-building
challenges for GDT.
International experience shows that the development of a market economy will not affect
the importance of large taxpayers for tax revenue collection. Even with a booming SME
segment of the economy, a small number of major enterprises remain the backbone of the
revenue system (figure 3.3). However, these large taxpayers will increasingly consist of
private businesses instead of SOEs. Ensuring the compliance of large enterprises, therefore,
must remain a top GDT priority. This requires building capacity to monitor large taxpayer
operations and reducing the tax compliance burden (box 3.2).
Figure 3.3: Taxpayer Segmentations: Population vs. Revenue Contribution
010%
Small taxpayers 7095%
82
T ax A dministration
Have decided not to comply Use the full force of the law
Business Industry
Industry
Dont want to comply, Deter by detection
Taxpayer
Taxpayer but will if we pay attention
83
T ax R eform in V ietnam
84
T ax A dministration
multiple steps between business households, local GDT departments, and the tax advisory
council.
The administrative processes for presumptive taxation result in extensive direct contact
between taxpayers and tax administrators, which increases compliance costs and the risk of
collusion and corruption. In Hanoi, for example, business households indicated that they
received on average of seven visits from tax officials, and 25 percent reported that they had
to go through a negotiation process with tax officers to determine the amount payable in the
settlement for 2006.20
To reduce administrative costs, unnecessary interactions, and the related compliance
burden for small businesses, the administration of the presumptive tax system needs to
be fully based on self-assessment of tax liabilities. This, of course, requires a dramatically
simplified calculation method for small businesses, which needs to be based on a size-based
differentiation of bookkeeping standards and tax treatment.
85
T ax R eform in V ietnam
A tax processes simplification program was launched as part of the governmentwide Master
Plan to Simplify Administrative Procedures in the fields of State Governance (Project 30)
(box 3.3).
A new Tax System Reform Strategy was approved by the Prime Minister for 201120.21
The strategy aims at continuing and deepening the tax administration reform process and
addressing on a priority basis remaining challenges and operational weaknesses. Three areas
are of particular importance in this respect:
Strengthening the tax audit function: Progress in modernizing the tax audit function has
been less impressive than reform results achieved in other areas. Expert staff allocated to
the audit function is insufficient, with only 20 percent of total staff working in the audit
area, which is far less than the international average of 30 to 35 percent. Audit selection
is largely done on a manual basis, with little consideration given to risk analysis. Pilot
exercises to develop a risk-based audit selection system have started and will result in the
implementation of an automated audit selection system. According to the performance
objectives of the reform strategy 20112020, at least 90 percent of audit cases willl be
selected based on an automated risk management system by 2015.
Ensuring sufficient IT support: The current tax administration IT system does not have
sufficient capacity to fully support and link all operational functions and procedures and to
process a rapidly increasing quantity of data. Consequently, a special initiative to develop
IT capacity for the processiong of PIT returns had to be launched in combination with the
introduction of a universal PIT. Realizing the full benefits of operational modernization
Decision of the Prime Minister No. 732/QD-TTg, May 17, 2011, approving the Tax System
21
86
T ax A dministration
and ensuring the availability of e-services for taxpayers will require the introduction of a
new Integrated Tax Administration Information System (ITAIS). The requirements for
such a system have been prepared and the system is expected to be fully operational by
2016. In combination with the ITAIS implementation, business processes will have to be
reviewed and reengineered.
Introducing a proactive compliance management approach: GDT has not yet started to
develop special targeted programs to address compliance risks and behaviors of individual
taxpayer segments. Filing and payment noncompliance, and deficiencies in keeping
proper documents and records, will have to be addressed more proactively. This requires
further strengthening the compliance analysis capacity and moving to a more taxpayer-
segment-oriented approach of compliance management.
The 10-year Tax System Reform Strategy set a target that taxes and fees (excluding revenues
from crude oil and export revenues) should constitute 70 to 75 percent of total state
budget revenues by 2015, and 75 to 80 percent by 2020, compared to the current level of
62.5 percent in 2010. An ambitious program supported by the World Bank to modernize
the tax administration is underway to improve governance in tax administration and raise
the level of voluntary compliance with the tax system. Through this program, government
resources, combined with financial and technical assistance from donors, will be mobilized
to implement a comprehensive reform plan to improve the business climate, sustain revenue
collection, and reduce corruption.
The tax administration modernization program is expected to be one of the key pillars of the
business-led development strategy and to provide a significant contribution to improving key
public services and state governance in the next five years. By 2015, Vietnams Tax Agency
is expected to rank among the top five in the East Asia region in key performance areas,
including but not limited to enhanced efficiency, increased transparency and accountability,
reduced compliance burden, increased taxpayer satisfaction, and improved compliance with
the legal system.
6. External Assistance
A Tax Administration Modernization Project (TAMP) was prepared in 2007, with technical
and financial support from the World Bank. The project aims to assist GDT to strengthen
its governance in tax administration and to increase the level of voluntary compliance with
the tax system by improving the effectiveness, efficiency, transparency, and accountability
of the tax administration. These reforms are expected to have a positive impact on the
business climate by developing predictable and enforceable tax administration processes and
procedures. The sustainability of revenue collection and greater equity will also be supported
through implementation of this project. The projects development objectives, therefore, are
complementary to two higher-level objectives of the countrys four reform pillars, namely
business-led development and strengthened governance systems (figure 3.4).
87
T ax R eform in V ietnam
Business-led Strengthened
Development State Governance
This reform program, technically and financially supported by the World Bank, consists of the
following four main components:
An institutional development component, which supports the implementation of the key
requirements and instruments for good governance in the tax administration. It will (i)
increase transparency and accountability of the administration, (ii) foster the efforts of the
GDT to counteract corruption, (iii) improve the legal framework for tax administration
operations, and (iv) build capacity to analyze compliance trends. Outreach activities
and the promotion of stakeholder participation in the reform process will create external
support for the institutional and operational modernization.
An operational modernization component, which will enable the GDT to ensure a high level
of voluntary compliance with the tax system. Streamlining and simplification of processes,
and the establishment of a service culture responding efficiently to taxpayer service and
information needs, will reduce the compliance burden for taxpayers. In parallel, the
activities under this component will increase the ability of the tax administration to detect
tax evasion and collect the full amount of taxes due. Introducing new functional business
processes will also increase transparency and reduce the discretion of tax officials.
IT development, which will primarily support the procurement and implementation of a
proven ITAIS. In addition, three other subcomponents would be carried out in parallel
with the activities and tasks for ITAIS, including procurement, installation, and testing
of IT hardware and system software, e-tax applications, and a pilot data warehouse (box
3.4). ITAIS is closely linked to project Component Two since it facilitates an automation
of the full functionality of tax administration, including but not limited to registration
and deregistration, returns processing, payment processing, refunds processing, taxpayer
accounting, assessments, compliance and delinquency control, collections, audit and
inspection, appeals (objections), revenue accounting, and taxpayer services. In addition,
88
T ax A dministration
ITAIS also supports all tax administration functions, including but not limited to case
management, taxpayer notices, workflow, document control, data entry and importing,
report writing, and system administration.
The project management component supports advisory services and the necessary office
infrastructure to assist the Project Management Unit in implementing all aspects of the
project. It will also support activities aimed at developing and implementing appropriate
strategies for managing organizational change at all levels of the tax administration, including
communication strategies to explain the rationale and potential impact of proposed changes
to all managers, staff, taxpayers, and other stakeholders.
Coupled with strong commitments to the reform program and World Bank support, GDT
also receives complementary assistance from other members of the donor community. The
International Monetary Fund (IMF) provides assistance in developing a self-assessment
scheme, improving the law on tax administration, and organizational restructuring of the
tax administration. With IMF technical assistance, GDT has piloted self-assessment since
2004 and is currently rolling it out to provinces. The IMF is also supporting the reform of
the PIT and the implementation of a strategic planning process. The International Finance
Corporation (IFC) is supporting GDT in reforming the tax system for micro and small
businesses, simplifying tax filing requirements, and introducing a risk-based approach for tax
audit selection. Japan, through the Japan International Cooperation Agency ( JICA), provided
a three-year program (200508) for the development of a training plan and training materials
to support tax administration reform. The program has been extended and is currently in
its second phase. A multidonor trust fund (MDTF) has been established to support basic
training on tax administration functions, provide consultancy services on taxpayer accounting,
conduct surveys on taxpayer service needs, and implement IT applications in self-assessment
implementation. In addition, the U.S. Treasury provides technical assistance for capacity
enhancement focusing on some tax administration functions, anticorruption strategy, and
human resource development and management.
This is a comprehensive modernization program, but not a risk-free one. A reform program
that aims at reducing the opportunity for part of the tax officers to charge discretionary rents
risks being seriously resisted. Implementing an overly ambitious multifaceted reform agenda,
with interdependence between a complex computerization component and other required
activities, and particularly given the current weak capacity of the system, would lead to delays
in responding to timeliness and meeting targets of the reform agenda.
89
T ax R eform in V ietnam
OECD Forum on Tax Administration, IT-related costs have increased from an average of 6.6
percent of aggregate administrative costs in 2005 to the OECD level of 11.7 percent in 2009.
90
T ax A dministration
91
T ax R eform in V ietnam
indicators has been developed within the framework of preparation of the TAMP and will be
used by GDT management to measure its performance (table 3.3).
Table 3.3: A New GDT Results Monitoring Framework
92
T ax A dministration
global picture, and recent surveys reveal a particularly high incidence of corruption in the
GDT (table 3.4).
Table 3.4: Which Are the Most Corrupt Government Agencies in Vietnam?
Reducing corruption in tax and customs administration requires addressing both motives
and opportunities for corrupt behavior (table 3.5).
Table 3.5: Key Issues in Anticorruption Strategy
94
T ax A dministration
union will be set up within the framework of the TAMP, and the TAMP will also support
improving the operation of a hotline in GDT for reporting allegations of corruption. A
short-term plan for the implementation of the governments national anticorruption strategy
in the tax administration was prepared by GDT and promulgated in September 2010. The
implementation plan focuses, in particular, on strengthening human resource management
aspects of counteracting corruption, such as developing codes of conduct for tax officers
and improving staff rotation policies; reducing the direct contact between taxpayers and tax
officials through, for example, expanding e-filing possibilities and revising procedures for tax
audits; and increasing transparency in tax laws and regulations.
95
T ax R eform in V ietnam
GDT has planned several initiatives to improve the relationship with taxpayers. Modernizing
taxpayer services is one of the main components of the Tax Administration Reform Plan.
Activities include the establishment of taxpayer service centers and the development of new
service and information programs. Regular collection of taxpayer feedback will occur through
periodic surveys. An area that will require further attention in the future is the participation
of taxpayers in the GDT strategic planning process. The development of a GDT reform
plan should ultimately lead to a regular strategic planning process with the development
of a multiyear strategic plan. The design of such a strategic plan, and its acceptance by
stakeholders, will benefit greatly from an extensive dialogue with all major stakeholders, in
96
T ax A dministration
particular, businesses, throughout the planning process. During plan implementation, regular
stakeholder forums are a useful way to maintain the dialogue with the taxpayer community.
Increasing the involvement of the private sector in the reform process will also have an impact
on the accountability arrangements of GDT. The recognition that taxpayers are an important
stakeholder of the tax administration logically leads to accountability relations between the
tax administration and taxpayers. These are substantially different from the accountability
relations between the tax administration and the MoF, since taxpayers expectations regarding
the performance of tax administrations differ from MoF priorities and, therefore, lead to the
development of a new set of performance benchmarks. In modern tax administrations, the
establishment of taxpayer charters has frequently been a visible outcome of this process. See
box 3.6 for how Singapore handles its accountability to its taxpayers.
97
T ax R eform in V ietnam
level of voluntary SME compliance. Despite the growth in the SME segment of the taxpayer
population, the major contribution to tax revenue collection will continue to come from
the limited number of large businesses. With the privatization of large business operations
and the decreasing role of SOEs for revenue generation, a highly specialized large taxpayer
administration must be established. A rather centralized large taxpayer office with industry-
specific information gathering and technical expertise would be an appropriate response to
ensure proper compliance from the major taxpayers.
Little information is currently available on compliance risks and the level of tax evasion in
the medium and small segment of the taxpayer population. Creating analytical capacity in
GDT to measure compliance behavior and the size of the tax gap should be envisaged. A
special analytical unit in GDT headquarters staffed with highly qualified experts could take
responsibility for such analytical work.
The level of voluntary tax compliance and the efficiency of revenue collection are strongly
influenced by the level of integrity in tax administration. Corruption surveys indicate that
major efforts are needed to improve integrity in tax administration. This requires the effective
implementation in the tax administration of the governments anticorruption strategy and a
commitment to stakeholders that integrity enhancement is a priority for GDT management.
The major elements of such an anticorruption strategy will be implemented within the
framework of the GDT reform plan and the TAMP. However, the human resources
management element of integrity enhancement requires further ongoing attention.
With the move to a modern, service-oriented tax administration, the accountability
arrangements for GDT need to change. Accountability and results agreements should
cover all operational areas and not be limited to monitoring collection results. Modern tax
administrations are accountable not only to the MoF, but also to other stakeholders. This
requires a transparent monitoring of service commitments to taxpayers.
Starting with the GDT five-year reform plan 20052010, a regular, ongoing strategic
planning process was introduced. The multiyear strategic plan outlining the challenges for
tax administration in Vietnam in the forthcoming years, and actions required to address these
challenges, should be regularly updated and supplemented with annual and departmental
plans. The involvement of all stakeholder groups in the planning process, and in the process
of monitoring the implementation of the plan, is of key importance.
Conducting regular feedback surveys from taxpayers and tax administration staff is both an
important step to deepen the accountability arrangements to taxpayers and a tool to support
further development of the GDT strategy. GDT implemented a pilot taxpayer survey in
2008, and a survey on the implementation of the Tax Strategy 20052010 was carried out in
2010. Taxpayer surveys should be conducted on a regular basis in the future and should use a
consistent methodology to ensure comparability of results. Results from the surveys should
be made available to stakeholders.
Dedicated administrative structures to deal with taxpayers outside the large taxpayer segment
are not yet available. Following the establishment of a special large taxpayer administration,
the potential benefits and risks of further taxpayer segmentation should be analyzed. This
process is linked to further reforms on the tax policy side, however. With the planned
98
T ax A dministration
99
Forecasting Tax Revenue:
Methods And Tools IV
By Gangadhar Prasad Shukla, Daniel Alvarez, Tuan Minh Le, and Duc
Minh Pham
1. Summary
Revenue forecasting is the starting point in the preparation of government budgets and is at
the core of public sector budgeting. Well before the budget cycle starts, the revenue forecast
should be in place. Revenue forecasts are crucial for avoiding unanticipated deficits that
would have to be filled by unplanned borrowing, expenditure cut backs and ad-hoc revenue
measures. Revenue forecasting has become all the more important since the advent of the
medium term expenditure framework (MTEF) which is based on revenue predictability over
the medium term.
In addition to forecasting future tax revenues for budget preparation, forecasting techniques
may be used for estimating revenue potential and monitoring tax collections, evaluation of
economic and structural aspects of fiscal policy and for purposes of tax expenditure analysis.
102
F orecasting T ax R evenue : M ethods A nd T ools
i. Impact Analysis: They enable us to analyze the impact of a tax policy change on taxpayers,
specific groups of stakeholders as well as the tax revenues. Thus one can have a scenario
analysis where alternative scenarios of tax policies are examined to see how they compare
and pick up the one that is best suited to the countrys needs.
ii. Forecast Revenues: The models also enable a more precise forecast of revenues with or
without any changes in tax policy or tax structure. For GDP based forecasting models,
the outcome of the tax intervention is hard to model.
(d) Timeframe
The entire process described above should take about 18 to 24 months provided it is done in
an organized and systematic manner. Once the models are ready, they should be calibrated
103
T ax R eform in V ietnam
to correct any special features of the forecast. It should be noted that while the first cut of
the forecasting models should be able to answer what if questions and yield a decent set of
revenue forecasts, the process would be far from complete. It will take at least two more years
before the database is complete and clean and the models have been fully refined.
2. Introduction
Revenue forecasting is the starting point in the preparation of government budgets and is at
the core of public sector budgeting. Well before the budget cycle starts, the revenue forecast
should be in place. The forecasts should also be revised during the course of the year if some
major parameters affecting tax revenues, for instance the GDP growth rate, undergo change.
Both the initial and revised forecasts are crucial for avoiding unanticipated deficits in the
course of a budget year that would have to be filled by unplanned borrowing, or expenditure
cut backs and emergency ad-hoc revenue measures.
Revenue forecasting has become all the more important since the advent of the medium
term expenditure framework (MTEF) which is based on revenue predictability. Changing
the psychology of the spending departments and agencies from a needs to an availability
mentality is the most important feature of the MTEF approach to budgeting and therefore
predictability of funds becomes significant to reallocating future budgets. A medium-term
perspective is necessary because the time span of an annual budget is too short to adjust
expenditure priorities, and uncertainties become larger over the longer term. Since MTEF is
a multiyear exercise, revenue forecasts two to three years forward are necessary.22
At present, the General Department of Taxation (GDT) in Vietnam maintains the database
and conducts the revenue forecasting exercise for total tax revenues, VAT revenue, excises,
personal income tax, corporate income tax and natural resource taxes including oil revenues.
The projections for the following year are based primarily on the growth rate of the current
year revenues over the previous years revenue collections during the same period of time.
This growth rate is suitably adjusted after looking at the growth rate of turnover of the relevant
industries and sectors of the economy and after due consultations with line ministries and
field offices of the tax department at the province and commune levels. Evidently, the present
forecasting exercise does not employ any kind of econometric or microsimulation modeling
for total tax revenues or for major individual taxes. It is simply the use of some sort of trend
analysis.
The quality of data is also poor. What is currently available from the GDT and its field offices
mainly comprises revenue collections by tax type and by months. Consistent and integrated
database on tax base and the underlying economic parameters that affect the performance of
the various bases is virtually nonexistent.
22 In developed countries, the time horizon for revenue forecasting may be extended substantially.
For example, the U.S. Congressional Budget and Impoundment Control Act of 1974 requires the
Congressional Budget Office (CBO) to prepare detailed budget projections in an annual report
to the Congress. The report contains projections of federal outlays and revenues for the current
year and the next 10 years to be used as a budget baseline against which the effects of any proposed
changes in tax and spending laws can be measured (CBO, 2001: Description of CBOs Models
and Methods for Projecting Federal Revenues).
104
F orecasting T ax R evenue : M ethods A nd T ools
Before getting into specific tax forecasting models, some general issues are examined concerning
the relevance of revenue forecasting, the details of the organizational structure and the specifics
of the revenue forecasting exercise in Vietnam compared with international practices.
Time Activity
9 months Macro forecast (medium term: years 1, 2, 3)
Revenue forecast and debt policy (medium term)
Determine revenue gaps: review and revise basic revenue policies based
on such fundamentals as the size of government (increase/decrease
expenditures), debt burden, and new tax policies (changes in structure,
administration, incentives, distribution of burden).
6 months Revised macro forecast
Revised revenue forecast including tax measure requirements to meet debt
policy
Medium term expenditure ceilings: detailed budgets prepared (updated)
for years 1, 2, 3 in consultation with the spending agencies
3 months Refined medium term forecast with emphasis on year 0 and 1 revenues with
final target for discretionary revenue measures for year 1. (The final forecast
for year 0 changes the basis of the forecast for years 1, 2, 3.)
0 months Present budget to the legislature
Source: Authors own study of best practices in sound budget making in developed economies.
105
T ax R eform in V ietnam
The budget making process should start with a macroeconomic and revenue forecasting
exercise. The macro forecast predicts the economic growth rate which feeds into the revenue
forecasting exercise by predicting the growth in tax bases (GDP, consumption etc.) and
indicates the level of the desired public sector investment and expenditures. These two
forecasts are also essential to determine the amount of debt required to meet the government
expenditures as well as the need for deficit financing, if any. Once the revenue envelope is
established, the expenditure limits should be decided, keeping in mind government policies
and demands from line ministries and spending agencies.
It is not uncommon, however, to see the entire process turned upside down in most developing
countries and transition economies, where expenditure demands are first collected and then
an effort is made to raise commensurate revenue amounts through different sources. This
practice has the potential of increasing the debt burden and also putting excessive pressures
on the tax administration to raise more and more revenues which may potentially destroy the
integrity of the tax system.
The suggested timeframe may vary somewhat from country to country but it is important to
follow this sequence and timetable as closely as possible to avoid any ad hoc decision making
in the budgeting exercise which may prove costly to the economy at a later stage.
The main tasks of revenue forecasting comprise:
Computation of elasticity (and buoyancy) of various taxes and the entire tax system;
Forecasting revenues for each type of tax and impact of policy change on major criteria of
a good tax system (for example, revenue adequacy, efficiency and equity); and
Evaluation of the effect of inflation and price changes on tax revenues.
106
F orecasting T ax R evenue : M ethods A nd T ools
Provision of a framework for analyzing policy options for income tax, using a
microsimulation model;
Assessment of the impact of changes in one tax regime on revenues from other taxes;
Scrutiny of nontax revenue sources and curtailment of nuisance taxes;
Estimation of tax revenue impacts of alternative premises about the economy such as
economic growth rate or growth rate of specific economic sectors; and
Measurement of the effect of macroeconomic policy changes on the tax regime and
tax revenues such as (a) removal of quantitative restrictions on imports, (b) trade
liberalization policies and their effect on import substitution and export promotion, (c)
deregulation of certain economic activities, (d) currency fluctuations, (e) outcome of
interest rate changes in terms of business profits.
107
T ax R eform in V ietnam
be assigned any task and responsibility by the Director. There is no specific distribution of
work among this group of officials.
At the provincial level, there is a forecasting unit manned by two to three people who are
engaged in the task of revenue projection for each type of tax. Thus, there are 63 field units
that work with the central revenue projection department. These units are equipped with
computers for data maintenance and revenue forecasting task.
Finally, GDT has its field offices at the commune or district level that support the tasks of tax
collection and data gathering.
108
F orecasting T ax R evenue : M ethods A nd T ools
apparatus to conduct a detailed revenue forecasting exercise. That task is performed by the
GDT and then the MoF makes only peripheral adjustments.
109
T ax R eform in V ietnam
of actual tax returns and data available for revenue projections has to involve the field offices
as well.
The present practice of revenue forecasting for different kinds of taxes is outlined below.
Value Added Tax: Step I, in the month of May, the preliminary projections for the
following year (say 2012) will be made. These projections are mainly based on the growth
rate of revenues between January to April that year (2011) over the collections of the
previous year (2010) in the same period. The growth rate of sales revenues of producer
companies and distributors between the previous year (2010) and this year (2011) are
also considered before arriving at the final growth rate figure. Then the same growth rate
is applied on that years (2011) estimated revenue figures to arrive at the following years
(2012) projections. The VAT revenue projections are made by sectors food, beverages,
pharmaceuticals, etc.
Step II, in the month of August, discussions are held with other ministries to get their feedback
and then the preliminary projections are sent to the field offices. The provincial offices assess
whether the projected figures seem to be of the right order and also if there is any gap between
the growth rate of revenue collections and the growth rate of sales, in which case they try to
find out the reason behind this divergence.
Step III is the finalization of the revenue forecast by the month of November for submission
to and approval by the National Assembly.
Retailers do not pay taxes based on their actual sales but rather on estimates made by the
tax department about their sales revenue, input costs and tax liability (using subtraction
method). Distributors and producer companies submit their returns which are maintained
by the field offices.
Special Consumption Tax: The revenue projection for Special Consumption Tax is done
separately for each category of goods subject to this tax. The revenue projection model is
almost the same as that for VAT.
Personal Income Tax: The revenue projection method is the same as in the case of other taxes
like VAT. The personal income tax data is maintained under two categories: foreigners and
Vietnamese nationals. Salaried persons do not submit tax returns but income tax is deducted
at source by the employers and they submit a return at the end of the year. Self-employed
professionals have to file tax returns and they are expected to file the returns at the field offices
of the tax department.
Corporate Income Tax: The revenue projection is done in a manner similar to other taxes
described above. The only additional point is that the figures for sales and profits are cross
checked with those for VAT to ensure that the two are consistent.
Natural Resource Taxes: The revenues are projected, enterprise by enterprise, based on their
production plans that outline the quantities of production and the expected prices in the
future and their actual performance in the previous years.
110
F orecasting T ax R evenue : M ethods A nd T ools
3. Conclusion
The following three aspects become clear from the above description.
i. The organizational set up for data collection and revenue forecasting and the necessary
staff are in place. However, the staff has not been properly trained either in tax analysis or
revenue forecasting. All the learning is on-the-job and therefore not methodical;
ii. The present practice of revenue forecasting is based only on the past years trends and
therefore is rudimentary and subjective in nature. Currently no macro or micro simulation
models are being used;
iii. Since the models are not in place, the necessary database has not been constructed in a
scientific way.
It should be noted that most developing countries are in a similar situation and are struggling
with the problems of a lack of a good database, the absence of revenue forecasting models and
a dearth of capacity and skill.
111
T ax R eform in V ietnam
proper concept of tax elasticity.23 Recently, some countries such as Russia and Bulgaria
have established and developed internal dedicated revenue forecasting functions within
tax administration agencies as part of comprehensive tax administration modernization
programs.
The IMF has done a comparative study of forecasting practices in developing countries and
transition economies based on a survey of 34 countries (Kyobe and Danninger 2005) from
Africa, Asia, Latin America, the Middle East, and the CIS. Some key findings of this study are
summarized below.
In most countries forecasting responsibilities are poorly defined and there are few formal
rules and regulations for the revenue forecasting exercise.
Estimation techniques are generally rudimentary in nature. About 85 percent of countries
use subjective assessment and simple extrapolation techniques as their main methods for
revenue forecast.
The process involves multiple executive agencies, outside the ministry of finance,
requiring a high degree of coordination.
It is common to have multiple competing forecasts.
Revenue forecasts are often produced late in the budget process, thus disabling the
important link between revenue forecasting and budget making.
Public accountability in terms of access to forecast results and participation of
nongovernmental agencies in the revenue forecasting exercise is limited.
These characteristics do not differ significantly along regional or per capita income levels.
More transparency and rigor are applied in Latin American countries which also have
higher income levels.
Higher formality and transparency are found in countries with a higher per capita income
and lower corruption levels.
23 Both tax buoyancy and tax elasticity are measured as the ratio between the real growth rate of tax
collection and the real growth rate of the tax base (for example, GDP). However, the elasticity of
a tax system or an individual type of taxes measures the efficiency of the underlying tax structure
and administration, excluding the impact of discretionary changes in tax rates and/or base. For
macro-based forecasting purpose, one should use the concept of tax elasticity.
112
F orecasting T ax R evenue : M ethods A nd T ools
of tax revenues with respect to the base. Thus, to apply this model, one has to estimate the
elasticity of overall taxes or a particular type of tax using a regression analysis on time series
data of tax revenue and tax base. The next step is to forecast the growth in tax base and finally
to forecast tax revenues with the help of tax elasticity and increase in the tax base. The forecast
of growth in the tax base (GDP, consumption) can be simply extracted from the real sector
(macroeconomic) forecasts.
If tax rates and/or the tax base also change (that is, there are discretionary changes in tax rate
and/or tax base), then the increase in revenue in that particular year would depend upon
the normal increase in the tax base plus the impact of the discretionary changes made. Now
one has to estimate the impact of the discretionary changes as well. If we were to try and
estimate the elasticity in the year of discretionary changes by the usual regression analysis, we
would end up calculating buoyancy which includes both the impact of economic growth and
discretionary changes. If T0 is tax revenue this year, then T1, the expected tax revenue next
year, can be linked to T0 and other factors as follows:
T1 = T0 + Growth in Revenue without Change in Tax Structure + Change Due to Discretionary
Measures
Thus to apply the macro model, one has to separate the growth in tax revenues coming purely
from an increase in the tax base and the increase coming from discretionary changes. While
it is called the GDP based model, the tax base need not always be GDP and it could be payroll
revenues or corporate profits or consumption or value/volume of imports.
113
T ax R eform in V ietnam
Clearly these models are more elaborate, and they need more extensive data. It also takes
time to build them. But the models can and should be built by users in a transparent way and
must not be black boxes where the user does not understand what is happening. Once the
models are functional, they are user friendly and effective.
Two crucial aspects need to be taken into account:
i. The time period analyzed: in an MTEF perspective, the time horizon for forecasting
should typically be around three years to ensure a proper overview of the macrorisks in
terms of budgetary outcomes (deficits) in the near future.
ii. Cash versus accrual basis: The essential question is whether one deals with cash receipts
or with the tax liabilities due in a given period. Typically, the macromodels forecast cash
receipts, whereas the microbased models study liabilities or tax potential which may
actually come to fruition some time (say, half a year) later and need to be corrected to
arrive at the actual cash collections which are of most interest to the government.
114
F orecasting T ax R evenue : M ethods A nd T ools
While the model building exercise can start even when the database is in the process of being created
and cleaned, it must be emphasized that the breadth, depth and integrity of the dataset will determine
the ultimate quality of the models and their results. So this exercise is important, should be first to
start and should be undertaken with due diligence and care.
115
T ax R eform in V ietnam
services are covered under the VAT, following current tax code rules in terms of base and rate
definition (The VAT Law 1997 and subsequent revisions as well as supporting legislation for
the implementation of the VAT Law). Then, using alternative VAT rates and the new base
with rationalized exemptions, an assessment of the impact on revenue collection was made.
Data requirement
The main data requirement is an updated Input-Output (I-O) table. If an I-O table is available
for previous years, it needs to be updated by augmenting the entries in the ratio of GDP of the
year of forecasting and the GDP of the I-O table year. If, however, the I-O table is more than
five years old, this ageing of data may not yield optimal results.
In addition, household surveys of income and expenditures and details of government
consumption are needed. Growth rates for consumption as a share of GDP can be extracted
from macroeconomic forecasts or simply by trend or regression analysis on past data.
Methodology
The tax base is estimated by employing the equivalence of the value added tax to a retail sales
tax levied on the final selling price of goods and services. Thus, the main concept in the Input-
Output revenue estimating model is to capture the total expenditure (personal, business,
and government) that is incurred in the economy. While constructing the VAT base, it is
necessary to arrange all the expenditures, including personal and government expenditures,
intermediate and capital expenses, under the current tax laws and rules.
The starting point for constructing the VAT base is the detailed information available for
domestic consumption. Only those domestic expenditures in the final demand categories
that are meant for personal and government use, not those for the further production of goods
and services for commercial purposes, are considered as final consumption for constructing
the VAT base.
The amounts of personal and government expenditures are depicted in the final demand
matrix of the I-O tables. Furthermore, since trade and distribution margins of retailers,
wholesalers and transportation operators are part of the expenditures eventually paid by the
final consumers; they should all be included in the VAT base.
As final government expenditure is generally subject to VAT, the VAT base is simply calculated
by adding domestic personal and government expenditures at retail prices. Wages and salaries
paid to public servants are, however, not subject to VAT and hence are excluded from the VAT
base.
From detailed final consumption expenditure by goods and services, adjustments are made to
subtract zero rated and exempt goods and services. When the VAT code zero-rates or exempts
certain commodities, the full value of zero-rated goods and services has to be removed from
the potential VAT base.
Capital expenditures (with the exception of residential construction) are not considered
part of the tax base, since they are used as business inputs into the production of final goods.
Further adjustments have to be made to the VAT base in order to take into account the effects
116
F orecasting T ax R evenue : M ethods A nd T ools
of those cases in which tax-exempt sectors are not allowed to claim taxes paid on their business
inputs and there is an element of cascading.
The expected VAT revenue by commodity at 100 percent compliance can be calculated by
multiplying the tax base with the taxable proportion and tax rates. In the case of Vietnam,
separate estimation needs to be made for those commodities subject to a standard rate (10
percent) and those subject to nonstandard rates or a preferable rate (5 percent). In general,
VAT revenues (R) at full compliance are estimated as follows.
R = Ci i + K j + i {Ei Si } m rm
i j i m
Where,
Ci = before-VAT final expenditures (private and nonwage government consumption) of
commodities in sector i;
117
T ax R eform in V ietnam
i. Base case scenario. Revenue collection for FY2007 is estimated, following the current
base and rate structure.
ii. Only one positive rate. A unified 10 percent tax rate substitutes the current structure of
two rates5 and 10 percentand allowing zero percent rate only for export transactions.
The current exemption structure is maintained.
iii. One-rate structure plus exemption rationalization. Starting from Scenario 2, preferential
treatment is allowed only for a few goods and service categories, specifically financial
services, primary education, basic healthcare (not higher education and specialized
health services), some cultural and merit goods and services, and aid financed activities.
The rest of the exempt items were moved to the taxed category.
The results are summarized in table 4.2.
Table 4.2. VAT Policy Impact Scenarios, 2007
Note: Estimated compliance rate of 74 percent for 2006 is applied in the 2007 VAT revenue projection.
Option 3 refers to the broadest revenue enhancement scenario. With a unique positive tax
rate, and a broad base, it is expected that an increase in the overall compliance rate will take
place due to efficiency gains in tax administration and reduction in compliance costs.
As can be seen, the rate of compliance is an important parameter in this model. This coefficient
was calculated using actual collection for FY 2006 and the revenue potential estimated for
that year. Using the dataset from private consumption expenditures from the latest I-O
Tables and the Household Income-Expenditure Survey, the current Vietnam VAT structure
was modeled for FY 2006 assuming full compliance by taxpayers. As a result, potential
collection is estimated to be VND 48,979.6 billion (5 percent of GDP). The compliance rate
is obtained from the ratio of actual VAT collection for FY2006VND 36,469 billionand
the revenue potential for 2006, thus resulting in a compliance rate of 74 percent. This is the
figure used in forecasts for FY 2007, assuming no major changes in the tax administration
effort during 200607.
Personal Income Tax
The microsimulation model is used as an analytical tool to evaluate the impact of the reform
proposals to the PIT structure in Vietnam at the individual level. The model aims to forecast
personal income tax revenues and to assess the distributional effect of a given policy proposal
on a particular segment of the population.24
Recently, a team of consultants applied an econometric model to analyze the revenue and incidence
24
impact of the new draft Law on Personal Income Tax. The study was funded by DANIDA and the
Tax Policy Department (TPD) was the coordinator as well as the beneficiary. The model uses two
main sources of survey data: the VHLSS-2006 and the 2005 General Department of Taxation survey
118
F orecasting T ax R evenue : M ethods A nd T ools
The main advantage of microsimulation modeling for personal income tax lies in its capacity
to estimate the distributional effect of a given policy proposal on particular sectors of the
population. Since these models are capable of dealing with distributional impacts by
identifying potential winners and losers in the society from a given policy proposal, they have
proved to be powerful tools in the hands of analysts (Gupta and Harding 2007).
Data Requirements
The following data are required for microsimulation modeling for personal income tax.
Individual or family-based annual income tax returns;
Household surveys, particularly the Household Income-Expenditure Survey (VHLSS)
to cover nonfilers;
Personal income tax regime; and
Basic social and economic data: consumer price index, growth rates of population, GDP,
and investment.
The process of collecting data is often complex, since the models require large representative
samples from the total population. Data is sometimes collected through surveys for specific
purposes. Analysts tend to use mostly administrative tax data or the censuses carried out by
government agencies.
Methodology
The following steps are used when constructing a micro simulation model for personal
income tax (Harding 1996).
(a) Database Construction
In practice we generally rely on a relatively small sample size. Typically, the sample size
ranges from 0.5 to 5 percent of the total filing population, depending upon the size of the total
potential taxpayers and the available resources to collect and analyze the data.
Different strata are established on the basis of income sources (for example, employees,
self-employed, investment, pensioners, etc.), place of residence (for example, urban, rural,
provinces, foreign residence, etc.), and income levels (for example, low, medium, high).
(b) Data Cleaning and Data Completion
Continuous cross-checking against external data sources is carried out at different stages of
the model development process. This includes filling in missing data on nonfilers on the basis
of household surveys.
of income and spending by high income individuals. While the model is elegant and well elaborated,
it does not escape from some typical self-selection and respondent biases inherent in survey data. In
this study, we follow the micro-simulation approach, applied by revenue administrations in developed
countries, that relies mostly on tax administration data (that is, the individual or household income
tax returns). For a discussion of the practicality and objectives of the microsimulation model, see
Gupta and Vishnu 2000. In addition, in the case of Vietnam, the microsimulation model approach
would enable the GDT to pinpoint the level of compliance by different groups of taxpayers and
thereby develop appropriate compliance for specific segments of taxpayers.
119
T ax R eform in V ietnam
120
F orecasting T ax R evenue : M ethods A nd T ools
formed by a stratified sample, each observation carries a weight corresponding to the number
of tax filers represented by that observation. Calculations of the model have to be adjusted
by this weight in order to obtain aggregate revenue estimates. Figures 4.1a and 4.1b provide
a conceptual framework for the construction of a typical taxpayer and aggregate model
structures as applicable in the case of Vietnam. Note in the figures, the Current Regime refers
to the provisions stipulated in the personal income ordinance of 2004 while the Proposed
Changes refer to the provisions in the personal income tax law enacted in 2009.
Figure 4.1a: Typical Taxpayer Model Structure
Aggregate Model
MacroAggregate
Module Model
Macro Module
Personal Information
Personal Information
Database
Database
Current Tax Proposed
RegimeCurrent Tax
ChangesProposed
Regime Changes
Impact Distribution
Impact Distribution
AnalysisAnalysis
Taxpayer
Tax Calculator
Taxpayer Personal
PersonalInformation Tax Calculator
Information
Personal Income Tax
Liability
Personal
CurrentIncome
Tax TaxProposed
Liability
Regime Changes
Current Tax Proposed
Regime Changes
121
T ax R eform in V ietnam
122
F orecasting T ax R evenue : M ethods A nd T ools
Table 4.3: Average Tax Rates for Different Levels of Income Tax Ordinance 2004 and
Current Law Enacted in 2009 (Million Dong per year)
Tax Payable Tax Payable Average Tax Rate Average Tax Rate
Total
(Ordinance (PIT Law Impact (Ordinance 2004) (PIT Law 2009)
Income
2004) 2009) (%) (%)
30 0.0 0.0 0.0 0 0
90 3.0 1.1 -1.9 3 1
180 12.0 8.3 -3.7 7 5
330 45.0 31.6 -13.4 14 10
540 114.0 76.2 -37.8 21 14
810 222.0 147.8 -74.2 27 18
1,200 378.0 273.5 -104.5 32 23
1,500 498.0 378.5 -119.5 33 25
2,000 698.0 553.5 -144.5 35 28
3,000 1,098.0 903.5 -194.5 37 30
Source: Own calculations, using Vietnam PIT Schedules.
Thus the PIT law of 2009 appears to be less progressive than the previous law of 2004 if only
wage incomes are taken into the tax base as was done in the law of 2004. Prima facie, it also
yields less revenue since the revenue collections for FY 2009 are estimated to be reduced by
21.7 percent in nominal terms.
Figure 4.2: PIT Progressivity (wage income)
40
35 37
33 35
30 32 30
Average Tax Rates
25 27 28
21 23 25
20
18
15 14 14
10 10
5 75
31
0 0
30 90 180 330 540 810 1,200 1,500 2,000 3,000
Annual Income (Million VND)
The progressivity and revenue potential of the PIT law 2009 need to be examined more
closely, however, because the tax base in this law is expected to expand by including other
forms of income.
123
T ax R eform in V ietnam
According to the VHLSS-2004, the average composition of income, the structure of income
per capita by source of income, is as follows: Salary or wage (32.7 percent), Agriculture (22.6
percent), Forestry (1 percent), Fishery (3.6 percent), Industry (5.4 percent), Construction
(0.4 percent), Trade (9.9 percent), Services (6.8 percent), and others (17.7 percent).
The PIT law 2009 exempts income from primary activities while other income sources are
taxed. Therefore, besides wages and salaries, which were practically the only source of income
taxed by the old PIT law, two new categories of income were introduced into the analysis: (i)
business income, which includes industry, construction, trade and services income sources,
and (ii) other incomes such as capital investment, royalties, capital gains, immovable property,
donations, inheritance, and capital transfer.
By using this new income source categorization, a new tax calculator was used to assess the tax
liability of a typical taxpayer under both PIT systems, for different annual income scenarios
(30 to 3,000 million dong). Considering an average of two dependents per household instead
of one, taxable income was estimated by deducting a yearly amount of 86.4 million dong
(67.2 million dong in the previous exercise) to reach taxable income.
Similar to the previous progressivity exercise, statutory marginal tax rates were applied under
both PIT laws. The only difference is that other income was taxed at a 13.33 percent flat rate,
which is the average tax rate applicable under the PIT law 2007 to sources of income such as
investment, royalties and alienation of immovable property.
Applying the PIT Micro simulation model to the data set of 3,000 individual tax returns available
for Vietnam, under the income tax law of 2007, the tax revenue collection is estimated to increase
by approximately 69 percent in nominal terms, to VND 1,853 billion from VND 1,098.7 billion.
Measured by average tax rates, the PIT law 2009 turns out to be more progressive at different
levels of income compared to the old system, when different sources of income are included
into the tax base. This is the outcome even when the taxpayers are given an allowance for two
dependents under the new law. The results are presented in table 4.4 and figure 4.4.
Table 4.4: Assessing Progressivity of Old (2004 and New (2009) PIT System
(VND million)
Total Wages Business Other Taxable income Tax Payable Average Tax
Income Income Income Rates
Old New Old New Old New
System System System System System System
30 14 9 7 14 30 0.0 1.0 0% 3%
90 41 27 22 41 90 0.0 2.9 0% 3%
180 81 55 44 81 180 2.1 8.3 1% 5%
330 149 100 81 149 330 8.9 26.2 3% 8%
540 244 164 132 244 540 24.8 60.9 5% 11%
810 366 246 198 366 810 55.8 115.8 7% 14%
1,200 542 365 293 542 1,200 114.8 210.2 10% 18%
1,500 677 456 367 677 1,500 169.0 292.3 11% 19%
2,000 903 608 489 903 2,000 259.3 440.8 13% 22%
3,000 1,355 912 733 1,355 3,000 440.0 737.9 15% 25%
124
F orecasting T ax R evenue : M ethods A nd T ools
30
Average Tax Rates 25 25
20 22
19
15 18 15
14
10 11 11 13
8 10
5 5 7
3 3 3 5
0 0 0 1
30 90 180 330 540 810 1,200 1,500 2,000 3,000
(5)
Annual Income (Million VND)
Data Requirements
(a) Tax returns filed by companies in the last two to three years: Stratified samples should
be selected from these to represent different categories of industries and amounts of assets/
investments. If only samples are available instead of the entire dataset, the selected samples
should represent different categories of industries and the amount of assets. Ideally, all the top
5 percent companies by assets should be included in the samples. To maintain confidentiality,
the company name and identification should be excluded from the file.
(b) Basic Data pertaining to the business sector: Such data related to operations, revenues and
expenditures of the business sector such as corporate investment (by sector and type of asset),
corporate income, profits, tax liability, dividend payments, and tax deductible expenses such
as interest payments, costs of goods sold, fiscal depreciation. In addition, if the tax agency
publishes any aggregate or disaggregate data on the corporate taxes that would be useful. The
sources for these business sector related data could be enterprise censuses or surveys.
Methodology
Conceptually, a microsimulation for CIT is similar to the one applicable to PIT. As part of the
model, a CIT calculator is to be constructed in such a manner that it is capable of computing
business income tax liability in Vietnam for a typical business. The existing tax law is applied
to a representative business to be compared with the tax liability of the same business under
a proposed tax policy change. Having developed the typical taxpayer model, it is possible to
simulate the impact on the tax liability of the proposed policy changes in the tax system, such as
the change on the CIT tax rate and the rationalization of incentives (Citro and Hanushek 1991).
The model reads individual sets of data from the microsimulation database and performs
individual tax calculations based on laws and regulations. As the database is formed by a
stratified sample, each observation carries a weight corresponding to the number of tax
125
T ax R eform in V ietnam
126
F orecasting T ax R evenue : M ethods A nd T ools
Model Results
According to the database available, three types of corporations are considered: state owned,
foreign invested, and nonstate owned. Also, the different CIT tax rates shown in the data
sample are considered. For the sake of simulation, a flat rate of 28 percent across corporations
was considered instead of the structure with differentiated rates. In any event, this table
provides flexibility as changes to the current and proposed tax rates can be easily done
without modifying the macro codes. The formulae in the Tax Calculator Model simulate the
tax calculation procedures as presented in the 91 tax return samples.
The Calculator Model presents the calculation of individual corporation tax liability under
the existing law and the proposed changes in the law, as well as the changes (impacts) on the
corporations tax liability due to the discretionary changes. The Tax Return Database stores
the tax returns data from national samples of selected corporations. The Macro Module
automates the process of simulating the calculation of tax liability for the corporations in
the database. The Module reads individual corporation records in the database, stores them
in the Tax Calculator Model and saves the calculated tax liability in the designated location
in the Tax Returns Database. This process continues until the last record in the database is
covered. The microsimulation performed for the 91 corporations tax returns data shows the
revenue impact from the 28 percent flat rate proposed.
Using the 91 CIT individual tax return database available, a policy analysis scenario was
constructed using a 28 percent CIT flat rate across the different corporation types. After
performing the tax calculator model, it is expected that the government would collect VND
105.8 billion, instead of the VND 71.2 billion estimated, which represents an additional CIT
collection of VND 34.6 billion, or an increase of 48.6 percent in nominal terms.
Note: a. Policy scenarios constructed under the following assumptions: VAT: 10 percent single rate and exemption
rationalization; CIT: Microsimulation using sample of 91 returns with a 28 percent flat rate across all corporations; and
PIT: Microsimulation using 3,000 returns and applying the PIT Law 2007.
b. The revenue figures are for the sample size of 91 corporations or 0.15 percent of the total companies in Vietnam. If
prorated, base case total revenue would be VND 46,466 and 70,000 billion dong.
127
T ax R eform in V ietnam
128
Tax Reform and Revenue
Assignments in Vietnam V
By Jorge Martinez-Vazquez and Minh Van Nguyen
1. Introduction
Tax reform and decentralization reform have some similar objectives but also their own
separately independent ones, and unless both policy processes are harmonized, decentralization
and tax policy reforms can work at cross purposes. The most important objectives pursued
by tax reform include those of simplicity, low economic distortions or efficiency, and equity
or fairness in the distribution of tax burdens. In the case of decentralization reform, those
objectives are also important but they necessarily take a second seat to the objectives of
revenue adequacy for subnational budgets and accountability of local government officials.
The attainment of these decentralization objectives requires of significant own taxes and fees
assigned to subnational governments.
However, pursuing a decentralized system with more tax sources assigned to subnational
governments will likely mean that the entire tax system for the country will be less simple,
and a bit more distortionary, and even perhaps less fair. But, it would be wrong to reach the
conclusion that we should not have fiscal decentralization on the revenue side of subnational
budgets. Instead, what is needed is the coordination of the tax reform effort with the
desired level of revenue decentralization for the country. This coordination needs to reach a
compromise between the decentralization objectives of accountability and revenue autonomy
for subnational governments and those of simplicity and efficiency for the entire tax system
at the national level. Even though subnational taxes may be less progressive than desired, it
needs to be kept in mind that what matters from the perspective of distributional equity is
the overall level of progressivity for national and subnational taxes combined. Subnational
taxes are better guided by the application of the benefit principle, that is, a correspondence
between the taxes paid and the value of the services received. The application of the ability to
pay principle and the objective of income redistribution are much more effectively pursued
at the national level. After all, potentially less progressive taxes at the subnational level can be
compensated with more progressive taxes at the national level.
In sum, it is important to revisit the issue of revenue assignment in decentralization design
at the time of major tax reform. Because Vietnam is involved in a major tax reform effort,
it is justified to take an in-depth view of its revenue assignment system. This is the major
motivation of this paper.
As part of the drive toward decentralization in Vietnam, budget allocation norms have been
established for recurrent and capital expenditures. Rules for tax revenue assignment are the
natural complement to those norms. However, the current criteria for revenue assignments,
in particular which taxes are retained and the sharing arrangement for each province,
fundamentally are still ad hoc. The time of fundamental tax reform is a good opportunity to
put the question of revenue assignments on a more sound footing in Vietnam. In this paper
129
T ax R eform in V ietnam
we review the current system and analyze these assignments against public finance principles
and best international practices. The paper also suggests several avenues for reform.25
Martinez-Vazquez (forthcoming).
The Budget Law leaves provincial governments with considerable discretion in the design of
26
revenue assignments to districts and communes within their borders. However, article 34 of the
2002 Law provides some general principles and minimum standards the provinces need to follow
in designing the revenue assignments for their local governments.
By far the most common approach internationally calls for sharing rates that differ by tax but that
27
bases of the actual revenue collections of the previous years, taking into account any tax policy
changes applicable in that year and the expected economic growth during the year. The minimum
expenditure needs of the provincial government are derived on the basis of the prevailing system
of expenditure norms, and cover all current expenditures (salaries, operation and maintenance,
and so on) and (some minimum amount of) capital expenditures. The norms are adjusted for
different regions depending on geography and remoteness.
130
T ax R eform and R evenue A ssignments in V ietnam
In addition to the shared revenues, as discussed below, the poorest provinces get an additional
equalization transfer. Relatively wealthier provinces get sharing rates between 100 and zero
percent. Table 5.1 shows the tax sharing rates for the provinces for the 200406, 200710 and
201115 stability periods. For the 200710 stability period, out of the 64 provinces, there
are 11 which are considered surplus provinces with sharing rates below 100 percent. This
represents a reduction by 4 provinces in the number considered as surplus units by comparison
to the 200406 stability period. Otherwise the changes in sharing rates between those two
stability periods were relatively small and generally downward. More recently, the MOF and
the provincial governments made a compromise to extend the stability period from three to
five years. The quid pro quo of the agreement was more stability for the provinces and the
ability of the center to get higher shares from the surplus provinces. The experience has been
that are the richer, more economically dynamic provinces that tend to benefit from the length
of the stability period. From the last column in table 5.1, it is apparent that most of the share
rates have been reduced. An exception is Hanoi, which because it has been merged with Hatay,
which was a poorer jurisdiction, it has been allowed to retain a higher revenue share.
Table 5.1: Tax Sharing Rates for the Provinces 200406, 200710,
and 201115 Stability Periods
131
T ax R eform in V ietnam
132
T ax R eform and R evenue A ssignments in V ietnam
Currently, Vietnam complements revenue assignments to the provinces with two types
of transfers: the balancing transfer and specific grants. The balancing transfer is an
equalization grant designed to increase the financing viability of poor provinces. These
grants are unconditional and calculated using a formula. In addition, these grants are
fixed in nominal terms during each stability period of three-to-five years.29 For specific
conditional grants, there are several types. An important type is those transfers for the
implementation of National Target Programs, such as those for various national health
programs. A second type is conditional transfers that are specifically designed for a
particular province. The third type is matching grants designed to implement important
projects for socioeconomic development. There are also emergency grants for help with
natural disasters and so on.30
In Vietnam, all tax collections are centralized. The General Taxation Department collects
all domestic taxes with offices that extend through the provinces and the districts, and the
Customs Department collects all taxes falling on imports and export tax. Only minor fees
and charges are collected by financial agencies and service providers.
29
The formula used for the determination of equalization transfers to the poor provinces is also the
same formula used to derive the customized tax sharing rate for the rich provinces discussed above.
For the most part, provinces implement with their lower-tier governments a system of balancing or
equalization transfers that is similar to that of the central government with the provinces. However,
between the provincial and lower levels, the provinces set their own norms and so effectively set
their own transfers.
30
Capital transfers in Vietnam are part of the equalization transfer and treated as a general transfer.
The capita transfer is now being determined by a formula that takes into account poverty, economic
development level, and other factors for each province. In the recent past there was no formula for
determining the amount of capital transfer; instead, the capital transfers were calculated by using
a grossing up factor for the expenditure needs determined for the calculation of the equalization
grant.
133
134
Table 5.2: Provincial Revenues by Revenue Sources and Summary Statistics, 19972010
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010e
TOTAL REVENUE
Total (Million VND) 30,360,918 31,875,486 40,122,103 47,508,691 58,842,075 65,822,571 85,850,251 105,039,065 124,193,506 146296,591 174810,837 235073,044 218711,704 273689,143
T ax R eform in V ietnam
Maximum (Million VND) 3,795,677 3,652,867 3,876,337 4,482,944 6,173,867 7,593,201 10,163,321 12,346,213 13,024,259 14439,759 15065,902 26684,144 23474,893 32830,801
Minimum (Million VND) 190,768 218,584 258,408 317,738 393,632 422,105 515,109 628,279 823,278 991,184 1076,351 1154,006 1411,484 1641,467
Average (Million VND) 497,720 522,549 657,739 778,831 964,624 1,079,059 1,407,381 1,641,235 1,940,524 2285,884 2731,419 3673,016 3471,614 4344,272
Standard Deviation 511,995 483,147 530,261 601,639 807,217 1,014,225 1,350,802 1,651,534 1,793,452 2086,635 2205,545 3693,735 3622,196 5065,666
Coefficient of Variation 1.03 0.92 0.81 0.77 0.84 0.94 0.96 1.01 0.92 0,91 0,81 1,01 1,04 1,17
Total (Million VND) 10,663,832 10,188,870 10,897,540 11,438,666 12,855,610 15,809,860 22,817,136 33,720,219 35,757,399 41827,070 43819,223 60477,542 52029,040 64455,744
Maximum (Million VND) 2,292,169 2,351,102 2,310,872 2,407,533 3,380,660 4,638,815 5,216,889 6,172,473 5,883,673 6389,724 4778,388 12033,857 9965,650 14031,610
Minimum (Million VND) 11,167 10,967 13,806 17,020 18,967 25,365 30,628 16,579 38,470 37,672 53,960 39,583 55,413 64,183
Average (Million VND) 174,817 167,031 178,648 187,519 210,748 259,178 374,051 526,878 558,709 653,548 684,675 959,961 825,858 1023,107
Standard Deviation 303,698 304,245 299,857 321,548 433,124 597,627 718,074 877,410 907,762 1008,224 937,854 1722,516 1454,326 2050,044
Coefficient of Variation 1.74 1.82 1.68 1.71 2.06 2.31 1.92 1.67 1.62 1,54 1,37 1,79 1,76 2,00
SHARED TAXES
Total (Million VND) 9,739,476 9,397,695 8,713,205 11,147,368 13,000,372 14,735,167 19,927,159 31,770,783 39,446,837 45463,862 55429,802 73488,958 82627,357 104928,015
Maximum (Million VND) 1,310,961 1,159,989 1,144,284 1,729,470 2,012,719 2,197,020 3,906,979 5,946,359 6,886,196 7778,749 10369,631 12752,406 14377,220 19542,484
Minimum (Million VND) 8,219 8,094 9,677 8,826 10,631 11,941 13,646 16,721 29,275 3,265 5,710 9,349 8,150 8,955
Average (Million VND) 159,664 154,061 142,839 182,744 213,121 241,560 326,675 496,418 616,357 710,373 866,091 1166,491 1311,545 1665,524
Standard Deviation 224,206 202,066 198,463 289,192 331,695 369,871 572,976 847,728 992,612 1148,415 1460,475 1901,030 2444,747 3343,462
Coefficient of Variation 1.40 1.31 1.39 1.58 1.56 1.53 1.75 1.71 1.61 1,62 1,69 1,63 1,86 2,01
TRANSFER
Total (Million VND) 9,957,610 12,288,921 20,511,358 24,922,657 32,986,093 35,277,544 43,105,956 39,548,063 48,989,270 59394,292 75645,100 95152,440 84055,307 104305,384
Maximum (Million VND) 461,193 567,523 863,980 1,060,089 1,362,348 1,798,383 1,913,027 1,971,745 2,625,883 4621,036 4074,251 10307,310 4815,793 6163,917
Minimum (Million VND) 8,796 29,626 153,418 61,008 242,473 215,737 264,586 61,724 44,342 141,446 58,131 87,863 56,133 56,000
Average (Million VND) 163,240 201,458 336,252 408,568 540,756 578,320 706,655 617,938 765,457 928,036 1181,955 1510,356 1334,211 1655,641
Standard Deviation 98,271 111,644 163,460 174,002 215,481 274,265 318,065 362,248 489,151 681,430 724,636 1406,175 871,686 1082,136
Coefficient of Variation 0.60 0.55 0.49 0.43 0.40 0.47 0.45 0.59 0.64 0,73 0,61 0,93 0,65 0,65
135
T ax R eform in V ietnam
The evolution of provincial revenues for the period 1997 to 2010 is shown in table 5.2 in
millions of VND. This table also shows the evolution of the three main components of:
own revenues or taxes assigned 100 percent to subnational governments, shared taxes, and
transfers. Total provincial revenues have increased steadily over the period quadrupling in
size. As illustrated in figure 5.1, a significant part of the growth in total revenues is explained
by the faster growth in transfers. However, with the beginning of the stability period in 2004,
revenues from shared taxes and, to a lesser extent, own taxes jumped quite significantly. The
jump in revenues from shared taxes was due to a large extent to the two additional taxes
(special consumption taxes and gasoline and oil fees) that were added to the list of sharable
taxes starting that year. In absolute size of collections, provinces can differ considerably; for
example in the distribution of revenues by province the difference between the maximum
and the minimum was twentyfold in 2010. Note also that even though this difference and the
coefficient variation decreased somewhat over time up to 200607, the tendency has been
again significantly upwards after that (figure 5.2).
Figure 5.1: Total Provincial Revenue, by Revenue Sources (Million VND)
300000000.0
250000000.0
200000000.0
150000000.0
100000000.0
50000000.0
0
1997 19981999 2000 200120022003 2004 2005 2006 20072008 2009 2010
1.5
0.5
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
136
T ax R eform and R evenue A ssignments in V ietnam
Table 5.3 shows the evolution of revenues (total and of the three main components) in
millions of VND per 1000 population, while the trends for the four categories of revenues
are illustrated in figure 5.3. Normalized by population, fiscal disparities have increased over
time in Vietnam up to recent times. For example, the differences between the richest and
the poorest province in revenues per capita jumped form fourfold in 1997 to fivefold in 2005
while the coefficient of variation also increased by 25 percent. However, in more recent years
(up to 2010) these trends have stabilized or decreased slightly.
As illustrated in figure 5.4, fiscal disparities in revenues per capita are created by the
distribution of revenues from shared taxes and also from own taxes. The distribution of
overall revenues shows smaller disparities than revenues from shared and own taxes because
of the compensating or equalizing effects of transfers. However, considerable differences
across provinces in their abilities to raise revenues remain in the system even after transfers.
Table 5.3: Provincial Revenue per 1,000 Population and Summary Statistics, by
Revenue Sources, 19972010
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010e
TOTAL REVENUE
Maximum 861 924 1,386 1,535 1,962 2,141 3,405 4,509 4,552 4277 7,556 6,932 5,695 7,083
(Million VND)
Minimum 228 240 286 329 392 446 576 540 814 760 1,026 1,088 1,394 1,647
(Million VND)
Average (Million 408 423 524 612 748 826 1,061 1,284 1,494 1859 2,258 2,866 2,698 3,281
VND)
Standard 155 168 220 254 335 398 543 669 720 751 1,036 1,319 1,032 1,311
Deviation
Coefficient of 0.38 0.40 0.42 0.41 0.45 0.48 0.51 0.52 0.48 0.40 0.46 0.46 0.38 0.40
Variation
TAXES 100%
ASSIGNED TO
SNG
Maximum 472 474 456 461 629 972 2,238 2,981 2,918 2,389 5,638 4,615 2,586 3,112
(Million VND)
Minimum 33 31 35 31 36 37 50 52 92 54 73 106 59 74
(Million VND)
Average (Million 143 135 142 147 163 198 282 412 430 432 499 585 509 594
VND)
Standard 89 83 83 94 110 170 306 406 395 367 724 645 438 549
Deviation
Coefficient of 0.62 0.61 0.58 0.64 0.67 0.85 1.08 0.99 0.92 0.85 1.45 1.10 0.86 0.92
Variation
SHARED
TAXES
Maximum 503 410 466 676 972 1,408 1,795 2,086 2,246 2,456 2,759 5,770 5,227 5,661
(Million VND)
Minimum 23 20 22 25 27 29 37 52 95 11 16 25 22 24
(Million VND)
Average (Million 131 125 114 144 165 185 246 388 475 485 567 774 804 987
VND)
Standard 91 82 83 127 155 204 263 351 416 492 524 902 828 1,001
Deviation
Coefficient of 0.69 0.66 0.73 0.88 0.94 1.10 1.07 0.90 0.88 1.02 0.92 1.16 1.03 1.01
Variation
137
T ax R eform in V ietnam
TRANSFER
Maximum 646 736 1,050 1,307 1,617 1,793 2,007 2,136 3,083 3,197 3,862 5,410 5,256 6,605
(Million VND)
Minimum 5 16 86 52 145 138 187 39 37 42 42 56 36 37
(Million VND)
Average (Million 134 163 268 321 419 442 533 483 589 944 1,192 1,507 1,385 1,700
VND)
Standard 148 165 210 252 307 337 404 504 693 755 898 1,281 1,137 1,393
Deviation
Coefficient of 1.10 1.01 0.79 0.78 0.73 0.76 0.76 1.04 1.18 0.80 0.75 0.85 0.82 0.82
Variation
Figure 5.3: Average of Provincial Revenue per 1,000 Population, by Revenue Sources
(Million VND)
3500
3000
2500
2000
1500
1000
500
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
1.6
1.4
1.2
0.8
0.6
0.4
0.2
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
TOTAL REVENUE SNG TAXES SHARE TAXES TRANSFER
138
T ax R eform and R evenue A ssignments in V ietnam
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010e
Revenues 65,352 70,612 78,489 90,749 103,773 121,716 175,908 170,375 202,683 241,117 341,057 430,549 454,766 559,170
Of Which:
Local budget 19,264 20,280 19,571 22,269 25,463 30,545 45,521 74,688 96,113
87,291 99,249 133,966 134,656 169,384
revenue
Share in total
state budget 29.5 28.7 24.9 24.5 24.5 25.1 25.9 43.8 47.4 36.2 29.1 31.1 29,6 30.3
revenue
Growth rate
(previous
year=100%)
State budget
4.8 8 11.2 15.6 14.4 17.3 44.5 -3.1 19.0 19.0 41.4 26.2 5.6 22.9
revenue
Local budget
8.5 5.3 -3.5 13.8 14.3 20 49.0 64.1 28.7 9.1 13.7 34.9 0.5 25.8
revenue
15
10
5
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
31
Of course, this figure excludes central government transfers to provincial governments.
139
T ax R eform in V ietnam
20%
10%
0%
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
140
T ax R eform and R evenue A ssignments in V ietnam
However, this revenue adequacy requirement does not offer a concrete guide for the reform
of revenue assignments; in fact, the same adequate levels of financing can be obtained from
many different tax assignment combinations or via intergovernmental transfers. But in the
case that all subnational government financing is from revenue sharing or other forms of
transfers from higher-level governments, there is a danger that subnational governments will
become spending agents of the center becoming less efficient,33 and that imposing a hard
budget constraint on subnational governments will become more difficult. 34
Besides enhancing accountability and providing revenue adequacy, there are several other
benefits from revenue assignments with tax autonomy. Subnational tax autonomy is the
best way, if not the only way, to address in a permanent way the difficult problem of vertical
imbalances, or mismatch of expenditure needs and revenue sources at different government
levels. Adequate revenue autonomy is also a key indicator of subnational governments
borrowing capacity and creditworthiness. There is also some evidence that countries with
more revenue autonomy at the subnational level are countries that over time have shown
greater macroeconomic stability.35
On the other hand, greater tax autonomy in revenue assignments can pose some additional
challenges. Depending on the geographical distribution of economic activity and tax bases,
greater subnational tax autonomy can lead to larger horizontal fiscal disparities across
subnational governments. Richer jurisdictions can have the ability to finance their expenditure
needs with relatively little effort while poorer communities may have to exert much greater
tax effort with their residents to provide for similar expenditure needs. However, this is not a
fatal consequence of subnational tax autonomy since these horizontal fiscal disparities can be
well addressed through the proper design of equalization grants.
With the realization that tax autonomy is paramount, we need to ask how much tax autonomy
is needed. Is it the case that subnational governments should finance themselves entirely
from autonomous tax sources? In reality, full own-financing by all subnational governments is
generally not feasible or even desirable. Instead, the generally accepted rule is that subnational
governments need to raise their own funds at the margin and operate with hard budget
constraints, which means that revenue sharing and grants should represent only inframarginal
33
A number of recent studies (for example, Ter-Minassian 1997; Ebel and Yilmaz 2002) suggest that
outcomes of decentralized spending depend on the form of financing used for these expenditures,
with a crucial aspect being the extent of control that local governments can exercise over the
sources of their revenues.
34
A hard budget constraint implies that those local governments given autonomy will be asked to
balance their budgets without recourse to any end-of-year assistance from the central government
and a clear understanding that there will be no-bailout at year-end or in case of debt default. See
Rodden et. al. (2003).
35
Traditionally, it has been thought that greater subnational revenue autonomy may compromise the
ability of the center to implement stabilization policies; in reality, the reverse seems to happen. It
could be that greater subnational revenue autonomy leads to more conservative budget policies
and lower deficits at all levels of government. See Martinez-Vazquez and McNab (2006).
141
T ax R eform in V ietnam
funding.36 Only the richest subnational governments should be close to financing their full
expenditure needs from their own revenue sources.37
Despite the well accepted consensus on the importance of subnational tax autonomy, in the
international practice generally we tend to observe low levels of tax autonomy. One reason
for this state of affairs, already mentioned above, is that central governments are reluctant to
devolve taxing powers for fear of competing with local governments for the same taxing base
and for fear of losing control. Ironically, many subnational governments are happy not to get
tax autonomy because quite simply they do not want to take on the responsibility of making
politically unpopular taxing decisions to meet their budget needs. Using intergovernmental
transfers as opposed to revenue autonomy appears to be a much easier path for all concerned.
To some degree, insufficient revenue autonomy can also be the result of the lack of
administrative capacity in some subnational governments. When low capacity is combined
with the desire to provide all subnational governments (regardless of size and capacity) with
the same autonomous taxing powers, low levels of tax autonomy can follow. This situation
(differing administrative capacities) poses a dilemma in decentralization design. A uniform
intergovernmental fiscal system under which all subnational governments must operate
has appeal. If all subnational governments have the same expenditure responsibilities and
revenue raising powers, management of the system and evaluation of its success are made
easier. Uniform treatment of all subnational governments also seems generally fairer. On
the other hand, a more effective route for effective decentralization may be the adoption of an
asymmetric tax assignment providing more tax autonomy to larger subnational governments
with more capacity, according to transparent objective criteria, and let the smaller ones grow
into this role over time.38
Although decentralized systems in some developed countries have high levels of tax autonomy,
in reality it is quite rare, especially among developing countries to find significant taxing
powers devolved to subnational governments at the onset of decentralization. Often, there
is considerable reluctance from central government to let go part of its authority and control
over taxes, which in turn is justified because of the need to facilitate attainment of proper
capacity at the subnational level. However, these stumbling blocks generally linger for many
years after the introduction of a decentralization program. With the passage of time a culture
of financial dependency takes hold with subnational governments becoming accustomed to
almost exclusively relying on central transfers for their financing needs.
Several things need to be done to get this right. First, there is a need to devise a sensible way
37
to measure the expenditure needs of subnational governments and to keep these measurements
reasonably updated. Next, there is a need to apply the golden rule for revenue assignment: tax
assignments should be sufficient to fund the expenditure needs (net of conditional grants) of the
wealthiest subnational governments. Sometimes, however, it may be advisable to break this rule
somewhat and to have even the wealthiest subnational governments partly financed by central
transfers. This may be because of vertical externalities in the use of tax bases, economies of scale in
the administration of some taxes, the need to maintain the integrity (harmonized nature) of some
taxes, and other considerations in tax administration, which are discussed below.
See Bird and Ebel (2007) on the possibilities and problems with asymmetric fiscal federalism.
38
142
T ax R eform and R evenue A ssignments in V ietnam
See Bird (2000); Bird and Ebel 2007; Boadway (1997); McLure (1998 and 2000a); and Musgrave
39
(1983).
The international experience shows that providing subnational governments with freedom to
40
select their own taxes (the open list approach) can easily backfire when subnational governments
introduce highly inefficient (distortionary) forms of taxation. A recent example is provided by
Indonesia, which adopted an open-list approach in the 2001 decentralization reform. See Alm,
Martinez-Vazquez, and Indrawati (2004). In the international experience, where subnational
governments are given more constitutional discretion, as in the case of some federal systems, open
lists with some general restrictions are common. Closed lists are used more frequently in unitary
systems of government.
See, for example, the discussion in Boadway, Marchand, and Vigneault (1998).
41
Sometimes the country Constitution, even in the case of some federal countries, is used to clearly
42
delineate what taxes can be used at different levels of government; for example, this is the case in
India, Pakistan or Switzerland.
143
T ax R eform in V ietnam
to use significant (either in size or elastic over time) tax bases, thus drastically reducing any
meaningful possibility of subnational tax autonomy. The imposition of exclusive tax bases
can also lead to cumbersome tax structures. All things considered, a closed list allowing for
the cohabitation of tax bases by different levels of government and using intergovernmental
transfers to correct for vertical externalities is the preferred approach.
The second dimension of tax autonomy relates to which level of government can legislate
over the structure of the tax bases and which level has discretion to set the tax rates. Of the
two types of autonomy for structuring subnational taxes, autonomy to define the tax base is
generally less desirable than autonomy to set tax rates.43 Variations in the definition of the tax
base, either through special exclusions from tax, deductions from the tax base, and credits
against the tax liabilities can more easily lead to complexity and lack of harmonization across
jurisdictions. The most important unwanted consequence of the lack of harmonization and
complexity is the higher tax administration cost for all the jurisdictions involved and higher
compliance costs for taxpayers who have tax obligations in several jurisdictions.
Autonomy in the form of setting tax rates generally tends to be more desirable because it is
simpler to deal with across jurisdictions for both tax administrators and taxpayers. Focusing
on autonomy over tax rate setting has the additional advantage of generating political
accountability because tax rates are the most visible element of a tax. Tax rate setting autonomy
also may be preferred because it has a more direct impact on revenues and spending ability
of subnational jurisdictions and because households and businesses have an easier time
comparing their taxes to the services they receive.
The third and last dimension of revenue autonomy refers to which level of government is put in
charge of administering the various taxes. Centralized tax administration even of subnational
taxes is likely to be more efficient because of economies of scale. However, administration
by subnational governments of their own taxes is likely to enhance accountability at the
subnational level if taxpayers are more aware of subnational taxes under this arrangement.
This efficiency-accountability tradeoff is likely to differ for different taxes. For example,
the efficiency gains from the centralized administration of subnational piggyback personal
income taxes may dominate any increase in accountability generated by decentralized
administration of those taxes. In contrast, there may be no significant efficiency gains in
the centralized administration of subnational property taxes by comparison to the losses in
local accountability implied by the centralization of the administration of these taxes. The
administration of subnational taxes or even shared taxes by the central administration can
present a problem with low incentives even for shared taxes when the central administrations
The ability to change either base or rate opens up the possibility of fiscal competition among
43
subnational governments. See, for example, Wilson (1999). Interjurisdictional fiscal competition
can have both good aspects, such as offering choices to taxpayers and keeping public officials more
accountable, and also bad aspects, such as a run to the bottom with subnational governments
competing with ever lower taxes to the point that they may lack sufficient revenues to provide
basic services. The ability to change tax bases or tax rates may also give rise to horizontal fiscal
externalities, whereby the policies of one jurisdiction (for example, raising tax rates) can have an
effect on the tax bases of other jurisdictions (raising their tax bases related to mobile taxpayers).
Intergovernmental grants and other policies can be implemented by the central government to
correct horizontal fiscal externalities
144
T ax R eform and R evenue A ssignments in V ietnam
share in the revenues is relatively small. What this means is that when cost advantages make
it desirable to centralize the administration, there is a need for setting incentive compatible
arrangements between levels of government for the collection of taxes.44
Finally, we need to mention the issue of whether certain forms of tax revenue sharing on a
derivation basis can contribute to the revenue autonomy of subnational governments. The
more generally accepted view is that tax sharing is not a form of revenue assignment because
subnational governments do not have a direct role in the structure and administration of the
tax; in this view, revenue sharing should be considered just another form of transfer. In the
minority view, shared taxes may be considered a form of tax assignment when the shared rates
are stable over a period of several years and especially when the subnational authorities can
influence the level of administration and affect the size of the tax bases. For these reasons, it
is customary in many transitional countries, especially those in the former Soviet Union, to
consider shared taxes as part of the own revenues of subnational governments.45
145
T ax R eform in V ietnam
arrangement provides provincial governments with budget flexibility and the ability to adapt
to the particular circumstances in the province. It may allow for higher levels of equalization
than would be possible were there a concretely specified revenue assignment. However, the
lack of concrete revenue assignments also imposes costs on lower-tier governments since it
can significantly limit revenue certainty and predictability.46 The consequences of the changes
in revenues assignments for local governments introduced in 2002 have not been studied in
any depth before 2007 although the casual evidence hinted at the existence of responsible
behavior by most provincial governments but also to the presence of some abuses. A study
of the Finance Academy (2007) reveals some issues related to real practice of expenditure
and revenue assignments by provinces after the 2002 revised Budget Law came into effect. In
the case of expenditure assignment the Finance Academy study found that some provinces
actually changed their expenditure assignments between stability periods. For example,
Ha Tinh province during the stability period 200406 assigned all education expenditures
at the provincial level but in the stability period 200710 all education expenditures have
been assigned to the district level. The study also highlighted that the shifting in expenditure
assignment has taken place more often at the commune level. In the case of revenue
assignments, the Finance Academy study also found that provinces tend to assign smaller
taxes to communes while keeping for themselves larger revenue. The study also found
instances of shifting in assignments during the stability period 200406; for example, that
behavior was observed for Bac Can province. Of course, some of these changes may also be
an indication of the drawback of using the policy of stability periods, especially at a time
when Vietnam is experiencing very rapid economic growth and frequent policy changes in
other areas (such as, raising minimum salary every year, higher expenditure demands due
to inflation, and so on). Districts and communes with different economic conditions end
up with significant disparities among them. Because of the limited revenue autonomy that
provinces currently enjoy, it is really hard for poor provinces to mobilize additional revenues
to meet their needs in between stability periods.47 On the other hand, surplus provinces are
generally in a better position to meet their expenditure needs because they are able to keep
larger amount of revenue collected during economic expansions (as the amount of revenue
they have to surrender to the center has been fixed at the beginning of the stability period for
the whole duration of this period). Of course, these are the provinces that will experience a
reduction in their tax sharing rates in the next round of stability periods. The findings in the
Finance Academy (2007) study would seem to point out a direction of reform to introduce
more stability and transparency in the assignment of expenditure responsibilities and
revenue sources at the subprovincial level. However, this will not be a cure for all problems.
For example, the 2007 CFAA (Country Financial Accountability Assessment) found that
in the case of some provinces, such as Ha Nam province, have experienced difficulties in
redistributing resources from high-revenue communes with real estate booms to poorer ones
This negative impact on certainty and predictability is partially offset by the practice of stability
46
periods.
Recently, the government of Vietnam abolished inflation and economic growth correction
47
allowances during the stability periods. However, there seems to be agreement within the Ministry
of Finance that these allowances should be re-introduced in the next round of revisions to the
Budget Law.
146
T ax R eform and R evenue A ssignments in V ietnam
because the current Budget Law mandates specific minimum rates the center has to assign to
the communes for the land and housing tax.
As pointed out above, in Vietnam the administration and collection of all practically all taxes
is centralized. A potential problem with centralized tax collections is the lack of incentives
that central government bureaucrats (within the tax administration) may have to mobilize and
collect subnational revenues. This problem is mitigated in Vietnam because there is de facto
dual subordination of tax administrators to the central administration and the subnational
authorities; this means that provincial and district officials can have a recognizable influence
on the decisions and activities of tax administrators. For example, it is not unusual for the
provincial authorities to provide bonuses to the tax administrators with a better performance
in collections. Provincial authorities also usually provide office facilities, housing and other
amenities to tax inspectors and other officials of the central government tax administration
agency. The de facto dual subordination and the role played by subnational incentives for
tax collection has been formalized in practice by letting subnational authorities retain a
share of the collections that are above the targeted amount for tax collection in the central
government plans.48 On the other hand, in some other countries, notably China (before the
1994 reform that created separate tax administrations for central and provincial taxes) but
also other transitional countries like Russia and other former Soviet Republics), de facto
dual subordination of central tax administrators was a source of problems as the subnational
authorities abused their position to push provincial interests over those of the central
government. This means that the practice of de facto dual subordination of central government
tax administrators to subnational officials is unlikely to be a good solution in the longer run in
order to address the issue of incentives, especially if after the reform of revenue assignments
in Vietnam subnational governments are provided with significant tax autonomy.
147
T ax R eform in V ietnam
Beyond the guidance provided by Musgraves governmental roles, there are some
characteristics of taxes that are commonly acknowledged as desirable regardless of whether
these taxes are to be assigned at the central or subnational levels. These include:
i. Revenue buoyancy, meaning that overall, revenues should change roughly in proportion
to the economic base;
ii. Equity, meaning that good revenue sources are fair or equitable in the sense of horizontal
equity under which taxpayers in similar circumstances should be treated similarly and
vertical equity under which taxpayers with different incomes should pay according to
their ability to pay;
iii. Efficiency, meaning that the tax should have relatively low administration and compliance
costs and create a minimum of distortion in the economy; and
iv. Political acceptance, meaning that taxes need to be sensitive to the historical and
institutional framework in a country.
There are in addition, several other features that are desirable for taxes to be assigned at the
subnational level.49 First, the benefit principle which relates revenue sources to the benefits
being provided should be implemented to the largest extent possible. Fees paid for water or
sanitation services are clear applications of the benefit principle. Second, subnational revenue
sources should have a tax base that is relatively evenly distributed across jurisdictions. This
helps to minimize fiscal disparities among subnational governments and reduces the burden
put on equalization grants to allow a more uniform quantity and quality of services. Third,
subnational tax sources should have immobile bases to minimize the likelihood and effects
of tax competition among jurisdictions in a race to the bottom. 50 Fourth, subnational taxes
should be geographically neutral in the sense that they do not interfere with domestic or
international commerce, they do not distort the location of economic activity across the national
territory, and they are not exported (so that the taxes levied by a subnational government are
not borne primarily by residents in other jurisdictions). Fifth, subnational taxes should be
implemented without undue costs of compliance and administration due the existence of
multiple jurisdictions. Sixth, subnational taxes should exhibit generally stable tax bases over
the economic business cycle; this means that revenue sources that are highly sensitive to general
economic conditions (for example, profit taxes) should be assigned to the central government,
which has greater ability to deal with cyclical fluctuations in revenues through borrowing
and other means. Seventh, subnational taxes should be highly visible so that tax burdens
are clearly perceived by local residents.51 And eight, subnational tax assignments need to be
stable over time; a typical problem of transitional countries has been unstable assignments,
with the assignments not being established in permanent laws but instead decided in annual
budgets. Ad hoc assignments decided on an annual basis may also result in a lack of uniformity,
unnecessary complexity, and perverse incentives toward revenue mobilization.
However, as we mentioned above, not all forms of tax competition are undesirable; moderate
50
levels of tax competition can provide for more accountability among politicians and bureaucrats.
Of course, subnational governments are likely to think quite differently about this and they would
51
148
T ax R eform and R evenue A ssignments in V ietnam
149
T ax R eform in V ietnam
assign the property tax to local governments as opposed to intermediate level (provincial or
regional) governments.52
Several features make property taxes especially attractive as a subnational tax. Most
important, the property tax is a visible tax and thus conducive to political accountability; in
addition the tax, for the most part, falls on an unmovable base. The more homogeneous are
the distribution of property values and the distribution of the population in a jurisdiction,
the closer the property tax comes to being a benefit tax.53 However, depending on how the
property tax is structured, it can easily move away from the benefit link; for example, when
the property tax burden falls just on a few classes of property, such as nonresidential property,
there is little to be said about the property tax as a benefit tax.
Other advantages of property taxes are their revenue potential and stability. Note also that
from a vertical equity viewpoint the property tax can be progressive, especially in developing
countries, and therefore can increase the overall vertical equity of the tax system. However, in
practice, the property tax can be made regressive by exemption policies that target wealthier
households.54 The property tax also has the desirable feature that much of the tax burden is
quite likely borne by residents in the jurisdiction where the services financed by property taxes
are provided. The property tax also imposes a relatively low compliance cost on taxpayers
because taxpayer intervention in terms of the determination of tax liability is minimal, except
in the case of appeals. Typically the property tax poses no significant problem of tax base
competition with the central government, basically because this is not a tax that central
governments tend to covet. 55 Finally, a part of property tax might be thought of as a charge
for land that can lead to significant improvements in the quality of land use.
The main drawback of the property tax is that, perhaps due to its visibility, it is likely unpopular
with taxpayers and, as a result, also with public officials. Other drawbacks include the fact
that it can lead to liquidity problems for homeowners with valuable real estate assets but low
incomes.56 In addition, the property tax administration requires costly revaluation of property
on a regular basis, and it is difficult to enforce, because the confiscation of property may be
However, despite the wide agreement on the advantages of the property tax as a subnational tax,
52
subnational governments in developing and transitional countries make relatively little use of the
property tax. On average, transitional and developing countries raise property tax revenues that
are equivalent to only about 0.6 percent of GDP. See Bahl and Martinez-Vazquez (2007) for an
investigation of this puzzle.
The balance between the services received by property owners and the property taxes they pay
53
on their real estate typically can be capitalized into property values. That is, property taxes do not
have to reduce the market value of dwellings if the general perception is that the quality of services
provided by the local government is good.
See Bahl and Linn (1992) and Sennoga, Sjoquist, and Wallace (2007).
54
Of course, low interest may also reflect the perception that the property tax is complex and has
55
low revenue potential vis--vis its associated political costs, although there are exceptions (for
example, China, Indonesia, and Jamaica.).
Being house rich and income poor can be a problem for elderly people. Some countries use
56
150
T ax R eform and R evenue A ssignments in V ietnam
considered too extreme because of the political fallout. Finally, the property tax lacks revenue
elasticity, meaning that the tax typically exhibits little automatic revenue growth.
In practice there are several forms of the property tax. For example, some countries separate
the taxation of land and improvements, or structures, and a few others tax only land values
or rents. Although a tax on land tends to be more efficient, it also has less revenue potential
and it is generally more difficult to administer properly, for example in terms of valuation or
assessment of properties. There is another type of property tax in the form of betterment
levies or lump-sum payments exacted up front by subnational governments from land and
housing developers and also from homeowners as a charge for public service improvements,
such as road paving, drain infrastructure, sidewalks, street lights etc, which all have a
visible benefit on property values. Betterment levies can be useful in providing subnational
governments with liquidity to invest in needed infrastructure; they also have the advantage
of being more directly contractual than property taxes and therefore reinforcing the benefit
principle feature in subnational government financing. There are different modalities for
the administration of the tax, including centralized or central oversight over cadastres and
reevaluation processes, which can make this type of tax even feasible in developing countries.
Note that tax autonomy is largely preserved as long as subnational authorities are given some
discretion over rate setting.57
Vehicle and Transportation Taxes:
These are generally attractive taxes at the subnational level because of a strong link between
the ownership of vehicles on the one hand, and the use of local services and infrastructure
(particularly roads) on the other hand. In addition, subnational taxes and charges on vehicles
can counteract the negative externalities associated with local traffic congestion and air
pollution in the local area. Vehicle and transportation taxes also tend to have elastic revenues.
It is perhaps for this reason that the central governments in some developing countries,
wrongly, tend to keep them fully centralized.58 An annual tax on the registration of motor
vehicles is often assigned at the subnational level. However, excise taxes levied one time on
the purchase of new motor vehicles and taxes on gasoline and other motor fuels are more
frequently assigned at the central level. It is also quite common that the revenues from taxes
on motor fuels are earmarked to a national road fund. In some cases, the revenues in the road
fund are shared between the central and subnational governments according to a formula
with the funds dedicated to road maintenance and construction.
Natural Resource Taxes (when resources are evenly distributed):
There is at least a partial link between taxes on natural resource extraction and the benefit
principle at the local level. Natural resource taxes can be justified at the local level to the extent
that extraction activities use local infrastructure (for example, roads needed to transport heavy
machinery and mined resources), place stress on other local infrastructure (temporary worker
camps, hospital facilities required to treat injuries incurred by those working in this industry,
57
For international experience with the property tax see Bird and Slack (2004) and Bahl and
Martinez-Vazquez (2007).
58
But there are some transportation taxes, such as in the case of air travel, which is rightly allocated
at the central level, since air traffic control and other similar services, should be centrally provided.
151
T ax R eform in V ietnam
and so on), and depending on the type of extractionmay pollute the environment or cause
other negative externalities increasing health costs of local residents. There has been growing
interest in the fiscal decentralization literature in the pros and cons of the assignment of natural
resource revenues to subnational governments.59 Notwithstanding the arguments for some
form of local taxation of natural resources, there are two major arguments against local taxation
of natural resources. First, in the case of geographically concentrated natural resources, local
taxation could cause extensive horizontal fiscal imbalances (for example, the recent cases of
Indonesia, Nigeria, and Russia). These fiscal disparities can lead to inefficient population
migration and location of business. Second, given the high volatility of world commodity
prices, local taxation of natural resources would not constitute a stable source of revenue.
Therefore, some balance must be reached, especially in the case of the uneven geographical
distribution of resources, between first, centralized taxation of natural resources to address
disparities and avoid or correct for negative economic externalities, and second, sharing some
of the revenues with subnational governments to compensate for the environmental damage
of the extraction process and so on.
Local Business Taxes:
Certain forms of business taxes or business license fees are justified at the subnational level
as an indirect but administratively easier way to tax income of business owners (especially
nonwage incomes), and as a benefit tax for the services and infrastructure provided by
subnational governments.
Where it is not feasible to recoup costs of local government services through user charges,
some form of broad-based levy on general business activity is warranted. To avoid economic
distortions, ideally the tax would apply equally to labor (payroll) and capital (assets) used by
businesses.60 Where this form of business taxation at the subnational level is not feasible it
may be possible to use business permit charges which may vary by type, size, or location of
the business.
Excise Taxes:
Subject to the area size, cross-border trade and smuggling limitations, excise taxes have good
potential at the subnational level, especially when they are designed as piggyback taxes on
the national taxes. An advantage is that excises tend to be more politically acceptable, can
be easily administered in coordination with national wholesalers as withholding agents, and
allow for rates differentiated by province.61 Moreover, the benefit principle accords well with
See, for example, McLure (1996) and Bahl and Tumennasan (2004).
59
An example of this is a business value tax (BVT) as discussed in Bird (2003) and calculated by
60
adding payroll, interest, rents, and net profits on the basis of annual accounts. An example of a
BVT is Italys regional business tax (known as the IRAP) prior to the elimination of payroll from
the tax base in 2003. The IRAP is discussed in Keen (2003).
For example, some OECD countries allow subnational government surcharges on excises. In
61
the Netherlands, provinces impose a surcharge on the motor vehicle tax levied by the central
government. Provinces are free to set the rate of the surcharge, subject to a ceiling imposed by the
central government.
152
T ax R eform and R evenue A ssignments in V ietnam
62
Generally speaking a local income tax should be levied at the place of residence because it is
there where most taxpayers consume subnational government services. However, because of
administrative convenience, subnational piggyback taxes are often withheld at source at the place
of work by employees. However, it is quite feasible to distribute the funds according to where
workers reside.
63
Other forms of tax autonomy that are less desirable but nevertheless practiced include the ability
to modify the base of the tax by providing more or less deductions and exemptions.
153
T ax R eform in V ietnam
154
T ax R eform and R evenue A ssignments in V ietnam
VAT is likely to take place in different subnational jurisdictions, which generally will imply
an arbitrary apportionment of VAT revenues across those jurisdictions.64 This also makes
it problematic to share VAT revenues with subnational governments on a derivation basis.
The problem here is, as in the case of the profit tax, is that the tax revenues get arbitrarily
allocated to the subnational government where the headquarters of the company are located
and not to where the economic activity actually takes place. However, in order to address that
arbitrariness, there are countries that use a formula to share VAT revenues with subnational
units.65
64
Revenue sharing on a derivation basis for the VAT also means that, as in the case for the sharing
of corporate income taxes, the tax tends to be paid according to the place of registration or the
location of the headquarters of business firms.
65
For example, the VAT can be shared on the basis of population (as in Belarus and Germany), or
on the basis of the regional shares in aggregate consumption (as in Canadas Maritime Provinces,
Japan, or Spain). In the case of Canadas Harmonized Sales Tax for the Maritime Provinces, all
three provinces have a uniform rate that piggybacks on the federal VAT.
155
T ax R eform in V ietnam
Good choices for enhancing revenue autonomy at the subnational level in Vietnam include:
Selectively making better use of some fees and charges for excludable services under
the benefit principle at the provincial and local; however, some exclusions may be used
for poor households or in areas with high incidence of poor households and where
equalization grants should be providing additional needed revenues.
The assignment of the real estate property tax to local governments with discretion to
set the tax rate between maximum and minimum values set in the national law. While
the assessment of property values could be left to the national tax administration, other
administrative functions (such as collections) could be decentralized asymmetrically
depending on the administrative capacity of the local government. Selected local
governments could also be allowed to charge betterment levies for the construction of
new housing and the urbanization of additional land for construction.66
The assignment of a motor vehicle tax with variable rates set in the national legislation to
the provincial governments. This would be an annual tax associated with the renewal of
the annual license to operate the motor vehicles.
The assignment of a piggyback flat rate personal income tax to provincial governments.
This tax would allow provincial governments to introduce a flat rate set between
a minimum and a maximum rate (for example, 1 percent and 4 percent) legislated in
national law. The tax would be charged on the same tax base as the national income tax
and paid on a residence basis.
The Government should also explore the feasibility of using piggyback taxes or surcharges
on the special consumption (excise) taxes on alcohol and tobacco. This would allow
provincial governments to add a surcharge to the national excise taxes and they would
be withheld by producers or distributors according to the destination (province) of the
merchandise shipments/sales to retailers.
A second thrust of the reform in revenue assignments would be the improvement of the
apportionment of shared taxes between central government and provincial governments. In
the case of the VAT, tax revenues would not be shared any longer on a derivation basis but
rather according to a formula. This formula could be based on either an equal per capita basis,
or in proportion to estimates of shares in final consumption for each province. In the case of
the profit tax, tax revenues would not be shared any longer on a derivation basis but rather
on the basis of a formula. This formula would be based on the geographical distribution of
the enterprise payroll, and to the extent possible also on the geographical distribution of the
enterprise assets and sales. The sharing of personal income tax revenues with subnational
governments would be according to the place of residence of workers, where most local
services get consumed, as opposed to the place of works as is now the case.
A third thrust of the reform in revenue assignments would be to introduce an explicit list
of tax and other revenue sources for local governments (districts and communes). The
current flexible unwritten approach to revenue assignments for districts and communes has
Betterment levies, like one-time or multiyear charges, are linked to infrastructure improvements
66
156
T ax R eform and R evenue A ssignments in V ietnam
advantages but its costs in terms of uncertainty and unpredictability for local governments
outweigh those advantages.
In terms of tax administration, we recommend that, for the time being, tax collections for
subnational taxes remain centralized as there are advantages of economies of scale and
information in the centralized administration of most of the taxes that have been suggested to
provide subnational governments with more tax autonomy and enhanced tax accountability.
However, at some point in the near future there should be an examination of how to introduce
incentive compatible contracts between the central tax administration and the subnational
governments for the collection of subnational taxes; this can be in the form of benchmark
payments linked to the performance of collections. In the longer term, for certain taxes, such
as the real estate property tax, some of the administration functions including registration
and collections could be devolved to subnational tax administrations. This devolution would
be phased out cautiously and in an asymmetric way taking into account the administrative
capacities of subnational governments.
157
Integration and Government
Revenues: Assessing the VI
Implications of Vietnams
Accession to the WTO
By Robert Warner
1. Summary
This paper presents an analysis of some of the implications for future tax policy of Vietnams
tariff reduction and binding commitments, entered into as one of the agreements associated
with accession to the World Trade Organization.
The main observations arising from the analysis are:
Implementing Vietnams tariff binding commitments will reduce the average and
dispersion of tariffs compared to the current rates, with a consequent improvement in
the incentives for investment resource utilization. But compared to some other countries,
Vietnam does not seem to have taken the opportunity of accession to lock in a tariff
regime more suitable for an open trading economy. Some bound rates are higher than
current applied rates, and over 50 percent of bindings are at rates of 20 percent or higher.
Vietnams phasing in period is quite long (12 years in total before all bindings are at their
final rates. However, over 99 percent of final bound rates will be in force after 7 years.
The tariff cuts that will follow from implementation of bindings look set to reduce tariff
revenues. But a large part of this revenue loss could be made up by reducing concessions
and exemptions and improving the efficiency of customs administration. Shortfalls
could also be made up by adoption of a different approach to taxing petroleum products.
More importantly, however, any assessment of the revenue effects needs to be carried
out in a way that takes account of the broader range of changes affecting the revenue
and expenditure system: growth should expand the base for some revenue sources, for
example, and reductions in subsidies will reduce demand for revenue. Some revenue
functions could be shifted from the tariff on the special consumption tax, and improving
the structure and administration of the VAT could generate significant yields.
The pattern of bindings seems to reflect a fairly typical trade negotiators approach to
tariff setting: most of the big cuts implied by the bindings seem to be in areas where the
tariff is largely redundant where Vietnam is a successful exporter or where import
competition is limited.
The cuts will reduce effective protection across the traded goods element of the economy:
but pockets of high protection will be allowed to persist if Vietnam maintains rates at
bound levels.
159
T ax R eform in V ietnam
The findings also suggest that it would be useful to look harder at ways of analyzing the impact
of changes in the structure and administration of other taxes and revenue instruments. The
analysis of the revenue impacts of tariff bindings presented in this paper is fairly rudimentary,
and has been constrained by lack of access to data that should inform a thorough assessment
of the revenue effects of large policy changes. It would be useful to try to develop a more
robust analytical framework for revenue analysis.
2. Introduction
Vietnams accession to the WTO will impact on revenues in a number of ways: there will
be first round effects as a result of implementation of the various commitments, and later
effects as the response of producers, traders, investors and consumers to the changes that
implementing the commitments will bring.
The first round effects will include:
an impact on the rate structure of some taxes (eg customs duties, some of which will have
to fall as a result of tariff bindings);
a consequent impact on the valuation base for other taxes on imports whose tariff rates
change (for example, the valuation base for VAT on imports will change as import duties
change);
changes in the way in which the value for duty of imports will be determined (application
of the GATT Valuation agreement);
other changes in administration of trade taxes and in charges levied on imports;
elimination of some concessions on import duties resulting from implementation of the
TRIMs agreement.
The second round effects will include the following.
A change in the trajectory of imports over time as traders respond to price reductions
consequent on tariff changes and other charges. This might in turn have some impact on
the exchange rate, affecting the value for duty.
A change in the transaction base for all other taxes if the economy grows more rapidly as
a result of implementation of the agreement.
This will depend to some degree on which parts of the economy respond most vigorously.
By no means all transactions and incomes are taxed under the current system of direct and
indirect taxes: small informal enterprises may escape some parts of the tax net, and significant
elements of consumption are exempted from the VAT.
Nontax revenues may also be affected: land fees and charges, for example may be affected by
changes in investment, and changes in the zoning of land and its valuation of land taxes.
This paper presents an assessment of some of these effects: primarily the immediate impact on
tariff revenues of the implementation of tariff bindings. It updates some modeling presented
in a paper prepared for the ADB (CIE 2007), and draws on earlier analysis carried out for the
World Bank during preparation of its support for Customs reform (CIE 2004).
160
I ntegration and G overnment R evenues
Import Tariffs
Vietnams import tariff has 3 schedules: normal, preferential and CEPT. There are now also
preferential rates applying to goods from countries with which Vietnam has a preferential
trade agreement, bilaterally or as a member of ASEAN.
The preferential schedule broadly corresponds to an MFN schedule, since the normal schedule
applies to imports from countries with which Vietnam does not have a trade agreement.
Table 6.2 summarizes the MFN tariff schedule for 2005 through to 2009. The data for 2007
and beyond includes the effects of changes in applied rates necessitated by implementation of
the tariff binding commitments made by Vietnam upon accession into the WTO. As the table
shows, the simple average of tariff rates has fallen from 18.4 percent to 11.3 percent.
Table 6.1: Composition of Revenues
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
In percent of total revenue
Tax revenue 73.1 74.1 77.5 72.2 76.6 79.6 82.1 84.7 84.8 86.0
Corporate 24.8 24.1 31.0 29.8 33.0 38.2 35.1 29.9 19.4 25.7
income tax
Individual 2.0 1.9 2.0 1.8 1.9 2.0 2.2 3 3.1 4.7
income tax
Capital user 1.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
chargesa
Land and 0.3 0.2 0.3 0.2 0.2 0.2 0.2 0.3 0.2 0.3
housing tax
Licence Tax 0.4 0.3 0.5 0.4 0.3 0.3 0.3 0.4 0.3 0.2
Tax on transfer 1.2 1.1 1.2 1.4 1.3 1.3 1.3 1.7 2.1 2.3
of properties
Tax on land use 0.3 0.2 0.3 0.3 0.5 0.5 0.4 0.7 0.1
rights
Value added tax 18.6 21.3 21.6 20.3 21.1 20.7 28.0 29.2 32.1 28.4
(VAT)
161
T ax R eform in V ietnam
The CEPT schedule applies to imports from ASEAN countries of goods on Vietnams inclusion
list. For those goods not on the inclusion list (for example sensitive agricultural products
which have a longer time period to be brought under the CEPT umbrella), the MFN rate
162
I ntegration and G overnment R evenues
applies. Table 6.3 summarises the CEPT schedule from 2005 to 2013. From 2006 onwards,
all products except those on the sensitive list are to be at rates between 0 and 5 percent.
Note that for goods from ASEAN countries to be eligible for the CEPT rate, they must meet
certain rules of origin requirements. Some imports (for example from entrept ports like
Singapore) will be subject to the MFN rate, because they do not meet the requirements under
the rules of origin framework specified for the agreement. There does not seem to be any
readily available information on the proportion of imports originating in ASEAN countries
to which the CEPT rates are applied.
Table 6.3: Summary of Vietnams CEPT tariff schedule
Tariff Bracket 2005 2006 2007 2008 2009 2010 2011 2012 2013
0 31.9 53.0 52.1 56.1 56.5 57.0 57.0 57.0 57.1
>05 50.8 46.5 45.7 42.8 42.4 41.9 42.0 42.1 42.4
>510 13.2 0.0 0.1 0.0 0.2 0.1 0.1 0.3 0.0
>1015 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
>1520 3.7 0.2 0.1 0.1 0.0 0.1 0.3 0.0 0.0
>20 0.2 0.3 2.0 0.9 0.9 0.8 0.6 0.6 0.6
Mean 4.5 2.5 2.5 2.8 2.7 2.7 2.6 2.5 2.4
CVa 108.4 144.1 133.7 255.8 256.2 258.7 240.5 230.2 205.8
Source: ASEAN Secretariat 2011.
Note: a. Coefficient of Variation (Standard Deviation as a Percentage of the Mean).
In recent years, the share of Vietnams imports coming from ASEAN has been declining, from
about 30 percent in 1998 to around 25 percent in 2006 and to 19.3 in 2010 (table 6.4.) Thus
at least 80 percent of Vietnams imports are now probably subject to MFN rates. (Vietnam has
preferential trade agreements with other partners, but in some cases, such as the agreement
with the United States, the preferential rates have largely been matched by the post WTO
accession MFN rate.)
Table 6.4: Imports by Region of Origin: ASEAN and the Rest of the World, 19952005
ASEAN Rest of World Total
US$m % US$m % US$m %
1995 2270.1 27.8 5885.3 72.2 8155.4 100.0
1996 2905.5 26.1 8238.1 73.9 11143.6 100.0
1997 3220.5 27.8 8371.8 72.2 11592.3 100.0
1998 3344.4 29.1 8155.2 70.9 11499.6 100.0
1999 3290.9 28.0 8451.2 72.0 11742.1 100.0
2000 4449.0 28.5 11187.5 71.5 15636.5 100.0
2001 4172.3 25.7 12045.7 74.3 16218.0 100.0
2002 4769.2 24.2 14976.4 75.8 19745.6 100.0
2003 5949.3 23.6 19306.5 76.4 25255.8 100.0
2004 7768.5 24.3 24200.3 75.7 31968.8 100.0
2005 9326.3 25.4 27434.8 74.6 36761.1 100.0
2006 12546.6 27.9 32344.5 72.1 44891.1 100.0
163
T ax R eform in V ietnam
There is an extensive set of exemptions from import duties provided for in current legislation.
Box 6.1 presents a summary of these exemptions from the most recent IMF summary of the
tax system and from information available from the Ministry of Finance.
164
I ntegration and G overnment R evenues
VAT
Vietnams VAT, which replaced a multirate turnover tax in 1999, was restructured in 2004 and
again in 2009, with the number of rates reduced from 4 to 3 (a previous top rate of 20 percent
was eliminated). Table 6.5 summarizes the structure of the VAT, as it applies to goods. A range
of commodities are exempted from the tax, including unprocessed products of agriculture
and fisheries. In addition, a set of commodities are exempted if imported for a particular use
(for example, machinery, equipment and means of transportation that are imported). Further,
imports funded by nonhumanitarian or nonrefundable aid are also exempt from the tax.
The VAT is levied on the duty paid price of imports (using the value that is used for
determination of the duty payable), so that changes in import duties impact on the amount of
VAT collected on imports, as do changes in the basis for determining value for duty.
An important feature of the VAT is that the system provides for either a credit or a subtraction
system to be used in for determining tax payments. The subtraction system cannot readily be
applied with a multiple rate tax, and the coexistence of the two systems can turn the VAT into
a variable turnover tax. (Most enterprises, however, use the credit method.)
Table 6.5: Application of VAT to Goods since 2009
Treatment / Commodity
Exempted
Unprocessed or semiprocessed products of cultivation, husbandry, aquaculture or seafood
Products which are animal breeds and plant varieties
Salt products
Irrigation and drainage, soil ploughing and harrowing; dredging of intrafield canals and ditches
for agricultural production
Certain specialised machinery, equipment or means of transportation that cannot be produced
locally for direct use in scientific research and technological development activities; for
prospecting, exploring and developing oil and gas fields
Leased airplanes, drilling rigs and vessels
Transfers of land use right, state-owned dwellings sold by the state to current tenants
Credit services and investment fund services, life and some other forms of insurance
165
T ax R eform in V ietnam
The special consumption tax is effectively an excise tax which in principal applies equally to
imports and locally produced goods, and in 2009 the tax was replaced by an excise tax. Goods
directly exported by production or processing enterprises are not subject to the tax, nor are
imports funded by humanitarian or nonrefundable aid, or goods imported by units of the
armed forces or donated to political and social organizations. The base for the tax when levied
166
I ntegration and G overnment R evenues
on imports is the value for duty plus the import duty paid. (The tax is also levied on certain
services, including dancing halls, massage parlors, casinos, lotteries and betting activities, and
golf course memberships and services). Table 6.6 summarizes the current coverage and rate
structure of the tax, and table 6.7 summarizes the excise rate structure.
Table 6.6: Special Consumption Tax on Goods (as at August 2007)
167
T ax R eform in V ietnam
168
I ntegration and G overnment R evenues
169
T ax R eform in V ietnam
170
I ntegration and G overnment R evenues
One problem is that notification of payment only goes to the port of entry. Customs Headquarters
does not receive notification of actual payments made, so is not able to implement effective
management control over the payment system and effect reconciliation of payments against
payment notices. Presumably, this also means that traders who have not paid accounts may be
able to bring shipments through other ports of entry: customs information systems would appear
to be unable to inform all customs houses of delinquency in payments.
This system is under revision: under new arrangements Customs will issue the tax notice
electronically to the Treasury, which will in turn notify Customs electronically when payment is
made.
The current payment and invoicing system, and lack of communication between computer
systems also create difficulties with respect accounting for traders VAT credits on tax paid on
imports.
Nontariff Barriers
The other aspect of Vietnams system for managing imports that has long had an impact on
revenue collections are the policies and processes for managing trade, including regulation of
entry into importing of particular commodities, prohibitions and quantitative restrictions on
imports, and regulation of transactions.
Import prohibitions: Most prohibitions appear to have been driven by concerns of
health, security and safety. However, imports of cigarettes and a wide range of second
hand goods were prohibited until Vietnams accession to the WTO.
Import licensing by the Ministry of Industry and Trade, typically used to manage
volumes of imports, to protect local producers, but also to manage demand and supply
of key commodities has been largely dismantled as a result of implementing WTO
commitments.
Specialised management by line agencies. A significant range of goods is subject to
management by line agencies. This management, which is largely aimed at quality and
safety issues can take the form of inspection of shipments, appointment of authorised
importers, import licensing, registration, or certification as to compliance with standards.
This form of management has in the past been used to protect state enterprises belonging
to the managing agency from competition, either from imports or from other traders.
The Government introduced a set of tariff rate quotas in 2003 for raw milk, condensed
milk, eggs, maize, raw tobacco, salt and cotton. These quotas, which allow a certain volume
of imports at a lower tariff than applies to over-quota imports frequently act as a straight
quantitative limit on imports. They are permitted, however, under the WTO
Agreement on Agriculture, and Vietnam has retained them in the postaccession period.
Trade in certain commodities (such as petroleum, aircraft, newspapers, cigarettes and
videotapes) remains restricted to state enterprises. As a consequence of WTO accession,
foreign firms and individuals are now allowed the same trading rights as domestic
enterprises.
171
T ax R eform in V ietnam
Proportion of
Tariff Lines Average Maximum Number of Standard
CVa
Reaching Final Rate Rate Rates Deviation
Bound Rate
% % % No No %
2007 59.5 17.5 200 37 22.0 114.4
2008 1.3 17.5 200 39 21.5 114.6
2009 3.9 17.3 200 39 20.0 115.8
2010 11.1 16.6 200 44 20.0 118.9
2011 1.1 16.4 200 45 20.0 119.5
2012 16.2 14.8 200 49 19.7 124.6
2013 0.4 14.7 200 49 19.6 124.4
172
I ntegration and G overnment R evenues
173
T ax R eform in V ietnam
Estonia 1999 100.0 6.6 30.0 99.3 100.0 17.7 59.0 68.0 85.5
Georgia 2000 100.0 5.8 20.0 100.0 100.0 12.1 30.0 84.5 96.5
Jordan 2000 100.0 15.0 30.0 77.2 100.0 25.0 200.0 39.8 95.0
Kyrgyz
1998 100.0 6.7 20.0 100.0 11.7 30.0 100.0
Republic 100.0 99.9
Latvia 1999 100.0 9.3 55.0 98.3 98.7 33.6 55.0 38.1 41.7
Lithuania 2001 100.0 8.2 30.0 99.2 100.0 15.6 100.0 80.1 91.4
Moldova 2001 100.0 5.7 20.0 100.0 100.0 12.4 25.0 99.9 100.0
Mongolia 1997 na 20.0 30.0 87.6 100.0 18.4 75.0 88.8 90.8
Nepal 2003 100.0 23.7 60.0 47.3 95.9 42.3 200.0 7.0 35.6
Oman 2000 100.0 11.0 20.0 100.0 100.0 30.5 200.0 83.4 87.6
Panama 1997 na 11.5 81.0 84.3 99.5 26.1 260.0 84.3 99.5
Vietnam 2007 100.0 12.5 100.0 70.9 92.8 21.0 135.0 48.5 68.3
Source: WTO (2001); Vietnam Law Data; authors calculations.
Note: a. Proportion of tariff lines bound individually. b. Of lines bound individually. c. Proportion of individually
bound tariff lines bound with rates below 20 percent. d. Proportion of individually bound tariff lines bound with rates
below 30 percent.
n.a. = Not available.
174
I ntegration and G overnment R evenues
25 20
Standard devn. Average
18
20 16
14
15 12
Per cent
No
10
10 8
6
5 4
2
0 0
1995 1996 1997 1998 2001 2003 2005 2007 2010 2012 2014 2019
Figure 6.1 summarizes the evolution of basic indicators of Vietnams MFN tariff rates since
1995, and compares them with indicators for the bound rates up to 2019. As the chart shows,
the bindings seem to continue a process of reducing the average and dispersion of rates that
began after 2003. This process followed a period of escalation in the tariff structure that began
after the 199798 Asian financial crisis, and accompanied a process of phasing out some the
key nontariff barriers.
Figure 6.2 summarizes the shifts in the distribution of rates that implementation of the bound
rates would bring in its train. It shows that the bindings could lead to a large reduction in the
number of zero-rated tariff lines, a significant increase in the number of tariff lines with rates
between zero and 5 percent, and a consolidation of rates in the range of 5 to 20 percent.
Figure 6.2: Distribution of Actual and Bound Tariff Rates, 2005 to 2019
35
30
25
20
15
Per cent
10
5
0
2005 2007 2010 2012 2014 2019
(In practice, as table 2.2 shows, during 200809, Vietnams applied tariff has been adjusted in
ways that have produced a lower average applied rate than the average bound rate, and a lower
dispersion of rates as measured by the standard deviation and coefficient of variation.)
175
T ax R eform in V ietnam
176
I ntegration and G overnment R evenues
1994
1995
1997
1999
2000
2001
2002
2003
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0
Rate
It is worth noting that this absence of a clear relationship between statutory tariff rates and
actual collections is not unique to Vietnam. Pritchett and Sethi examined this relationship for
a number of countries implementing tariff changes in the late 1980s and 1990s, and found
that:
The collected rate of duty for any given item in the tariff code is only weakly related to the
official rate for that item;
The variation of collected rates around the official rate increases with the level of the
official rate;
Collected rates, on average, increase much less rapidly than official rates
When the official rate is high, the rate at which collected rates increase as official rates
increase falls.
From this evidence they argue that in reforming the system of tariffs and tariff revenue
collection, the change in official rates, especially at the high levels, is likely not to be the most
important element of reform affecting revenue. They also argue that assuming that tariff
revenues fall one-for-one with rates overstates the impact of rate reductions on revenues,
because it assumes constant collection performance.
The General Department of Customs collects data on the value of imports that enter
under different types of processing regime. As table 6.10 shows, over 60 percent of the
value of imports in 2002 and 2003 were classed as imports of goods destined for domestic
consumption, as opposed to further processing or temporarily admitted. Customs sources
have suggested that a large proportion of imports for further processing of cleared by foreign
invested enterprises are exempt from duties under the various provisions determining
exemptions and concessional treatment.
177
T ax R eform in V ietnam
2002 2003
Customs processing treatment
$US mil. % $US mil. %
Consumer goods 11908.3 60.8 14508.6 63.7
Imports by foreign invested enterprises 2767.3 14.1 3097.4 13.6
Temporary admission 374.2 1.9 609.5 2.7
Imports for further processing 4533.2 23.1 4551.6 20.0
Total 19582.9 100.0 22767.2 100.0
Source: CIE 2004a.
How would Tariff Reductions and Bindings Affect Revenue Collections Implied
by Statutory Duty Rates?
Analysing the effects that tariff binding commitments may have on revenue involves:
conjecturing how Vietnams tariff bindings may affect applied duty rates;
estimating how these changes in duty rates would affect notionally collectible revenues
for a given level and structure of imports;
adjusting these estimates for the impact of concessions and exemptions and other factors
affecting collection efficiency; and
factoring in how the volume of imports of different commodities might change in
response to the price effects of duty changes.
The scope for implementing these steps is seriously constrained by data problems. The main
constraints are due to problems of aggregation and of tracking of revenue assessments. These
problems can be summarised as follows.
The aggregation problem
Tariff rates are specified at the eight-digit level of the HS tariff nomenclature. However,
comprehensive import data is only reported at the six-digit level of the HS (and this only for
2003), so some way is required of creating a representative tariff rate to apply to the import
data. Whatever method is used will lead to inaccuracies in the estimates of implied revenues.
The problem of revenue assessments
The other main problem isas discussed abovethat no data seems to be available on actual
duty assessments at any level of the HS. Nor is information readily available on the incidence
of concessions and exemptions, or on the proportion of imports from ASEAN countries that
is eligible for CEPT treatment. This means that it is not possible to develop any systematic
way of assessing the impact of these exemptions and concessions on revenue collections, and
how it might vary across different commodities and over time.
178
I ntegration and G overnment R evenues
superimposes estimates of estimated tariff changes associated with the bindings on a scenario
of projected growth in imports. The model takes no account of how imports may respond
to price changes caused by tariff reductions. But it does allow consideration of the impact
of rationalising the current system of exemptions and concessions and so increasing the
efficiency of duty collections (as measured by the difference between notional and actual
revenues raised on imports.) The model can be extended to take account of the effect on
revenues of duties collected on imports under the CEPT but for now the focus is on
imports subject to MFN rates.
The Growth Scenario
The analysis has been carried out assuming that Vietnams GDP will continue to grow at and
average 7.5 percent per annum over the period 2007 to 2019. A set of consistent growth rates
for imports of consumer goods, petroleum and production inputs has been constructed, by
assuming a relationship between GDP and imports. Table 6.11 shows that over the period
1995 to 2006, a 1 percent change in GDP has been associated on average with a 2.7 percent
increase imports. For the purpose of this analysis, it has been assumed that the elasticity of
imports to GDP will average 1.6 over the projection period, with real imports of consumer
goods and petroleum products projected to grow at 9 percent per annum and real imports of
production inputs (raw materials, intermediates and capital goods) projected to grow at 13.5
percent per annum. (Because the key data set, the disaggregated information on imports, is
for 2003, the analysis generates estimates in 2003 prices.)
Table 6.11: Real Imports, GDP, Growth, and Elasticities
179
T ax R eform in V ietnam
6000
Base scenario
5000
Scenario 2
Scenario 3
US$ billion
4000
Scenario 4
3000
x
2000
1000
0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
As the chart shows, under scenario 1, nonpetroleum tariff revenues are projected to grow
steadily from US$1.6 billion to US$5.5 billion (in 2003 prices) between 2006 and 2019.
Under the second scenario, these revenues increase, but are affected by the tariff reductions,
and reach US$3.1 billion in 2019. If rates were set to the bound rate in each year, revenues
would grow to US$3.7 billion. If action were taken to reduce the incidence of concessions and
exemption according to the trajectory in table 6.12, revenues would dip slightly below those
of the first scenario until 2010, and catch up with those revenues in 2018.
180
I ntegration and G overnment R evenues
One critical element that shapes overall tariff revenues is the taxation of imports of petroleum
products. As indicated in the previous chapter, tariff rates on these products are frequently
adjusted to balance revenue and pricing objectives. Figure 6.5 shows what could happen to
petroleum revenues raised on imports from sources subject to MFN rates, and prices remained
unchanged from 2003 levels. If 2006 rates continued to apply, real revenues would grow to
around US$630 million by 2019. If the upper bound rates were applied, revenues would be
some US$1.3 billion higher in 2019. This would come close to completely compensating for
any reduction in nonpetroleum revenues under the various scenarios.
Figure 6.5: Petroleum Revenue Scenarios
2500
1500
1000
500
0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
181
T ax R eform in V ietnam
Policy changes with respect to the extensive and complex concessions and exemptions
can readily compensate for the effect of rate reductions, and could allow, from a revenue
perspective, more radical reductions in tariff rates.
There is considerable scope to make up for revenue shortfalls by increasing taxation of
petroleum products.
It is also worth noting that the foregoing analysis takes no account of what tariff reductions
would do to the base for other taxes on imports (primarily the value added tax) and for the
base for income taxes. Reduction in the price of duty paid price of imports could reduce
the value basis for the VAT, but might also lead to higher volumes of imports. Similarly,
growth and increased formalisation of the economy would increase the base for other taxes,
including income and property taxes. However, this may also require further attention being
paid to the structure and administration of these taxes, especially the VAT. (Recent analysis
of the VAT suggests that there are still some serious problems with the VAT, in terms of the
base, rate structure and calculation methods (CIE 2004a). On balance, it is unlikely that the
Government would suffer a revenue shortfall in the medium to longer term, although there
may be some issues to deal with in the short term.
182
I ntegration and G overnment R evenues
a set of standard concordances linking HS tariff lines to the commodities classified as the
products of each ISIC industry. This is not a good indication of the actual level of protection
accorded these industries in Vietnam. (To do this would require weighting tariff lines by the
value of production of commodities that compete with imports to which the tariff would
apply. Unfortunately, there is not as yet data on production of commodities based on an
international commodity classification that can be brought into concordance with the
tariff nomenclature system.) However, it is reasonable to surmise that big variations in the
calculated tariff averages for these economic activities signifies variation in the protection
available to import-competing production under these activities.
Table 6.13: Vietnams Tariffs and Tariff Bindings by Economic Activity
183
T ax R eform in V ietnam
The table suggests a number of observations about the current tariff structure and possible
impact of Vietnams binding commitments.
There is considerable variation of average tariffs across economic activities. Import
competing production in some activities (such as manufacture of tobacco products
[ISIC 16], transport equipment [ISIC 34 and 35]) would seem to be much more heavily
protected than others (such as agriculture (ISIC 1) and manufacture of chemicals, basic
metals office and other machinery and precision instruments (ISIC 24,27,30,33).
Average tariffs seem surprisingly high for some industries where Vietnam is successfully
exporting a considerable volume of production (fishing (ISIC3) and ISIC 18, manufacture
of apparel). Activities that are efficient enough to compete on international markets would
normally be competitive with imports on the domestic market without any protection
Some of the larger reductions in tariffs that application of tariff bindings implies are
for activities where Vietnam is a successful exporter. This suggests that, perhaps not
surprisingly given the way that tariff bindings are negotiated, some of the reductions will
remove water in the tariff rather than reduce levels of protection actually used by some
industries.
It also appears that the bindings provide room for Vietnam to significantly increase tariffs
for some activities where large investment is under way such as petroleum refining
184
I ntegration and G overnment R evenues
Employment
Gross output
35.0
30.0 30.0
25.0 25.0
2005
Value
2019
15.0 15.0
10.0 10.0
5.0 5.0
0.0 0.0
0 >0-5 >5-10 >10-15 >15-20 >20-25 >25-30 >30-35 >35-40 >40-45 >45-50 >50-75 >75 0 >0-5 >5-10 >10-15 >15-20 >20-25 >25-30>30-35>35-40 >40-45 >45-50 >50-75 >75
While these data suggest that a high proportion of manufacturing employment may be
dependent on high levels of protection, some important caveats have to be made. First, as
table 6.14 shows industries falling in the two-digit activity covering manufacture of textiles
(ISIC 17) and manufacture of wearing apparel; dressing and dyeing of fur (ISIC 18)) together
account for over 23 percent of manufacturing employment. While tariffs on imports of the
products of these industries are high, much of their output is exported with no or limited
assistance. This means that much of the employment may not benefit from the high tariffs.
Second, many industries involved the manufacture of food products and beverages (ISIC 15,
which accounts for 16 percent of manufacturing employment and has an average MFN tariff
of 35.7 percentsee appendix table A.1) may not face much competition from imported
products. This is because local production is very competitive, or involves goods that do not
readily enter into international trade. There may well be water in the tariff for parts of this
industry and producers may not need the full extent of the available tariff protection. (This
is not true of all production in this sector: sugar production, for example is not competitive
and may well be relying heavily on the average tariff of around 70 percent that applies to this
industry.)
185
T ax R eform in V ietnam
The structure of manufacturing activities varies across the regions of the country, and so the
dependence of employment and production on tariff protection also varies. Table 5.2 shows
the average tariff protection available to the manufacturing given the structure of industry in
each region. It shows that the manufacturing sector in the Red River Delta, Mekong River
Delta and South Central Coast regions has a larger proportion of activities with high tariff
protection than in the North East and Central Highland regions. (This could suggest that
industry in the latter regions might be less negatively affected by tariff reductions than the
counterparts in the more developed parts of the country.)
The table also shows that the average tariff on manufacturing weighted by gross output in
four-digit industries is quite a lot higher (29.5 percent) than the simple average of rates on
manufacturing tariff lines.
186
I ntegration and G overnment R evenues
Agriculture
Mining & quarrying
Food beverages & tobacco
Building materials
Paper and wood products
Chemicals
Precision equipment 2005
Household appliances
Transport equipment
Non-electrical machinery 2007
Electrical machinery
Metal products 2019
Textiles, clothing & footwear
Other
Figure 6.8 provides an attempt to take account of the fact that there is a fair degree of
export-oriented production in some of these industry groups, for which the tariff provides
no protection. The figure suggests that application of the bindings does little to change the
impost that tariffs place on export-oriented activities.
187
T ax R eform in V ietnam
Agriculture
Mining & quarrying
Food beverages & tobacco
Building materials
Paper and wood products
Chemicals
Precision equipment 2005
Household appliances
Transport equipment
Non-electrical machinery 2007
Electrical machinery
Metal products 2019
Textiles, clothing & footwear
Other
Note: The chart indicates the average effective rates of protection for exporting and import competing
production in broad industry classes (aggregates of industries identified in the Vietnamese Input-Output
industry classification) using the 2005 applied MFN tariff rates and the 2007 and 2019 bound rates.
As table 6.16 shows, implementation of the bound rates would reduce the average ERP across
all traded goods producing activities identified in the Input-Output tablefrom 47.1 to 34.5
percent for import substituting production, and from 32.9 to 24.2 percent across importing
competing and exporting activities.
Table 6.15: Summary Measures of Changes in ERPs
The table suggests that the dispersion in effective rates would also fall. But as chart 6.9
shows, the distribution of value added in these industries across different levels of effective
protection suggests that the occurrence of pockets of quite high and very low protection
will persist.
188
I ntegration and G overnment R evenues
Figure 6.9: Distribution of Industrial, Mining, and Agricultural Value Added in I/O
Industries, by ERPa, 2005 and 2019
50 120
Distribution Cumulative distribution
45
40 100
35
80
30
Percent
25 2005 60
Percent
2005
20
15 2019 40 2019
10 20
5
0 0
0
-1 0
0
5
10
15
20
25
30
35
40
45
50
60
70
80
0
>1 0
50
0
-1 0
0
5
10
15
20
25
30
35
40
45
50
60
70
80
0
>1 0
50
10
15
<-2
10
15
<-2
Per cent Per cent
189
Value-Added Tax
By Gangadhar Prasad Shukla, Tuan Minh Le, and Duc Minh Pham VII
1. Introduction
Value Added Tax (VAT) has been an important source of revenue in Vietnam. It presently
contributes about 30 percent of the total tax revenues. Except for a few years in the past,
it has been generally a buoyant tax, the revenues growing at a rate greater than the growth
rate of the GDP. Prior to the introduction of a VAT, Vietnam had adopted a turnover tax in
October 1990 in its first round of tax reform. In an effort to mitigate the shortcomings of this
tax, several rounds of amendments have been carried out since then. Finally, the turnover
tax was replaced by a VAT in 1999. Subsequently, several amendments have been made
with some major changes adopted in 2003, 2006, and the last VAT Law revision conducted
in 2008.
191
T ax R eform in V ietnam
Central North
Sub-
Asia & Europe Africa & Small
Saharan EU America Average
Pacific & Middle Islands
Africa
Russia East
Average 16.4 10.9 19.8 18.6 16.3 13.9 16.1 16.0
Standard
Rate (%)
VAT revenue 3.9 3.3 7.0 6.4 5.7 4.9 4.7 5.1
as % GDP
VAT revenue 28.4 21.7 20.7 27.8 28.1 33.0 18.0 26.5
as % of tax
revenues
Source: IMF estimates (Ebrill et al. 2001); International Bureau of Fiscal Documentation (IBFD) 2005; Worldwide
Summaries (PricewaterhouseCoopers) and Consumption Tax Trends (OECD 2006).
192
V alue - A dded T ax
It is clear from table 7.3 that the standard tax rates in Asian countries are more moderate at
around 10 percent. While the Philippines has a low threshold of US$10,600, Singapore has a
comparatively high threshold of US$620,100. The contribution of VAT to revenues is around
3 to 4 percent of GDP in most of these countries. While the VAT system and its performance
in Vietnam may look very different from most of the OECD countries, it compares well with
the Asian countries.
Table 7.3: VAT in Asian Countries
Threshold
VAT Rates (percent) VAT Revenue
(US$)
Date General (percent
Standard (percent
VAT First Rates Other Rates Thresholds of tax
Rate of GDP)
Introduced Basic revenue)
Cambodia 1999 10 10 61,000 (services) 35.5 2.9
125,000 (goods)
China 1994 13,17 17 13 224,000 27.9 3.9
(wholesale)
3,000 & 1,200
(goods &
services)
Indonesia 1985 10,15,20,25,35 10 15,20,25,35 109,200 19.3 2.7
Philippines 1988 12 12 10,600 22.1 3.0
Singapore 1994 5 5 620,100 8.4 1.2
Taiwan 1986 5 5 ... ...
Province of
China
Vietnam 1999 5,10 10 5 No threshold 24.5 4.0
Thailand 1992 7 7 30,900 20.8 2.8
Mongolia 1998 15 15 8,300 29.9 5.3
Korea 1977 10 10 12,500 19.2 4.6
Source: IMF World Economic Outlook; IBFD; Ernst and Young; VAT and Sales Tax Worldwide various years.
In order to assess the VAT efficiency, two efficiency ratios are generally used: production
efficiency ratio and consumption efficiency or C-efficiency ratio. The production efficiency
193
T ax R eform in V ietnam
ratio of VAT is measured by the ratio of VAT revenues to GDP divided by the standard VAT rate
expressed as a percentage. A low ratio indicates erosion of the tax base due to poor tax policy
such as zero-rating and reduced rates or poor enforcement of the law. Consumption efficiency
or C-efficiency ratio is a similar measure where the share of VAT revenue as a percentage
of consumption divided by the standard rate is estimated. If a uniform consumption tax is
imposed on all consumption, its efficiency will be 100 percent. Zero-rating of some items
will reduce the ratio below 100 and taxation of investment goods will raise it to more than
100 percent. Alternatively, a high C-efficiency could be an indicator of multiple breaks in the
VAT chain due to excessive exemptions resulting in taxation of both the final consumption
and some of the constituent intermediate goods. Both these ratios show the same thing but
one is normalized with reference to an income type VAT while the other to a consumption
type VAT.
Clearly the three factors that would determine the revenue productivity of VAT in a country
are: (a) The rules that prescribe the rates, the base, the threshold and other issues related to tax
structure; (b) The scale of taxable consumption in the economy that is, the final expenditure
made by the consumers on taxable items; and (c) The effectiveness of the tax administration
and the level of compliance. Since VAT on tradable goods are collected by the customs
department, the third factor related to tax administration would also depend to some extent
on the share of trade in the economy that is, the trade/GDP ratio, because it is easy to collect
VAT on the imports.
Table 7.4 shows an estimate of productivity and consumption efficiencies across the
geographical regions. The estimates (for the year 2001) indicate that both production
efficiency and consumption efficiency ratios for Asia are comparable to the ones in the
EU, Central Europe, North Africa and the Middle East, and Latin America and they are
significantly higher than in Sub-Saharan Africa.
Table 7.4: Production and Consumption Efficiency Ratios for VAT by Region
A comparative picture of production and consumption efficiencies for Vietnam and some
of its neighboring countries is presented in table 7.5. Both the productivity efficiency and
consumption efficiency in Vietnam compare well with its neighboring countries. In fact,
these are significantly higher than the regional averages presented in table 7.4. In addition to
many other factors, this could also be the result of a series of cascading-induced exemptions
in the VAT law.
194
V alue - A dded T ax
To ensure that Vietnam has a robust and clean VAT law and yields sufficiently high productivity
and consumption efficiencies, the logical sequence to proceed would be to first fix the tax
structure and then focus on the tax administration and tax compliance issues. Eliminating
or at least minimizing the exemptions would be necessary to give a realistic picture of the
performance of the VAT. Once these issues are tackled successfully, the expected revenues
can be ascertained by forecasting VAT revenue with the help of forecasting models in order
to see the level of compliance in the economy and then try to improve it. Some typical
features of the VAT in Vietnam such as tax structure, exemptions, computation method for
tax liability, refund and equity are examined below, bearing in mind international practice of
VAT regimes.
195
T ax R eform in V ietnam
purpose machinery and equipment for agricultural production and newsprint; cultural
exhibits and sports activities.
Finally, a tax rate of ten (10) percent is applied to the remaining items under VAT.
A major change compared to the earlier VAT law has been that VAT now applies to all goods
and services subject to the Special Consumption Tax (SCT). It is an important change and
ensures that Vietnam law conforms to international practice.
Twenty five categories of goods and services are VAT exempt under the new law.
Rate Structure
As pointed out above, Vietnam uses two VAT rates in addition to a zero rate. Currently, of
all the countries that have adopted VAT, about 54 percent of countries have a single rate in
addition to a zero rate, while about 23 percent of countries have two rates, and a small number
have more than two rates.67
The issue of multiple VAT rates may be analyzed on the dual principles of efficiency and
equity. Efficiency gains could be realized by following the inverse elasticity rule according
to which goods/services with low elasticity of demand should be taxed at a higher rate and
goods/services with high demand elasticity should be taxed at a lower rate. This would
justify higher VAT rates for the so-called sin goods like alcohol and tobacco products in
countries applying a differentiated rate structure.68 It could also be argued that some other
goods such as gasoline and some luxury items have low elasticity of demand so far as the rich
are concerned. A higher level of taxation for all such goods, however, can be easily ensured by
imposing excises on those items in addition to the normal VAT.
Another aspect of the efficiency point could be made on the grounds that higher rates would
result in higher tax revenues. There is, however, no strong evidence to support this hypothesis.
On the other hand, it is a well known fact among tax administrators that higher tax rates are
more likely to encourage tax evasion and lower tax revenues.
The rationale for multiple rates is rooted primarily in an equity concern: the poor should not
be taxed heavily and therefore some goods and services that are predominantly consumed
by them should be lightly taxed. The list of goods and services with a 5 percent VAT rate
in Vietnam mostly includes items related to agriculture and animal husbandry, education,
medicine, science and technology.
Studies of tax systems where rate differentiation has been implemented, however, tend to
show that in the majority of such cases, the tax benefits largely go to the rich who are likely to
spend larger sums on those items compared to the poor. This is also true of food items that
are often taxed lightly on the grounds that the poor spend a greater share of their earnings on
The data about tax rates, exemption levels, etc. are drawn from a 2001 study of VAT practices
67
of 10 percent and an excise (or special consumption tax). This conforms to international best
practice in VAT design.
196
V alue - A dded T ax
food in comparison to the rich. Even if it is accepted that distributional gains may be realized
by rate differentiation in VAT, such gains are likely to be quite modest.69 Also, there are very
few items that can be differentiated clearly on the basis of consumption by different income
groups. Therefore, for equity or distributional considerations, income taxes and expenditure
policies are preferred policy instruments compared to the use of different rates of indirect
taxation on different commodities.
Even though there may be some marginal benefits of rate differentiation in VAT, several costs
imposed by this kind of policy are well known and would generally outweigh such benefits, if
any. Some of these costs are as follows (Ebrill et al. 2001):
a. A single positive rate lowers the cost of compliance by simplifying the requirements
of keeping records and invoices. It also simplifies tax forms which goes a long way in
promoting self assessment.
b. Rate differentiation creates opportunities for misclassification of items, thereby raising
the cost of administration. On the other hand, a single positive rate removes the
confusion that might arise in the treatment of border line cases and thus lowers the cost
of administration.
c. The use of a single rate helps limit the numbers of refunds. If there is a low tax rate along
with a zero rate, some taxpayers particularly exporters and importers of capital goods
could be in a sustained credit position vis--vis the tax department. This would result in
additional refund claims.70
In the case of Vietnam, a lower VAT rate on several input items is sometimes justified on
the grounds that it will reduce the cash flow burden of the VAT taxpayers. The VAT payers,
however, typically enjoy a cash flow benefit from collecting higher VAT on their outputs so
this cannot be a serious problem.
Thus, there is hardly any justification for maintaining multiple rates on a long term basis.
Vietnam should therefore consider moving to a rate structure with zero and one positive VAT
rate. This implies that the 5 percent rate should be eliminated and only the 10 percent rate be
applied to all goods and services subject to VAT.
The introduction of a single VAT rate should be part of an overall tax policy reform rather
than implemented in isolation. For example, it may have to be synchronized with the reform
in excises where some luxury items are brought in the tax net so that the elimination of the
lower VAT rate is not seen as a step unfavorable to the poor.
Future Outlook
Out of the 115 countries for which data is presently available, the VAT rate is less than 10
percent in 25 countries, more than 10 percent and below 15 percent in 24 countries, more than
69
For evidence to support this argument, see Sah (1983, 89101).
70
Sijbren Cnossen (1994) estimates that for a simple VAT regime in New Zealand (with a single
positive rate), the administration cost of the VAT is about US$50 per registrant per year, but the
cost quadruples for the case of the U.K., where the VAT is structured with two positive rates and
multiple zero rating (cited in Ebrill et al. 2001).
197
T ax R eform in V ietnam
15 percent but below 20 percent in 46 countries and more than 20 percent in 15 countries. As
shown in table 7.5 above, many of the low rate countries are in Asia Singapore, Indonesia
and Thailand for instance. On the other hand, the Philippines, China and Mongolia have
rates varying between 12 and 17 percent.
Also, with the gradual reduction of trade taxes under the provisions of WTO (for example,
tax revenues from crude oil tend to be shrinking), the question of recouping the loss of
revenue may arise for Vietnam in the future. The revenue from the corporate income tax
is currently at the level of just under 6 percent of GDP and it will be unrealistic to expect a
phenomenal increase in this. Personal income tax cannot be expected to yield substantial
amount of revenues, at least in the short term, because of the cautious approach adopted by
the Government toward this tax, which is justified to some extent given the present stage of
the growth of the economy and the labor market. The scope of additional revenues from
property and land related taxes is limited not only in Vietnam but generally in most countries.
Thus VAT has to play a dominant role in any increase of revenues in the future.
At some later stage in the future, the Government may therefore consider enhancing the VAT
rate in parallel with improvement of compliance management.
2. Exemptions
The list of exempt goods and services includes 25 categories of items. Compared to the
international best practices, this is obviously a long list of exemptions and needs to be
examined carefully. But first, a brief note on some aspects of exemptions is presented below.
198
V alue - A dded T ax
199
T ax R eform in V ietnam
charge. Defense, other not-for-profit public services, some categories of health and education
services fall in this category. It is hard to tax a service that is given away free except by taxation
of inputs. In fact, the public sector may be regarded as the final consumer of noncommercial
services that it produces and gives away free of charge. This is the practice currently adopted
in the EU countries as well.
As regards health and education, the standard advice and practice is to exempt basic services
primary education, basic healthcareand tax specialized services at the normal tax rate.
One relevant question is: does it make a difference whether such public sector services
that are provided free of cost are taxed or exempt? Since the final price is zero, taxation or
exemption apparently does not make any difference. As regards tax revenue on inputs,
that is also ineffective because that will increase the budgetary costs to the government for
providing those services. The only difference is that exemption would break the chain with
its accompanying implications.
Some countries (Canada, some EU countries) have therefore adopted the practice of taxing
these services and then rebating the VAT to public bodies engaged in these kinds of activities.
This virtually converts the exemption to zero-rating. The problem of assessing the sales value
for such services is often solved by looking at the user fee collection, if there is a user fee, or by
the amount of budgetary subsidy provided to the agency in question. In any case, the overall
revenue implication for the government would be zero.
Financial services such as banking and insurance are typically VAT exempt because it is not
easy to value their output and also because these are internationally mobile and putting them
under VAT may drive them out of the countrys boundary while they can still do business
from off shore locations. In that case, the country will not only lose VAT revenues but also
personal income tax and corporate income tax revenues. Taxation of financial services is
further discussed in a subsequent section.
Taxing of Real Estate and Construction
The ideal treatment of real estate and construction would be to tax the services that flow from
them. Credit may, of course, be given if services are used as business input. The leasing of
real estate for commercial purposes can be easily subject to VAT but this would be difficult in
the case of owner occupied houses. To avoid distorting choices between renting and house
ownership, the commercial leasing of residential property is also generally exempt.
One way to tax residential property under VAT would be to tax at the time of house purchase.
Taxing the subsequent resale does not make a lot of sense as the tax revenues paid by the buyer
and the refund claimed by the seller would cancel out. The construction activities should be
taxed in the normal way and credit given for construction as a business activity. This practice
would ensure that those who construct houses for their own occupation, and are therefore
VAT exempt on the final value of the house, at least pay VAT on their inputs. Those who buy
houses from contractors would, of course, pay VAT on their purchases while the contractor
can claim the refund on inputs.
200
V alue - A dded T ax
201
T ax R eform in V ietnam
202
V alue - A dded T ax
low as US$1,643 while the threshold in Chile is US$477,633. The exemption levels vary
considerably among Vietnams neighbors as well. Cambodia has a threshold of US$61,259
for services and US$122,520 for goods, while Indonesia has a threshold of US$109,165,
Philippines US$10,612 and Thailand US$30,964.71 The level of threshold for Vietnam
should be chosen so that it eliminates the bulk of those small VAT payers who do not make
substantial contributions to revenues while imposing a large administrative burden.
Most countries start with a high threshold and gradually lower it. In the case of Vietnam, the
threshold has been virtually zero and this approach cannot be applied. It is, however, possible
to examine, with the help of econometric modeling, several alternative thresholds in terms of
their revenue implications and the resulting number of taxpayers. The one which is found to
be most appropriate and cost effective should be selected.
From July 1, 2007, when the new Tax Administration Law was in effect, the GDT would be
72
203
T ax R eform in V ietnam
accounting, invoices, and vouchers to serve as the bases for tax calculation under this method
and to businesses involved in gold, silver and gems trading activities.73
Once a suitable threshold for exempting small traders is chosen and implemented, the need
for this dual computing system may be abolished and only the credit or tax deduction method
should remain applicable.
Many developing countries have adopted levying of a turnover tax in the range of 2 to 3 percent,
73
called a unified tax, for unincorporated business in lieu of both the VAT and income taxes.
Vietnam, however, has gone beyond that stage and its current dual system can easily move to a
pure credit invoice method.
204
V alue - A dded T ax
The present practice in Vietnam generally fits into the strategy presented above and this may
be allowed to continue for some time to come. The first two categories of items in the list of
exempt goods/services pertain to the agriculture sector and related activities. However, the
large farmers may be considered for inclusion in the VAT system. Also, small farmers may be
given the option of voluntary registration under VAT.
Financial Sectors
Financial services are treated differently under VAT for a variety of reasons. First, it is difficult
to identify on conventional invoice computation the value added on each service provided
by financial institutions. There is seldom a market price for these services that can be used
for computing the VAT liability. If normal interest rates were taxed, that would be taxing a
combination of inflation and return on savings. Thus, instead of taxing consumption it would
become taxation of income. Second, this taxation would be mostly applicable to households
while for businesses it would be deductible as credit. As a result, the tax base tends to exclude
the business community and essentially remains narrow. Third, putting them under VAT
would increase the cost of borrowing to the private sector which becomes a politically
sensitive matter. Finally, both capital and financial services are internationally mobile and
they can easily move offshore and operate from there.
Keeping these special features in mind, the preferred approach has been to exclude financial
services from VAT. Under this arrangement, at least they keep paying taxes on all their inputs:
buildings, computers, office equipments, stationery and these are not credited back. That is
how most European VATs treat financial services.
Vietnam has adopted a similar approach and has exempted insurance, credit provision
services, capital transfer, derivative financial services and forward and future contracts. This
seems to be a reasonable approach.
Tourism
Typically all services should be taxed under VAT except for a few exempt services such as
healthcare, education, social and financial services. Therefore, tourism should be under
normal VAT. It is true that as of now tourism is a budding industry in Vietnam and the
industry may be seeking tax concessions from the Government. Often the labor content as a
proportion of the final price in tourism is high and, as a result, these services are provided by
small scale establishments.
Therefore, the best approach may be to keep the tourism industry under the VAT net but
exempt the small tourism operators and travel agents under the exemption clause for small
traders if their turnover is below the prescribed threshold.
Real Estate
Taxation of real estate and the construction sector has been dealt with in section 6.4 above in
the context of the exempt items.
205
T ax R eform in V ietnam
206
V alue - A dded T ax
the VAT system. The registration staff should carefully assess and verify initial information
presented for registration.
The level of threshold for compulsory registration is relevant not only for effective
administration of VAT but also for reducing refund abuse. If the threshold is too low,
the tax administration resources become over stretched in managing the VAT registered
businesses and monitoring refund claims. Vietnam, with zero threshold, faces this
problem. So an appropriate level of threshold is important for effective monitoring and
management of VAT refund.
An effective risk-based audit system plays an important role in reducing VAT refund
fraud. Exchange of VAT and income tax information and VAT and customs information
are important in this regard. Where direct and indirect taxes are administrated together
in a unified tax administration, exchange of VAT and income tax information is routine.
Where these taxes are administered separately, however, exchange of information is
generally limited. In some countries, this kind of exchange is legally prohibited. As
a substantive share of VAT revenues are collected through the customs agency, the
cooperation between VAT and customs also becomes important.
207
T ax R eform in V ietnam
credibility of tax administration. Thus clearly there is a great deal of variation among
countries. Also, the actual refund practices are found to be quite different from what is
laid down in the law.
Two-thirds of the surveyed countries undertake some sort of risk assessment in processing
VAT refunds. Sometimes it is quite sophisticated (UK) while in many countries it is
often rudimentary and may amount to verifying each refund claim. A quarter of surveyed
countries have a statutory requirement to verify every refund claim.
It is also found that a lack of an appropriate audit program in many countries contributes
to the existence of a poor refund system. A third of the surveyed countries report that
they do not have a VAT audit program. These countries end up carrying out extended
prepayment checking of claims rather than relying on a sound postpayment audit. This
naturally delays the entire process. It is important for the tax administration to have a
sense of the level of refunds payable so that financial resources are in place in a timely
fashion and also as a warning signal against attempted fraud.
Exchange of information about VAT registered traders and the value of supplies across
countries has become important for international transactions. For instance, EU member
states exchange information via a VAT information exchange system (VIES) about VAT
traders and sales of goods and services.
There are two methods of budgeting for VAT refunds: (a) making payments from gross
VAT revenues, (b) paying from budget expenditure appropriations. Irrespective of the
method used, it is necessary to have a forecasting and monitoring system to estimate VAT
collections and level of refunds. Actual collections and refunds should be tracked against
forecasts and variances should be explained and reconciled.
Some countries have adopted innovative practices to reduce refund fraud. For instance,
Kenya provides for mandatory certification of VAT claims by certified public accountants
(CPAs) for claims exceeding specified amounts. Sanctions are also imposed on CPAs
who knowingly certify false claims. A number of countries give preferential treatment to
taxpayers with sound compliance histories (gold status in Pakistan). Some countries
(Bulgaria) have introduced a VAT bank account system where a VAT registered business
must open a VAT account for depositing VAT payments and spending money on VAT
purchases. This is meant to reduce cash transactions and thus reduce VAT fraud.
The following lessons with significant policy implications for Vietnams VAT system can
be learned from analysis of international experience in VAT refunds.
The number of taxpayers should be kept at a level that can be realistically managed by
choosing the right threshold and verifying the VAT registration applications to prevent
fictitious traders entering the system.
Suitable forecasting and monitoring systems should be established to anticipate refund
levels and make provisions for refund payments.
Refunds should be processed in a reasonable period of time and interest paid on late
payments. Excess VAT credit should be offset against VAT and other tax arrears.
Exporters should be paid refunds promptly. In a functioning risk management, the
tax department should make a clear distinction between claimants with a history of
compliance and those who are new. The Tax department should maintain historical
profiles for each refund claimant.
208
V alue - A dded T ax
209
T ax R eform in V ietnam
Also, in Vietnam some other complications related to VAT refunds need to be remedied.
There is no justification for distinguishing between the VAT refund on capital goods versus
the VAT refund on other inputs. A carry forward period of three months may be applied to
every business and a threshold for obtaining the current period refunds may also be fixed but
it should not be excessive and should remain uniform across different businesses. This would
render the system simpler and more transparent.
All new businesses must be audited for a couple of years before being given refunds to make
sure they are genuine units and not a scam. Otherwise, a new unit may come up and claim a
lot of refunds based on false invoices and then disappear. Older, more established units are
less likely to commit fraud.
To deal with the concerns of investors about late refunds, some countries have adopted the
practice of allowing deferment of VAT payments on specific capital imports until VAT input
deductions are claimed. This approach requires effective customs control and coordination
between customs and domestic VAT administration.
Sometimes, zero-rating of capital equipment items has been tried in a few countries but this
may result in a net revenue loss to the extent that capital equipment is used by unregistered
businesses as well.
210
V alue - A dded T ax
enforcement of VAT. In fact, close cooperation among the different parts of tax administration
is helpful for enforcement of all taxes income tax, VAT and custom tariffs. However, this
kind of cooperation in developing and transition economies is more of an exception than a
rule. Clearly, a separate VAT department independent of the domestic tax department that
administers income taxes is bound to make this kind of cooperation harder to achieve.
In the case of Vietnam, since VAT is administered by the GDT, which is responsible for all
domestic taxes, cooperation and coordination between GDC and GDT would achieve the
objective of enhancing VAT enforcement. The question is what kind of arrangement between
the two agencies would ensure this cooperation? Does it have to be some sort of formal
arrangement or would an informal arrangement do?
In most countries, the problem arises due to two reasons. First, the confidentiality guaranteed
to the taxpayer by the law may be a hurdle in sharing of information. In most countries,
however, the law does not preclude sharing of information within the tax administration
bodies and the ministry of finance. For instance, in Vietnam article 6 of the Tax Administration
Law mandates that the taxpayer information should be kept secret by the tax administration
bodies. But this cannot be interpreted as stopping the GDC and GDT from information
sharing between them or with the Ministry of Finance (MoF).
Second and more importantly, it is administrative apathy and compartmentalization that is
the main problem. Vietnam may not be an exception to this problem and historically there
may be a general lack of coordination among GDC, GDT and the MoF (tax policy unit). This
may, however, be remedied through an administrative arrangement where clear instructions
are laid down for periodical sharing of information on prescribed formats. If it is deemed
necessary, the task of coordination and cooperation among the various tax agencies may be
supervised by a senior officer or deputy minister nominated by the Minister of Finance.
211
T ax R eform in V ietnam
An input-output VAT base modeling is applied to assess the revenue impact of rate
unification and/or rationalization of exemption. The main source of data is Vietnams I-O
Tables 2005, supplemented by the results of the 2004 household expenditures survey and
other macroeconomic data (for example, economic growth rate, growth rates of various
components of GDP, in particular, private and government consumption and investment).
Details of modeling methodology and projected results are presented in Chapter 4. The
results are summarized as follows:
a. Unification of the rate structure (at the standard rate of 10 percent) with existing
exemptions would generate expected revenue gains of about 5.8 percent.
b. Unification of the rate structure combined with rationalization of exemptions would
expect to generate revenue gains of about 18 percent. In both reform scenarios, it is
assumed that the existing compliance rate remains unchanged. The estimates do not
take into account the savings in reduced compliance and administration costs due to the
simplification of the VAT regime.
Small traders
a. The current practice of having a zero threshold of exempting small traders should be
abolished and a suitable threshold adopted. The final choice should be based on the
tradeoff between the cost of administration and revenue implications.
b. For the small traders below the threshold some sort of taxation should be envisaged.
Applying a turnover tax of 2 to 3 percent which is then tax deductible by the next
registered buyer may be considered. It should be administered by a special division within
the GDT that is familiar with the problems of dealing with small traders. Alternatively
this responsibility may be handed over to the local authorities.
Method of computing VAT liability
a. With the adoption of the credit method across the board, the subtraction method of VAT
calculation for some traders who are unable to observe accounting regulations may be
abolished.
Exports and VAT refund
a. The current VAT refund system needs to be rationalized. There are different regulations
for different kinds of businesses and this makes the system complex and opaque. While
the long-term solution of this problem is linked to improved administrative capacity and
practices, some short-term measures are required. While deferment on imported capital
goods or zero-rating their sales may offer some relief, these measures have their own
problems. At least, some uniform system of waiting and threshold should be mandated
to mitigate this problem. For instance, a carry forward period of three months may be
applied to every business and a threshold for obtaining the current period refunds may
also be fixed but it should not be excessive and should remain uniform across different
businesses.
b. All new businesses should be audited before paying refund to make sure these are genuine
units.
212
V alue - A dded T ax
213
Taxation In Vietnam:
Who Pays What? VIII
By Jonathan Haughton
1. Introduction
This paper has two purposes. The first is to document the incidence of taxation on households
in Vietnam, using recent data, from 2006. The most recent publicly available study uses
data from 1998 (Haughton, Quan, and Bao 2006), and so does not take into account the
rapid changes in the tax system and in household spending and income patterns since then.
Information on who bears the burden of different taxes is helpful to policy makers when
they consider changes in the structure of taxes, and so will be useful as Vietnam reshapes and
modernizes its tax system over the coming few years.
The second, and more important, purpose of this study is to trace the implications of tax
incidence for the design of the personal income tax in particular, and the tax system in general.
In December 2007, the National Assembly approved a major reform of personal income
taxation, to come into effect in January 2009. Here we evaluate the revenue and incidence
effects of this change, and situate the discussion in the wider context of tax reform.
The next section summarizes the current structure of taxes in Vietnam. Direct taxes fall on
income, and indirect taxes are imposed on expenditures, and so an understanding of tax
incidence calls for an appreciation of how the different components of income and expenditure
vary as one goes from poor to rich households, or from region to region; this is addressed in
section 3. In section 4 we then present detailed information, both graphically and in tabular
format, on the incidence of several important taxes, and compare the relative progressivity of
each. We provide an analysis of the revenue and incidence effects of proposed changes in the
personal income tax in section 5; in the concluding section we assess the implications of our
findings for the direction of tax reform.
215
T ax R eform in V ietnam
The third major source of tax revenue is income tax. The revenue collected by the corporate
income tax peaked at over 10 percent of GDP in 2006; much of this revenue comes from
the state-owned industrial sector, whose share of GDP is now shrinking (EIU 2008). The
personal income tax yields just over 2 percent of government revenue, a share that has changed
little over the past decade.
In table 8.1, the numbers for 2007 are preliminary and the figures for 2008 are estimates.
However, if they can be believed, Vietnam will face some difficulty in maintaining its current
level of revenue mobilization (relative to GDP), with anticipated relative declines in revenue
from corporate income tax, and taxes on trade. The decline of oil production in the middle
of the decade did not help tax revenue attributable to oil (as a fraction of GDP) peaked in
2005, although some recovery has occurred since.
Table 8.1: Revenue from Taxes and Fees, 19972008
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 (e)
trillions of VND, current prices
Total revenues and grants 65.4 73.0 78.5 90.7 103.9 121.7 153.0 191.3 217.1 264.2 287.9 323.0
of which:
Taxes
Corporate income tax 11.6 13.1 14.5 22.2 25.8 29.3 47.4 57.0 71.7 100.8 98.9 108.1
Individual income tax 1.5 1.8 1.9 1.8 2.1 2.3 3.0 3.5 4.2 5.2 6.9 8.1
Capital user charge 1.5 1.7 1.5 1.6 1.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Land and housing tax 0.3 0.3 0.3 0.4 0.3 0.3 0.4 0.4 0.5 0.6 0.6 0.7
Licence tax 0.4 0.3 0.4 0.4 0.4 0.4 0.8 0.7 0.7 0.8 0.8 0.9
Tax on the transfer of properties 1.0 1.0 1.0 0.9 1.2 1.3 1.8 2.6 2.8 3.4 4.5 5.2
Tax on land use right transfer 0.3 0.4 0.3 0.2 0.3 0.3 0.4 0.6 1.0 1.3 1.7 2.0
Value added tax 11.8 11.8 17.2 17.2 19.3 25.9 33.1 38.8 45.7 54.8 78.6 96.7
Special cons. tax for domestic 4.6 5.6 4.5 5.3 6.2 7.3 8.9 12.8 15.7 17.1 17.0 19.9
Natural resouces tax 3.4 3.3 4.6 7.5 8.4 8.5 9.7 17.4 21.9 20.2 19.1 19.6
Agricultural tax 1.7 2.0 2.0 1.8 0.8 0.8 0.2 0.1 0.1 0.1 0.1 0.1
Imp - Exp. tax, special cons. tax on imports 13.5 14.9 14.4 13.4 17.5 21.9 22.4 21.6 23.6 26.3 25.0 26.2
Other taxes 1.5 2.9 2.2 0.3 0.3 0.4 0.2 0.0 0.0 0.0 0.0 0.0
Fees, charges and non-tax
From discrepancy of import prices
Fees and charges (include gasoline fee) 7.1 7.0 7.3 12.3 13.6 15.8 16.6 23.4 22.1 24.5 9.0 9.9
Rental of land 0.5 0.5 0.6 0.6 0.6 0.5 0.5 1.0 1.0 1.6 1.8 1.7
Others 2.1 4.3 3.4 2.8 3.4 4.4 4.6 8.5 3.8 3.9 2.7 2.9
Capital revenues (revenues from sale of State
- owned houses, land use right assignment) 17.7 17.6
Grants 2.6 2.1 2.4 2.0 2.0 2.3 3.0 2.9 2.3 3.6 3.4 3.6
Memo items:
GDP, in current prices 313.6 361.0 399.9 441.6 481.3 535.8 613.4 715.3 839.2 973.8 1143.6 1358.7
GDP, in 1994 prices 231.3 244.6 256.3 273.7 292.5 313.2 336.2 362.4 393.0 425.1 461.0 502.5
Imports, in current prices 142.5 159.7 164.7 226.7 238.8 301.7 391.7 503.2 583.0 718.0 977.5 1459.9
as percentages of GDP
Total revenue and grants 20.9 20.2 19.6 20.5 21.6 22.7 24.9 26.7 25.9 27.1 25.2 23.8
of which:
Corporate income tax 3.70 3.63 3.63 5.03 5.36 5.47 7.73 7.97 8.54 10.35 8.64 7.96
Individual income tax 0.48 0.50 0.48 0.41 0.44 0.43 0.49 0.49 0.50 0.53 0.60 0.60
Value-added tax 3.76 3.27 4.30 3.89 4.01 4.83 5.40 5.42 5.45 5.63 6.87 7.11
Excise taxes 1.47 1.55 1.13 1.20 1.29 1.36 1.45 1.79 1.87 1.76 1.49 1.46
Natural resources tax 1.08 0.91 1.15 1.70 1.75 1.59 1.58 2.43 2.61 2.07 1.67 1.44
Taxes on trade 4.30 4.13 3.60 3.03 3.64 4.09 3.65 3.02 2.81 2.70 2.19 1.93
Fees, charges, and non-tax revenues 3.09 3.27 2.83 3.55 3.66 3.86 3.54 4.60 3.21 3.08 2.73 2.36
Grants 0.01 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
All other taxes 2.96 2.96 2.52 1.72 1.45 1.08 1.10 1.02 0.88 1.00 0.98 0.91
Memo item:
Revenues from oil n.a. 4.0 4.6 6.5 7.4 6.8 7.0 8.0 8.7 8.2 6.3 4.8
Non-oil revenues n.a. 16.2 15.0 14.1 14.2 15.9 17.9 18.8 17.1 18.9 18.9 18.9
Memo item:
Oil revenue as % of total revenue 20.0 23.6 31.4 34.4 29.8 28.2 29.7 33.8 30.3 24.9 20.3
Import duties as % of total import value 9.5 9.3 8.7 5.9 7.3 7.3 5.7 4.3 4.0 3.7 2.6 1.8
Sources. IMF (2003, 2006, 2007) for figures to 2006; Ministry of Finance web site for 2007 and 2008(e) figures; EIU (2008) for growth rate for 2007.
Notes. Figures for 2007 are revised estimates. Excise taxes are those levied on domestically produced goods; excise collected on imports is included in
trade taxes. GDP figures for 2008 assume 11% inflation and 7% growth in real GDP. Estimates of GDP for 2007 assume 8.3% inflation between 2006 and
2007. Revenue figures for 2007 are revised estimates.
216
T axation I n V ietnam : W ho P ays W hat ?
217
T ax R eform in V ietnam
The most important direct tax is the value-added tax (VAT), which is levied at a rate of 10
percent on most items; however, some goods face a lower rate of 5 percent, others (such as
exports) pay 0 percent, and some important items are exempt (such as some basic unprocessed
foods). We assume that the VAT is passed on to consumers in the form of higher prices;
thus, if we know how much households consume of each good or service, we can apply the
appropriate VAT rate and infer how much tax the household is effectively paying.
In addition to the VAT, there are excise (special consumption) taxes on a number of important
goods, including cigarettes (now taxed at 65 percent), beer (75 percent on canned beer, 40
percent on draught beer), automobiles (50 percent on sedan cars, for instance), gasoline, air
conditioners, dance halls, and betting. Here again we assume that the effective incidence of
the tax is on the consumer, even when the tax is paid formallythe statutory incidenceby
the vendor.
Table 8.2: Summary of Main Taxes in Vietnam
218
T axation I n V ietnam : W ho P ays W hat ?
Import duties Levied on cif price, average tariff is Rates vary from 0% to 60%, with most
about 8%. Some exemptions for aid, in the 1%, 3%, 5%, 10%, 15% brackets.
goods in transit, education, research,
for export processing, and certain
machinery & equipment.
Export duties Levied on a few items only. Oil: 4%. Wood: 5 to 20%. Cashews:
4%.
Property taxes Agriculture and land use tax is being Computed as 50 to 650 kg. paddy/ha
phased out; exempts barren land, but paid in cash equivalent.
reclaimed land, and households in
difficulty.
Land and housing tax. As for
agriculture and land use tax, but higher
rates.
Land use right transfer tax. 2% for agricultural and fishing land,
4% otherwise.
Registration fees Applied to boats, cars, motorbikes, and 0.5% to 2%.
guns.
Sources: Haughton, Quan and Bao 2006; IMF 2007; Shukla 2006.
In this study we do not attempt to track the incidence on households of import duties, export
taxes, resource taxes, property transfer taxes, land taxes, or taxes levied on incorporated
businesses. In most cases the reason is pragmatic: adequate data are not available.
We are thus able to track the incidence on households of slightly over half of all tax revenue.
But before presenting these results, it is helpful to set out in more detail the components of
income and expenditureand hence, the tax bases of the various taxes under consideration.
219
T ax R eform in V ietnam
has considerable experience in collecting good quality survey data. The response rate to the
survey is believed to be over 90 percent, even in urban areas, but detailed information on this
aspect of the survey was not collected.
Table 8.3 provides a breakdown of the sources of income by income per capita deciles. The
procedure that we followed was first to create a measure of income based on the survey data.
Most of the sources of income are straightforward; however, the methods used to measure
income from durable goods housing, merit a few further comments. Although there are a
few renters, most Vietnamese households own their homes. This should be seen as an asset
that generates income; the homeowner is paying rent to himself, and the imputed value of
the services of the housewhich we took to be 2 percent of the reported valueis included
both as expenditure and income. The inclusion of imputed rent in household income has
been recommended by the United Nations since 1977 (Yates 1994); the need to include it as
a component of household expenditure is also widely recognized (Deaton and Zaidi 2002).
Similar reasoning applies to durable goods. A motorbike, for instance, is used up
(consumed) over a number of years, and so rather than include the full purchase price as
part of expenditure in the year in which it was bought, it is more appropriate to spread the
value of the consumption over the expected useful life of the asset. Thus, for each major asset
we compute average depreciation rates from the available survey data, adjust for inflation,
and assume a 3 percent real interest rate, to arrive at the rental equivalent of durable goods. It
might be argued that this rental equivalent should also be counted as part of the households
incomeit is paying the rental of the motorbike to itselfbut this is not quite so: the
income is the rental receipt less the depreciation of the asset, so only the interest component
of the rental should be counted in income. By analogy, if the money had been put into an
interest-bearing bank account, we would count that interest in income.
Our results are not particularly sensitive to the choice of parameters here; if housing services
are imputed using a rate of 3 percent instead of 2 percent, and if durable goods use an interest
rate of 5 percent rather than 3 percent, then total expenditure would be about 8 percent higher
and measured income would be 10 percent higher, and both would be distributed somewhat
more unequally.
Since prices vary by region and by month of the year, we adjusted income using price deflators
supplied by the General Statistics Office (GSO). Unless otherwise indicated, all numbers
related to the survey are expressed in average prices for 2006; they thus can be expected to
differ slightly from the GSO figures, which are expressed in the prices of January 2006. The
price-adjusted measures of household income were then divided by household size to obtain
income per capita, and households were then sorted from poorest to richest into ten deciles,
each of which contains an equal number of individuals.
It is not surprising that as one household moves from low- to high-income household, the
share of income from agriculture, livestock, and forestry falls, from 53 percent in the bottom
decile to 10 percent in the top decile. This pattern has a simple corollary: taxes on the
agricultural sector will tend to fall more heavily on poorer households.
220
T axation I n V ietnam : W ho P ays W hat ?
Table 8.3: Sources of Household Income for Vietnam Households, by Income Per
Capita Decile, 2006
Income per capita deciles
1 (poor) 2 3 4 5 6 7 8 9 10 (rich) All
percentage of total income from each source of income
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
of which
Wages and salaries 23.8 27.4 27.6 28.0 29.3 28.5 30.8 30.3 30.6 27.4 28.4
Agriculture 36.0 31.3 28.0 23.6 20.5 19.3 15.3 14.4 10.1 7.2 20.6
Livestock 9.1 9.0 8.0 7.9 7.1 6.0 5.3 4.3 3.2 2.2 6.2
Forestry 7.5 3.1 2.3 1.5 1.2 0.8 0.6 0.4 0.2 0.4 1.8
Aquaculture 2.0 3.2 3.2 4.0 4.1 3.0 3.2 2.4 2.9 3.2 3.1
Household enterprises 3.6 7.7 11.3 14.4 15.7 18.9 20.3 20.6 20.3 21.4 15.4
Durable goods 0.8 0.9 1.2 1.1 1.2 1.2 1.4 1.4 1.4 1.5 1.2
Housing 5.0 5.5 6.0 6.5 7.1 7.8 8.6 10.1 13.0 15.8 8.5
Subsidies, ed & hlth 1.8 1.3 1.2 1.3 1.0 1.1 0.9 0.9 0.9 0.9 1.1
Other income 10.5 10.6 11.3 11.7 12.7 13.2 13.4 15.0 17.3 18.4 13.7
of which:
foreign remittances 0.2 0.4 0.7 0.8 1.0 1.0 1.6 2.0 3.9 5.8 1.7
domestic remittances 6.4 6.7 6.3 6.3 6.8 7.0 6.2 6.0 5.4 5.8 6.3
Notes: Agriculture includes agricultural services; forestry includes hunting; "other income" includes rent, remittances, interest, and transfer payments such as pensions.
The average exchange rate in 2006 was VND15,994/US$.
Notes: Agriculture includes agricultural services; forestry includes hunting; other income includes rent, remittances,
interest, and transfer payments such as pensions.
The average exchange rate in 2006 was VND15,994/US$.
More unexpectedly, the importance of wages and salaries in total income does not vary
systematically with per capita income. This is in contrast with income from household
businesses, which constitutes just 4 percent of the income of those in the poorest decile
but a fifth of the income for those in the top four deciles. A flat tax on household business
income, such as the corporate income tax, is therefore likely to be progressive, in the sense of
representing a proportionately greater tax burden on rich households than on poor.
A breakdown of household expenditure is shown in table 8.4. In this case, households are
sorted by real expenditure per capitathat is, expenditure deflated for differences in the
prices of main commodities across regions and over months of the year. Engels Law, which
states that as real incomes rise, the share devoted to buying food falls, is clearly evident: the
poorest households devote over three fifths of their spending to food, compared with about a
quarter for those in the top quintile.
The other striking differences in the consumption patterns of the poor and the rich is that
more affluent households spend relatively more heavily on housing, utilities, and durable
goods; thus, flat taxes on these groups of goods and services will be inherently progressive in
nature.
221
T ax R eform in V ietnam
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
of which:
Food 60.8 55.2 51.7 47.7 46.4 43.0 40.4 37.8 33.4 26.9 44.3
Non-food items 21.6 22.7 23.2 24.1 24.4 25.2 25.9 26.5 25.8 26.2 24.6
Durable goods 3.9 5.6 6.8 8.0 8.3 9.5 9.6 9.9 11.7 12.3 8.6
Housing 5.2 6.0 7.0 7.9 8.6 9.1 10.6 12.2 16.3 22.7 10.6
Utilities 1.8 2.1 2.3 2.3 2.3 2.4 2.5 2.5 2.9 3.0 2.4
Health 3.3 4.0 4.6 4.9 5.1 5.4 5.5 5.8 5.0 4.4 4.8
Education 3.4 4.5 4.4 5.1 4.9 5.3 5.4 5.2 5.0 4.4 4.8
Also:
Home production 34.1 24.7 19.7 16.6 13.5 10.1 7.4 5.1 2.8 1.4 13.5
Purchases 65.9 75.3 80.3 83.4 86.5 89.9 92.6 94.9 97.2 98.6 86.5
Notes: Non-food items include daily non-food purchases, annual purchases, and "other household expenditure". Durable goods refer to the rental value of durable
goods. Housing measures actual rent paid, or imputed rent (in the case of owners), plus home repairs. Utilities comprise spending on water, electricity, and garbage
collection. Average exchange rate in 2006 was VND15,994/US$.
For completeness, a breakdown of income and expenditure by region is shown in table 8.5,
along with a comparison between urban and rural areas. Region 1 is the Red River Delta,
centered on Hanoi, while Region 7 is the Southeast, centered on Ho Chi Minh City. These
two relatively affluent areas stand out: there is relatively less spending on food, and more
on housing; and income is more likely to come from household business and less likely to
originate in agriculture. The urban-rural split shows a similar pattern. The Central Highlands
(region 6) is noteworthy in that income is heavily dependent on agriculture, but unlike the
other poor hilly areas (regions 2 and 3 in the northeast and northwest), households in region
6 do not rely much on home-produced food. This reflects the importance of cash cropping,
particularly coffee, in the Central Highlands.
222
T axation I n V ietnam : W ho P ays W hat ?
Table 8.5: Breakdown of Household Expenditure and Income, by Region and Urban
Rural, 2006
Regions
1 2 3 4 5 6 7 8 Urban Rural All
% of total expenditure devoted to each expenditure heading
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
of which:
Food 40.5 49.3 54.9 46.5 44.0 47.0 37.8 47.0 36.5 47.2 44.3
Non-food items 22.0 25.1 23.7 23.0 24.3 25.0 25.2 27.5 23.7 24.9 24.6
Durable goods 9.6 7.9 6.9 8.6 9.6 8.3 8.7 7.6 10.1 8.0 8.6
Housing 14.4 7.5 7.1 8.2 9.1 7.5 16.1 7.4 17.3 8.1 10.6
Utilities 2.7 1.9 1.3 2.2 2.4 1.8 3.1 2.3 3.3 2.1 2.4
Health 5.7 4.4 2.9 6.6 5.9 5.3 4.5 2.8 5.0 4.7 4.8
Education 5.0 3.9 3.1 5.0 4.6 5.1 4.3 5.5 4.2 5.0 4.8
Also:
Home production 13.3 26.4 36.0 18.9 10.3 14.1 3.5 8.5 2.7 17.5 13.5
Purchases 86.7 73.6 64.0 81.1 89.7 85.9 96.5 91.5 97.3 82.5 86.5
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
of which
Wages and salaries 29.0 23.0 17.8 21.9 34.0 23.1 36.9 29.1 36.1 25.5 28.4
Agriculture 16.4 25.6 39.3 23.3 15.3 41.4 10.6 21.3 5.0 26.2 20.6
Livestock 6.8 10.7 9.2 9.3 5.6 5.6 3.0 3.4 1.5 7.9 6.2
Forestry 0.0 5.9 8.4 3.0 1.8 2.0 0.3 0.6 0.2 2.4 1.8
Aquaculture 1.6 1.9 2.0 3.3 3.1 0.6 1.5 7.5 1.2 3.8 3.1
Household enterprises 16.4 10.7 6.7 12.8 17.9 10.8 19.7 17.1 23.9 12.3 15.4
Durable goods 1.3 1.1 1.1 1.3 1.4 1.1 1.2 1.1 1.4 1.1 1.2
Housing 12.0 5.9 5.7 6.6 7.5 5.9 13.1 5.7 14.5 6.4 8.5
Subsidies, ed & hlth 1.2 1.1 1.2 1.3 1.4 1.4 0.9 0.9 1.0 1.2 1.1
Other income 15.2 14.0 8.7 17.1 12.0 8.0 13.0 13.4 15.3 13.1 13.7
of which:
foreign remittances 1.3 1.2 0.3 1.9 1.7 0.8 2.9 2.1 2.4 1.5 1.7
domestic remittances 6.2 5.3 4.0 7.3 6.1 4.4 6.2 7.4 5.6 6.6 6.3
Notes: Non-food items include daily non-food purchases, annual purchases, and "other household expenditure". Durable goods refer to the rental value of durable
goods (for expenditure) or the interest on durable goods (for income). Housing measures actual rent paid, or imputed rent (in the case of owners), plus home repairs.
Utilities comprise spending on water, electricity, and garbage collection. Average exchange rate in 2006 was VND15,994/US$. Agriculture includes agricultural
Note: The regions are: 1: Red River Delta, 2: Northeast, 3: Northwest, 4: North Central Coast, 5: South Central Coast, 6: Central Highlands, 7: Southeast, 8: Mekong
River Delta.
A technical point is worth noting at this point. The measures of expenditure used in this study
were built using the detailed data of the VHLSS-2006. This was necessary in order to be able
to calculate the incidence of taxes, where one often has to apply different tax rates to detailed
expenditure (or income) data. But there are some differences between our measures, and the
summary measures produced by the GSO, as set out in table 8.6. Our measures of nonfood
spending, the services of durable goods, and housing are all substantially larger than the GSO
numbers, and the difference is too large to be due to the increase in consumer prices between
January 2006 (the GSO price base period) and the average for 2006 (our base price period).
These discrepancies are probably mainly due to differences in the assumptions used to impute
the values of housing and durable goods services; clearly even the esoteric assumptions about
interest rates and depreciation play a role here. However, the net effect of these differences on
tax incidence is relatively modest.
223
T ax R eform in V ietnam
Table 8.6: Spending Comparison (in VND 000 per Household per Year, 2006)
4. Tax Incidence
The essential results on tax incidence are shown in tables 8.7 and 8.8, and the associated
figures 8.1 and 8.2. As explained above, we have assumed that the ultimate weight of all these
taxes falls on households.
For the value-added tax, excise taxes, and personal income tax, the amounts reported in tables
8.7 and 8.8 are obtained by multiplying the tax rates by the relevant bases. Thus these are
measures of the hypothetical amount of tax paid; the actual amount of tax collected may be
higher (if households underreport income or spending) or lower (if there is tax evasion).
On the other hand, the amounts paid in fees and in household enterprise taxes are the actual
values reported by households in the VHLSS-2006.
The difference between table 8.7 and table 8.8 is that the former ranks households by real
expenditure per capita, while table 8.8 sorts households by real income per capita. It turns out
that in the case of Vietnam it does not make a lot of difference which one chooses, a reflection
of the relatively close correlation between household income and expenditure in the country;
this is in contrast with what has been observed in Latin America or the United States, where
the pattern of tax incidence can look very different depending on whether households are
ranked by income or expenditure (see Haughton, Quan, and Bao 2006 for an example).
The most important finding is that the tax system is progressive overall: more affluent
households, whether measured by income or by expenditure, generally face a proportionately
higher tax burden than poorer households. This may also be seen at a glance in the top left
panels of figures 8.1 and 8.2; as one moves from left to right, equivalent to moving from poor
to rich, the proportion of spending (or income) going to pay taxes generally rises.
This pattern differs somewhat from that found by Haughton, Quan, and Bao (2006) for 1998
in two respects: the tax burden is now heavier, and it is more progressive. The earlier study
224
T axation I n V ietnam : W ho P ays W hat ?
found that taxes represented about 8 percent of spending for all but the top quintile, where
it stood at just over 10 percent. The numbers for 2006 show a tax burden of about 8 percent
only for the bottom two deciles, rising steadily to reach a plateau of about 14 percent in the
four top deciles. The higher, more progressive, tax structure is likely due in large part to the
introduction of the VAT in 1999, when it replaced a complex system of turnover taxes.
Of the taxes considered in tables 8.7 and 8.8, by far the largest is the VAT. This tax is generally
progressive, representing 4.5 percent of the expenditure of households in the bottom
expenditure decile, compared to 7.0 percent for those in the top decile. The main reason
appears to be that richer households are more likely to buy the goods and services they need
incurring VAT in the processin contrast to poor households; in the poorest decile, as much
as a third of household spending is on home production, which is untaxed, compared to little
more than 1 percent for those in the top decile. On the other hand, the VAT rate drops off in
the highest decile, due mainly to the relatively high savings rate for this group, which means
that a substantial fraction of income escapes all taxes on consumption spending. Although
VAT is not levied on unprocessed agricultural products, it is applied to some inputs such as
fertilizers, and to processed products; we were able to take such details into account by using
the disaggregated expenditure data from the VHLSS survey.
The excise tax is also progressive, and this is also somewhat surprising, as Haughton, Quan, and
Bao (2006) found this tax to be approximately proportional in 1998. The main explanation
seems to be the rise in spending on transportation, most notably motorbikes and the gasoline
required to run them, both of which are subject to excise taxes and are disproportionately
owned by richer households. For instance, although 59 percent of households owned at least
one motorbike in 2006, the proportion varied from 21 percent in the poorest (expenditure)
decile to 93 percent in the top decile. Households in the top tenth of the expenditure
distribution owned an average of 1.7 motorbikes, compared to 0.22 per household in the
poorest tenth.
Not all taxes are progressive. Agricultural fees, while modest overall, are relatively burdensome
for the lower-middle part of the expenditure distribution. Educational fees exceed 1 percent
of spending for all groups, but come close to 2 percent of spending for those in the middle-
upper expenditure brackets.
225
T ax R eform in V ietnam
Household expenditure per capita 2,000 2,938 3,640 4,324 5,077 5,981 7,149 8,811 11,593 22,681 7,417
Total tax paid 156 252 372 458 557 709 984 1,151 1,679 3,146 946 100.0
of which:
value-added tax 90 144 212 263 319 420 525 662 1,007 1,577 522 100.0
excise taxes 13 23 30 38 47 60 73 93 141 333 85 97.3
educational fees 22 37 58 73 88 112 130 168 191 264 114 62.1
agricultural fees 12 22 31 37 33 34 28 28 23 12 26 40.2
other fees 16 22 29 34 41 48 65 83 133 247 72 99.0
taxes on household enterprises 3 3 12 12 30 35 163 116 184 651 121 15.4
personal income tax - - - - - - - - 1 61 6 0.2
as percentage of expenditure
Total tax paid 7.8 8.6 10.2 10.6 11.0 11.9 13.8 13.1 14.5 13.9 13.4
of which:
value-added tax 4.5 4.9 5.8 6.1 6.3 7.0 7.3 7.5 8.7 7.0 7.0
excise taxes 0.6 0.8 0.8 0.9 0.9 1.0 1.0 1.1 1.2 1.5 1.1
educational fees 1.1 1.3 1.6 1.7 1.7 1.9 1.8 1.9 1.6 1.2 1.5
agricultural fees 0.6 0.8 0.9 0.9 0.7 0.6 0.4 0.3 0.2 0.1 0.4
other fees 0.8 0.8 0.8 0.8 0.8 0.8 0.9 0.9 1.1 1.1 1.0
taxes on household enterprises 0.1 0.1 0.3 0.3 0.6 0.6 2.3 1.3 1.6 2.9 1.6
personal income tax - - - - - - - - 0.0 0.3 0.1
Notes. Revenue from value-added tax, excise taxes, and personal income tax, are estimated based on spending and income patterns. Results are
preliminary, and based on the VHLSS survey of 2006. All figures are in prices of 2006.
Table 8.8: Tax Paid by Vietnamese Households in 2006, by Income Per Capita Deciles
Income per capita deciles
1 (poor) 2 3 4 5 6 7 8 9 10 (rich) All % hh
paying
Income per capita 2,370 3,511 4,387 5,279 6,268 7,508 9,076 11,118 14,660 29,343 9,351
Total tax paid 174 290 372 460 558 684 888 1,098 1,528 3,409 946 100.0
of which:
value-added tax 96 153 201 257 309 376 494 627 886 1,818 522 100.0
excise taxes 15 25 33 40 48 57 83 100 130 319 85 97.3
educational fees 30 54 68 84 99 113 130 162 185 217 114 62.1
agricultural fees 14 26 29 29 29 31 29 26 22 26 26 40.2
other fees 18 25 29 33 44 54 69 90 137 220 72 99.0
taxes on household enterprises 1 6 11 17 28 54 83 93 167 749 121 15.4
personal income tax - - - - - - - - - 62 6 0.2
as percentage of income
Total tax paid 7.4 8.3 8.5 8.7 8.9 9.1 9.8 9.9 10.4 11.6 10.1
of which:
value-added tax 4.0 4.4 4.6 4.9 4.9 5.0 5.4 5.6 6.0 6.2 5.6
excise taxes 0.7 0.7 0.8 0.8 0.8 0.8 0.9 0.9 0.9 1.1 0.9
educational fees 1.3 1.5 1.6 1.6 1.6 1.5 1.4 1.5 1.3 0.7 1.2
agricultural fees 0.6 0.7 0.7 0.6 0.5 0.4 0.3 0.2 0.2 0.1 0.3
other fees 0.8 0.7 0.7 0.6 0.7 0.7 0.8 0.8 0.9 0.7 0.8
taxes on household enterprises 0.0 0.2 0.3 0.3 0.4 0.7 0.9 0.8 1.1 2.6 1.3
personal income tax - - - - - - - - - 0.2 0.1
Notes. Revenue from value-added tax, excise taxes, and personal income tax, are estimated based on spending and income patterns. Results are
preliminary, and based on the VHLSS survey of 2006. All figures are in prices of 2006.
The two direct taxes are highly progressive, as the bottom panels of figures 8.1 and 8.2 shows.
The tax on business income mainly hits the top four deciles; and the personal income tax is
borne almost entirely by households in the top decile. A word of caution is in order here;
226
T axation I n V ietnam : W ho P ays W hat ?
despite a sample of 9,189 households, only 14 people reported paying personal income tax.
This represents just 0.15 percent of the sample.
Figure 8.1: Incidence of Taxes by Expenditure Per Capita Deciles in Vietnam, 2006
Tax as a % of expenditure, by decile, 2006 VAT as a % of expenditure, by decile, 2006
Taxes paid as % of expenditure
Overall VAT
Excise Tax as a % of expenditure, by decile, 2006 Educational Fees as a % of expenditure, by decile, 2006
Taxes paid as % of expenditure
1.6 2.5
1.4 Taxes paid as % of expenditure 2.0
1.2
1.0 1.5
0.8
0.6 1.0
0.4 0.5
0.2
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Expenditure per capita decile (poor = 1, rich = 10) Expenditure per capita decile (poor = 1, rich = 10)
Agricultural Fees as a % of expenditure, by decile, 2006 Other Fees as a % of expenditure, by decile, 2006
Taxes paid as % of expenditure
1.0 1.4
0.9 1.2
0.8
0.7 1.0
0.6 0.8
0.5
0.4 0.6
0.3 0.4
0.2 0.2
0.1
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Expenditure per capita decile (poor = 1, rich = 10) Expenditure per capita decile (poor = 1, rich = 10)
3.5
Taxes paid as % of expenditure
0.3
3.0
0.3
2.5
0.2
2.0
1.5 0.2
1.0 0.1
0.5 0.1
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Expenditure per capita decile (poor = 1, rich = 10) Expenditure per capita decile (poor = 1, rich = 10)
227
T ax R eform in V ietnam
Figure 8.2: Incidence of Taxes by Expenditure Per Capita Deciles in Vietnam, 2006
Tax as a % of income, by decile, 2006 VAT as a % of income, by decile, 2006
Taxes paid as % of income 14.0 7.0
Overall VAT
Excise tax as a % of income, by decile, 2006 Educational Fees as a % of income, by decile, 2006
Taxes paid as % of income
Income per capita decile (poor = 1, rich = 10) Income per capita decile (poor = 1, rich = 10)
0.6 0.8
0.7
0.5 0.6
0.4 0.5
0.3 0.4
0.2 0.3
0.2
0.1 0.1
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Income per capita decile (poor = 1, rich = 10) Income per capita decile (poor = 1, rich = 10)
2.5
0.2
2.0
0.2
1.5
1.0 0.1
0.5 0.1
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Income per capita decile (poor = 1, rich = 10) Income per capita decile (poor = 1, rich = 10)
228
T axation I n V ietnam : W ho P ays W hat ?
Table 8.9: Tax Paid by Vietnamese Households in 2006, by Region and Urban/
Rural Residence
Tax paid Expenditure Tax as % of Memo: VAT as Income # of households
per capita expenditure % of expend. per capita sampled
Note: The regions are: 1: Red River Delta, 2: Northeast, 3: Northwest, 4: North Central Coast, 5: South Central
Coast, 6: Central Highlands, 7: Southeast, 8: Mekong River Delta.
It is helpful to summarize the progressivity of taxes, and this can be done with the help of
table 8.10. The first two rows of that table measure the inequality of per capita expenditure
and income respective, using the Gini coefficient. This is obtained by sorting individuals
from poorest to richest along a horizontal axis, and graphing their cumulative expenditure
(or income) on the vertical axis, to create the Lorenz curve. Denote by A the area between
this curve and a line of equality, and by B the area under the Lorenz curve. Then the Gini
coefficient is given by A/(A+B); it varies between 0 (perfect equality) and 1 (complete
inequality);
For expenditure, the Gini coefficient is 0.390, while for income per capita it is 0.403. It is
common for income inequality to be greater than expenditure inequality. These numbers
reflect moderate inequality. Official calculations show an expenditure (per capita) Gini
coefficient of 0.358; the discrepancy appears to be due to differences in the assumptions made
when constructing expenditure (see table 8.7), most notably in housing, durable goods, and
nonfood spending.
The bottom part of table 8.10 reports concentration ratios (or quasi-Gini coefficients) for the
various taxes. These are constructed in a manner similar to the Gini coefficient: households
are still sorted by per capita expenditure on the horizontal axis, but now the vertical axis
graphs the cumulative proportion of tax paid, forming a quasi-Lorenz curve. Unlike the
Gini coefficient, a concentration ratio could take on a negative number.
A high concentration ratio, meaning a value close to one, indicates that tax payments are
highly unequalindeed, they are especially concentrated among higher-income individuals.
Such a tax is therefore progressive. Kakwani has suggested that the difference between the
concentration ratio and the Gini coefficient serve as a measure of the progressivity of a tax:
if tax payments are more unequal than per capita expenditure, then the tax is progressive and
the Kakwani measure is positive.
229
T ax R eform in V ietnam
The Kakwani measure for all the taxes we have considered here is 0.071, which denotes a
moderate degree of progressivity. In table 8.10 we display the remaining taxes in order, starting
with the most progressive and ending with the most regressive. These numbers confirm the
patterns seen in figures 8.1 and 8.2: the income taxes are highly progressive, excises and VAT
are somewhat progressive, and educational and (especially) agricultural fees are regressive.
Table 8.10: Measures of Tax Progressivity
It is not easy to reconcile the tax revenue estimated on the basis of actual or imputed data
from the VHLSS-2006, and the data provided by the national budget, but an attempt is made
in table 8.11. The first column of numbers shows the estimated tax payments per capita,
for the major tax categories. When multiplied by the population, which was 84 million in
2006, one gets the estimated levels of revenue shown in the middle column. These may be
compared with the budgetary data in the final column.
The VAT numbers based on the VHLSS-2006 data are understated. One plausible reason is
that substantial amounts of sales, which are subject to VAT, are made to outsiders, including
tourists and foreigners resident in Vietnamgroups that are not covered by the VHLSS.
230
T axation I n V ietnam : W ho P ays W hat ?
Another possibility is that households understate their actual spending, which is a common
feature of expenditure surveys. Similar reasoning applies to the understatement of excise
tax revenues, much of which consists of sin taxes on goods such as alcohol and tobacco,
notorious for being underreported in household survey data.
Based on the VHLSS-2006 numbers, the personal income tax shows a very modest amount
of total revenuejust VND500 billioncompared to a budget total of VND5.2 trillion. We
argue below that this is just about plausible, and reflects the fact that most of the revenues from
this tax come from collections on the wages and salaries of foreigners working in Vietnam. It
is also quite possible that the VHLSS-2006 did not adequately capture information on the
incomes of the very top earners in Vietnam.
Taxes collected from household enterprises contribute an estimated one tenth of total revenue
from the corporate income tax, which is plausible given that the latter mainly collects revenue
from larger corporations.
Perhaps the most interesting entry is the last one, which projects fees to yield a total of VND
17.8 trillion, compared to VND 3.9 trillion in the budget. The explanation is that most of
the revenue from fees and contributions goes to local bodiesschools, municipalities,
health clinics, and the likerather than into the central budget. These leviesand they are
numerouslack transparency, and little is known about the extent to which they are true user
fees, or how they are determined.
Table 8.11: Reconciling Survey with Budget Data
231
T ax R eform in V ietnam
We would like to know what effect the new income tax would have on revenue and on incidence,
and we approach this problem in two ways. First, we use the data from the VHLSS-2006: we
strip out the business income tax and current personal income tax, and then apply the rules
of the new tax. The results are shown in table 8.13, where we see that the estimated revenue
would fall from VND 127,000 per person to just VND 21,000 per person. The new tax would
fall almost entirely on those in the top decile, and so would be highly progressive.
However, this cannot be the final word on the matter, if only because two thirds of personal
income tax revenue is collected from foreigners, and the effect on them of the new tax
structure is not included here.
Some further insight into the problem may be gained by turning to data from another source.
In late 2005, the General Department of Taxation undertook a survey of high income
individuals. The sample of 15,500, representing about 5 percent of its tax rolls, included 3,200
foreign taxpayers, 7,200 Vietnamese paying the personal income tax, and 5,100 Vietnamese
paying business income tax. There were 11,532 responses to the survey, representing a 74
percent response rate. The survey used a relatively short questionnaire that asked about
income and expenditure, but not about taxes paid. An effort was made to assure respondents
that their answers to the questionnaire would not be used for tax collection purposes, but
it is not clear how effective this assurance was. Response rates were relatively modest for
foreigners, and in Ho Chi Minh City, but were high elsewhere; the results reported below
represent weighted averages that adjust for the differential response rates.
232
T axation I n V ietnam : W ho P ays W hat ?
Table 8.13: The Distributional Effects of Revising the Personal Income Tax Law
Expenditure per capita deciles
1 (poor) 2 3 4 5 6 7 8 9 10 (rich) All % hh
paying
thousands of VND p.a.
Household expenditure per capita 2,000 2,938 3,640 4,324 5,077 5,981 7,149 8,811 11,593 22,681 7,417
Total tax paid 156 252 372 458 557 709 984 1,151 1,679 3,146 946 100.0
of which:
taxes on household enterprises 3 3 12 12 30 35 163 116 184 651 121 15.4
personal income tax - - - - - - - - 1 61 6 0.2
Total new tax paid
on income 8 1 13 190 21 1.0
as percentage of expenditure
Total tax paid 7.8 8.6 10.2 10.6 11.0 11.9 13.8 13.1 14.5 13.9 13.4
of which:
taxes on household enterprises 0.1 0.1 0.3 0.3 0.6 0.6 2.3 1.3 1.6 2.9 1.6
personal income tax - - - - - - - - 0.0 0.3 0.1
Total new tax paid
on income - - - - - - 0.1 0.0 0.1 0.8 0.3
Notes. Revenue from value-added tax, excise taxes, and personal income tax, are estimated based on spending and income patterns. Based on the VHLSS
survey of 2006. All figures are in prices of 2006.
Income per capita 2,370 3,511 4,387 5,279 6,268 7,508 9,076 11,118 14,660 29,343 9,351
Total tax paid 174 290 372 460 558 684 888 1,098 1,528 3,409 946 100.0
of which:
taxes on household enterprises 1 6 11 17 28 54 83 93 167 749 121 15.4
personal income tax - - - - - - - - - 62 6 0.2
Total new tax paid
on income 212 21 1.0
as percentage of expenditure
Total tax paid 7.4 8.3 8.5 8.7 8.9 9.1 9.8 9.9 10.4 11.6 10.1
of which:
taxes on household enterprises 0.0 0.2 0.3 0.3 0.4 0.7 0.9 0.8 1.1 2.6 1.3
personal income tax - - - - - - - - - 0.2 0.1
Total new tax paid
on income - - - - - - - - - 0.7 0.2
Notes. Revenue from value-added tax, excise taxes, and personal income tax, are estimated based on spending and income patterns. Based on the VHLSS
survey of 2006. All figures are in prices of 2006.
The key data are shown in table 8.14, which divides respondents into three groups
Vietnamese and foreigners who pay personal income tax, and Vietnamese who pay business
income tax (here labeled household enterprises). The reported annual incomes of those
surveyed are relatively high, especially for foreigners, where the total came to VND 632
million (about US$40,000) in 2005. For foreigners, the most important source of income
by far was wages and salaries, while other household members added relatively little to total
household income.
In contrast, the Vietnamese taxpayers relied on income from a greater variety of sources, with
large contributions from business income and, in the case of those who paid personal income
tax, other family members.
The reported levels of spending in table 8.14 are implausibly low; even so, the low proportion
of income that expatriates spend inside Vietnam is quite striking, and plausible enough. The
Vietnamese taxpayers appear to spend heavily on assets rather than consumption. This is
233
T ax R eform in V ietnam
consistent with the ideawhich is at the heart of the permanent income hypothesisthat
many of those who are caught in the tax net in a given year have temporarily high incomes; in
order to smooth their consumption patterns, they spend a high proportion of the transitory
income on acquiring assets that will allow them to maintain higher levels of consumption into
the future.
At the bottom of table 8.14 we estimate the amount of revenue that should, in theory, have
been collected from these taxpayers in 2005. The personal income tax revenue should have
been VND 15.5 trn (= VND 14.849 trn + VND 0.648 trn), when it actually came to VND 4.2
trn. This is evidence that evasion is widespread under the current income tax.
The new income tax arrangements came into effect in 2009, at which point incomes and
prices (but not the tax brackets) were substantially higher than they were in 2008, or than
they were in 2005 at the time of this survey. So in order to simulate the effects of the new
income tax regime, we first project income out to 2009 for every taxpayer in the survey, and
then apply in turn the current tax rules, and the proposed tax rules. Note that the personal
income tax would be very elastic; although nominal GDP rose by an estimated 83 percent
between 2005 and 2009, personal income tax from Vietnamese taxpayers was expected to
rise by 253 percent over the same period even under current rules, making this tax very elastic
(see table 8.14).
The most important results are displayed near the bottom of table 8.14, and shows that the
theoretical amount of tax revenue collected from these high-income tax payers would rise
by 8 percent with the change in the income tax regime. This is an underestimate, because
it is computed on the basis of those who paid these taxes in 2005; by 2009 the number of
households caught in the tax net would be higher, reflecting growth both in incomes and the
total working population.
Although total revenue would rise, the amount paid by foreigners would fall by 15 percent
while the tax paid by Vietnamese would rise sharplyby 178 percent for personal income
taxpayers, and by 37 percent for those who pay household business tax. These higher tax
payments result mainly from the provision in the new law that would tax global income; when
wages and salaries, capital income, and business income are added together, many households
find themselves pushed quickly into relatively high tax brackets. These proportionate changes
assume that the degree of tax evasion does not change between 2005 and 2009, despite major
alterations in the tax code.
234
T axation I n V ietnam : W ho P ays W hat ?
Table 8.14: Income and Estimated Tax for High-Income Households, 2005
Vietnamese Foreigners Household Full sample % foreign
enterprises
thousands of VND per year
Memo items:
Total household income 370,619 671,127 266,619 399,276
Household current spending 79,273 116,266 39,741 74,014
Household asset purchases 255,995 38,100 59,914 145,971
If the simulation based on the VHLSS-2006 data is to be believed, the new income tax regime
was expected to reduce tax revenues sharply, by about 5 percent overall. But when we use the
information on high-income households, we conclude that the tax change would increase
government revenue. An attempt at reconciling these two contradictory conclusions is
summarized in table 8.15. Fewer people would pay personal income tax (table 8.13), and
the amount paid by high-income taxpayers would fall slightly, essentially because foreigners
would pay less (table 14), for a net loss of revenue, had the new system been in place in 2006,
of VND 0.5 trillion dong, as shown in the top row in table 8.15.
It is more difficult to reconcile the numbers for tax on household enterprise profits. Over
half of these profits are collected from households in the top decile, and fewer of these are
expected to have to pay the tax under the new integrated income tax regime (table 8.13);
business tax revenue from households in the lower deciles will almost evaporate (table 8.13),
as they are subject to the more generous personal income tax regime.75 On the other hand,
high-income households who pay business income tax would generally have to pay more, as
the base on which they have to pay tax is broadened (table 8.14). The net effect of all these
changes would be a reduction in tax revenue of VND8.2 billion, or about 3 percent of total
tax revenue in 2006.
75
The drop in revenue from taxes on household enterprise income is probably overstated, because
some of the taxes reported by households as business income tax may in fact represent other
business taxes, which would remain in effect. Unfortunately, we are unable to quantify this
effect.
235
T ax R eform in V ietnam
Table 8.15: Estimating the Net Effect on Tax Revenue of Personal Income Tax
Changes, 2006
5. Property Tax
Vietnam does not currently have a property tax. Calls for the introduction of such a tax are
becoming more common; proponents argue that such a tax would be both efficient (especially
the part that falls on land), and equitable. In this section we examine the effects of introducing
a property tax on nonagricultural real estate owned by households, levied at a rate of 1 percent
of the property value.
Our procedure is straightforward: we assume that there would be no evasion, and that the
property values reported by households are not systematically over-or understated. Then
we apply the tax rate to the values of real estate reported in the VLHSS of 2006, assume that
the tax is borne by owners, and derive the incidence numbers shown in tables 8.16 and 8.17.
We also compare these results with the case of a property tax that is levied at 1 percent but
would exempt the first VND100 million (US$6,250) worth of household property from the
tax.
The first point to note is that a 1 percent tax on nonagricultural household-owned real estate
is a heavy tax. It would represent, on average, 7.8 percent of household expenditure, or 6.2
percent of household income; that is about three-fifths of the total amount currently collected
directly and indirectly from households. It also represents a tax of perhaps 20 percent on
the income stream from property, assuming the latter yields a return of 5 percent. A tax of
this size would meet strong resistance, and so is best thought of as an upper bound to the
236
T axation I n V ietnam : W ho P ays W hat ?
(politically) feasible rate. Moreover, this burden supposes that revenue collection is efficient
and that there are no evasions which are rather optimistic assumptions.
Table 8.16: The Distributional Effects of Introducing a Tax on Nonagricultural
Real Estate
Expenditure per capita deciles
1 (poor) 2 3 4 5 6 7 8 9 10 (rich) All % hh
paying
thousands of VND p.a.
Household expenditure per capita 2,000 2,938 3,640 4,324 5,077 5,981 7,149 8,811 11,593 22,681 7,417
Total tax paid 50 82 119 168 210 277 384 545 1,028 2,944 580 97.9
of which:
taxes on residence 49 80 113 157 200 257 357 500 922 2,587 522 97.5
taxes on other property 1 2 5 10 10 20 26 46 106 358 58 8.9
Total tax paid
with VND100m threshhold per household 1 4 12 44 62 109 197 335 798 2,677 424 35.8
as percentage of expenditure
Total tax paid 2.5 2.8 3.3 3.9 4.1 4.6 5.4 6.2 8.9 13.0 7.8
of which:
taxes on residence 2.4 2.7 3.1 3.6 3.9 4.3 5.0 5.7 8.0 11.4 7.0
taxes on other property 0.1 0.1 0.1 0.2 0.2 0.3 0.4 0.5 0.9 1.6 0.8
Total tax paid
with VND100m threshhold per household 0.1 0.1 0.3 1.0 1.2 1.8 2.8 3.8 6.9 11.8 5.7
Notes. Revenue from value-added tax, excise taxes, and personal income tax, are estimated based on spending and income patterns. Based on the VHLSS
survey of 2006. All figures are in prices of 2006.
Income per capita 2,370 3,511 4,387 5,279 6,268 7,508 9,076 11,118 14,660 29,343 9,351
Total tax paid 56 96 132 178 230 315 430 626 1,038 2,702 580 97.9
of which:
taxes on residence 54 92 126 166 219 298 404 562 955 2,343 522 97.5
taxes on other property 2 5 6 12 10 17 26 64 83 359 58 8.9
Total tax paid
with VND100m threshhold per household 2 9 23 51 81 149 247 419 813 2,443 424 35.8
as percentage of expenditure
Total tax paid 2.4 2.7 3.0 3.4 3.7 4.2 4.7 5.6 7.1 9.2 6.2
of which:
taxes on residence 2.3 2.6 2.9 3.2 3.5 4.0 4.5 5.1 6.5 8.0 5.6
taxes on other property 0.1 0.1 0.1 0.2 0.2 0.2 0.3 0.6 0.6 1.2 0.6
Total tax paid
with VND100m threshhold per household 0.1 0.3 0.5 1.0 1.3 2.0 2.7 3.8 5.5 8.3 4.5
Notes. Revenue from value-added tax, excise taxes, and personal income tax, are estimated based on spending and income patterns. Based on the VHLSS
survey of 2006. All figures are in prices of 2006.
A property tax would be highly progressive, representing a burden of less than 3 percent
of income or expenditure for the poorest fifth of the population, but over 7 percent for the
richest fifth. Nine-tenths of the revenue would come from taxing principal residences, with
the remainder coming from property that households rent out to others. A property tax that
is only levied above a threshold would be even more progressive, collecting just 0.1 percent of
spending from the poorest decile and 11.8 percent from the top decile. The burden of the tax
would be four times as high (as a percentage of household spending)
237
T ax R eform in V ietnam
Note: The regions are: 1: Red River Delta, 2: Northeast, 3: Northwest, 4: North Central Coast, 5: South Central
Coast, 6: Central Highlands, 7: Southeast, 8: Mekong River Delta.
This analysis has a number of limitations, due largely to data imperfections. Because it is based
on household survey data, it does not include the value of property held by corporations or
the government. It assumes that property values in 2006 were realistic, correctly reported by
respondents, and would not fall if a property tax were introduced. This last assumption is not
particularly compelling; with the introduction of a property tax we would expect the value
of land to fall by the capitalized value of the stream of expected property tax payments. In
response to a 1 percent property tax, property values could easily fall by 10 percent.
Nonetheless, the results appear robust enough to have confidence in the conclusion that such
a tax would be progressive, and could have very substantial yield.
238
T axation I n V ietnam : W ho P ays W hat ?
net effect is a modest reduction in measured tax progressivity: the Kakwani measure falls from
0.071 in the baseline case to 0.056 with the new assumptions (see table 8.10).
Figure 8.3: Checking the Robustness of the Incidence Results
16.0 16.0
14.0 Using higher cost of
14.0
Fees as payments for services durables & housing
Taxes paid as % of expenditure
4.0 4.0
2.0 2.0
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Expenditure per capita decile (poor = 1, rich = 10) Expenditure per capita decile (poor = 1, rich = 10)
Distribution of taxes for baseline (left) and Distribution of taxes for baseline (left) and if
if fees are considered as payments for measure of expenditure uses higher costs for
services (right) durables and housing (right)
16.0 20.0
8.0 10.0
8.0
6.0
6.0
4.0
4.0
2.0 2.0 10
- -
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9
Expenditure per capita decile (poor = 1, rich = 10) Expenditure per capita decile (poor = 1, rich = 10)
Distribution of taxes for baseline (left) Distribution of taxes for baseline (left)
and if taxes on alcohol and tobacco and if VAT is levied at a uniform
are doubled (right) 10 percent (right)
One can also explore the effects of other tax changes; the lower left panel of figure 8.3 shows
the effect of doubling excise taxes on alcohol and tobacco, and leads one to conclude that the
effects on progressivity would be very modest although the change may be understated
here, since spending on alcohol and tobacco are typically substantially underreported in
household surveys. Finally, in the lower right panel of figure 8.3 we show what would happen
if the VAT were levied at a single rate of 10 percent (while leaving zero-rated and exempt
goods unchanged). This is a rough and ready estimate that does not account for VAT that
is partly collected on intermediate goods only, but is nonetheless a reasonable first pass at
the problem. The overall burden of taxation would rise, but progressivity would essentially
remain unchanged the Kakwani measure would only change from 0.071 to 0.072 (which, if
anything, represents greater progressivity, relative to expenditure). At least in the context of
current-day Vietnam, a higher VAT rate would not disproportionately burden the poor.
239
T ax R eform in V ietnam
The key finding is that the Vietnamese tax system is relatively progressive, in that it imposes
a higher relative burden on the rich than on the poor. However, it must be remembered
that only about half of all taxation can be attributed to households; for now, we have not yet
allocated to households the taxes that are levied on business corporationsa controversial
issueor the tariffs collected on imports (for which more data would be needed). Nor do we
have information on the extent of in-kind contributions make in lieu of taxes that are required
of all citizens. And the model we use is a mechanical one; ultimately it would be desirable to
build a full computable general equilibrium (CGE) model to track the effects of tax changes
on revenue, income distribution, and output (see for instance Shoven and Whalley 1984).
The VAT and excise taxes are somewhat progressive. This is not the case in every country, but
occurs in Vietnam mainly because poor people provide for many of their own needs rather
than purchasing them, which in turn reduces the VAT they have to pay. Over time this will
change, as home production becomes unimportant, and gradually this tax will become less
progressive. As Vietnam considers its options for raising more tax revenue, changes in rates
and base of the VAT are well worth considering: it is administered effectively, raises a lot
of revenue, could be changed at minimal expense and, as we have shown, does not unduly
burden poorer households. On the other hand, if expanded too much, the VAT too could
discourage the development of the formal sector, especially in retailing.
In an effort to help the poor, it would be tempting to reduce or even abolish agricultural and
educational fees, which are regressive. There is a recent precedent: the unpopular agricultural
tax was essentially abolished, which provided substantially more relief to poor than to rich
households. However, even if the size of local fees is only loosely related to the services
they finance, one should be cautious about reducing important sources of local revenue,
especially if the result is diminished local servicespoorly maintained irrigation systems,
weaker schools, or understocked health clinics. As a general proposition, it would be more
desirable to ensure that fees are set so they correspond more closely with the services they pay
for. Tax exemptions could be considered for the very poorest households; there is already an
infrastructure for determining who is poor in setting educational scholarships.
The personal income tax has received more attention over the past year or two than its feeble
contribution to state revenue would appear to justify. This still begs the question of what
the future direction of public policy should be: should the personal income tax be expanded
robustly, or should it remain an unimportant tax. Those who argue for expanding the personal
income tax emphasize its revenue potential and its strong progressivity; opponents worry
that it will hinder the development of the formal sector, and it is expensive to administer
relative to the revenue it yields.
The new personal income tax has one major advantage over the current version: it would tax
fewer households, essentially because many of the households who currently pay the business
income tax levied at a flat 28 percent once profits exceed the minimum wagewould not
have to pay tax when the new personal income tax comes into effect. This should make it
more attractive, initially at least, for people to create formal (rather than informal) institutions,
such as registered businesses. Our best, although admittedly rough, estimate is that the
new income tax, had it been in place in 2006, would have reduced government revenue by
240
T axation I n V ietnam : W ho P ays W hat ?
about 3 percent; almost all of this effect would be through the reduction in revenue from the
household business income tax.
If the personal income tax continues to raise little revenueand in the Asian countries
surveyed by Alm and Wallace (2004), it raised just a sixth of all tax revenueshould it be
abolished entirely? Certainly, a case for this can be made, principally on the grounds that
expenditure taxation is more efficient that direct taxation (see for instance Tuerck et al.
2006) because it creates fewer distortions in the decisions to work and save. On the other
hand, the personal income tax, as currently constituted and as proposed, is the most effective
mechanism for ensuring that expatriate workers pay a reasonable amount of tax to Vietnam;
that alone would be reason enough to keep the tax. But if this is the main rationale for it, then
the tax should continue to apply just to those in the highest earnings brackets. By 2009 the
threshold for paying personal income tax will be about four times the level of average income,
and the number of people subject to the tax will begin to rise very rapidly. This would probably
imply substantial extra collection costs for limited additional revenue, which suggests that at
a minimum the tax brackets should be adjusted for inflation over time. It would probably
also be desirable to match the top income tax rate with the corporation income tax rate of 28
percent (to ensure that people do not rearrange their business activities simply to avoid taxes),
and simplify the structure of the tax by reducing the number of brackets. The implication is
that the personal income tax should continue to play a limited role over the coming decade;
if more tax revenue is needed, it will be necessary to look elsewhereto the VAT, a property
tax, and perhaps excise taxes.
A property tax looks particularly attractive as an option: it would be highly progressive, and
would yield substantial revenue. However, the administrative costs of such a tax can be high,
especially at first (establishing ownership, creating maps, and the like); and it can create
serious burdens on the cash flows of households and businesses, especially in a time of rapidly
inflating house prices.
Finally, it is worth noting that the discussion in this paper has focused entirely on the
revenue side of government. It would be desirable to measure the incidence of government
expenditures as well as revenue, as proposed by Demery (2000) and done by Haughton
(2005) in the context of Peru. The reason is that even a regressive tax might be desirable if
the ensuring spending were returned even more heavily to the poor.
241
Property and Land Taxes
By Bruno Moser and Nicolaus Tideman IX
1. Property and Land Taxes: Legal Background and Revenue
The basis for holding land in private possession is expressed in the constitution and is further
regulated in the land law 2003.76 The drafters of this law made an effort to define, foresee, and
regulate every variety of land use (and the amount of money that should be paid to the state
for the privilege of each type of use).77 Nevertheless, total revenue from land taxes (apart
from royalties from natural resources) was less than one million dollars in fiscal year 2006.
However, provinces and municipalities derive significant income from allocating (selling)
land use rights. For example, in recent years Ho Chi Minh City has derived more than 40
percent of its annual budget from this source. This practice is not sustainable.
A summary of laws related to land taxes is provided below.
243
T ax R eform in V ietnam
The tax rate per acre is prescribed for each type of street in each type of urban center. Urban
land has been classified into 5 grades, and the rates have been linked to the tax rate for
agricultural land use. For instance, for grade-I urban centers (highest grade), the land tax
rates have been fixed between 9 and 32 times that on the use of agricultural land of the highest
grade in the region, and for grade-V urban centers (lowest grade), the land tax rates are fixed
at 5 to 13 times that on the use of agricultural land of the highest grade in the region. For
townships, the land tax rates are between 3 and 13 times that on the use of agricultural land
of the highest grade in the region.
For residential land and construction land in suburban areas, areas adjacent to traffic hubs
or along main traffic routes, the tax rates are prescribed as follows. For land in population
quarters and construction land in the suburban areas of grade-I urban centers, the land tax
rates is 2.5 times that on the use of agricultural land of the highest grade in the communes.
For suburban areas of grade-II, -III, -IV, and -V urban centers, the land tax rates is 2 times that
on the use of agricultural land of the highest grade in the villages. Finally, for residential land
and construction land in rural, delta, midland and mountainous areas, the land tax rates is
equal to the average agricultural land use tax rates in the communes.
Tax calculation
The farmland-use tax is to be calculated on the basis of the area of land, land categories, and
tax rates determined on the basis of kilograms of rice per unit of land. The land category is
dependent on the nature of the land, its location, climate, irrigation conditions and average
productivity obtained in normal conditions.
Tax rates are determined on the basis of kilograms of rice per hectare in case of cultivated and
aquacultural land; for perennial land the rates are slightly higher, and for perennial trees on
cultivated land rates are related to the rate applied to cultivated land. Wood trees and one-
harvest perennial trees are subject to a tax rate of 4 percent of the output value. The price of
rice is determined by the peoples committee.
244
P roperty and L and T axes
Exemptions
The law did not apply if the State allotted or leased out land, if the transfer took place among
family members or when land use rights were donated to the Peoples Committees or to
political organizations.
Tax calculation
The land use right transfer tax was based on the land area, land prices and the transfer tax
rates. The land prices were the prices set according to the Governments price bracket for
different land categories. In case of land use rights that were transferred by auction, it was the
auction price. If land was transferred by agreement between two parties, the agreed price was
the basis. The tax rate was two percent for land used for agricultural production, forestry, and
aquaculture and four percent for residential land and land for project construction.
Repeal
The land use right transfer tax was repealed in 2009, on the ground that it was sufficient to
collect personal and corporate income taxes on land price gains.
245
T ax R eform in V ietnam
Households and individuals that are assigned residential land; economic organizations
that are assigned land for the purpose of building dwelling houses for sale or lease or for
the purpose of production and business or for agricultural, forestry, aquaculture are liable to
pay this tax. Persons who lease land from the State and those using land for construction of
infrastructure for common use in industrial parks are exempt from the land use levy.
The land use levy depends upon land area, land prices and land use terms. The price of land is
determined by the Peoples Committees or the auction price as the case may be.
The land use levy is not collected in cases of investment promotion, construction of public
works for commercial or social purposes in the fields of education, healthcare, culture,
physical training and sports.
246
P roperty and L and T axes
According to the State Budget Law, land taxes have been found to be an excellent
source of revenue for local governments in a large number of countries, provinces, and
municipalities.
78
VN-Constitution Article 17: The land, forests, rivers and lakes, water supplies, wealth
lying underground or coming from the sea, the continental shelf and the air [] come under
ownership by the entire people.
VN-Constitution Article 18: The State manages all the land in accordance with the plan and
the law, and guarantees that its use shall conform to the set objectives and yield effective results
[] organizations and individuals are responsible for the protection, enrichment, rational
exploitation and economical use of the land; they may transfer the right to use the land entrusted
to them by the State, as determined by law.
VN Land Law 2003 Article 1: This Law makes provisions for powers and responsibilities of the
State as a representative of the entire peoples ownership of land and exclusive manager of land; the
management and use of land and rights and obligations of land users.
VN Land Law 2003 Article 5-1: Land is under the ownership of the entire people, and the State is
the representative of the owner.
247
T ax R eform in V ietnam
Thus, rather than considering land to be private property, it might be better to consider it as
subject to private possession or secure tenure.
There are also important differences between taxes on land and taxes on improvements in
terms of administration. The proper taxation of improvements requires continual on-site
evaluations that are somewhat subjective and open to corruption. Land values on the other
hand, because they are generally smooth functions of location, are much more objective,
transparent, and easy to assess.
For all of these reasons, an economic analysis of property taxes should be divided into an analysis
of a tax on natural opportunities (urban sites, agricultural land, fisheries, electromagnetic
spectrum, landing slots, etc.) and a tax on humanly produced capital improvements.
3. Land Tax
The efficiency of land taxes
Taxing land values is one of the most efficient ways of raising substantial government revenue.
When properly administered, a tax on land, unlike most other sources of public revenue,
introduces no inefficiency, or excess burden into an economy. Therefore, whenever a nation
or a commune replaces a revenue source that has an excess burden with a tax on land, the
overall efficiency of the economy improves.79
There are two requirements for a tax on land to have no excess burden. First, the tax must be
independent of how productively the land is used. Second, the tax must not be greater than
the value of using the land. If these two conditions are met, then all land that can be used
productively will continue to be used when land is taxed, and those using land will continue
to have an incentive to use it as productively as they can. A tax that is independent of how
efficiently land is used can collect up to the full rental value of land without creating any excess
burden.
Furthermore, enforcement and compliance costs of land taxes are minimal. Enforcement costs
involve the maintenance of a cadastral system, frequent and transparent assessments, sending
out bills, and dealing effectively with delinquencies (of which there will always be a few).
The compliance costs of a land tax are comparable to the costs of paying bills such as electricity.
For urban and commercial land, a monthly payment would probably be most convenient. This
could even be done by an automated debit from the possessors bank account. For agricultural
land, it might be best for payments to be due shortly after the harvest. For southern provinces
this could mean up to three payments per year, while in mountainous regions in the center
and the north one single annual payment would be appropriate.
The excess burden of a tax that has an excess burden is roughly proportional to the square of the tax
79
rate. It is also roughly proportional to the reciprocal of the sum of the reciprocals of the elasticities
of supply and demand.
248
P roperty and L and T axes
While there are other ways of raising public revenue without excess burden,80 the
economically most important way of raising public revenue without excess burden is by
taxing land.81
80
Other ways of raising public revenue without excess burden are taxes that compensate for the
harm caused by pollution, congestion, and global warming, and prices (no greater than marginal
cost) for public services.
81
From an economic perspective, taxes on land include taxes on other scarce natural opportunities
such as fishing and the electromagnetic spectrum (used for mobile phones, radio, and TV).
249
T ax R eform in V ietnam
The laundering of illegitimate funds is one of the most extreme examples of enjoying low interest
82
rates.
250
P roperty and L and T axes
efficiency. To the extent that governments refrain from spending on wasteful projects because
of reduced rent-seeking, there is a further improvement in social efficiency.
4. Equity
Land as the common heritage of all persons of all generations
As already mentioned a land tax gives expression to the constitutional declaration that land
belongs to all. Since an equal physical division of a nations land is nearly impossible and
would be economically disastrous, an equal financial division, through a tax on land, is an
appropriate expression of equal rights to land.
Collecting the annual rent of land for public purposes also reflects Adam Smiths first
cannon of taxation, The subjects of every state ought to contribute toward the support of
the government in proportion to the revenue which they respectively enjoy under the
protection of the state.
251
T ax R eform in V ietnam
worthwhile investment, the total increase in land rent that it generates will be greater than the
cost of the bridge, so that if the bridge is financed by the increase in land rent it generates there
will be no need for the bridge to be subsidized by people in unaffected regions.83
The same principle applies at the next level of government. Thus if a province constructs a
large bridge that benefits several communes, land values will rise in those communes. The
province can pay for the bridge by requiring the communes to pay in proportion to the
increases in the rental value of land that they receive. The province would therefore not
need to charge citizens who do not benefit, and would not need a subsidy from the national
government either.
A similar analysis applies to the national government. If the national government provides
bridges or other items of infrastructure to individual regions (access to national highways or
high-speed trains, for example), then fairness requires the regions that benefit to compensate
the national government according to the benefits that they enjoy.84 A practice of having
substantial taxes on land would provide the funds with which individual provinces and
communes could pay for national infrastructure in proportion to the benefits that they receive.
In sum, capturing the annual rental value of land allows individuals and communities to pay
for the benefits that they receive from government infrastructure projects, in the way that
fairness requires.
5. Tax on Improvements
Unlike a tax on land, the component of a property tax that falls on improvements85 can have
a significant excess burden. Therefore, when there is a substantial tax on land, it is important
to tax improvements at a much lower rate, if they are to be taxed at all.
While taxes on improvements are generally inefficient, there are at least two reasons why
a small tax on improvements might be included in a package of efficient sources of public
revenue. First, the more improvements to land a nation has, the more it will need to spend on
police, courts, national defense and diplomacy to protect its property. It is reasonable to assign
some portion of these costs to improvements and therefore to levy a tax on improvements
to collect that much revenue. If a tax on improvements simply compensates for the cost of
protecting additional improvement, then there is no excess burden of the tax.
The second reason that one might wish to levy a tax on improvements is to erode the wealth of
the wealthy. The efficiency rationale for eroding the wealth of the wealthy is that it is difficult
to maintain the cohesion of a society if inequality in the distribution of wealth is too extreme.
The extension of the Jubilee line in Londons rapid transit system has dramatically increased
83
neighboring land values. The increases around just two of the six new stations would have
generated enough revenue to pay the total construction cost of 3.5 billion. It is estimated that the
total increase in land value along the extension was three times as much as the construction cost.
Similarly, negative externalities of national projects, such as fenced in high ways with no nearby
84
exit or even crossing, should be compensated according to the losses in land value.
Improvements are not only buildings but also private investments in irrigation ditches, private
85
roads and bridges, continual individual efforts to maintain rice paddies, etc.
252
P roperty and L and T axes
Thus a tax on the value of property for this purpose would have a significant exemption, so
that people would not be penalized for providing for their basic needs and retirements. If
a society has a consensus that an extreme concentration of wealth is costly, then a tax that
compensates for this cost has no excess burden.
If a tax on improvements exceeds the marginal cost of protecting improvements and the cost
of more extreme inequality in the society, then there is an excess burden to the tax, because
the more improvements are taxed, the fewer will be built and the smaller will be expenditures
on maintaining improvements. The excess burden of a tax on improvements will be roughly
proportional to the square of the amount by which the tax on improvements exceeds the sum
of the marginal costs of protecting improvements and the marginal costs of inequality.
86
In November 2007 the National Assembly has voted for the introduction of a 25 percent on capital
gains on real estate transfers. This is on the top of the 2 percent transfer tax already in place and
brings it almost in line with the current income tax on rent which is at 28 percent.
87
This tempting practice is not sustainable: it deprives future generations of such benefits, it often
converts land to higher uses prematurely (thereby disrupting unnecessarily well established
communities and livelihoods), and it invites for corruption. Furthermore, the taxes, fees, and
bribes that must be paid for land use conversions can easily bankrupt poor households, as they
lack the resources to pay upfront the unexpected charges that are out of line with the current use
and rental revenue, reflecting actual local market demand.
253
T ax R eform in V ietnam
7. Policy Implications
In Vietnam more than 86 million persons compete with thousands of state owned enterprises
and private companies for the use of land. In addition to this private demand for land, there
is communal demand for land, for such uses as parks, streets, waterways, and natural reserves.
One of the central aims of land legislation in Vietnam has been the creation of an efficient land
market, designed to promote equitable outcomes and sustainable development. However,
even though many legislative measures have been taken to eliminate inefficiencies and
inequalities, these measures have not managed to control price inflation, land speculation,
encroachment onto public lands, persistent resistance to formalization, and corruption. And
the SEMLA reports argue that the Land Law and associated decrees have reached such a level
of complexity that they are difficult for even the most sophisticated citizen to understand.
In this section we show how taxing (collecting the rent of) land according to its full market rental
value could help resolve, in a efficient and equitable manner, a number policy questions that
have been discussed repeatedly in the five year socio-economic development plans (SOEDs).
Advantages of Formalization of the Land Market88
Land deals between two private parties do not necessarily need to be registered formally.
People sometimes consider it sufficient to transfer land use rights informally, by a handshake
or a simple rental agreement. However, if there is no legal formalization, there remains a
risk that investments by the person who does not have legal tenure might be taken without
compensation. Thus it is valuable, for both efficiency and equity, to have a simple and
inexpensive process by which land use rights can be legally transferred.
Keeping track of land title information does not request a large bureaucracy and can be
financed by fees equal to marginal costs of the service or by general public revenues. The
public has an interest in a transparent and efficient system of land titles, to avoid the social
costs of idle land and to promote the efficient and sustainable use of land.
Managing land title registration and tax collection, once these are well established, is like
managing the front desk and back office of any hotel: Room allocation, proper registration,
and full rent collection. The fee for land registration, as an adjunct to a system of land value
taxation, should be no more than the cost of a clerks time. Any reduction in transfer and
registration costs will be reflected in increased land prices. Hence, by taxing land values those
value increases would be recovered for the public good.
8. Land Recovery
Land use rights are allocated to land users by the state, and the state has power to recover
these rights in prescribed circumstances.89 Article 38 of the Land Law of 2003 lists the
Land markets remain largely informal. Recent figures from the ADB suggest that only about 50
88
percent of transactions of agricultural land are formally registered. In urban areas estimates vary,
suggesting that between 60 percent and 85 percent of land is transacted informally (CIEM 2006;
Gillespie 2002).
This is equivalent to eminent domain, a nearly universal practice around the globe.
89
254
P roperty and L and T axes
reasons for which the state is authorized to recover land use rights. The first clause is quite
comprehensive, allowing recovery when the state needs to use the land for the purpose
of national defense, security, national benefit, public benefits or economic development.
Notice must be given by the state prior to recovery: 90 days for agricultural land and 180
days for nonagricultural land. In most cases, households from whom land is recovered are
eligible to receive compensation from the state, either in form of similar land elsewhere or
in cash. Some are eligible for additional support to help redress the effects of disruption on
their livelihoods.90
While there have been problems in the land recovery process, which have caused complaints,
resistance, conflict, and significant delays in land clearance and public works, the main cause
of public outcry has been the compensation scheme. Official prices do not match market
prices for land. This makes it almost impossible for compensated title holders to gain the
use of land comparable to what has been recovered from them. Further fuel is added to
sensed injustice when land is recovered, not for basic infrastructure needs but for economic
development projects and industrial zones. In these cases the difference between the
compensation and the value in the new use value is drastically widened.
Under a regime of full land value taxation this issue disappears because public collection
of the rental value of land drives the selling price of land toward zero, no matter what the
designated current or future use is. Land recovery under land value taxation would cause
only the loss of tenure, and not a financial loss. Since the selling price of land would be close
to zero, no compensation would be needed for the loss of land. Under land value taxation,
compensation would be limited to the financial and psychological costs of dislocation and the
loss of capital improvements attached to the site.
It might be suggested that a compensation scheme should be compassionate. Under a regime
of public collection of the full rental value of land, compassionate compensation is more
feasible since no funds are needed to compensate for land, which belongs to the entire people
in the first place.
90
Decree 197/2004/ND-CP on compensation, support, and resettlement.
91
Gillespie (2001) reports that resolution of land and housing disputes can be time consuming,
expensive and uncertain, with over 70 percent of unresolved civil law cases concerning housing
disputes. The district courts of HCMC hear about 600 cases a year with one-third going to appeal
at higher courts and taking about an average of up to three years for settlement.
255
T ax R eform in V ietnam
Land value taxation also reduces the stakes in land disputes by reducing the value of land. (It
is also possible to have disputes over the value of improvements, and land taxation does not
reduce the magnitude of these disputes.)
Average plot size is also much smaller and plots are much more widely dispersed than in the south,
making land consolidation among households very difficult. Furthermore, in an effort to expedite
land distribution following the 1993 Land Law only single land use rights certificates have been
issued for all plots of land. Many households are thus forced to reregister their land on separate
certificates at additional costs, resulting in public reluctance toward the efforts.
For example, farmland close to unsealed landfill sites in Da Nang has become polluted and
93
unproductive resulting in protests and compensation claims by the local population. Tu Liem
district of Hanoi provides another example. There, a common complaint amongst farmers is
that they are compensated only for the land taken for construction. Left over, adjacent land in
small, often awkward-shaped plot corners which no longer have irrigation or other agricultural
infrastructure is not compensated, despite its loss of productive potential.
256
P roperty and L and T axes
94
Depending on the revenue estimates made and the time a land tax has been in place and well
established it is fair to assume that total revenue exceeds government needs. The excess revenue
could be used to establish a citizens dividend and/or finance a system of social security for all.
95
However, many originally registered land holders are ghost-owners. The elderly person holding
the land title might have been moved by the family, the old hold demolished and a new one
constructed at its place, rented out at full market value to whomever is in need of real estate.
Furthermore, current land use fees (for example, within the city of Hanoi) are as low as VND
40,000 per month, not even enough to fill the gas tank of a motorbike. Migrant workers and
students, on the other hand, pay VND 200,000 per month and more, for as little as a few square
meters of housing.
257
T ax R eform in V ietnam
First, allow any poor person to defer his or her land taxes until death, if there is enough value
in the property to pay the taxes. At death, sell the property and pay the taxes, with interest.
Second, when it is apparent that there will not be enough value in the property to pay the
taxes at death, make explicit, case-by-case decisions about whether to grant tax waivers on a
compassionate basis. Because cities need to be redeveloped and because all redevelopment
could be blocked if no poor person was required to pay their land taxes, it may be reasonable
to put pressure on people to give up their valuable land. One possible compromise is to
help people with valuable land exchange their land use rights for high-rise condominiums in
their neighborhoods, so they can live in familiar surroundings, in apartments equipped with
modern conveniences, without occupying a significant amount of valuable land.
As of 2002, SOEs held up to 95 percent of land in Hanoi under lease to organizations, leaving
96
only a relatively small share available for private use. Furthermore, data from the 2006-PCI found
that of the firms currently with a lease, 13.7 percent lease informally from an SOE. It should be
no question that the SOEs should be liable for all implicit (collected from third parties or not)
rent and pass it on to the state as part of their total rent payment for the privilege of monopolized
control of a location. Furthermore, leasing out land by SOEs is currently illegal (Land Law 2003,
Article 109/2). However, in economic terms leasing out land (a simple sublease) would not cause
any market distortions were they forced to pay the market rent just as everyone else does. Thus,
this could be another law that could be nullified in the future.
The Vietnamese government ordered privatization of State Owned Enterprises in 1992. More
97
than 3,500 SOEs have been privatized by the end of 2005. Another 1,505 are planned to be
privatized by 2010 (among them all the national banks).
258
P roperty and L and T axes
pockets. The companies will be sold at below-market prices, since it is impossible to predict
future land prices. Thus, many companies will be bought up simply for the purpose of
speculating on their land. Factories, even profitable ones, will be closed and the land held
until the public has made major infrastructure investments.
Taxing land values would prevent the damage from speculative purchase of SOEs, since
it takes the profit out of land speculation. Taking land speculation out of the equitization
equation would also provide greater opportunities for groups with less deep pockets (such as
the workers) to bid on SOEs, since less capital would be required.
Taxing the land of SOEs, whether privatized or not, would ensure that the use of this land,
this part of the Vietnamese homeland, would provide a steady stream of income for future
generations.
259
T ax R eform in V ietnam
throughout the nation represents such a pilot project already, based on the reasoning that
lower taxes and duties will spur economic growth.98
An interesting project is emerging on the eastern banks of Hanois Red River, a new multi-
billion-dollar city center along the eastern bank of Hanois Red River has been agreed upon by
Vietnam and Korea. The large and modern commercial and residential area would require the
resettlement of 170,000 people and is subject to approval from the national assembly of the
communist country. The urban development would by 2020 run along a 40-kilometer stretch
of waterfront, across the Red River from the ancient centre of Hanoi. The 4,200-hectare
urban project would require about US$7 billion in investment for new high-rise buildings,
residential blocks, parks and dykes to be built along the flood-prone Red River.
By applying the principles of land value taxation this mammoth project could move forward
more quickly, more smoothly, and in a more equitable way than under the current tax,
investment, and land laws. Turning the 4,200 hectare area into on single tax-free zone today,
charging annual market based land use fees, would put all stake holders on a level playing
field and assure transparency in the process. However, as the project is only supposed to be
finished by 2020, Vietnam would lose the opportunity to benefit on a large scale from the
growth and sustained prosperity that would stem from a switch to land value taxes envisioned
within the current five year socioeconomic development plan.
Note the strong relationship of a given tax regime and rent: where taxes are lower, rent is commonly
98
higher and vice versa. Thus, without the full public collection of land rent, the lower tax rates will
be absorbed by the landlords.
260
P roperty and L and T axes
99
As strange as the idea sounds at first hand, anytime someone leases an office space or rents an
apartment, a basic self assessment takes place: Can I afford the location? What is the value to me?
Am I willing and able to make regular payments? For a discussion of the theory of self-assessment,
see Plassmann and Tideman (forthcoming).
100
The International Association of Assessing Officers IAAO acknowledges an error-bandwidth of
+/- 20 percent as generally satisfactory. However, this applies to real estate appraisals, including
the hard-to-assess improvements. If it were only land values to assess, the accuracy would be
greatly increased.
261
T ax R eform in V ietnam
Public rental auctions can be used to determine which potential user of the land would
place the greatest value on the opportunity to use the land. In such auction, each potential
user submits a single, sealed bid, representing how much he is willing to pay for the use of the
land for the first year. The land goes to the highest bidder, but the price that the highest bidder
is required to payand the price that becomes the initial assessed valueis the second-
highest bid. As a consolation prize, the second highest bidder receives 1 percent of the first
years rent. The results of such public rental auctions provide data for a land value map. Other
points on the map are derived by interpolation.
There are two reasons for requiring a payment of only the second highest bid. First, the
second-highest bid represents the opportunity cost of the land, what the land would have been
worth to someone else if the highest bidder had not wanted it. It is fair that the user of land be
charged no more than this. Second, if the price that must be paid is the second-highest bid,
then a bidder will not be concerned about the possibility that he will be the highest bidder
and the second-highest bid will be much lower, implying that he could have obtained the land
for much less than he bid. When bidders have this concern, they have an incentive to bid less
than the land is actually worth to them, and it will be possible that the land will not go to the
person who can make the best use of it.
It can be shown that when the amount that must be paid is the second-highest bid, no bidder
can do better with a false, strategic bid than he does by reporting honestly what the land is
worth to him (see Vickrey 1961, 837). The payment of 1 percent of the first years rent to
the second-highest bidder provides an incentive to knowledgeable people to participate in
the process and reveal the potential value of the available land, thereby providing important
information about land values.
The highest bidder would be entitled to continue using the land for as long as he was willing
to pay the (regularly revised) rental value of his land in an unimproved state. In future years,
the rental value would be interpolated from the results of public rental auctions of nearby
sites.
Because people at the local level have relevant local knowledge, assessments should be done
on local level. Because the locality may be tempted to try to prove that its land has little
value, to get greater support from the province, provincial government should oversee the
assessment process and adjust assessments across local jurisdictions. Similarly, the central
government should oversee the assessment activities of provinces to achieve comparability
across provinces.
262
P roperty and L and T axes
improvements at the most efficient level, so as to continually achieve the greatest return on
the use of the site.
rr 11 rr 22AP
AP
SS 22
00
APeffdd 22AP
APe
ff ff 00
AP[([( 22))eeff]] 22 [[11((11 frfr))eefrfr]]
ff
For
ForaaFor
city
cityawith
with aasize
size
city with greater
greater
a size than
greaterthan about
about
than 55km,
about km, the effect
thethe
5 km, effect of the
ofof
effect theradius
the radiusof
radius ofthe
of thecity
the cityis inconsequential
inconsequentialand
isisinconsequential and
can
canbe treated
beandtreated as
can beasinfinity,
infinity,asso
treated that
thatthe
thesoformula
soinfinity, that thebecomes
formula becomes
formula becomes
22AP
AP
SS
ff22
To To translate
Totranslate
translate from
fromthe from thevalue
thesale
sale sale value
value of
ofland
landoftoto
landtheto
the the rental
rental
rental valuevalue
value of of one
ofland,
land, land, one employs
oneemploys
employs the
thefact thethat
fact factthe
that that
the rental
rental
price the
price of rentalas
of land,
land, asprice of land, as
aa proportion
proportion ofathe
of proportion
the selling
selling priceof the
price of selling
of land price
land will
will be of land will be Iapproximately
be approximately
approximately I ++ TT G,
G, so I + the
so that
that the
total T
totalrental G, so
rentalvalue that
valueof the total
of theprivaterental
privateland value
landin inthe of
thecity the private
cityisis land in the city is
22AP
AP
RR ((II TT GG))
ff22 263
Some
Someplausible
plausiblesuggestions
suggestionsfor
forthe
theparameters
parametersin
inthe
theformula
formulaare:
are:
To translate from the sale value of land to the rental value of land, one employs the fact that the rental
Tprice
ax Rof land, asin
eform a proportion
V ietnamof the selling price of land will be approximately I + T G, so that the
total rental value of the private land in the city is
2AP
R ( I T G)
f2
A A fraction
fraction of athat
of a city cityisthat is private
private land land
P $17 billion/sq. km. (Ho Chi Minh City); $8 billion/sq. km (Hanoi) peak price
P $17 billion/sq. km. (Ho
of land (divide by 1Chi MinhtoCity);
million $8 billion/sq.
get price per squarekm (Hanoi) peak price of land (divide
meter)
by 1 million to get price per square meter)
f 0.7 exponential rate of decline of land prices (factor of per kilometer)
f I 0.7 15 percent rate
exponential interest rate inof
of decline theland
economy
prices (factor of per kilometer)
T 0.1 percent current annual rate of taxation on sale price of land
I 15 percent interest rate in the economy
T G 0.1 percent
8 percent rate annual
current of growth
rateofofthe economy
taxation on sale price of land
G 8 percent rate of growth of the economy
With these parameter values, the estimated total rental value of land is US$7.7 billion per year
With for Hoparameter
these Chi Minhvalues,
City andtheUS$3.6 billion
estimated totalper yearvalue
rental for Hanoi.
of landThe introduction
is US$7.7 billionofper
landyear
value
for Ho
taxation
Chi Minh Citywould generatebillion
and US$3.6 one force that would
per year push The
for Hanoi. theseintroduction
numbers upofand another
land value that would
taxation would
pushone
generate them down.
force The greater
that would efficiency
push these that up
numbers comes with removing
and another taxespush
that would from commerce
them down. The
greater efficiency
would thatrents
push land comesup.with
Theremoving taxes of
unprofitability fromlandcommerce
speculationwould
whenpush
landland rentswould
is taxed up. The
unprofitability
mean thatofa substantial
land speculation
amount when landwould
of land is taxed wouldavailable
become mean that
forause,
substantial
pushingamount of land
land rents
woulddown.
become available for use, pushing land rents down. Allowing for the greater
Allowing for the greater influence of the second factor, we estimate the total land rent influence of the
second factor, we estimate the total land rent in Ho Chi Minh City and Hanoi to be $9 billion per year
in Ho Chi Minh City and Hanoi to be $9 billion per year (allowing for a 20 percent negative
(allowing for a 20 percent negative error margin). We estimate the total land rent for the rest of
errorconservatively,
Vietnam, margin). We estimate the billion
to be US$1 total land
per rent
year,for
forthe rest of
a total of Vietnam, conservatively,
US$10 billion per year intoland
be rent
US$1country.
the whole billion per year, for a total of US$10 billion per year in land rent the whole country.
The total amount of rent in the country would include not only land rent, but also the rent
The total amount of rent in the country would include not only land rent, but also the rent from oil
from oil
extraction, extraction,
coal extraction,coal extraction,
spectrum spectrum
rental, fishingrental,
rights,fishing
and otherrights, and other
sources sources
of rent. of rent. was
Oil revenue
US$8.0Oilbillion
revenue inwas
2006US$8.0 billion inbillion
(plus US$3.8 2006 in(plus US$3.8income
corporate billiontaxes
in corporate
from theincome
oil andtaxes from
gas industry);
the oil and
with todays gas oil
higher industry);
prices itwith todays
should now higher
be much oilgreater.
prices itThus
should now
it is be much
likely greater.
that rent Thus
is great enough
that, with a stroke
it is likely thatofrent
a pen, Vietnam
is great enoughcould
that,scrap
withall taxes on
a stroke of acommerce
pen, Vietnam and replace themallintaxes
could scrap revenue-
neutral
onfashion
commerce withand
a free market
replace based,
them efficient and equitable
in revenue-neutral fashion withsystem of taxing
a free marketland values,
based, and thus
efficient
unleash a sustainable growth potential without precedent.
and equitable system of taxing land values, and thus unleash a sustainable growth potential
without precedent.
There are two philosophies that can be employed in designing a system for sharing the revenue
from rent collection among levels of government. One can design a simple system or one can
design a system that is intended to reward jurisdictions according to the efficiency with which
they manage their affairs.
A Simple System
One simple system would use a rule of revenue sharing in thirds (1/3). In terms of
transparency, local knowledge, and accountability land rent is best collected at local level, so
that is the level that would collect rent. The locality would keep 1/3, which would be shared
264
P roperty and L and T axes
between communes and other subprovincial governments and pass on 2/3 to the province
(=33 percent of total for the local community). The province would keep one third of what it
received from the local governments and pass on two thirds to the national government (= 22
percent of the total for the province and 44 percent of the total for the national government).
Each level of government would pay its expenses from its share of rent and distribute the
excess revenue back to the population as an equal dividend among its citizens. In this way
the poorest farmer in the most distant village in the most distant province would receives the
same national dividend as the richest rich person in central Hanoi. To the one it might be a
years worth of labor; to the other it might not pay the maid.
In addition to the revenue from a traditional land tax, the central government would distribute
revenue from other rent generating privileges such as mineral royalties and licenses from
such things as the electromagnetic spectrum, hydropower licenses, airport landing slots, and
fisheries. (Where the extraction of mineral deposits, such as coal, means having ones province
turned upside-down, polluted, and degraded, additional compensation for the local people
who endure those costs is appropriate.) Therefore, the national divided might reasonably be
expected to contribute the biggest share in a citizens dividend, further emphasizing the equal
rights of all citizens to the benefits of the nations natural wealth. Good governance includes
an understanding that land is under ownership by the entire people.101
A table according to the plan outlaid above would look like this:
Three-Level Government
Rent Local Province National
Share Share Sharea
Initial Share 100% 67% 15%
Share to keep and dispose of according to
principles of good governance 33% 22% 44%
a. Not including revenues from natural resources.
The only way to sabotage the egalitarian properties of a land tax would be by bad
assessment. Therefore assessments should be public and frequent. Tax delinquency
should be dealt with by levying penalties for late payment and by taking property and
selling it at auction when payment is late enough to raise reasonable doubt as to whether
the title holder will pay.102
Vietnamese Constitution, articles 17 and 18; and Land Law 2003, articles 1 and 5.
101
Making land tax delinquency public makes use of the shame principle to reduce the need to use
102
force. However, there are cases when it is unreasonable to dislocate a person, even though he or
she is unable to support the community in the same way as others. Such persons should be dealt
with on a case-by-case, compassionate basis. Tax deferral until death provides a compromise that
is often satisfactory. If no compassionate relief has been granted, a payment delinquency of 90
days should raise a red flag. After 180 days the title should be auctioned by a Vickrey (second
price) auction. Any improvement would be sold in the auction. The proceeds of the auction
would be used to pay back taxes and administrative costs. Any remainder should be reimbursed
to the delinquent former possessor.
265
T ax R eform in V ietnam
266
P roperty and L and T axes
that one or the other is more appropriate. But each designation has a different set of political
and psychological consequences. Ownership suggests a greater degree of individual control
and responsibility. These qualities make ownership attractive. On the other hand, if those
who use land are called its owners, there will be a greater tendency to believe that it would
be unjust to increase the amounts that owners pay for the use of land (taxes) simply because
the rental value of land has increased. But if the taxes are to not increase as rental values
increased, then the economy will become burdened with other taxes, those who own land
will gradually become a privileged class, and future generations will lose the opportunity to
have access to land on the same terms as the current generation.
Calling the users of land renters emphasizes the idea that the land of the nation is the common
heritage of all generations, and makes it easy to understand why the users should pay more
when rental values rise. But to make rental work effectively, those who have land use rights
must have options of indefinite duration to continue using the land they rent, provided that
they are willing to pay up-to-date rental values. It will also be necessary to have security
bonds against land abandonment, so that people will have an incentive to maintain the value
of the land they use.
Since neither rental nor sale has connotations that fit the best way for a nation to manage
its land, a new term is needed to describe appropriate land relations. The term in English
that seems most appropriate to us is private possession of land. Whatever term is selected,
it will be important for the public to acquire an understanding that a person with land use
rights:
Has the right and responsibility to decide how land is to be used
Is entitled to continue using the land for as long as he or she wishes, upon continuing
payment of the up-to-date rental value of similar unimproved land
Is entitled to transfer the right to use the land to anyone else on whatever terms are
mutually agreed.
24. Dealing with Land that Has Been Sold by the State
Some title holders have bought land from the government and therefore have already paid at
least some of the land rent the land can command today.
Out of fairness, and to avoid double taxation, a credit in the amount of the payment already
made should be given to those who have bought land use rights from the government. To
avoid major revenue shortfalls and to soften the transition it might be suggested that the
credits could only be applied to a maximum of 20 percent of the new annual, market based
tax that is due. The remainder of the credit would be carried forward with interest to future
years.
If land use rights were bought from private parties, the state would not have a responsibility
for the possibility that the current possessor had paid too much to the previous possessor, but
only for any original purchase from the state. With a record of the amount paid to the state, a
corresponding credit would be granted to the current possessor.
267
T ax R eform in V ietnam
This is not only a problem where land is held in common by the state. In fact, it is probably
103
much worse in the West where many industrial sites have been abandoned. For example, there
are about 100,000 contaminated land sites in the U.K. (1993). A detailed survey by the (then)
Department of Environment identified 39,600 hectares of land so damaged by industrial or
other development that it is incapable of beneficial use without treatment. Half of this land is in
urban areas. Total contaminated land, according to the Royal Commission on Environmental
Pollution, could cover up to 200,000 hectares. The cost of dealing with this toxic legacy could be
as high as 10 billion. Many cities in the United States are also obstructed to rejuvenate derelict
sections (Detroit, Philadelphia, the Bronx, etc.), as the land does not automatically come back
into the possession of the state when tax payments default. Crumbling buildings and brownfields
thus sit afoul for many years, further impacting negatively the area, depriving the community of
rapidly gaining access for clean-up and reconstruction.
268
P roperty and L and T axes
For agricultural uses of land, the equivalent of leaving remnants of use on the land is to deplete
the soil or to use pesticides and chemical fertilizers to such a degree that the land requires extra
effort before it yields as much as land in the condition provided by nature. Bonds required of
agricultural users of land would ensure that users were not motivated to deplete or spoil the
soil and then abandon the land.
A user of land would have the option, at any time, of selling the improvements on the land
and the security bond, for whatever price someone else was willing to pay. If the user wished
to cease using the land without finding another user, that too would be permitted. The
user would simply be required to give some specified advance notice and leave the land in a
condition no worse than unimproved land.
Upon relinquishing the land in adequate condition, the user would be entitled to the return
of his security bond.
Sir Winston Churchill, in an address to the House of Commons on May 4, 1909, said: Nothing
104
is more amusing than to watch the efforts of land monopolists to claim that other forms of
property and increment are similar in all respects to land and the unearned increment on land.
[[http://www.grundskyld.dk/2-churchill-uk.html]] In the 21st century, global tax consultants
and a myriad of other representatives of privilege have joined hands with land monopolists to
shift the source of public revenue from land to labor and capital.
Unfortunately, most former communist and socialist countries have simply copied the public
105
finance practices of western nations. While their market liberalizations have freed entrepreneurial
forces and stimulated commerce, thereby greatly increasing the productivity of their economies,
these nations have also privatized land and other natural resources, creating privileges without
proper public compensation. Passing up the chance to get public revenue from land, they have
focused taxes on labor and capital, limiting the benefits that they can get from their market
reforms.
269
T ax R eform in V ietnam
of low taxes on commerce. The steps that must be taken to introduce land value taxation are
listed below. Once the political leadership has announced its intention to introduce land
value taxation, the remaining points can be addressed simultaneously.
Bring the land value tax option to the table.106
Declare the political will to introduce land value taxation, display leadership for its
introduction, and immediately enact supporting decrees.
Start recovering land rent on the basis of current available assessment records.
Complete cadastral records, including updated assessments and cadastral maps, and the
issuance of LURCs. Ensure the data access to all stakeholders on equal terms, taking
advantage of the opportunities the web has to offer, including the use of Geographical
Information Systems (GIS).107
Simplify, alter, or nullify obstructing and obsolete laws, decrees, and regulations.108
UN Habitat Action Agenda Section B. 56(h) to which all UN member states agreed: Consider the
106
adoption of innovative instruments that capture gains in land value and recover public investments. The
Vancouver Action Plan the 1976 founding document for UN Habitat states: Social justice,
urban renewal and development, the provision of decent dwellings and healthy conditions for
the people can only be achieved if land is used in the interests of society as a whole. Excessive
profits resulting from the increase in land value due to development and change in use are one of
the principal causes of the concentration of wealth in private hands. Taxation should not be seen
only as a source of revenue for the community but also a powerful tool to encourage development of
desirable locations, to exercise a controlling effect on the land market and to redistribute to the public at
large the benefits of the unearned increase in land values The unearned increment resulting from
the rise in land values resulting from change in use of land, from public investment or decision
or due to the general growth of community must be subject to appropriate recapture by public
bodies (the community).
Taking advantage of numerous past and ongoing government- and ODA-funded programs
107
in the area of land administration and forming the foundation for the effective and equitable
operation of the land/property taxation system, despites of numerous efforts, by 2006 only about
63 percent of parcels of eligible agricultural land, 49 percent of parcels in eligible forest land
and 70 percent of eligible urban households have been granted with LURCs. Cadastral records
including mapsthe basis for Vietnams torrent-based land administration is incomplete and
not updated.
In addition to the nullification of the conflicts between the Land and Housing Laws which the
108
recent Session of the National Assembly has paved the way for, the current land law could be
simplified to half of its length. (For example, the differentiation of land classes and lease terms
are no longer important, other than for general land use planning efforts. Lease durations would
not need to be differentiated as leases would be permanentprovided the annually adjusted,
market value reflecting land tax is being paid. Thus, for example, Decree 84/07, which states that
foreign investors can now lease land for 70 years, with extensions available at no extra cost could
be scraped. Similarly, many other laws and decrees referring, for example, to the current tax and
investment laws could be eliminated, as land rent will substitute for current revenue and foreign
direct investment would transcend onto the nation naturally). However, it is generally easier to
introduce a law than to repeal one, as vested interests have gotten accustomed to their privileges.
Thus, even greater leadership is required to induce drastic simplification of current law.
270
P roperty and L and T axes
Encourage provincial and local government to shift to land value taxes to finance their
duties under decentralization. Let them start collecting land value taxes while lowering
other taxes and tariffs in a revenue-neutral fashion.109
Turn land registration, public assessment, and tax payment into a One-Stop-Shop (OSS)
experience. Establish and operationalize a system for the resolution of land-related
complaints and disputes, especially related to assessment.110
Absorb feedback so that every step can be adjusted and fine-tuned, based on that feedback.
Declare the whole nation a tax-free zone, except for socially harmful things such as
pollution, congestion, and other privileges.
Deal with cases of personal hardship on a compassionate basis. Where applicable, offer
payment deferment.
Develop a citizens dividend system, financed by surplus land rent.
Start reaping the benefits of public rent recovery combined with low or no taxes on
income and commerce and witness the nation prosperity in a efficient, equitable, and
sustainable fashion.
Of course there will always remain some open challenges and questions, but unless the initial
steps are taken they will not even have a chance to arise. Instead, the possibility of real reform
will remain buried in every days emergency plans of standard political procedures to deal
with the symptoms of a deeply flawed, inefficient and inequitable system of funding public
revenue, of a continuous fight of corruption and poverty, and of endless efforts of economic
micromanagement.
The biggest flaw in any open market economy, committed to the principles of good governance,
remains the failure to recover the full economic rent for the benefit of the public.
The central government could encourage provinces and local jurisdictions to shift to a revenue
109
system based on land value, by permitting them to declare their constituencies free of all income
taxes as long as they provided the central government the same amount of tax revenue (adjusted
for GDP growth) as in the year before the shift, using the revenues they now retrieve in a efficient
and equitable fashion from the full market value land rental collection
This push-and-pull approach would encourage individual provinces to move quickly, without
the central government having to appear overly authoritarian. By inviting provinces to embrace
the potential for the socioeconomic benefits of taxing land values, the central government could
induce provinces to implement and benefit from it faster than any compromise-seeking central
rule could possibly achieve.
In addition, localities or provinces that would make use of it would serve as pilot programs for the
ultimate national implementation, inducing others to copy their policies and delivering valuable
information for fine-tuning the system as its benefits become more and more evident.
This is critical as feedback to the assessed value will improve overall market accuracy and thus
110
limit future complains. Furthermore, the implementation of collecting the annual land value
goes in parallel steps with the improvement of the land administration system. Some complaints
and disputes are inevitable. However, as the winning landowner then owes the annual rent of
the sitenullifying any future speculative gainsand limiting the interest in possession to the
economic feasibility of the projected return on the use, a noticeable decrease in title claims can
be expected. This would improve the situation with the number of complaints to the government
currently encounters, whereof reportedly 80 percent are related to land.
271
T ax R eform in V ietnam
As demonstrated, the implementation of full rent recovery supports many current policy
goals, all complementing each other. Thus, land value taxation cannot be analyzed based on
static economic observations but has to be seen in the full light of its dynamic compilation,
rising simultaneously in an upward spiral economic efficiency and equity.
Thus, requesting a definite answer to all possible and impossible questions without daring
to embrace the principles and to take the initial step is like asking on how to reach personal
enlightenment without heartfelt commitment. For any nation, province, or locality that is
committed to reform, liberty, and pervasive sustainable prosperity, raising revenue by taxing
the private use of the common heritage in land is the economic imperative.
Social Reform is not to be secured by noise and shouting; by complaints and denunciation;
by the formation of parties, or the making of revolutions; but by the awakening of thought
and the progress of ideas. Until there be correct thought, there cannot be right action; and
when there is correct thought, right action will follow.111
Henry George, Social Problems, 1883, Chapter 22. This time might have arrived now, since in the
111
United States the debate around the constitutionality of (income) taxation is currently taking the
form of a popular movement similar to the times of the Boston Tea Party (for more information
google and YouTube, for example, Aaron Russo and Ron Paul).
272
Annex
Annex 1: Vietnam: Key Economic Indicators
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
273
T ax R eform in V ietnam
A Total revenues and 21.6 22.7 25.8 27.8 28.4 27.1 27.6 28.1 28.1 26.7
grants
I Current revenues 21.0 22.1 23.8 25.2 26.1 25.0 24.7 25.4 25.4 24.0
I.1 Taxes 19.1 19.8 20.9 21.7 22.8 23.7 23.2 24.2 24.2 22.3
1 Corporate income tax 6.9 6.9 7.7 8.0 9.0 10.3 9.0 9.1 9.1 6.8
2 Individual income tax 0.4 0.4 0.5 0.5 0.5 0.5 0.6 0.9 0.9 0.9
3 Land and housing tax 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
4 License tax 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
5 Tax on the transfer of 0.2 0.2 0.3 0.4 0.3 0.3 0.5 0.5 0.5 0.6
properties
6 Tax on land use right 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.0
transfer
7 Value added tax 4.0 4.8 5.4 5.4 5.5 5.6 6.1 6.1 6.1 6.4
8 Special consumption 1.3 1.4 1.4 1.8 1.9 1.8 1.5 1.5 1.5 1.8
tax
9 Natural resources tax 1.7 1.6 1.6 2.4 2.5 2.1 1.7 1.8 1.8 1.1
10 Agricultural tax 0.2 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
11 Export & import tax 3.6 4.1 3.6 3.0 2.8 2.7 3.4 4.0 4.0 4.6
12 Other taxes 0.03 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
I.2 Fees, charges, and 1.9 2.3 2.9 3.4 3.3 1.4 1.5 1.2 1.2 1.7
nontax
13 Revenue from 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
discrepancy of import
prices
14 Fees and charges 1.1 1.1 1.1 1.1 1.0 0.8 0.7 0.8 0.8 1.0
15 Rental of land 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.2
16 Others 0.7 1.0 1.8 2.2 2.2 0.4 0.5 0.3 0.3 0.5
II Capital revenues 0.2 0.2 1.5 2.2 1.8 1.7 2.5 2.2 2.2 2.3
III Grants 0.4 0.4 0.5 0.4 0.5 0.4 0.4 0.5 0.5 0.4
Source: MoF 2010.
274
A nnex
Total Tax 0.22 1.34 1.92 1.98 1.58 1.75 1.60 1.68 1.46 0.77 1.68 -0.51
Revenues
Corporate 0.51 1.83 20.92 1.93 0.91 2.77 1.43 2.81 2.85 -0.67 1.20 -4.04
Income Tax
Individual 1.82 -0.23 -0.63 1.53 1.30 2.45 1.32 1.34 1.69 3.84 6.81 0.84
Income Tax
Land and -2.64 1.12 0.53 -1.90 -0.24 0.05 1.65 1.03 0.87 1.30 0.61 9.83
Housing Tax
Licenses Tax -2.62 0.59 1.18 0.39 -0.25 10.50 -2.85 0.22 0.26 0.29 -0.67 1.48
Transfer of -0.61 0.87 -1.93 3.90 1.03 3.72 4.21 -0.15 1.46 6.70 1.04 4.27
Properties
Land Use 2.22 1.42 -5.79 5.76 0.79 2.27 5.81 5.19 2.20 8.60 0.95 3.12
Right
Transfer
Value Added -0.60 -0.46 -0.58 1.65 3.97 2.65 1.07 1.10 1.36 2.12 0.90 2.12
Tax
Excise 1.42 0.21 1.95 2.49 1.71 1.89 4.31 1.66 0.23 -0.71 0.17 5.03
Natural -2.00 6.46 8.48 1.53 -0.28 0.91 8.46 1.55 -1.29 -1.08 1.53 -6.29
Resources
Tax
Agricultural 1.02 -0.84 -1.83 -8.72 -1.15 -10.77 -2.65 -0.82 -1.77 -1.55 -4.62 -6.55
Tax
Import- 0.15 -1.55 -1.41 4.23 2.85 -0.54 -1.39 0.10 0.45 4.14 4.43 3.95
Export Tax
Source: Own calculation based on MoF data, 2010.
275
T ax R eform in V ietnam
276
A nnex
taxes makes the system transparent and easy to comply with. However, over simplification is not
desirable otherwise some sectors of economy, specially the self-employed, will remain out of the
tax net.
Low Collection Cost: High cost of collection reduces net tax revenues available to the government.
But some costs may have to be legitimately incurred in bringing the hard-to-tax sectors within
the purview of taxation. Expenditures are also justified for taxpayer education as it ultimately
reduces cost of compliance.
Neutrality: A tax incentive must not distort investment decisions across sectors by favoring
particular sectors and in the process promoting low return investments over higher return
investments.
277
T ax R eform in V ietnam
278
A nnex
ii. Decree No. 26/2009/ND-CP of March 16, 2009, detailing a number of articles of the
Law on Excise Tax;
iii. Circular No 64/2009/TT-BTC of March 27, 2009 guiding the governments decree No
26/2009 ND-CP of March 16,2009, which detail a number of articles of the Law on
Excise Tax
Trade taxes
i. Law No. 45/2005/QH11 of June 14, 2005 (Law on Import and Export Tax)
ii. Decree No./87/2010/ND-CP of August 13,2010 detailing a number of articles of the
Law on Import and Export Tax
iii. Circular No 194/2010/TT-BTC of December 02, 2010 guiding the customs procedure,
customs monitoring , import and export tax and tax administration on export and
import goods
Natural Resource tax
i. The law on Natural Resource tax ( No.45/2009/QH 12)
ii. Decree No./50/2010/ND-CP of May 14, 2010 detailing a number of articles of The law
on Natural Resource tax
iii. Circular No 105/2010/TT-BTC of July 23, 2010 guiding the governments decree No
50/2010 ND-CP of May 14, 2010, which detail a number of articles of the law on Natural
Resource tax
iv. Petroleum Law, (Amended and supplemented in 2000)
v. Decree No. 48/2000-CP of December 17, 1996 of the government detailing the
implementation of the petroleum law
vi. Petroleum Law, (Amended and supplemented in 2008)
vii. Circular No.32/2009/TT-BTC of February 19, 2009 guiding the implementation of tax
provisions applicable to organizations and individuals conducting oil and gas prospection,
exploration and exploitation activities under the Law on Petroleum
viii. Decree No. 115/ 2009 of the government amended and supplemented some articles of
Decree 48/2000 detailing the implementation of the petroleum law
Nontax Regime (Fees and Charges)
i. Circular No. 63/2002/TT-BTC of July 24, 2002
ii. Circular 71/2003/TT-BTC of July 30, 2003
iii. Circular 45/2006 of May 25, 2006
Tax Administration
i. Tax Administration Law
279
T ax R eform in V ietnam
280
A nnex
Dumping
While countries are allowed to impose antidumping duties, such action is based on the member
being able to prove dumping, calculate its impact, and show injury to domestic industry.
Customs Valuation
This basically amounts to accepting the transaction value as the basis for valuation, with the
GATT valuation code as backup.
Rules of Origin
These rules are the basis for assessing duty, imposing antidumping and countervailing duties,
implementing safeguards and origin marking. All members are obliged to frame well-publicized
rules of origin that specify how the country of origin of imported goods is determined, with the
ultimate objective of having a harmonized set of rules in place for all countries.
Preshipment Inspection
Some countries prefer to insect imported goods before they are shipped from the exporting
country to the importing country to ensure quality control. This practice, however, is becoming
increasingly rare. Recognizing the practice among developing country members, the agreement
calls on developed countries to provide developing country members technical assistance.
Import Licensing
This agreement requires members to have a set of simple, transparent and predictable
procedures. The agreement also requires governments to publicize the basis for granting such
licenses.
Investment
This applies to trade in goods, and contains prohibitions on applying certain trade-distorting
measures, subject to relaxations for developing countries for protecting infant industries or
balance of payment considerations.
Safeguards
Members may restrict temporarily certain imports if a domestic industry is threatened with
injury due to a surge in imports, subject to an agreed definition of surge. The focus is on
transparency and established, predictable rules, as in the case of dumping
Information Technology
The agreement provides for members to eliminate duties on IT products, developing signatory
countries were to do so by 2005.
State Trading Enterprises
Article XVII of GATT requires that state trading enterprises (STEs) function on the basis of
nondiscrimination, and requires members to notify STEs to the WTO.
Source: website www.taxindiaonline.com.
281
T ax R eform in V ietnam
These examples are based on flow diagrams from Daniel Johnston (1994).
112
282
A nnex
Total contractors take $31.72 or 63 percent Total contractors take $26 or 52 percent
283
T ax R eform in V ietnam
284
A nnex
285
T ax R eform in V ietnam
Human Resources
The tax administration should develop a comprehensive human resource program to support
the functionally oriented organization. Such an organization requires people with the
specific skills necessary to promote voluntary compliance through the provision of services
to taxpayers and the effective detection of noncompliance.
Taxpayer Education and Assistance
A comprehensive taxpayer education and assistance program is an essential element of
a modern tax administration system. Such a program promotes voluntary compliance by
making taxpayers aware of their rights and responsibilities under the law and by making it as
easy as possible for taxpayers to comply with the law.
Audit
The tax administration should develop a comprehensive audit program that maximizes
voluntary compliance for a given amount of audit resources by establishing a credible
expectation on the part of taxpayers that noncompliance will be detected and appropriate
penalties and interest will be assessed.
Appeals
In order to build a relationship of mutual trust, the tax administration should develop an
administrative appeals process that is fair, expeditious, and transparent. There should be a
single avenue of appeal, the first level of which is administrative, and the second level of which
is judicial.
Collection
The tax administration should establish a collection function to collect taxes that are due but
not paid and to identify those who have stopped filing returns (stop-filers) and those who
have never filed returns (nonfilers).
Processing and Information Systems
The tax administration should develop or acquire and modify existing computer processing
and information systems to streamline certain tax administration activities.
Autonomous Revenue Authority
In due course, there should be a move toward establishing a semiautonomous revenue
authority as a practical means of achieving the goals of revenue administration reform. This
gives the administration adequate autonomy and administrative tools to perform its duties
efficiently, fairly and effectively.
286
A nnex
Source: World Development Indicator 2005; International Country Risk Guide (ICRG), available at [http://www.
countrydata.com/].
a. GDP per capita, PPP (constant 2000 international US$).
b. The corruption index from ICRG ranges from 0 (most corrupt) to 6 (least corrupt). To make it consistent and
easier to compare with Tanzi, Davoodi (1997), we have rescaled the ICRG index by multiplying it by -10/6 so that
the index ranges from -10 (least corrupt) to 0 (most corrupt).
c. Revenue, excluding grants (percent GDP).
287
T ax R eform in V ietnam
288
A nnex
289
T ax R eform in V ietnam
290
A nnex
In Fisher equation, M is denoted for money demand, V is for velocity, p is for prices, and T is for
113
291
T ax R eform in V ietnam
292
A nnex
293
References
Academy of Finance. 2007. On the Real Power of Sub-national Governments in the Current Inter-
governmental Fiscal Arrangement in Vietnam. Hanoi.
Alm, James, and Sally Wallace. 2004. Can Developing Countries Impose an Individual Income
Tax? Atlanta, Georgia State University.
Alm, James, Jorge Martinez-Vazquez, and Sri Mulyani Indrawati. 2004. Reforming
Intergovernmental Fiscal Relations and the Rebuilding of Indonesia: The Big Bang Program and
its Economic Consequences. London: Edward Elgar Publishing.
Arnott, Richard J., and Joesph Stigliltz. 1979. Aggregate Land Rents, Expenditure on Public
Goods, and Optimal City Size. The Quarterly Journal of Economics 93 (4) (November): 471
500.
ASEAN Secretariat. 2011. http://www.aseansec.org/19585.htm.
Bahl, Roy, and Bayar Tumennasan. 2004. How Should Revenues from Natural Resources Be
Shared in Indonesia? In Reforming Intergovernmental Fiscal Relations and the Rebuilding of
Indonesia The Big Bang Program and its Economic Consequences, ed. James Alm, Jorge
Martinez-Vazquez, and Sri Mulyani Indrawati. London: Edward Elgar Publishing.
Bahl, Roy, and Jorge Martinez-Vazquez. 2007. Property Taxes in Developing Countries: Where
Will We Be in 2015? In Roy Bahl, Jorge Martinez-Vazquez, and Joan Youngman (eds.), Making
the Property Tax Work in the Developing World. Atlanta: Lincoln Institute.
Bahl, Roy, and Linn, Johannes F. 1992. Urban Public Finance in Developing Countries.
Washington: Oxford University Press. Washington DC.
Bird, Richard M. 2000. Rethinking Subnational Taxes: A New look at Tax Assignment. Tax
Notes International 8: 206996.
Bird, Richard M. 2003. A New Look At Local Business Taxes. Tax Notes International 30:
695.
Bird, Richard M., and Enid Slack. 2004. International Handbook of Land and Property Taxation.
London: Edward Elgar Publishing.
Bird, Richard M., and Milka Casanegra de Jantscher, eds. 1992. Improving Tax Administration
in Developing Countries. Washington, DC: International Monetary Fund.
Bird, Richard M., Jorge Martinez-Vazquez, and Benno Torgler. 2004. Societal Institutions and
Tax Efforts in Developing Countries. Working Paper 04-06. Andrew Young School of Policy
Studies, Georgia State University, Atlanta.
Boadway, Robin. 1997. Tax Assignment in the Canadian Federal System. Reshaping Fiscal
Federalism in Australia, ed. Neil A. Warren. Sidney: Australian Tax Foundation, pp. 6190.
295
T ax R eform in V ietnam
Boadway, Robin, Maurice Marchand, and Marianne Vigneault. 1998. The Consequences of
Overlapping Tax Bases for Redistribution and Public Spending in a Federation. Journal of Public
Economics 68 ( June): 45378.
Bourguignon, Francois, and Luiz Pereira da Silva, eds. 2003. The Impact of Economic Policies on
Poverty and Income Distribution. Washington DC: World Bank and Oxford University Press.
Booz Allen Hamilton. 2007. Business Process Re-engineering Baseline Findings. Internal
Document.
Centre for International Economics (CIE). 2004a. Revenue impacts of WTO accession:
Vietnams customs modernisation. Canberra: Centre for International Economics.
. 2004b. WTO Accession and Industry: Assessing the Implications of Tariff
Commitments. Canberra: Centre for International Economics.
. 2007. Vietnams WTO Accession: Assessing the Implications of Tariff Binding
Commitments. Report prepared for the Asian Development Bank, Centre for International
Economics, Canberra.
Casanegra de Jantscher, Milka, and Richard Bird. 1992. Improving Tax Administration in
Developing Countries. Washington DC: International Monetary Fund.
Choi, Jay Pil, and Marcel Thum. 2005. Corruption and the Shadow Economy. International
Economic Review 46 (3) (August).
Citro, C. F., and E. A. Hanushek, eds. 1991. The Uses of Microsimulation Modeling. Vol. 1,
Review and Recommendations. Washington, DC: National Academy Press.
Das-Gupta, Arindam, Engelschalk, Michael, Mayville, William. 1999. An Anti-Corruption
Strategy for Revenue Administration. World Bank PREM Note 33, World Bank, Washington,
DC.
De Jantscher, M. Casanegra. 1990. Administering a VAT. In Value Added Taxation in Developing
Countries, M. Gillis, C. S. Shoup, and G. P. Sicat, eds. World Bank: Washington, DC.
Deaton, Angus, and Salman Zaidi. 2002. Guidelines for Constructing Consumption Aggregates
for Welfare Analysis. Living Standards Measurement Surveys Working Paper 135, World Bank,
Washington DC.
Demery, Lionel. 2000. Benefit Incidence: A Practitioners Guide. Washington, DC: World Bank.
Ebel, Robert D., and Serdar Yilmaz. 2002. On the Measurement and Impact of Fiscal
Decentralization. World Bank Policy Research Working Paper No. 2809, World Bank,
Washington, DC, March.
Ebrill, Liam, Michael Keen, Jean-Paul Bodin, and Victoria Summers. 2001. The Modern VAT.
International Monetary Fund, Washington, DC.
EIU (Economist Intelligence Unit). 2008. Vietnam: Country Report April 2008. London.
296
R eferences
Engelschalk, Michael, Dana Weist, and Samia Melhem. 2000. Computerizing Tax and
Customs Administrations. World Bank PREM Note 44, World Bank, Washington, DC. General
Statistics Office. 2008. http://www.gso.gov.vn/default_en.aspx?tabid=472&idmid=3&Item
ID=6540.
Gordon, Richard K. 1996. Law of Administration and Procedure. In Tax Law Design and
Drafting, Vol. 1, ed. V. Thuronyi. International Monetary Fund, Washington, DC.
GSO (General Statistical Office of Vietnam). 2001. Some Issues in Compliance of Main Accounts
by International Sectors in Vietnam. Country Report for Interregional Workshop on the 1993
System of National Accounts, May 711, Manila, the Philippines.
Gupta A., and A. Harding, eds. 2007. Modeling our Future, Population Ageing, Social Security and
Taxation. International Symposia in Economic Theory and Econometrics. Amsterdam: Elsevier.
Gupta, A., and K. Vishnu. 2000. Micro-simulation in Government Policy and Forecasting. Amsterdam:
North-Holland: Elsevier Science Ltd.
Harrison, Graham, and Russel Krelove. 2005. VAT Refund: A Review of Country Experience.
IMF Working Paper, International Monetary Fund, Washington, DC, November.
Haughton, Jonathan. 2004. An Assessment of the Tax System of Lebanon. For office of Prime
Minister, Beirut.
. 2005. An Assessment of Tax and Expenditure Incidence in Peru. For the Inter-American
Development Bank.
Haughton, Jonathan, Nguyen The Quan, and Nguyen Hoang Bao. 2006. Tax Incidence in
Vietnam. Asian Economic Journal 20 (2): 21739.
IFC (International Finance Corporation). 2007. Tax Compliance Cost Survey. Washington,
DC.
IMF (International Monetary Fund). 2006. Vietnam: Statistical Appendix, IMF Country
Report No. 06/423. International Monetary Fund, Washington, DC.
. 2007. Vietnam: Statistical Appendix, IMF Country Report No. 07/386. International
Monetary Fund, Washington, DC.
Investment and Trade Promotion Centre (ITPC). 2007. Vietnams Custom Tariff Schedule
2006. http://www.itpc.hochiminhcity.gov.vn/en/trade_guide/vietnam_customs_tariff_
schedule_2006
ITD. 2007. Taxation of Small and Medium Enterprises. Background paper prepared for ITD
Global Conference, Buenos Aires, Argentina, October.
http://www.itdweb.org/smeconference/documents/itd%20global%20conference%20-%20
background%20paper.pdf>.
Jenkins, Glenn P., Chun-Yan Kuo, and G. P. Shukla. 2000. Tax Analysis and Revenue Forecasting
Issues and Techniques. Cambridge, Massachusetts: Harvard Institute for International
Development, Harvard University.
297
T ax R eform in V ietnam
Johnson, S., D. Kaufmann, and A. Shleifer. 1997. The Unofficial Economy in Transition.
Brookings Papers on Economic Activity, Brookings Institution, Washington, DC.
Johnson, Simon, Daniel Kaufmann, and Pablo Zoido-Lobaton. 1998. Regulatory Discretion
and the Unofficial Economy. American Economic Review 88 (2).
Keen, Michael. 2003. Tax Reform in Italy. Tax Notes International 17 (February): 66582.
Kiefer, Donald W. 2001. Distributional Tax Progressivity Indexes. National Tax Journal 37:
497513.
King, John. 1995. Alternative Methods of Revenue Forecasting and Estimating. In Tax Policy
Handbook, ed. P. Shome. Washington, DC: International Monetary Fund.
Kyobe, A., and S. Danninger. 2005. Revenue Forecasting How is it Done? Results from a
Survey of Low-Income Countries. International Monetary Fund Working Paper, International
Monetary Fund, Washington, DC.
Le, Minh Tuan. 2007. Estimating the VAT Base: Methodology and Application. Tax Notes
International 46 (2) (April).
Le, Minh Tuan Le et al. 2008. Expanding Taxable Capacity and Reaching Revenue Potential:
Cross-Country Analysis. World Bank Working Paper No. 4559, World Bank, Washington, DC,
March.
Linn, Johannes F. 1983. Cities in the Developing World: Policies for Their Equitable and Efficient
Growth. Oxford: Oxford University Press.
Lopez Laborda, J., Jorge Martinez-Vazquez, and C. Monasterio. Forthcoming. The Practice of
Fiscal Federalism in Spain. In The Practice of Fiscal Federalism: Comparative Perspectives, ed. A.
Shah. The Forum of Federations. Atlanta, USA.
Martinez-Vazquez, Jorge. Forthcoming. Revenue Assignments in the Practice of Fiscal
Decentralization. In Fiscal Federalism and Political Decentralization, ed. Nuria Bosch and Jose
Maria Duran. Atlanta: Edward Elgar, U.K.
Martinez-Vazquez, Jorge, and Jameson Boex. 2001. Russias Transition to a New Federalism,
with, World Bank Institute Learning Resources Series. Washington, DC: World Bank.
Martinez-Vazquez, Jorge, and Robert McNab. 2006. Fiscal Decentralization, Macroeconomic
Stability, and Economic Growth. Hacienda Pblica Espaola-Revista de Economa Pblica, no.
179.
Martinez-Vazquez, Jorge, Andrey Timofeev, and Jameson Boex. 2006. Reforming Regional-
Local Finance in Russia. World Bank Institute Learning Resources Series. Washington, DC:
World Bank.
Martinez-Vazquez, Jorge, Charles E. McLure, Jr., and Francois Vaillancourt. 2006. Revenues
and Expenditures in an Intergovernmental Framework. In Perspectives on Fiscal Federalism, ed.
R. Bird and F. Vaillancourt. Washington, DC: World Bank.
298
R eferences
McLure, Charles E., Jr. 1996. The Sharing of Taxes on Natural Resources and the Future of the
Russian Federation. In Russia and the Challenge of Fiscal Federalism, ed. Christine I. Wallich.
Washington, DC: World Bank.
. 1998. The Tax Assignment Problem: Ends, Means, and Constraints. Public Budgeting and
Financial Management 9 (4): 65283.
. 2000. Tax Assignment and Subnational Fiscal Autonomy. Bulletin for International Fiscal
Documentation (December): 62635.
Musgrave, Richard A. 1959. The Theory of Public Finance. New York: McGraw-Hill.
. 1983. Who Should Tax, Where, and What? In Tax Assignment in Federal Countries, ed.
Charles E. McLure, Jr. Canberra: Center for Research on Federal Financial Relations, pp.
219.
Nellor, David. 1987. The Effect of Value Added Tax on the Tax Ratio. IMF Working Paper
87/47. International Monetary Fund, Washington, DC.
Norregard, John. 1997. Tax Assignment. In Fiscal Federalism in Thoery and Practice, ed. Teresa
Ter-Minassian. Washington, DC : International Monetary Fund, pp. 4972.
OECD (Organisation for Economic Co-operation and Development). 2001. Compliance
Measurement Practice Note. Prepared by the OECD Committee of Fiscal Affairs Forum on
Strategic Management. OECD, Paris. http://www.oecd.org/dataoecd/36/1/1908448.pdf.
. 2001. Principles of Good Tax Administration. OECD, Paris.
. 2004. Compliance Risk Management: Managing and Improving Tax Compliance. OECD,
Paris.
. 2007. Taxation and Governance. Policy Paper (draft of October 31). OECD, Paris.
. 2010. Tax Administration in OECD and Selected Non-OECD Countries: Comparative
Information Series. OECD, Paris.
. 2011. Cutting Red Tape Administrative Simplification in Viet Nam, 2011. OECD, Paris.
Plassmann, Florenz, and T. Nicolaus Tideman. Forthcoming. Accurate Valuation in the Absence
of Markets. Public Finance Review. Available as a working paper at: http://econ.binghamton.
edu/wp01/WP0101.pdf.
Sah, Raaj K. 1983. How much Redistribution is Possible through Commodity Taxes? Journal of
Public Economics 20 (1): 89101.
Sanford, C. T. 1995. Tax Compliance Costs: Measurement and Policy. Bath: Fiscal Publications.
Schneider, Friedrich, and Dominik Enste. 2000. Shadow Economies: Size, Causes, and
Consequences. Journal of Economic Literature 38 (March): 77114.
Schneider, Friedrich, and Robert Klinglmair. 2004. Shadow Economies around the World:
What Do We Know? IZA, Bonn.
299
T ax R eform in V ietnam
Schneider, Friedrich, Andreas Buehn, and Claudio E. Montenegro. 2010. New Estimates for
the Shadow Economy all over the World. International Economic Journal 24 (4): 44361.
Sennoga, Edward, David Sjoquist, and Wallace. 2007. Incidence and Economic Impacts of
Property Taxes in Developing and Transitional Countries. In Making the Property Tax Work in
the Developing World, ed. Roy Bahl, Jorge Martinez-Vazquez, and Joan Youngman. Georgia
State University, Atlanta, Georgia.
Shome, P. 1988. On the Elasticity of Developing Country Tax Systems. Economic and Political
Weekly XXIII (34) (August 20).
Shome, Parthasarathi, ed. 1995. Tax Policy Handbook. Washington, DC: Fiscal Affairs
Department, International Monetary Fund.
Shoven J. B., and J. Whalley. 1984. Applied General Equilibrium Models of Taxation and
International Trade: An Introduction and Survey. Journal of Economic Literature 22: 100751.
Shukla, G. P. 2006. Tax Policy Analysis and Legal Framework: Taxation Reform and Modernization
Program, Vietnam. Duke Center for International Development, Duke University, Durham,
North Carolina.
. 2007. Tax Policy Analysis and Legal Framework: Taxation Reform and Modernization
Program, Vietnam. Duke Center for International Development, Duke University, Durham,
North Carolina.
Tait, Alan A. 1988. Value Added Tax: International Practice and Problems. International
Monetary Fund, Washington, DC.
. 1991. Value Added Tax: Administrative and Policy Issues. International Monetary
Fund, Washington, DC.
Tanzi, Vito, and Hamid Davoodi. 1997. Corruption, Public Investment, and Growth. Working
Paper/97/139, International Monetary Fund, Washington, DC.
Tanzi, Vito, and Hamid Davoodi. 2000. Corruption, Growth & Public Finances. Working
Paper/00/182, International Monetary Fund, Washington, DC.
Tenev, Stoyan, Amanda Carlier, Omar Chaundry, and Quynh-Trang Nguyen. 2003. Informality
and the Playing Field in Vietnams Business Sector. International Finance Corporation, World
Bank and Mekong Private Sector Development Facility.
Ter-Minassian, Teresa. 1997. Fiscal Federalism in Theory and Practice. Washington, DC:
International Monetary Fund.
Tuerck, David, Jonathan Haughton, Keshab Bhattarai, Phuong Viet Ngo, and Alfonso Sanchez-
Penalver. 2006. The Economic Effects of the FairTax: Results from the Beacon Hill Institute CGE
Model. Boston MA: Beacon Hill Institute, Suffolk University.
United Nations. 1999. Handbook of Input-Output Table Compilation and Analysis. New York:
United Nations.
300
R eferences
301
S ng k k hoch xut bn: 733-2011/CXB/02/04-05/H
In 600 cun (300 ting anh + 300 ting vit), kh 20,5x28cm
Chu trch nhim ni dung: Ngn hng Th gii
Nh xut bn Hng c