This Selected Issues paper on Bangladesh underlies the export performance of readymade garment industry and inflation dynamics. Bangladesh has demonstrated that it is highly competitive in the world’s major garment markets. Inflation inertia, monetary factors, and exchange rate fluctuations are the main determinants of inflation in Bangladesh. Despite adoption of numerous tax policy measures during the past few years, policies implemented by the Bangladesh authorities have not been fully successful in lifting the revenue ratio to a level warranted by developmental objectives.

Abstract

This Selected Issues paper on Bangladesh underlies the export performance of readymade garment industry and inflation dynamics. Bangladesh has demonstrated that it is highly competitive in the world’s major garment markets. Inflation inertia, monetary factors, and exchange rate fluctuations are the main determinants of inflation in Bangladesh. Despite adoption of numerous tax policy measures during the past few years, policies implemented by the Bangladesh authorities have not been fully successful in lifting the revenue ratio to a level warranted by developmental objectives.

IV. Government Revenue Performance and Reform Potential 1

A. Introduction

1. In Bangladesh, improving government revenue performance has been one of the key objectives of the country’s poverty reduction and growth strategy. Having targeted a significant improvement in the revenue-to-GDP ratio, the government has implemented a number of reform measures in the past several years—mainly focusing on the administration side. Owing to these efforts, the revenue-to-GDP ratio increased modestly, but the improvement has fallen short of the initial target. The lack of buoyancy in revenues could be attributable to fundamental structural problems in the tax system, including pervasive tax incentives and exemptions that have eroded the tax base. Both the authorities’ recent tax expenditure study and regression analysis indicate that there is a significant potential gain from comprehensive tax reforms.

B. The Macro-Fiscal Context

2. Bangladesh needs to significantly scale up government revenue if it is to ensure fiscal sustainability while supporting growth and poverty reduction. Fiscal policy has generally been managed prudently with the overall deficit and domestic borrowing being contained at modest levels and with external financing largely limited to concessional terms. However, there is a pressing need for a higher level of spending to promote growth and accelerate poverty reduction. In this context, the most recent DSA conducted by the IMF and World Bank indicates that there is only a moderate risk of debt distress for public debt, but also that efforts to mobilize domestic revenues are the key to ensure improvement in the main debt indicators.2 In recognition of the fact that the availability of concessionary external financing is insufficient to finance needed spending, a significant increase in revenue—percentage point of GDP annually—was targeted under the PRGF-supported program. In the end, the target was not met, with only modest improvement in revenue taking place.

C. Assessing Revenue Performance

3. Tax revenue in Bangladesh has recorded a gradual modest increase, and the tax-to-GDP ratio has shown a small improvement over the last decade (Figure 1). Only in recent years has the tax revenue ratio exhibited some buoyancy owing mainly to administrative efforts (most notably, an introduction of the large taxpayers unit for income tax in 2004). One of the major tax policy changes in recent years has been to limit tax holidays to 18 industries starting from the FY2005/06 budget and to eliminate them entirely by end-June 2008, but its revenue impact has not yet realized. The collection of international trade taxes has been largely unchanged relative to GDP since the beginning of the 1990s, and it is expected to remain at present levels or decline as the move toward trade liberalization continues.

Figure 1.
Figure 1.

Bangladesh: Central Government Tax Revenue

(In percent of GDP)

Citation: IMF Staff Country Reports 2007, 230; 10.5089/9781451804232.002.A004

Sources: Bangladesh authorities; and IMF staff estimates.

4. A cross-country comparison shows that when the level of development, as measured by per capita GDP is taken into account, Bangladesh’s tax revenue ratio is significantly lower than the average for low-income countries. Figure 2 plots the tax-to-GDP ratios (average for 2000–04) for 30 PRGF-eligible countries and 16 Asia and Pacific countries against their per capita GDP in 2000. The regression line indicates that there is a positive correlation between per capita income and the tax-to-GDP ratio, a relation that could be used to gauge a country’s overall revenue position relative to that of other countries at a similar level of development (see Section C).

Figure 2.
Figure 2.

Bangladesh Per Capita GDP and Tax-to-GDP Ratio

Citation: IMF Staff Country Reports 2007, 230; 10.5089/9781451804232.002.A004

Sources: IMF, Government Finance Statistics IMF, International Financial Statistics; and IMF staff estimates.1/ The sample includes Bangladesh, Bhutan, China, Fiji, India, Indonesia, Malaysia, Maldives, Mongolia, Nepal, Pakistan, Papua New Guinea, Philippines, Sri Lanka, Thailand, and Vietnam.

5. Bangladesh’s revenue level and its recent trend generally compare unfavorably with other Asian and Sub-Sahara African countries (Table 1). Revenue trends in Asian countries during the early 2000s were mixed, with some countries (Bhutan, Philippines) experiencing declines. However, on average, these countries maintained their revenue-to-GDP ratio at a significantly higher level than Bangladesh, and the majority managed to increase the ratio. Similarly, the trends among selected Sub-Saharan African countries are also mixed, but the average ratio is generally higher than in Bangladesh.

Table 1.

Bangladesh: Government Revenue—Selected Asia-Pacific and African Countries, 2000–06 1/

(In percent of GDP)

article image
Sources: Bangladesh authorities;IMF Government Finance Statistics; IMF International Financial Statistics; IMF World Economic Outlook; and various IMF country reports.

Excluding grants. Italics indicate figures taken from IMF country reports. GDP per capita is estimates for 2005.

Fiscal year.

General government.

Budgetary central government.

Central government, provinces, municipalities, and countries.

For each revenue classification, only countries for which data are available are included in the calculation.

6. Bangladesh’s relatively low tax-to-GDP ratio reflects mainly a low level of domestic taxes (Annex Table 2). The collection of both direct and indirect taxes relative to GDP in Bangladesh is significantly lower than in the comparators, while the collection of international trade taxes relative to GDP exceeds that of most comparators in the Asian region. The reliance on international trade taxes is also very pronounced in terms of the composition of tax revenue (Table 3): international trade taxes accounts for more than 30 percent of the total tax revenue in Bangladesh, higher than in most of the comparator countries and significantly above the average for the Asian countries.

Table 2

Bangladesh: Tax Levels for Selected Asia-Pacific and African Countries, 2004

(In percent of GDP)

article image
Sources: Bangladesh authorities; IMF Government Finance Statistics; IMF International Financial Statistics; IMF World Economic Outlook.

Fiscal year.

Budgetary central government.

General government.

For each revenue classification, only countries for which data are available are included in the calculation.

Table 3.

Bangladesh: Tax Structure for Selected Asian and African Countries, 2004

(In percent of tax revenue)

article image
Sources: Bangladesh authorities; IMF Government Finance Statistics; IMF International Financial Statistics; IMF World Economic Outlook.

Including taxes on specific services.

Budgetary central government.

General government.

For each revenue classification, only countries for which data are available are included in the calculation.

7. While the level of tax collection is low, statutory nominal tax rates in Bangladesh are generally high relative to the comparators (Figure 3). The corporate income tax rates applied to unlisted companies (40 percent) and financial institutions (45 percent) in Bangladesh are significantly above those in the comparator countries.3 Among the Asia and Pacific comparators, China is the only country that has a higher VAT rate than Bangladesh.

Figure 3.
Figure 3.

Selected Asian Countries: Corporate Income Tax and VAT Rates

Citation: IMF Staff Country Reports 2007, 230; 10.5089/9781451804232.002.A004

Sources: International Bureau of Fiscal Documentation (IBFD); Worldwide Summaries (PricewaterhouseCoopers); and IMF country documents.

8. With comparatively high nominal tax rates but low revenue yields, Bangladesh’s revenue productivity for CIT and VAT is far behind the comparators (Table 4. and Figure 4).4 Relative productivity vis-à-vis comparators are much lower for CIT than for VAT.

Table 4.

Bangladesh: Revenue Productivity for Selected Asia-Pacific and African Countries

article image
Sources: International Monetary Fund, Fiscal Affairs Department, VAT productivity database, and Fund staff estimates.

Only countries for which data are available are included in the calculation.

For 2004. Calculated by dividing revenue as percent of GDP by the highest statutory CIT rate.

For the most recent year between 2001-2003 for which data are available. Calculated by dividing revenue as percent of GDP or consumption by the standard VAT rate.

Figure 4.
Figure 4.

Bangladesh: Selected Asian and Pacific Countries—CIT and VAT Revenue Productivity

Citation: IMF Staff Country Reports 2007, 230; 10.5089/9781451804232.002.A004

9. The low level of revenue collection in Bangladesh is also reflected in the low level of average effective tax rates (AETRs)—especially on capital and consumption—compared to other Asian countries (Figure 5). AETRs are standard indicators of the effective tax burden on categories of income or consumption and are measured as the ratios of tax collections to notional tax bases derived from national accounts (e.g., taxes on labor as percentage of labor income; indirect taxes as percentage of consumption). Because of data limitations, the calculated AETRs for Bangladesh are not fully comparable to those of the comparators, but this is believed not to significantly affect the outcome of this analysis.5 While the limited availability of time series data from the sample countries precludes a more systematic analysis, it is evident that the tax burden in Bangladesh according to these measures is also low relative to the comparator countries.

Figure 5.
Figure 5.

Bangladesh: Selected Asian Countries—AETRs on Labor, Capital, and Consumption

Citation: IMF Staff Country Reports 2007, 230; 10.5089/9781451804232.002.A004

Sources: Poirson (2006); and the author’s estimates.

10. Three interrelated factors provide a highly plausible explanation for low tax productivity:

  • Continued excessive use of a variety of tax incentives erodes the tax base, and strains an already heavily challenged tax administration. Incentives of all sorts are provided in the tax acts, combined with a high degree of discretion to produce nontransparency with regard to exactly who benefits, by how much, and at what cost to the budget. A culture of negotiated tax payments has developed.
  • There are structural flaws in the main taxes, which—even in the absence of tax incentives—would seriously impede their revenue raising capabilities. The current VAT represents a pervasive excise-cum-turnover tax system with over 6,500 product definitions, tax-relevant distinctions between manufacturers, wholesalers and retailers, separate definitions and treatment of goods and services, and presumptive valuation methods. Numerous amendments to the Income Tax Ordinance over the past two decades have made it complicated and hard to administer.
  • While some progress has been achieved in recent years, tax administration remains quite weak and in need of modernization.

D. Scope for Increasing Revenue

11. The preceding analysis clearly indicates a significant scope for Bangladesh to increase the total tax-to-GDP ratio in general and the domestic tax-to-GDP ratio in particular. In this section, we attempt to quantify the room to increase revenue. First, we briefly review the authorities’ estimates of tax expenditure contained in a recent study.6 We then conduct regression analysis of various tax ratios and use the results to measure the potential revenue impact of comprehensive tax reforms.

Estimates of tax expenditure

12. In their tax expenditure study, the authorities reviewed existing tax incentives, exemptions, and their associated revenue losses, with a view to identifying the scope for broadening the tax base. Tax expenditure consists of revenues foregone due to provision of tax holidays, exemptions and deductions, rate reduction, deferrals, and tax credits. Effects of eliminating tax expenditure have been identified using data for FY2005/06 (Table 5). According to the study, tax expenditures amounts to 0.28 percent of GDP for direct taxes, 1.60 percent of GDP for domestic VAT, and 0.65 percent of GDP for other indirect taxes. Among direct taxes—for which the total estimate is about twice as large as in the April preliminary report—tax holidays account for a quarter of the revenue losses. Exemptions and deductions of import VAT (mostly for capital machinery) account for the bulk of the revenue losses in indirect taxes. As for domestic VAT, exemptions on services (in particular, mechanized road transport and bank credit) account for a larger share of revenue losses than exemptions on goods. It should be noted that the potential revenue gain could be larger in the case of direct taxes due to pervasive income tax incentives, and smaller for indirect taxes since the estimate includes VAT exemption on bank interest (that are normally not subject to VAT in other countries) and on intermediate inputs (that are creditable at a later stage).

Table 5.

Bangladesh: Tax Expenditure, FY 2005/06

(In percent of GDP)

article image
Source: Bangladesh authorities.

Potential impact of comprehensive tax reforms

13. Bangladesh needs to undertake comprehensive and far-reaching reforms of the domestic tax system to broaden the tax base and enhance revenue productivity. To gauge the full impact of such wide-ranging reforms, item-by-item estimates would not be available or appropriate. Therefore, we use an aggregative approach taking the revenue level of comparator countries as a reference point. The results suggest, albeit in a highly aggregative manner, the revenue raising potential of comprehensive tax reforms in Bangladesh over the medium term, after these reforms have been fully implemented. Our analysis indicates, furthermore, that the scope for base broadening is potentially much larger than estimated in the authorities’ tax expenditure study.

Total tax revenue

14. The simple correlation between income levels and the tax-to-GDP ratios as shown in Figure 2 suggests that Bangladesh’s revenue ratio would be significantly higher if brought to the level implied by the regression lines. Expanding the framework to conduct a more elaborate regression analysis we use panel data for the 16 low-income countries that are listed in Tables 14. As explanatory variables, we take not only per capita income, but also the share of agriculture in GDP, the share of imports in GDP, and the share of broad money supply in GDP. We estimated a panel regression with fixed country effects. Agriculture is considered to be a difficult sector to tax, so we would expect the coefficients to be negative. On the other hand, we would expect that the higher is the share of imports in GDP, the higher the tax-to-GDP ratio because imports are generally easier to tax than domestic goods. We also expect that the higher the share of broad money supply in GDP, the higher the tax-to-GDP ratio because monetization would make collection of domestic taxes easier. The results are presented in Table 6. Equation (1) includes all four variables; in Equation (2), we dropped the share of agriculture in GDP because its sign is opposite to what was expected; in Equation (3), we further dropped the share of imports in GDP because the estimated coefficient was not statistically significant. We should note that the coefficients on per capita income are quite robust to alternative specifications. Using Equation (3), it is predicted that Bangladesh could increase its tax revenue ratio by 3½ percentage points of GDP.

Table 6.

Determinants of Tax-to-GDP Ratio 1/2/

article image

Samples include Bangladesh, Bhutan, China, CÔte d’Ivoire, Ghana, India, Indonesia, Kenya, Malaysia, Nepal, Pakistan, Philippines, Sri Lanka, Thailand, Uganda, and Vietnam.

Standard errors in parentheses. *, **, and *** indicate that the estimated coefficients are statistically significant at 10, 5, and 1 percent level respectively.

For Bangladesh, authorities data on fiscal year basis; for all the other countries, data are taken from IMF Government Finance Statistics and IMF International Financial Statistics.

Data are taken from World Bank World Development Indicators.

VAT and corporate income tax

15. An alternative regression analysis based on revenue productivity suggests that the collection of VAT and CIT relative to GDP could both rise by 1.8 percentage points. Regressing VAT and CIT productivity on per capita income shows the level of VAT and CIT productivity on average given an income level (see Figure 4). In other words, the estimates indicate the achievable revenue level assuming tax rates constant. One interpretation would be that the estimates show the potential impact from base broadening and improved compliance.

E. Conclusion

16. Despite adoption of numerous tax policy measures during the past few years, policies implemented by the Bangladesh authorities have not been fully successful in lifting the revenue ratio to a level warranted by developmental objectives. The tax ratio in Bangladesh appears stuck at one of the lowest levels among developing countries and has improved only marginally over this period. The limited revenue improvement most likely results from the combined effects of serious base erosion, deficient tax design, weak tax administration, and low compliance. To realize the potential for revenue improvement suggested by various estimates, widening the tax base through the phasing–out of tax incentives, improving structural tax characteristics, and improving administration and compliance including through simplification of the tax system are all important. To succeed, this strategy requires strong and continuous political commitment, in the context of countervailing political lobbies and vested interests.

References

  • Bangladesh Bank, 2006, “Tax Expenditure in Bangladesh: An Introductory Analysis,” Policy Note No. 0706, December 2006 (Dhaka).

  • International Monetary Fund, 2006, Bangladesh: Fifth Review Under the Poverty Reduction and Growth Facility, and Request for Waiver of Performance Criteria, Extension of the Arrangement, and Rephasing, IMF Staff Country Report, No. 06/406 (Washington).

    • Search Google Scholar
    • Export Citation
  • Poirson, Hélène, 2006, “The Tax System in India: Could Reform Spur Growth?” IMF Working Paper, No. 06/93 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
1

Prepared by Noriaki Kinoshita (FAD).

2

IMF Country Report, No. 06/406, November 2006, Annex I.

3

Using the data from the LTU, the implied weighted average of CIT rates is estimated at 44 percent.

4

The revenue productivity refers to the tax-to-GDP ratio divided by the applicable nominal tax rate. It should be noted that low tax productivity can follow from a low share in GDP of the tax base in question, before account is made of tax relief (for example, business profits or consumption as a share of GDP). But not all tax bases can have a low share of GDP, and it is noteworthy that tax productivity is relatively low for all types of taxes in Bangladesh compared to the sample countries.

5

The methodology applied for comparator countries is described in IMF Working Paper No. 06/93. For Bangladesh, the ratio of individual income tax to compensation of employees is used as an indicator for the AETR on labor, the ratio of corporate income tax to the net operating surplus of the whole economy for the AETR on capital, and the ratio of taxes on domestic goods and services and international trade taxes to final consumption (excluding government wage consumption) as a proxy for the AETR on consumption.

Bangladesh: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Bangladesh: Central Government Tax Revenue

    (In percent of GDP)

  • View in gallery

    Bangladesh Per Capita GDP and Tax-to-GDP Ratio

  • View in gallery

    Selected Asian Countries: Corporate Income Tax and VAT Rates

  • View in gallery

    Bangladesh: Selected Asian and Pacific Countries—CIT and VAT Revenue Productivity

  • View in gallery

    Bangladesh: Selected Asian Countries—AETRs on Labor, Capital, and Consumption