Information about Asia and the Pacific Asia y el Pacífico
Journal Issue
Share
Article

Malaysia: Selected Issues

Author(s):
International Monetary Fund. Asia and Pacific Dept
Published Date:
March 2014
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

A Medium-term Fiscal Strategy1

Malaysia has embarked upon an ambitious consolidation strategy: the authorities aim to reduce the federal fiscal deficit from about 4.0 percent of GDP in 2013 to about 3 percent in 2015, and about zero by 2020.2 Under staff’s baseline projections, this medium-term fiscal consolidation would lower the federal debt-to-GDP ratio to about 40 percent by 2020, consistent with staff’s recommended medium-term debt target. Several measures have been announced in recent months, including gradual subsidy rationalization starting in September 2013 and the introduction of a Goods and Services Tax (GST) in April 2015. Implementation of these measures will play a key role in facilitating near-term fiscal adjustment.

This paper provides a quantitative assessment of additional measures that will be needed during the medium term in order to achieve Malaysia’s fiscal targets. It takes into account the projected decline in oil-related revenues; higher non-discretionary spending on pensions; and other spending pressures. It first identifies adjustment needs in the medium term. It then takes stock of potential revenue and expenditure measures that can be used in designing a balanced, growth and equity-friendly consolidation strategy.

A. Background

1. Overview. The Malaysian authorities have taken important steps in recent months to reaffirm their commitment to sound public finances. Taking advantage of solid non-inflationary growth, they adopted a gradualist strategy that combines multi-year fiscal consolidation with decisive strengthening of macro-fiscal institutions. Their intention is to help restore fiscal buffers that have diminished following their successful countercyclical responses to the global crisis. They are cognizant of the need to gradually reduce reliance on declining (and volatile and pro-cyclical) oil and gas revenues, introduce broad-based taxes, and improve the efficiency and equity of spending.

2. Fiscal targets. The authorities aim to lower the budget deficit to about 3 percent of GDP by 2015, down from 4.0 percent in 2013, and to balance the budget by 2020. In this context, the 2014 budget takes action along a number of fronts, including reduction in fuel subsidies, the reform of social safety nets, and the introduction of a Goods and Services Tax (GST) in April 2015. Complementary reforms include improved top-down macroeconomic control and the strengthening of macro-fiscal management. Most notably, they created a high-level executive Fiscal Policy Committee (FPC), supported by a new Fiscal Policy Office (FPO), to guide fiscal decision making.

3. Considerations. The authorities are aware of Malaysia’s relatively high public debt level, the large share of foreign participation in Malaysian government securities markets, and of the importance of sound public finances in shaping investor confidence. They plan to further elaborate, quantify and communicate their medium term fiscal consolidation strategy. This paper aims to make a modest contribution in the design of Malaysia’s multi-year fiscal strategy. It discusses the growth and equity effects of different fiscal measures and tradeoffs involved. Depending on the authorities’ preferences, different choices are possible in designing a multi-year fiscal strategy.

4. Pace of adjustment. The same considerations over growth and equity of different fiscal measures also help to shape policymakers’ choice of the pace of fiscal adjustment. An excessively front-loaded consolidation can have significant growth and equity implications in the short run; on the other hand, the large adjustment needs and shrinking fiscal buffers may question the credibility of a predominantly back-loaded strategy. Some front-loading of measures may be necessary to signal the authorities’ commitment to fiscal prudence. More generally, the outlook on external conditions will be important in Malaysia’s consolidation strategy. Large share of foreign investors in the government securities market leave Malaysia’s public finances and the financing conditions more exposed to external developments.

5. The role of fiscal reforms. The credibility of a gradual fiscal consolidation strategy can be reinforced by the announcement of concrete structural fiscal reforms to be implemented in the medium-term. Anchoring fiscal consolidation on structural fiscal reforms can help make the reforms irreversible. For instance ad-hoc cuts in development spending or current spending could easily be reversed in future years, whereas reforming social safety nets by introducing better targeted systems could help deliver more effective social protection while also producing sustained fiscal savings. Similarly ad-hoc fuel price adjustments are less sustainable than automatic, depoliticized price adjustment mechanisms.

B. Adjustment Needs

6. Fiscal targets. Malaysia’s medium term fiscal adjustment needs are pinned down by its fiscal targets and rules. The objective of achieving a balanced budget by 2020 implies a consolidation of about 4.0 percentage points of GDP in the headline federal deficit during 2014-2020, or 4.0-4.1 percent of GDP in terms of the cyclically- adjusted primary balance. The authorities have also reiterated their commitment to keep gross debt below 55 percent of GDP-their self-imposed debt ceiling. Malaysia also adheres to a current balance rule, which requires that current balance (revenues minus current spending) is always positive. Based on staff projections, the government can meet its fiscal targets and fiscal rules by gradually reducing the deficit to zero under its baseline growth and interest rate assumptions. The baseline growth assumptions reflect the effect of fiscal consolidation on growth, which are expected to be mild3, with growth averaging 5 percent in the medium-term. Growth projections should be revised appropriately if the fiscal measures are expected to have a larger growth impact

7. Protecting development spending. Adjustment will need to rely on higher revenues and consolidation of current spending rather than cuts in development spending for several reasons:

  • Malaysia has limited scope to cut public investment: development spending has declined considerably as a share of GDP over the past two years, but Malaysia needs to continue investing in infrastructure and other development priorities if it is to achieve its growth and development goals by 2020.

  • As discussed later in the paper, the relatively high fiscal multipliers associated with public investment spending make it a less desirable instrument of fiscal consolidation4.

  • Malaysia’s current fiscal balance has shrunk—it is expected to be only about 0.4 percent of GDP in 2013, meaning that shocks to revenues or spending can easily pull the current balance into negative territory.

Consolidation Path in the Baseline

8. Adjusting current spending and raising revenues. These considerations suggest a fiscal consolidation that relies more on cuts in current spending and higher revenues. This strategy will also help ensure that the current balance increases to a more comfortable level going forward.

9. The role of structural factors. If unchecked, a number of factors will help drive up Malaysia’s fiscal deficit going forward. These factors call for the adoption of measures (above and beyond the announced adjustment) in order to secure fiscal targets for 2020.

  • Declining oil-related revenues, driven primarily by the projected medium-term decline in oil prices, will affect budgetary revenue from corporate income taxes, royalties, and export taxes. Based on staff’s projection of oil and gas production, and WEO projections for international oil prices and assuming that PETRONAS dividends are maintained at constant levels, oil and gas related revenues may decline by 1.5-2.0 percentage points of GDP during 2013-2020. The dividend paid by PETRONAS is not determined based on a pre-determined formula and is not linked to profits, which has to some extent helped shield the budget from movements in international oil prices. Going forward, if PETRONAS dividend is linked more closely to profits, the budget would have higher exposure to movements in international oil and gas prices.

  • Significant spending pressures from higher non-discretionary outlays on pensions; higher spending on social security arrangements and healthcare as Malaysia reaches high income status; obligations under PPP contracts; repayment of off-budget stimulus spending; and other debt-like federal obligations, such as spending on the MRT, are expected to materialize in the medium term. Although precise estimates of future spending pressures associated with these obligations are not available, they can be important. For the purposes of illustration, we assume that these obligations will increase spending by about 0.7-1 percentage points of GDP during 2013-2020. Together with the 4.0-4.1 percentage points of adjustment in the primary balance, the size of total measures to be identified is about 6.3-7.3 percent of GDP

C. Revenue Mobilization

In this section we analyze the structure of government revenues and concentrate on the tax structure to assess the scope to raise revenues using different taxes. The results from this analysis are used as input to determine the set of revenue side measures that can be used to support fiscal consolidation.

10. Tax revenues. Malaysia’s tax revenues are low (15-16 percent of GDP in 2011-2012) relative to other higher-middle income and high income countries, and have been declining. Excluding oil-and gas-related tax revenues, Malaysia’s federal tax ratio is about 12.2 percent of GDP, compared with the average non-oil tax revenue of 22 and 33 percent of GDP in a sample of higher-middle income and high income countries respectively in 2012. These comparisons point to the scope for raising tax revenues, which puts Malaysia in a relatively more advantageous position relative to its European peers, which already have high level of taxes. In other Asian countries, including Korea, Thailand, Indonesia and Vietnam, we observe that tax revenues are also relatively low, which reflects in part the smaller size of government in these countries relative to their European counterparts.5

Malaysia: Tax Revenues

(In percent of GDP)

Sources: CEIC Data Co. Ltd.; and IMF staff calculations.

Non-Oil Tax Revenue and Expenditure

Source: Torres, 2013.

11. The role of direct income taxes. Malaysia’s tax structure relies heavily on direct income taxes, especially on corporations, while indirect taxation is underutilized. The ratio of direct (excluding oil and gas income taxes) to indirect taxes in Malaysia is 2.4, while the median in upper-middle income and high income countries are 1 and 2 respectively. Malaysia used to rely more on indirect taxes, especially from international trade, but these declined considerably during 1980-2000 following trade liberalization and a gradual decline in imports as a share of GDP.6 While the decline in international trade taxes was common across many peer countries following trade liberalization, it was typically offset, though only partially, by the introduction of income and consumption taxes. 7 In Malaysia, by contrast, lower trade revenues were partially offset by higher corporate income taxes, especially on oil and gas companies, which probably reduced the urgency of raising domestic indirect taxes. As described below, this was reinforced by the increase in rents from hydrocarbon wealth (non-tax revenues, including dividends from PETRONAS and royalties), which rose substantially in the 2000s.

Malaysia: Tax and Nontax Revenues

(In percent of GDP)

Sources: CEIC Data Co. Ltd.; and IMF staff calculations.

Malaysia: International Trade Taxes

(In percent of GDP)

Sources: CEIC Data Co. Ltd.; and IMF staff calculations.

Selected Countries: Change in Taxes During 1990–2011

(In percent of GDP)

Sources: IMF, FAD Revenue Database; and IMF staff calculations.

12. Desired taxation levels. The appropriate level and composition of taxation depends on a wide range of country-specific factors. These include societal choices over the size of government, its political economy, equity and efficiency principles, capacity to administer taxes, and the availability of alternative sources of revenues. The size of tax revenues depend on the size of government through the budget constraint and the size of government varies considerably across countries and regions. Tax revenues may also depend on the capacity of the government to collect taxes. Tax administrations are typically stronger in higher income countries and the revenue base also tends to be larger8. There might also be different political and societal preferences over the level and composition of taxation that reflect different preferences over equity and efficiency. Finally, some countries may have large non-tax revenues, for instance due to natural resource revenues, which may reduce the need to rely on taxes.

13. Tax potential. In this context, it is useful to distinguish between the potential to raise certain types of tax revenues and whether it is optimal to do so. The starting point of our analysis is to identify, based on different methods, the scope or potential to raise additional revenues in Malaysia from different types of taxes. We first consider an assessment of potential tax revenues based on a peer-based analysis of tax gaps. We supplement this with a direct look at the current level of tax rates and the tax base in Malaysia. We then discuss the growth and equity implications of using alternative taxes and the trade-offs involved.

14. Tax gap analysis. Cross-country tax gap analysis is used to assess the potential to raise revenues. We compare Malaysia’s tax revenues with the average of its peers, controlling for a range of characteristics relevant for tax revenues. We use the cross-country database on different tax revenues in 2012 constructed in Torres (2013)9 and a linear regression model for different types of taxes. We compare the predicted tax revenues from the regression model with Malaysia’s actual revenues in 2012. The difference between the predicted and actual taxes is interpreted as a “tax gap.” The regression controls for income per capita, the old-age dependency ratio, exports of oil and gas,10 and also includes dummies for geographical regions to capture other unobserved regional characteristics.11 The control variables have the expected sign and are included in the regression if they are significant at the 10 percent level. We use two different samples, one only higher-middle income and high income countries and the other all countries, to check robustness.

Tax Gaps for Malaysia

(In percent of GDP)

The tax gap analysis suggests that there is considerable scope to raise Malaysia’s indirect tax revenues. The predicted norm for Malaysia’s consumption taxes is 6-6.3 percent of GDP, which suggests a gap of 2.7-3 percentage points. The predicted norm for direct income taxes, on the other hand, are lower than the current level of income taxes in Malaysia, suggesting that Malaysia does not have much scope to raise income taxes. The relatively high level of income taxes in Malaysia is due to corporate income taxes, while personal income taxes are relatively low. The regression model for other taxes, which mainly captures property taxes, does not have a high explanatory power and predicts a gap of only 0.1-0.2 percent of GDP for Malaysia. Given that property taxes are typically underutilized across all countries, the peer-based analysis may also underestimate the potential for using these taxes for revenue mobilization.

15. Analysis of tax potential. We also consider the potential to raise revenues from different taxes based on an assessment of current tax rates and the tax base in Malaysia. A country can raise higher revenues by increasing tax rates, broadening the tax base, or improving tax administration. Given that the latter is hard to estimate, we concentrate on the first two

  • Consumption taxes. Starting with the GST, we can project its revenue yield using a range of reasonable C-efficiency ratios and tax rates12. Based on a range of 0.5-0.7 for the C-efficiency ratio, the GST under the proposed rate of 6 percent would yield 1.9-2.6 percent of GDP, for a net revenue gain of 0.3-1 percentage points (the GST would replace current sales and service taxes yielding 1.6 percent of GDP). If the GST rate is set higher, at say10 percent, the net revenue gain can be 1.5-2.8 percentage points instead. While the set of exempt and zero-rated goods is not unusually large; it is still significant and increasing the revenue base through reducing exemptions and multiple rates can be considered going forward. The proposed GST rate of 6 percent is relatively low and is not likely to have a large initial revenue impact; however, instituting the GST is, itself an important step, since its parameters can be adjusted over time once the tax is operational. In fact many countries have increased VAT rates after its introduction.

  • Corporate income taxes. Malaysia’s tax system features a relatively high corporate income tax (CIT) rate and a range of tax incentives which narrow the tax base considerably. The authorities are currently conducting a cost-benefit analysis of existing incentives, which can inform future policy decisions on the granting of incentives and potentially help rationalize inefficient incentives. Although an estimate of total tax expenditures is not available, reducing some exemptions can certainly help raise revenues going forward. The relatively high CIT rate makes it difficult to raise it further and, in fact, the 2014 Budget announced that the CIT rate will be reduced by 1 percentage points in 2016, as an offsetting measure for the GST. We assume that rationalizing tax incentives could deliver up to 0.5 percent of GDP in fiscal savings, while recognizing the inherent uncertainty in these estimates. The authorities could also consider reducing the CIT rate further while rationalizing tax incentives, in a revenue-neutral manner that improves the efficiency of the corporate income tax system and raises investment and economic growth.

  • Personal income tax. Revenues from personal income taxes (PIT) in Malaysia are relatively low at 2.4 percent of GDP. This seems to reflect relatively low tax rates, in particular at the top of the income distribution. Based on the 2012 Household Income Survey, the top decile of households earns 32 percent of total household incomes and pays an effective income tax ranging between 15-25 percent.13 The 2014 budget announced that the PIT rates will be reduced across the board, including for the top of the distribution, by 1-3 percentage points. The new measure will also add new tax brackets at the top of the distribution, potentially allowing for more progressivity. Although the new schedule of PIT rates gives a larger tax relief (in percent) to lower and middle-income households, the improvement in progressivity is not very significant and the bulk of the revenue foregone will be due to lower taxes on high income households. Given Malaysia’s relatively low PIT rates, these could be further raised for the top income brackets going forward.

  • Property and wealth taxes. These are relatively low in Malaysia (1.1 percent) of GDP and are transaction-based (the property gains tax and stamp duties). Low yields are due to the fact that there are no recurrent taxes on property or other wealth. From an efficiency point of view, it is not desirable to use transaction taxes, although there might be other considerations—for example transaction taxes on property sales could be used to cool down a frothy property market. Instituting recurrent taxes on wealth would be a progressive and more growth-friendly measure that could be used in Malaysia. Although it is hard to estimate the potential revenue yield from such a measure, we assume that property taxes can generate 0.3-0.5 percent of GDP in additional revenues, which is a conservative assumption based on median recurrent taxes on property in OECD countries (at 1 percent of GDP).

  • Non-tax revenues. While Malaysia’s revenue effort relies mainly on tax revenues, non-tax revenues are also sizable and have increased over the past decade. The share of non-tax revenues has increased since 2000, driven mainly by rising oil prices,14 and has stabilized at around 30 percent of total revenues during 2006-2010. Since then its share has declined and in 2012 it was about 25 percent. The significant role of non-tax revenues is also consistent with the dependence on oil-and-gas related revenues in Malaysia. Oil and gas related revenues is about one-third of total and is composed mainly of the company income tax and the PETRONAS dividend. Although there may be scope to also increase revenues through adjusting oil and gas related taxes, in this paper we concentrate on the non-oil tax revenues given that one of the overarching goals of the consolidation strategy should be to reduce reliance on oil and gas related revenues. Secondly, the taxation of oil and gas sector requires a more detailed and customized analysis and is beyond the scope of this paper.

Gross Revenue Yield From GST Under Different Parameters

OECD and Selected Asian Countries: VAT Rates

(In percent)

Sources: International Bureau of Fiscal Documentation, 2013; and Deloitte, Global Indirect Tax Rote, 2013.

Tax Revenue to GDP Ratio in Upper-Middle Income and High Income Countries and Other Asian Countries 1/

(In percent)

Source: Torres (2013).

1/ Tax revenue excludes taxes from oil and gas companies.

Selected Countries: Top Corporate Income Tax Rate, 2013

(In percent)

Sources: University of Michigan, The World Tax Database; KPMG Global Corporate Tax Rate Tables; and IBFD Comparative Tables.

Selected Countries: Top Personal Income Tax Rates, 2013

(In percent)

Sources: KPMG Global Corporate Tax Rate Tables; IMFD Comparative Tables; World Tax Indicators Data Portal; OECD Top Statutory Marginal Tax Rates.

Personal Income Tax Rate For Different Income Groups—Current versus New Schedule

OECD Countries and Malaysia: Average Property/Wealth Taxes, 2000–2011

(In percent of GDP)

Share of Nontax and Oil- and Gas-Related Revenues

(In percent of GDP)

Sources: Malaysian authorities; and IMF staff calculations.

D. Expenditure Measures

16. Subsidy rationalization. This is acknowledged as having the greatest potential to generate large and immediate fiscal savings to support Malaysia’s consolidation strategy. The subsidy bill increased significantly during the 2000s, driven by rising international fuel prices; and reached 4.7 percent of GDP in 2012-13. The authorities have already increased fuel prices in September which is expected to reduce subsidies by 0.4-0.5 percent of GDP in 2014. Part of the savings will be used to fund direct cash transfers to low and medium-income households. Gradually raising domestic prices to reach parity with international prices and adopting an automatic pricing mechanism would ensure that the fiscal savings from subsidy rationalization is locked in and remains sustainable. Gradual phasing of fuel subsidy rationalization as opposed to an immediate one-shot adjustment has been more effective in other countries through spreading the impact of reforms on growth and inflation over a longer period of time. Therefore ensuring that the subsidy rationalization is gradual would be important to ensure the successful adoption of subsidy reform. A recent IMF paper (IMF (20013a) also highlights the importance of a comprehensive energy sector reform plan which entails clear and long-term objectives for the broader energy sector. In Malaysia, this entails designing its subsidy reform strategy to also address gas and electricity subsidies. Malaysia already took measures to reduce electricity subsidies, by increasing tariffs for electricity effective January 2014 and it plans to reduce gas subsidies as well. While these subsidies are mainly implicit subsidies (i.e. not fully reflected on the government’s fiscal accounts), reducing them would have a positive impact on revenues through improved profitability of PETRONAS. Fiscal savings from fuel subsidy rationalization could be in the range of 2.2-2.7 percent of GDP, taking into account the offsetting direct cash transfers. The authorities have also fully eliminated sugar subsidies. Other potential measures include rationalizing other subsidies such as subsidies on cooking oil, rice and flour.

17. Public sector wage bill. Containing growth in the public sector wage bill can also support the consolidation effort. The wage bill increased by 0.9 percentage points since 2008 to 6.2 percent of GDP, reflecting improvements in the salary scheme, a time-based promotion scheme for teachers, and additional bonus payments. Restricting hiring to essential personnel and limiting the establishment of new posts (as already adopted by the authorities) can help going forward. One area where it may be feasible generate savings is emolument spending in education, which accounts for close to 50 percent of total federal government spending on emoluments. Malaysia’s Education Blueprint notes that the student to teacher ratio in Malaysia is high compared to other countries in the region and highlights the importance of the quality of teachers as opposed to class size in improving educational outcomes

18. The efficiency of spending across the board. Broad-based efficiency gains in public expenditure can also be a source of fiscal savings. The authorities have already adopted measures to improve public procurement which can lead to important fiscal savings (about 0.2-0.3 percent of GDP). Assuming an administrative productivity tax of 1.5 percent implies savings of about 0.1-0.2 percent of GDP over 5 years.

19. Social safety nets and other transfers. There are multiple social assistance programs in Malaysia that operate separately and thus there could be efficiency gains from consolidating the various social assistance programs under one roof, allowing for a better targeting of these transfers. Under other transfers, there can also be savings. For instance, there is scope to reduce grants to public universities through higher tuition payments and rationalizing student loan subsidies which are not targeted.

20. The role of development spending. As discussed above, development spending has declined in recent years and thus offers limited scope for fiscal savings. Development spending is at its lowest levels since the 1990s relative to GDP, and further cuts can have adverse effects on short-term and long-term growth prospects. Although the authorities can make a more informed assessment of the scope for cutting development spending, we use 0.2-0.4 percent of GDP as a reasonable assumption. As mentioned before, productivity of investment matters significantly for the long-term growth and equity implications of investment.

E. Putting It Together

21. The principle. The composition and timing of tax and expenditure measures that can be adopted as part of Malaysia’s consolidation strategy should be guided by their growth and equity implications. There has been a growing literature on the growth implications of different taxes and expenditure measures; however, there is considerable uncertainty in these empirical and theoretical estimates. Our knowledge on instrument-specific fiscal multipliers is even more uncertain. In addition, the design of measures, and the broader macroeconomic environment, can play a significant role in how fiscal policy affects growth and equity. Estimates of fiscal multipliers for Malaysia point to a relatively low impact of fiscal policy on growth, with the exception of public investment, which is estimated to have relatively high fiscal multipliers (Table 2) 15. With respect to distributional implications of fiscal policy, it is not possible to rely on country-specific estimates. We therefore rely on the assessment used in several OECD studies on fiscal consolidations and our own judgment based on the design and nature of specific measures considered for Malaysia16.

Table 1.Malaysia and Selected Emerging Markets: Fiscal Multiplier Estimates
CountryStudyRevenueExpenditure
Malaysia
Rafiq and Zeufack (2012)About zero within one year, peak multiplier is 0.1/0.2 in normal times/recessions.2/2.7 for investment for normal times versus recessions, government consumption multiplier is negative in normal times, peaks at 2 in the very short-run and goes to 0 in one year for recessions.
Rafiq (2013)0.3/0.4 peak, 1.28/0.85 cumulative by year 4 for (pre/post Asian crisis)
Tang and others (2010)0.2-0.5 depending on identification method0.2-0.3 depending on identification method
IMF’s “Bucket Approach”0–0.20.15–0.45
Other Emerging Markets
Panel EMsIlzetki (2011)0.30.2
ThailandTang and others (2010)1.0−0.4
South AfricaJooste (2012)0.70.3
SingaporeTang and others (2010)0.5−0.2
PhilippinesTang and others (2010)0.10.4
PeruAnos-Casero and others (2010)0.00.0
IndonesiaTang and others (2010)0.4−0.3
Table 2.Malaysia: Potential Measures for Fiscal Consolidation
Type of MeasuresCommentsTime FramePotentialGrowthEquity
YieldShortLong
2014–2020termterm
Revenue
GSTBudget 2014 announced the introduction of the GST in 2015 with a 6 percent rate. Base broadening through reviewing exempt and 0-rated goods, and raising the standard rate can be used to increase the revenue yield.Given that the GST will be introduced in 2015, initially the revenue base can be widened, followed by a rate hike later in 2017–2018 if needed.0.5–3.0-~/-~/-
Property taxesBroaden recurrent taxes on property/wealthCan be introduced gradually during 2015/16–2020.0.3–0.5~/-~++
Personal income taxesBudget 2014 announced the reduction in various marginal rates including the top rate effective 2015. T he top PIT rate in Malaysia is relatively low and could be raised going forward, while leaving the rates for low and middle-income households intact.Although there is scope to raise the top rate going forward, this can only be done after 2017–2018.0.2–0.5~/-−−++
Corporate income taxesBudget 2014 announced the reduction in CIT rate effective 2016, which is estimated to reduce revenues by about 0.2 percent of GDP. The authorities can rationalize incentives and this may yield additional revenues, however, the revenue impact is uncertain and would depend on the type of incentives. Alternatively the authorities could choose to reduce rates while rationalizing incentives, this could be designed as a revenue-neutral strategy to improve efficiency and to the extent that tax incentives and holidays have not had the desired results (for example on growth) this strategy may have positive growth effects.An assessment of the cost and benefits of tax incentives needs to be conducted to formulate a strategy to rationalize incentives that are not effective. This could be done in the short to medium-run.0.2–0.5+
Expenditure
Wage billWage bill has increased considerably in recent years, and emoluments at the level of basic education remain relatively high. The government can radualy reduce the number of teachers per student while instituting a performance-based pay system for teachers that aims to improve the quality of teachers.The adjustment could be implemented gradually during 2014–20200.3–0.6-/−−~~
Other public consumptionAn administrative productivity tax of 1.5 percent on whole of government as a proxy for efficiency gains and improvements in procurement.The adjustment could be implemented gradually during 2014–20200.3–0.5−−~~
SubsidiesRationalizing fuel and non-fuel subsidies and partially offsetting these cuts with direct cash transfers. Adopting an automatic pricing mechanism for domestic fuel prices would be needed to sustain fiscal savings.The authorities already made a price adjustment in September and further hikes can be adopted in 2014–2015, phasing the adjustment gradually over time.2.5–3.0-++++
Public investmentGovernment’s direct development spending is projected to decline to its lowest level (relative to GDP) in 2014 since the1990s, and there is not much scope to reduce public investment further.Public investment cuts can be quicly implemented as they are more discretionary but given the high multipliers, it woud not be advisable to use investment cuts when the economy is slowing down.0.2–0.4−−−−−−−−
Social safety nets and other transfersImproving the targeting of social transfers can yield fiscal savings. Cash transfers benefit a significant proportion of the population and can thus be targeted more to low-income households. Putting together a comprehensive database for social assistance beneficiaries would be a good first step to consolidate different social assistance programs. Reducing grants and subsidies to public universities through increased tuition is another area that can generate fiscal savings.Reforming social safety nets can take time, so potential fiscal savings may be realized in the medium-term. Reducing other transfers gradually should be feasible.0.4–0.5-~~/+

22. Bottom line. It is useful to summarize how different fiscal instruments perform along the growth and equity dimensions (Table 3). More negative signs suggest a larger negative impact on growth or equity; the converse is true for the positive signs. These rankings can guide policy makers in selecting the composition of fiscal consolidation, based on the preferences and weights they assign to different goals.

Table 3.Ranking of Different Measures
MeasuresYield (% of GDP)Short-term growthLong-term growthEqualEquity Weight
Subsidies33111
Property taxes0.51212
Social safety nets and other transfers0.53253
Personal income taxes0.51814
Corporate income taxes0.53745
Wages0.67266
GST33687
Other public consumption0.58267
Public investment0.49999

The ranking based on equal weights are computed using numerical scores for each instrument for different policy goals, ranging between -3 (−−−) to +2 (++), representing the range of scores presented in Table 2. We compute the weighted score for each instrument based on equal weights for all policy goals and then use these scores to rank policy instruments.

The ranking based on equal weights are computed using numerical scores for each instrument for different policy goals, ranging between -3 (−−−) to +2 (++), representing the range of scores presented in Table 2. We compute the weighted score for each instrument based on equal weights for all policy goals and then use these scores to rank policy instruments.

23. On the revenue side, we consider the relative performance of income, consumption and property taxes.

  • Growth. All revenue measures have negative, short-term effects on growth (represented by a single negative sign), but income taxes also have large negative impact on long-term growth. This follows from the distortions introduced when raising the tax burden on variable and mobile factors of production, which have a negative supply-side impact on output (represented by double negative sign). The GST and other consumption taxes are less distortionary, but can still reduce output in the short-run depending on the design of the tax. The GST that Malaysia plans to introduce in 2015 has some exempt and zero-rated products to protect low-income households and will also be implemented together with a package of offsetting tax and transfer measures. These measures will thus help reduce the impact of the GST on growth. However, this also means that the revenue yield from GST will be low. For GST to generate a higher revenue yield, a combination of base broadening and rate hikes could be considered. These measures could in return imply a relatively higher impact on growth (represented by a single negative sign for long-term growth impact). Finally, there is broad consensus in the literature that wealth taxes, such as recurrent taxes on property, are generally more growth-friendly. Furthermore, to the extent that property and personal income tax measures are designed to be progressive, their impact on growth in the short-term could be relatively benign (represented by ~/- to signal no effect or a small negative impact).

  • Equity. Personal income taxes and property taxes are typically the most progressive tax instruments, followed by corporate income taxes (represented by double and single positive signs). Consumption based taxes, such as the GST, are typically more regressive as higher-income households consume a smaller share of their income. To contain the adverse effects of consumption taxes on equity, exemptions and zero-rating of certain consumption products are commonly introduced. However, there is a trade-off between equity and efficiency in the design of the GST. Base broadening improves efficiency but reduces equity. Higher rates intensify the regressive nature of the GST, but also have negative effects on efficiency. As discussed above, the GST that will be implemented in 2015 is likely to have a mild impact on equity given the exemptions and zero-rated products.

24. On the expenditure side, we consider a range of different instruments, including cuts in government consumption, subsidies and public investment.

  • Growth. Subsidy reform ranks high in terms of its growth implications, by improving the efficiency of resource allocation throughout the economy and enhancing output, especially in the long run (represented by small negative effect in the short-run and positive effect in the long-run). The impact of cuts in the wage bill can also have fairly large negative effects on growth in the short-run. Cuts in government consumption have a relatively large negative growth effect in the short run. Consistent with previous studies on Malaysia, cuts in productive public investment are assessed to have the largest adverse growth effects, in both the short and long runs (represented by triple negative signs).

  • . Equity. Subsidy rationalization ranks high in terms of its equity implications. Through improving the targeting of fiscal transfers, subsidy rationalization can serve a significant role in improving the equity of fiscal policy in Malaysia. Improving the efficiency of spending on social safety nets, improving the targeting of subsidies on education loans and reducing grants and subsidies to public universities through increased tuition are other options that could be designed in a progressive way and improve equity. Cutting investment spending can on the other hand have adverse effects on equity. Consistent with evidence reported in Seneviratne and Sun (2013) for ASEAN-5 countries, infrastructure spending can have significant negative impact on income inequality.

25. Ranking of instruments and priorities. The rankings of different revenue and expenditure measures with respect to growth and equity are summarized in the figures above. Preserving short-term growth calls for relying more on rationalizing subsidies, improving the targeting of social safety nets and relying on revenue measures. Subsidies and property taxes rank high in terms of their impact on long-term growth. Subsidy reform also ranks high with respect to its beneficial effect on equity. Property taxes and income taxes are among other measures that rank high in improving equity. These rankings reflect some element of judgment and the impact of these measures on growth and equity would ultimately depend on their design and implementation. The analysis presented in this paper can therefore be revised to reflect different views on the relative trade-offs among different goals of fiscal policy

Trade-Offs for Different Tax Measures

Trade-Offs for Different Expenditure Measures

26. Composition of fiscal adjustment. Ranking fiscal instruments under different fiscal policy goals can help policy makers identify the composition of fiscal adjustment based on their preferences. For instance if the policymaker assigns equal weight to all goals, then one can compute a weighted average of scores for different instruments. Combining this ranking with the instruments’ potential yield, one can identify the optimal set of measures required to achieve the needed fiscal adjustment. For illustration, we consider a set of preferences whereby the policymakers give equal weight to all three objectives: short-term growth, long-term growth and equity. The ranking of instruments is provided below, with the set of measures needed to meet the consolidation needs highlighted in bold and italics. Subsidy reform and property taxes are ranked at the top, followed by improving the targeting of social safety nets, personal income taxes and corporate income taxes. Although not reported here, the ranking of the top three instruments are robust to different preferences or weights. The GST does not rank very high due to its relatively benign effect on equity, whereas other measures typically improve equity. Although it ranks lower than most other measures under this set of preferences, the consolidation strategy calls for using the GST given its high potential yield which is critical to meet the government’s adjustment needs. Other important considerations not reflected in the relative rankings also call for the implementation of the GST. First, the GST should help reduce reliance on volatile oil and gas revenues and broaden Malaysia’s revenue base, thereby reducing fiscal volatility and risks. Secondly, instituting the GST also constitutes an important structural reform, which provides the authorities with an important lever to raise additional revenues in the future; and thus should have a positive impact on Malaysia’s debt sustainability.

Appendix 1. Tax Revenue Gap Regressions
Table 1.1.Tax Revenue Gap Regressions (Full Sample)
Dependent Variable
VariableTotal RevenuesConsumptionTaxes Income TaxesPayroll Taxes Other Taxes
Commodity exports−0.15−0.06−0.05
−(5.4)−(5.0)−(3.4)
Old-age dependency ratio0.590.200.170.270.05
(5.2)(3.7)(2.6)(6.4)(3.5)
Per capital income0.07−0.070.130.01
(1.9)−(4.3)(7.0)(1.8)
Europe Dummy6.413.82−2.405.95
(3.3)(4.4)−(2.0)(7.6)
Middle East Dummy−4.58−2.96
−(3.1)−(3.5)
Asia and Pacific Dummy−2.84−2.18
−(1.8)−(2.3)
Western Hemisphere Dummy−2.190.85
−(2.6)(1.5)
Constant13.735.525.03−1.290.37
(12.1)(10.8)(8.1)−(3.2)(2.3)
Adjusted R-squared0.700.540.470.770.16
Number of Observations157157157157157

t-values are reported under the coefficients in parantheses.

t-values are reported under the coefficients in parantheses.

Table 1.2.Tax Revenue Gap Regressions (Higher-Middle and High Income Countries Sample)
Dependent Variable
VariableTotal RevenuesConsumption TaxesIncome TaxesPayroll TaxesOther Taxes
Commodity exports−0.13−0.06−0.05
−(4.4)−(3.8)−(2.3)
Old-age dependency ratio0.590.160.130.340.08
(4.9)(2.3)(1.5)(5.8)(3.3)
Per capital income0.07−0.070.140.01
(2.3)−(4.0)(6.5)(1.7)
Europe Dummy5.654.15−3.394.74−0.86
(3.1)(4.1)−(1.8)(4.9)−(2.1)
Middle East Dummy−7.30−5.93
−(3.8)−(3.5)
Asia and Pacific Dummy−5.53−3.90
−(2.6)−(2.0)
Western Hemisphere Dummy−3.55
−(2.2)
Constant14.486.156.64−1.550.12
8.837.034.36−2.140.39
R-squared0.830.610.530.730.15
Number of Observations8787878787

t-values are reported under the coefficients in parantheses.

t-values are reported under the coefficients in parantheses.

References

Prepared by Elif C. Arbatli (APD). The author would like to thank Martine Guerguil (FAD), Irineu de Carvalho Filho (RES), Kenji Moriyama (SPR), Christian Saborowski (SPR) and Sampawende Jules Tapsoba (FAD) for very helpful suggestions and comments.

Staff estimates the fiscal deficit in 2013 to be about 4.3 percent of GDP based on the GFSM 2001 standard. Based on the authorities modified cash-based accounting system, the 2013 deficit is below 4 percent of GDP. The difference is attributable to differences in methodology/basis of recording and the treatment of certain items.

Fiscal multipliers are assumed to be about 0.3.

Estimates of public investment multipliers are in the 2-2.5 range.

A fuller analysis of the optimal mix of oil- and non-oil taxes in Malaysia is beyond the scope of this paper.

Indirect taxes in Malaysia declined from about 9.7 percent of GDP on average during 1980-1984 to 3.7 percent of GDP by 2012, driven entirely by a 6 percentage points of GDP decline in international trade taxes since 1980.

See Baunsgaard and Keen (2010) for an analysis of how and to what extent developing and emerging market economies have replaced the decline in trade taxes with other taxes.

For instance, due to differences in the level of informality.

This database was also used in the 2013 Fiscal Monitor to estimate tax gaps.

We control for exports of oil and gas to account for the fact that countries with resource revenues tend to rely less on non-resource tax revenues. Our results for Malaysia, therefore already account for the fact that it also relies on resource revenues.

See Tables 1.1 and 1.2 in the Appendix for detailed regression results.

The c-efficiency ratio is the ratio of actual VAT revenues to the potential revenue base which is total consumption times the standard VAT rate. The c-efficiency ratio would be 1 if there were no exempt goods or multiple rates and if there were no tax avoidance. The c-efficiency ratio therefore reflects the productivity of the VAT.

In this calculation, we assume that there is a single income earner in each household, which is unlikely to be true in particular for the higher income household groups. This effective income tax rate can therefore be interpreted as a higher bound.

The largest component of non-tax revenues is PETRONAS dividends, followed by licenses and permits and oil and gas royalties.

The empirical estimates and a rule-of-thumb approach yield values for consumption and revenue multipliers for Malaysia that are similar to other EMs. The rule-of-thumb approach is developed by IMF staff to help desk economists make appropriate assumptions on fiscal multipliers based on structural characteristics of the country such as openness and level of debt. See the “IMF’s bucket approach” in Table 2.

Other Resources Citing This Publication