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

Thailand: Selected Issues

Author(s):
International Monetary Fund
Published Date:
August 2001
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

VI. Medium-Term Debt Outlook1

A. Introduction

1. In the decade preceding the financial crisis, Thailand’s public debt ratio fell steadily, to around 5 percent of GDP in 1995/96 (Text Figure).2 The decline in debt was driven by a combination of sizeable primary surpluses, which peaked at more than 6 percent of GDP, and high economic growth rates (in excess of the real interest rate). These factors were mutually reinforcing, as the high real growth rates created an economic environment conducive to running fiscal surpluses.

Thailand: Debt Ratio, Central Government and FIDF Debt,

1970/71 - 1999/00 (In percent of GDP)

Sources: GFS. IFS. Thai authorities, and staff estimates

2. Since the crisis, however, the debt ratio has risen sharply. Thus, from around 5 percent of GDP before the crisis, the debt ratio increased to 39 percent of GDP ($48 billion) by end 1999/2000. The increase in debt is notable not only for the pace with which it has accumulated, but also for its level, which is high by historical standards. The increase is even larger when measured using a broader (and more commonly cited) debt aggregate that includes non-financial public enterprises (NFPEs). On this basis, the debt ratio has risen from 14 percent of GDP before the crisis to 57 percent of GDP ($70 billion) at end 1999/2000 (Text Table).

Thailand: Public Debt, 1995/96–1999/2000(In percent of GDP)
1995/961996/971997/981998/991999/00
1. Central govt. + FIDF Liabilities5.224.131.838.638.8
Central government3.95.110.720.922.6
Domestic1.10.84.913.014.6
Foreign2.84.45.87.88.0
FIDF liabilities1.319.021.117.816.2
2. Non-financial public enterprises9.312.113.417.218.5
Of which: Foreign5.07.28.010.59.5
3. Total (1+2)14.536.345.255.957.3
Sources: Country authorities; and staff estimates.
Sources: Country authorities; and staff estimates.

3. The increase in the debt ratio is due largely to the costs of financial sector restructuring. Indeed, over two-thirds of the increase in debt is directly attributable to the financial sector restructuring costs (see below). The remainder is mainly due to expansionary fiscal policies in support of the economic recovery.

4. Even though Thailand’s government debt ratio is high by its own historical standards, the debt ratio does not compare unfavorably with other countries (Text Figures). Differences in data coverage complicate comparisons, especially with regard to the accounting treatment of the costs of financial sector restructuring. However, based on a broadly uniform definition, the relative size of the debt ratio in Thailand is estimated to be comparable to that of Korea and Malaysia, and below that of Indonesia and the Philippines. Thailand’s debt ratio also does not compare unfavorably with OECD countries, although important differences between levels of development and structures of OECD economies relative to Thailand must be borne in mind (for example, the OECD countries typically have deeper tax bases and financial markets than in the case of Thailand).3

Asia Region Pabllc Debt, 1996–2000

Excludes non-financial public enterprises, (percent of GDP)

OECD: General Govt. Gross Financial Liabilities

2000 (In percent of GDP)

5. Nevertheless, Thailand’s debt ratio has risen steeply in a short period of time. Any static comparison of debt levels must be interpreted carefully as these are not indicative of future debt-dynamics and mask underlying differences in levels of economic development and relative importance of the public sector in the economy. A better indicator is instead the change in the debt ratios. Under this metric, Thailand does not compare favorably with other OECD countries, as the increase in Thailand’s debt ratio occurred during a time when nearly all of the OECD countries were pursuing policies that succeeding in substantially reducing their debt ratios. In this regard, the magnitude of Thailand’s increase in government liabilities since 1996 has been among the highest in the crisis-hit Asian countries, and is on par with the increase in Japan (Text Figures above).

OECD: Increase In General Govt. Gross Financial Liabilities,

1996–2000 (In percent of GDP)

B. The Costs of Financial Sector Restructuring

6. As noted above, the largest single contributor to the increase in government debt has been the cost of supporting the ailing financial system. By the end of fiscal year 1999/00, the stock of debt for financial sector assistance, including cumulative principal costs plus interest costs, had reached about 30 percent of GDP (or $33 billion, at the then prevailing exchange rate). Of this: (i) roughly 14 percent of GDP was related to the resolution of the 56 closed finance companies; (ii) 10 percent was mainly due to the recapitalization and resolution of the state and intervened banks, and (iii) the remaining 6 percent was due to the cumulative carrying cost of these principal injections.

7. The financial sector restructuring costs are associated with the considerable public sector liquidity and capital support provided to the financial system. Beginning in early-1997, the FIDF4 provided liquidity support to the finance company sector, peaking in mid-year with the suspension of 56 finance companies (Text Figure). As the crisis unfolded and confidence in the overall financial sector waned, additional support was provided to a number of weak banks that were eventually intervened. Ultimately, the bulk of the overall liquidity injections were essentially transformed into capital support, while state-owned financial institutions also received large capital injections necessitated by their extensive credit losses. The FIDF’s overall balance sheet increased substantially as a result of these activities, and its accumulated losses have resulted in a large negative net worth position on the basis of currently recognized cash and contingent liabilities (Text Figure).5

FIDF Support ID Financial Institutions, 1997–2001

(baht trillions)

FIDF Balance Sheet, 1994–2000

(baht trillions)

8. To date, the financing of the public sector costs of financial sector restructuring has come entirely from the domestic market (Text Figure). The main sources have included borrowings from the money included borrowings from the money market, including through repurchase operations, issuance of government bonds, and recoveries from FRA’s sale of assets from the intervened finance companies. Especially during 1997–98, the FIDF financed its operations through large borrowings in the repo market, where it effectively recycled liquidity that had been built-up in stronger financial institutions as depositors fled weak finance companies and banks.

Structure of FIDF Funding, 1994–2000

(baht trillions)

9. Over time, the FIDF has decreased its reliance on short-term money market borrowing. The government issued domestic bonds totaling B 500 billion (equivalent to about 10 percent of GDP) during 1998–99 as a first step towards fiscalizing the costs incurred by the FIDF in supporting financial sector restructuring. Meanwhile, with the winding up of FRA operations, approximately B 175 billion of its auction proceeds have been placed on deposit at the FIDF, further contributing to the reduction in borrowing from the repo market. The small switch back into repos in 2000 has been matched by a reduction in money market and other borrowings. In recognition of the maturity risks inherent in short-term borrowing from the repo and money markets, the authorities have also decided to allow the FIDF to issue up to B 200 billion in longer dated instruments during 2001, guaranteed by the Ministry of Finance. The FIDF has thus far auctioned such bonds worth about B 85 billion, mostly in the 2–5 year duration.

10. Meanwhile, the process of liquidating the assets of the 56 closed finance companies by the FRA is coming to an end. The last major Financial Sector Restructuring Agency (FRA) auction was completed in early-2000. Thus far, the FRA has auctioned about 85 percent of the roughly $22 billion of core and non-core assets acquired from the closed finance companies, with an average recovery rate of 35 percent of the principal amount. About $5 billion (or 23 percent of the total assets on sale) have been acquired by the state-owned AMC.

11. Interest payments on the component of public sector debt associated with financial sector assistance have ranged in recent years between 1½-2½ percent of GDP. The interest costs can be divided into two categories :fiscalized and non-fiscalized. The first refers to the interest cost on bonds issued by the government to compensate FIDF for part of its losses. These interest expenses are “fiscalized” as they are paid out of the central government budget. The “non-fiscalized” component of interest costs is imputed on the basis of FIDF’s financing in the (short-term) repo and money markets. These imputed costs also include interest payments made by the FIDF to the holders of promissory notes issued to depositors and creditors of the closed finance companies (under the so-called “note-exchange program”), as well as yield maintenance payments on the nonperforming loans of the intervened banks that have been merged with other state-owned institutions. They are however net of interest incomes received by the FIDF but exclude receipts accruing to FIDF from the deposit guarantee fee, which are instead counted towards FIDF’s capital accounts.

C. Projections and Sensitivity Analysis

12. In the near-term, Thailand’s government debt is expected to continue to rise. The debt to GDP ratio is projected to peak in year 2005/06, driven by a continuation of fiscal deficits, and the realization of additional costs for financial sector restructuring (see below). Any privatization receipt would reduce the debt level below the estimates presented here.

13. However, over the medium term, fiscal consolidation and a return of robust economic growth could put Thailand’s public debt ratio on a declining path (Text Figure). The key assumptions underlying this outlook are: (1) a reversion of the VAT rate to 10 percent in 2002/03, (2) a gradual increase in GDP growth to 5–5½ percent; (3) from 2002/03 onward, constant non-interest expenditure at the 2000/01 ratio to GDP; and (4) some modest buoyancy in revenue as the output gap is gradually eliminated (see the Technical Annex for details). Reversion of the VAT to 10 percent is projected to increase revenue by around 1¼ percent of GDP a year. Together with the general revenue buoyancy, the primary balance would improve by around 3 percentage points of GDP.6 However, the main contributor to the falling debt ratio would be GDP growth (see Text Table on next page). As mentioned above, the decline in the debt ratio would not begin for several years: the debt ratio would start declining only after reaching a peak of around 45 percent of GDP in 2005/06. This highlights the risks to the scenario, and the dependence of the debt dynamics on strong economic growth.

Thailand: Central Government and FIDF Debt-Contributions to Changes in Debt/GDP, 2001/02–09/10
2001/022002/032003/042004/052005/062006/072007/082008/092009/10
Fiscal adjustment scenario (in percent of GDP)
Debt43.743.943.045.844.842.740.638.736.9
Change in debt2.10.2–0.82.8–1.1–2.1–2.1–1.9–1.8
Contribution to debt change (+ is increase in debt)
Primary balance1.60.60.2–0.1–0.4–1.1–1.1–1.1–1.1
Interest payments2.72.72.52.52.42.32.22.12.0
Principal costs of fin. restruct.0.40.1–0.23.70.4–0.2–0.20.00.0
Nominal GDP growth–2.6–3.0–3.3–3.2–3.4–3.2–3.0–2.9–2.7
Source: Staff projections
Source: Staff projections

Thailand: Projections of Central Government and FIDF Debt

(In percent of GDP)

Sources: Thai authorities and staff estimates

14. Without fiscal adjustment, on the other hand, the debt ratio would remain relatively flat. Indeed, in the absence of the VAT reversion, the debt ratio would be projected to remain above 45 percent by 2009/10. This underscores the centrality of fiscal adjustment for reducing the debt ratio. It also highlights that even under relatively favorable interest rate and growth assumptions, Thailand cannot simply grow out if its debt.

15. Financial sector restructuring costs will continue to have a large impact on the debt ratio for the next several years. This is seen by the large hump in the ratio in 2004/05. The increase largely reflects additional principal costs associated with the final resolution of intervened and state banks.7 The key assumptions underlying these projections are (i) recovery rates of between 40–45 percent; (ii) a gradual pattern of asset resolution; and, (iii) and a pick-up in interest costs. The other main contributors to the evolution of debt are an improvement in the primary balance stemming from the assumed revenue buoyancy, and GDP growth in excess of real interest rates (see Text Figure). Fiscal adjustment could be also be complemented by other measures to improve the medium-term efficiency of the tax system (see Box VI. 1).

Thailand: Contributions to Increase in Debt

(In percent of GDP)

Note: Baseline growth and VAT at 7%

Sources: Thai authorities and staff estimates

Box VI.1.Medium-Term Tax Reforms

Overall, the tax system in Thailand is modern, reasonably efficient, and in broad conformity with international good practices. The key challenges over the medium term, therefore, are to pursue structural reforms directed at improving some of the weaker aspects of the tax system, and to resist pressures for policies that would weaken the tax system. In addition to the reversion of the VAT—which is relatively low but efficient by regional standards (see Text Table)—the priority areas for reform include the following:

  • Simplify, improve the fairness, and enhance the revenue buoyancy of the personal income tax (PIT). Reforms could be aimed at broadening the base by scaling back allowances and deductions; improving the fairness by replacing retained allowances and deductions with tax credits; and lowering the marginal rates of taxation. In addition, the system of capital taxation should also be modernized. In a regional context (Asian PIT yields tend to be lower than in non-Asian OECD countries), the PIT in Thailand yields relatively little revenue, and its buoyancy over the medium term could be increased.

  • Modernize the tax treatment of financial services by replacing the specific business tax (levied on gross turnover or transactions) with a VAT-type tax, or by simply eliminating it and exempting financial services from the VAT. Such a modernization would also likely reduce the tax burden on the financial sector.

  • Eliminate or replace Board of Investment (BOI) tax incentives. The present system of incentives offers significant tax holidays and rate reductions, is difficult to administer, not very transparent, and expensive in terms of foregone revenue. One option would be to completely eliminate the present incentives. Another, and second-best, option would be to replace the present incentives by a system of investment tax credits based on investment cost recovery, with the corresponding lost revenue recorded in the budget as a tax expenditure. Such a system would be more transparent, and easier to administer.

  • Reform the social security system. The recently established old age pension system (OAPS) is most likely not financially viable over the long run as benefits would eventually exceed contributions. Although this problem would not emerge for some time (perhaps 40 years or so), it would be beneficial to implement the necessary reforms now while the system is still at an early stage of implementation and there are not many beneficiaries. Options for improving the long-run viability of OAPS include increasing the retirement age and lengthening the qualification period.

Selected Countries: Comparison of Tax Rates and Revenue
VATTotal and personal income tax
YearMain rate

(1)
Revenue

(In percent of

GDP)

(2)
Efficiency

(2)/(1)

(30)
YearTotal tax

(In percent of

GDP)

(4)
Personal

(In percent of

GDP)

(5)
Personal/

Total

(In percent)

(6)
Thailand1999/20007.03.30.481999/200013.61.813.2
Australia10.0199729.812.541.9
Cambodia10.01997
China199817.05.30.311997
Indonesia199810.02.70.27199715.61.49.1
Japan19985.02.10.41199718.25.932.4
Korea199910.04.20.42199719.53.618.5
Malaysia1999199916.02.113.1
New Zealand1999/200012.58.40.67199736.415.743.1
Philippines199910.03.20.32199716.82.213.0
Singapore1997/983.01.40.471997/9816.32.012.3
OECD Average199727.610.235.2
Sources: Country authorities, OECD; and staff estimates
Sources: Country authorities, OECD; and staff estimates

Sensitivity analysis

16. The medium-term debt dynamics are highly sensitive to the underlying assumptions, including with regard to economic growth, interest rates, buoyancy of the tax system, and the costs of financial sector restructuring.

  • A prolonged period of slow growth could result in a continuously rising debt ratio (Text Figure).8 Slower growth increases the debt ratio by reducing nominal GDP (the denominator of the debt ratio), and also by adversely affecting revenue buoyancy. Under a slower growth scenario, even with fiscal adjustment of the magnitude contemplated earlier in this paper, the debt ratio would continue to grow and thus be on an unsustainable path. Without fiscal adjustment, slow economic growth would make the debt ratio grow even more sharply, exceeding 65 percent of GDP by 2009/10. This sensitivity analysis highlights the risk of waiting too long to begin the process of consolidation.

  • The debt-dynamics are also sensitive to interest rate assumptions. An increase in the assumed average interest rate on outstanding debt by 3 percentage points per year (from 6½ to 9½ percent) from 2002/03 onwards would raise the 2009/10 debt ratio by about 10 percent of GDP (see figure).9 in a “worst case” scenario (not shown) with slow growth, no VAT reversion, and the higher interest rate, the 2009/10 debt stock would exceed 80 percent of GDP. Even with robust growth and the VAT at 7 percent the debt-dynamics would be explosive (mainly because interest rates would exceed growth rates). Whereas under the baseline interest rate assumption the debt ratio falls even though there is a small primary deficit in the outer years (around ½ percent of GDP), the reversal of the beneficial relationship between interest rates and growth (through higher interest rates) results in even a small primary deficit leading to explosive debt dynamics.

  • Revenue buoyancy also contributes significantly to the debt dynamics. Revenue buoyancy comes from the closing of the output gap and the projected increase in the ratio of consumption to GDP (which boosts the VAT and excise revenue). In the baseline growth scenario, the effect of holding revenue buoyancy at zero is to increase the 2009/10 debt ratio by around 10 percent of GDP (the majority of which is attributable to holding the output gap constant). For comparison, in the low growth scenario the 2009/10 debt ratio would be around 20 percent of GDP higher than in the baseline scenario. This 20 percent of GDP increase could thus be decomposed into the impact of revenue buoyancy and the relationship between interest rates and growth. Specifically, half of the increase arises from shutting down the revenue buoyancy in the low growth scenario and the other half from altering the relationship between interest rates and growth, as highlighted above in the discussion of the higher interest rate scenarios.

  • The projections are also sensitive to the costs of financial sector restructuring. On the up-side, the current projections assume no privatization proceeds, although the FIDF and MOF have currently projected that sale of some of the state-owned banks could generate as much as B 200 billion in privatization revenues (about 4 percent of GDP). Further, if the TAMC is highly effective, recovery rates could be somewhat higher than those currently assumed. For example, if rates similar to those registered by Danaharta are achieved (about 65 percent of the face value of assets acquired), by end 2004/05 the stock of debt could be reduced by about 3½ percent of GDP. On the other hand, the risk remains that recovery rates could be lower than assumed, reflecting the poor quality and hard to manage (fragmented) nature of state bank assets. In this case, if recovery rates were instead 20 percent on average, this would add about 5 percent of GDP to the 2009/10 debt stock.

Thailand: Projections of Central Government and FIDF Debt

(In percent of GDP)

Note: Low growth = 3 percentage points below baseline.

Sources: Thai authorities and staff estimates

Thailand: Projections of Central Government and FIDF Debt

(Baseline growth, in percent of GDP)

Note: High interest rate = 3 percentage points above baseline.

Sources: Thai authorities and staff estimates

D. Conclusion

17. Fiscal consolidation and a resumption of strong economic growth are needed to put the government debt ratio on a declining path in the medium term. The sensitivity analysis presented above illustrates that debt projections are highly sensitive to assumptions about medium-term macroeconomic developments, especially GDP growth and interest rate levels. Should developments not be as favorable as currently projected, the debt ratio would remain high. This underscores the importance of taking action over the next few years, particularly with respect to returning the VAT to its previous level of 10 percent, to consolidate the deficit.

Technical Annex

1. The main definition of debt used in this study is the sum of central government debt and the FIDF liabilities. This differs from the Government Finance Statistics (GFS) definition of government debt, which excludes the FIDF liabilities as they are considered to be part of the financial public sector liabilities. For analytical purposes, however, it is useful to include these liabilities in the definition of public debt, since the FIDF financial sector restructuring activity is of a fiscal nature and the FIDF liabilities are conceptually interchangeable with government debt (indeed around 10 percent of GDP in the FIDF losses have already been fiscalized). Consistent with GFS, other financial sector public debt, such as state bank liabilities or BOT debt, is excluded. Non-financial public enterprise (NFPE) debt is also excluded on the grounds that the NFPEs reportedly have substantial positive net worth, are largely profitable, and thus are unlikely to present a future fiscal burden. Moreover, depending on privatization developments, NFPEs could actually contribute to a reduction in government debt. The remainder of this annex describes the methodology used to project the various fiscal variables that contribute to the evolution of the debt ratio.

Central government

2. The revenue projections are based on a model linking revenue to changes in the output gap (the difference between potential and actual GDP). The projections are done separately for each major component of revenue, for which a different tax base and elasticity are assumed. Changes in the revenue to GDP ratio for a given component are thus driven by (1) policy changes; (2) changes in the revenue base as a share of GDP; or (3) movements in the output gap. The intuition behind the latter is that, setting aside policy or revenue base changes, the structural budget balance (at least from the revenue side) should be constant. Structural revenue is defined as,

where rs is the structural revenue to GDP ratio, rt is the revenue to GDP ratio in period t, ytp is potential GDP in period t, yt is GDP in period t, and α is the elasticity. Holding rs constant across time implies that the revenue to GDP ratio in t+1 is given by,

To account for the fact that each revenue component may have a different base, the actual formula for revenue item X is given by,

where B is the relevant base, and y is nominal GDP. While the selection of the base is relatively straightforward, the choice of elasticity is more subjective and is based on an assessment of historical performance (see Text Table). The results, however, are not highly sensitive to the choice of elasticity.

Thailand: Revenue Elasticity Assumptions
TaxBaseElasticity
Personal incomeNominal1.20
Corporate incomeNominal1.80
VATPrivate1.30
ExcisePrivate1.00
Specific businessNominal1.00
CustomImport1.00
Other taxNominal1.00
NontaxNominal1.00
Source: Staff
Source: Staff

3. The expenditure projections are more straightforward. Non-interest expenditure is held constant as a share of GDP at the projected 2000/01 level. Unlike revenue, expenditure is not directly linked to the business cycle, justifying holding it constant as a share of GDP. Nonetheless, a case could be made for both increasing the ratio over time (for example due to the costs of fulfilling education and other development objectives) or decreasing it (for example due to declining foreign financed expenditure). Interest expenditure is calculated by multiplying the beginning of period debt stock by the assumed nominal interest rate, which is best interpreted as the average interest on outstanding debt. This interest rate is assumed to stay at around 6.5 percent, or 4 percent in real terms (see below).

4. Extra-budgetary funds are excluded from the projections. The main reason is that their activities by and large do not directly affect the debt stock. The largest of the extrabudgetary funds, those managed by the social security office (such as for the old age pension system), have accumulated financial assets, are expected to continue to run surpluses throughout the projection period, and therefore would not be expected to borrow or otherwise impact the debt-dynamics. Consistent with this, holdings of government securities by these funds are not subtracted from the outstanding debt stock.

5. The main macro economic assumptions for the baseline scenario are summarized in the following Text Table. A summary of the fiscal projections for the baseline and adjustment scenario are reported in the attached tables (Tables VI.1 and VI.2 respectively).

Thailand: Medium-term Macroeconomic Assumptions(In percent)
2000/012001/022002/032003/042004/052005/062006/072007/082008/092009/10
Real GDP growth2.13.74.75.45.55.55.05.05.05.0
Inflation (GDP deflator)2.42.92.72.52.52.52.52.52.52.5
Nominal GDP growth4.66.87.58.08.18.17.67.67.67.6
Output gap–6.8–6.7–5.9–4.6–3.6–2.70.00.00.00.0
Avg. interest rate6.46.56.56.56.46.46.46.46.46.4
Priv. Consumption/GDP56.756.857.658.358.859.459.559.559.559.5
Imports/GDP62.663.163.163.563.964.664.864.864.864.8
Source: Staff projections
Source: Staff projections

Financial sector restructuring costs

6. The estimates for the resolution of the state and intervened banks are currently based on the establishment of individual asset resolution vehicles for each bank. (See Table VI.3 for details, including historical costs.) These would offer yield maintenance and gain-loss sharing arrangements for the individual banks. Such arrangements have already been established for the management of assets at KTB-SAM and Bank Thai, though they have lain dormant in anticipation of the TAMC. The yield maintenance is projected as a mark-up over the average of savings deposit rates of the four largest private banks, with a 200 basis point pick-up over the life of the gain-loss sharing agreement of 5 years beginning 2001. It is assumed that the AMC/CAP cash flows are broadly in balance through the first

4 years of the arrangements, with remaining principal costs recognized as a lump sum at the end of the period. The interest rates on the debt are assumed to tend towards VA percent in real terms in the medium-term. However, overall interest costs deviate periodically, reflecting the gradual pick-up assumed in the yield maintenance rates.

7. The establishment of the TAMC is not projected to change substantially the overall estimated resolution costs for the state and intervened banks. While the exact modalities of financing and the timing of recognition of losses may change, it does not yet appear likely that the TAMC will lead to a substantially different resolution path and outcome than currently assumed. First, the TAMC will not take on additional state bank assets relative to what has been assumed in the current estimates presented here. Also, this reflects in part an assumption that the TAMC could continue with some of the same asset resolution models already established, particularly for KTB-SAM and Bank Thai. More generally, it would appear reasonable to assume that the existing overall parameters, such as the recovery rate and the interest rate on TAMC bonds (also based on banks’ average deposit rates), will continue to apply.

8. Some additional costs could arise with regards to the resolution of the private bank assets transferred to the TAMC. However, reflecting the small size of the private bank assets expected to be transferred to the TAMC (assuming there are no further rounds) as well as the existing high level of loan loss reserves set aside by most banks, the TAMC should not be exposed to significant down-side risk on this front. Indeed, it is expected that within the confines of the gain-loss sharing arrangements with private banks, the TAMC’s maximum exposure would not exceed 2 percent of GDP.

Table VI.1.Thailand: Medium-term Fiscal Projections, Central Government and FIDF(Baseline, VAT at 7 percent)
2001/022002/032003/042004/052005/062006/072007/082008/092009/10
(B billions)
Total revenue and grants8178919841,0841,1921,3291,4321,5421,660
Tax revenue7478169019931,0931,2191,3121,4121,519
Personal income tax105113125136149166179192207
Corporate income tax151165182201221249268289311
VAT155169188208230257276297320
Excise196215238263290321346372400
Specific business tax131415171820222425
Custom duties97105116127140156168180194
Other tax323438414550545862
Nontax revenue7076839199109118127136
Grants000001234
Expenditure and net lending9941,0921,1851,2841,3911,5001,6141,7371,869
Non-interest9039971,0771,1641,2591,3551,4581,5691,689
Interest9196108120132145156167180
1. Cent. gov. (CG) balance–177–201–201–200–199–170–182–195–209
2. FIDF interest535956636364656769
3. CG + FIDF balance (1–2)–231–260–257–263–262–234–247–262–278
4. FIDF pincipal costs234–1524930–16–1600
5. Augmented balance (3+4)–254–264–242–512–292–218–231–262–278
Government debt2,3572,6182,8553,3623,6513,8694,1014,3634,641
Central government1,4711,6681,8642,0602,2562,4262,6092,8043,013
FIDF8869509911,3021,3951,4431,4921,5591,628
(In percent of GDP)
Total revenue and grants15.115.415.716.016.316.916.916.916.9
Tax revenue13.814.114.414.714.915.515.515.515.5
Personal income tax1.92.02.02.02.02.12.12.12.1
Corporate income tax2.82.82.93.03.03.23.23.23.2
VAT2.92.93.03.13.13.33.33.33.3
Excise3.63.73.83.94.04.14.14.14.1
Specific business tax0.20.20.20.20.30.30.30.30.3
Custom duties1.81.81.81.91.92.02.02.02.0
Other tax0.60.60.60.60.60.60.60.60.6
Nontax revenue1.31.31.31.31.31.41.41.41.4
Grants0.00.00.00.00.00.00.00.00.0
Expenditure and net lending18.418.818.918.919.019.019.019.019.0
Non-interest16.717.217.217.217.217.217.217.217.2
Interest1.71.61.71.81.81.81.81.81.8
1. Cent. gov. (CG) balance–3.3–3.5–3.2–3.0–2.7–2.2–2.1–2.1–2.1
2. FIDF interest1.01.00.90.90.90.80.80.70.7
3. CG +FIDF balance (1–2)–4.3–4.5–4.1–3.9–3.6–3.0–2.9–2.9–2.8
4. FIDF pincipal costs0.40.1–0.23.70.4–0.2–0.20.00.0
5. Augmented balance (3+4)–4.7–4.6–3.9–7.5–4.0–2.8–2.7–2.9–2.8
Government debt43.745.145.649.649.849.148.347.847.2
Central government27.328.729.730.430.830.830.730.730.6
FIDF16.416.415.819.219.018.317.617.116.6
Memorandum items:
Primary balance (percent of GDP)–1.6–1.8–1.5–1.2–0.9–0.3–0.3–0.3–0.3
Nominal GDP (fiscal year)5,3975,8026,2676,7777,3287,8878,4889,1369,832
Source: Staff projections.
Source: Staff projections.
Table VI.2.Thailand: Medium-term Fiscal Projections, Central Government and FIDFAdjustment, VAT at 10 percent from 2002/03)
2001/022002/032003/042004/052005/062006/072007/082008/092009/10
(B billions)
Total revenue and grants8179641,0651,1731,2901,4391,5501,6691,797
Tax revenue7478889821,0821,1921,3291,4301,5391,657
Personal income tax105113125136149166179192207
Corporate income tax151165182201221249268289311
VAT155241268297328367395425457
Excise196215238263290321346372400
Specific business tax131415171820222425
Custom duties97105116127140156168180194
Other tax323438414550545862
Nontax revenue7076839199109118127136
Grants000001234
Expenditure and net lending9941,0921,1801,2741,3741,4761,5811,6951,816
Non-interest9039971,0771,1641,2591,3551,4581,5691,689
Interest9196104109116121123125127
1. Cent. gov. (CG) balance−177−128−116−101−84−36−31−26−19
2. FIDF interest535956636364656769
3. CG+FIDF balance (1–2)−231−188−171−164−147−100−97−93−88
4. FIDF pincipal costs234−1524930−16−1600
5. Augmented balance (3+4)−254−192−156−412−177−84−81−93−88
Government debt2,3572,5452,6973,1053,2803,3643,4453,5373,625
Central government1,4711,5951,7071,8031,8851,9211,9531,9781,997
FIDF8869509911,3021,3951,4431,4921,5591,628
(In percent of GDP)
Total revenue and grants15.116.617.017.317.618.218.318.318.3
Tax revenue13.815.315.716.016.316.816.816.816.8
Personal income tax1.92.02.02.02.02.12.12.12.1
Corporate income tax2.82.82.93.03.03.23.23.23.2
VAT2.94.24.34.44.54.64.64.64.6
Excise3.63.73.83.94.04.14.14.14.1
Specific business tax0.20.20.20.20.30.30.30.30.3
Custom duties1.81.81.81.91.92.02.02.02.0
Other tax0.60.60.60.60.60.60.60.60.6
Nontax revenue1.31.31.31.31.31.41.41.41.4
Grants0.00.00.00.00.00.00.00.00.0
Expenditure and net lending18.418.818.818.818.818.718.618.518.5
Non-interest16.717.217.217.217.217.217.217.217.2
Interest1.71.61.71.61.61.51.51.41.3
1. Cent. gov. (CG) balance−3.3−2.2−1.8−1.5−1.1−0.5−0.4−0.3−0.2
2. FIDF interest1.01.00.90.90.90.80.80.70.7
3. CG + FIDF balance (1–2)−4.3−3.2−2.7−2.4−2.0−1.3−1.1−1.0−0.9
4. FIDF pincipal costs0.40.1−0.23.70.4−0.2−0.20.00.0
5. Augmented balance (3+4)−4.7−3.3−2.5−6.1−2.4−1.1−1.0−1.0−0.9
Government debt43.743.943.045.844.842.740.638.736.9
Central government27.327.527.226.625.724.423.021.720.3
FIDF16.416.415.819.219.018.317.617.116.6
Memorandum items:
Primary balance−1.6−0.6−0.20.10.41.11.11.11.1
Nominal GDP (fiscal year)5,3975,8026,2676,7777,3287,8878,4889,1369,832
Source: Staff projections.
Source: Staff projections.
Table VI.3.Thailand: Estimated Fiscal Costs of Financial Sector Restructuring, 1996/97–2004/05
1996/971997/981998/991999/002000/012001/022002/032003/042004/05
I. FINANCIAL SECTOR ASSISTANCE, PRINCIPAL PAYMENTS (l+2+3+4)5682965−711723231−15249
In percent of GDP12.16.30.1−1.5030.44.0−0.23.7
In billions of US dollars11.88.10.1−1.60.40.55.5−0.46.0
1. Finance Companies42161−125−771902280−16
a. Fifty-six suspended finance companies394120058022900
Liquidity support39420000000
Note exchange program, principal repayments0II0058022900
b. Twelve intervened finance companies1682−70000000
Liquidity support1654−70000000
Long-term investments [including debt-equity conversions)0280000000
c. Remaining finance companies, liquidity support11−50000000
d. Asset recoveries, FRA/AMC0−29−63−77−390−10−16
e. Recapitalization of Krunghai Thanakit (prior to merger)009000000
2. Commercial Banks152252106−81843233282
a. Seven intervened banks137246−118−S33604020146
i. Liquidity support13756−206000000
ii. Long-term investments (including debt-equity conversions)019043000000
iii. Resolution cost of intervened banks, of which: 1/2/0045−S33604020146
Initial recapitalization0045−15−290000
Loan losses000762604020146
b. State-owned banks (Krung Thai and Radanasin)00390−390000
Liquidity support00390−390000
Recapitalization021185025−17−17−17136
c. Remaining commercial banks, liquidity support14−140000000
3. Recapitalization Schemes00463500000
a. Tier-100463500000
b, Tier-2 (net injection)000000000
Gross placement of government bonds006000000
Bond-for-debenture swap00−6000000
4. Other Items−4−17−21−21−20−19−19−18−18
a. Deposit guarantee fee−4−18−21−21−20−19−19−18−18
b. Capitalization of FRA/AMC, etc.020000000
II. INTEREST COST (i+ii+iii+iv)27117768110210010299109
In percent of GDP0.62.51.71.72.01.91.81.61.6
In billions of US dollars0.63.22.01.92.42.32.42.42.6
i. Imputed debt servicing cost44151696766698590101
ii. Interest payments under the note exchange program (closed finance companies)12416131112400
iii. Yield payments on nonperforming loans of intervened banks003525201399
iv. Interest income of FIDF−IS−58−12−300000
Memorandum items:
Stock of debt for financial sector assistance (end of period, in billions of baht) 3/6511,3491,4191,4391,5001,6241,7311,8152,172
In percent of GDP13.828.931.129.829.730.129.829.032.1
In billions of US dollars143738333438414352
Average cost of Binding (annualized, in percent)12.517.25.15.56.16.06.06.05.9
Real average cost of funding9.46.97.35.43.73.03.33.53.4
Sources: Bank of Thailand; FIDF; FRA; and staff estimates.

Increased by the liquidation cost of Bangkok Bank of Commerce and decreased by the receipts for initial privatization of Nakorthon and Radanasin Banks.

Includes loan losses and gain loss sharing, but excludes yield maintenance payments. No receipts from future divestiture of government banks is assumed here.

Includes off-balance sheet liabilities such as the promissory notes issued under the note exchange program for the closed finance companies. These obligations have no immediate cash impact.

Sources: Bank of Thailand; FIDF; FRA; and staff estimates.

Increased by the liquidation cost of Bangkok Bank of Commerce and decreased by the receipts for initial privatization of Nakorthon and Radanasin Banks.

Includes loan losses and gain loss sharing, but excludes yield maintenance payments. No receipts from future divestiture of government banks is assumed here.

Includes off-balance sheet liabilities such as the promissory notes issued under the note exchange program for the closed finance companies. These obligations have no immediate cash impact.

Prepared by Steven Barnett and Vikram Haksar.

For purposes of this study, the debt ratio is defined as the sum of central government debt and FIDF liabilities divided by nominal GDP (see Technical Annex for a further discussion).

With the exception of Thailand, the data in the figure are based on OECD definitions. For Korea this yields a debt ratio lower than estimates based on a definition more comparable to that being used to calculate Thailand’s debt (the OECD definition differs because it excludes the equivalent of FIDF liabilities and consolidates inter-governmental holding of debt).

The FIDF was established as a vehicle for channeling public support to the financial sector during previous crises, and after years of dormancy became active again at the onset of the crisis in 1997. It is an independent body within the framework of the BOT, and maintains separate financial accounts. It is overseen jointly by the BOT and MOF.

More details are provided in the Technical Annex.

The assumption that growth rates exceed real interest rates implies that the debt ratio would fall even if the government runs small primary deficits.

These projections assume that the distressed assets in the state sector are resolved over the next five years, including within the TAMC framework. However, the TAMC is envisaged to have an operational life of up to 10 years, such that the disposal process could in principle take longer than assumed here. As discussed in the Technical Annex, this would mostly affect the timing of the realization of costs, and not the long-term level of the public debt.

On the assumption that real GDP growth was lower by 3 percentage points per year relative to the baseline scenario (i.e., from 5–5½ percent to 2–2½ percent).

Since the interest rates in this exercise are expressed as averages applied to the stock of outstanding debt, a 3 percentage point increase is rather significant, especially in the near term, as it would imply a sharp up-tick in the interest rate on new debt.

Other Resources Citing This Publication