A. The Present Fiscal Situation

94. Since the 1998 crisis, Ukraine has experienced a sharp decline in the general government debt-to-GDP ratio. The general government debt ratio has fallen from 54 percent of GDP to a projected end-2004 value of about 25 percent of GDP. The steep reduction was driven by a sharp reduction in the expenditure-to-GDP ratio between 1997-99 (which helped generate small primary surpluses of 0-2 percent of GDP in 1999-2003); the receipts from the privatization program (which averaged 1¼ percent of GDP per annum); an ex-post negative real interest rate on the government debt stock (-6¾ percent); and high economic growth (6¾ percent per annum).

95. Stabilizing the public debt ratio at the present level of about 25 percent of GDP is appropriate for Ukraine’s situation. At 25 percent, the debt is comfortably below the 40-50 percent of GDP threshold identified in the recent debt intolerance literature,²⁹ and well below the 41 percent of GDP debt level where Ukraine was forced to restructure its debt in 1998. Some comfort zone is necessary: there are suggestions in the debt tolerance literature that strong fiscal institutions may help countries to carry relatively higher debt ratios; Ukraine does not fit the profile of an institutionally strong country.³⁰

96. Ukraine has proposed a 2005 general government budget deficit of 1.9 percent of GDP, a level which would not stabilize the debt ratio in the medium to long term.³¹ This would produce a primary deficit of close to 1 percent of GDP. Given the present debt level of about 25 percent of GDP, and assuming medium-term growth of about 5 percent of GDP, and a real interest rate of about 5¼ percent (above recent levels, and consistent with the present 325 basis points spread on Ukrainian bonds), the debt stabilizing primary surplus would be about ¼ percent of GDP.

97. However, due to several temporary factors, limiting the deficit to 1.9 percent of GDP would lead the debt ratio to decline through 2010 (to about 18 percent of GDP).³²

First, the carry-over of historical debt at lower interest rates would ensure that the real interest rate remains lower than the projected growth rate through 2010. Second, real exchange rate appreciation (see Chapter III) will either lead to revaluation of the two-thirds of debt which is foreign (in the event of nominal appreciation), or even lower real interest rates (in the event of higher domestic inflation). Third, asset sales are likely to continue at a brisk, albeit declining pace. Indeed, the state still owns some 4,000 public enterprises. Transactions in the pipeline for 2005, including Telekom and the Odessa-by-Sea fertilizer plant, will likely generate some 1½ percent of GDP in proceeds alone.

B. Medium-Term Fiscal-Structural Challenges and Their Cost

98. The medium-term fiscal baseline for Ukraine thus appears very positive, but this abstracts from the cost of many needed fiscal structural reforms. While details and plans are still being developed in many areas, in this section, some attempt is made to collect what is known about the impact of tax rate reductions, the replacement of obsolete infrastructure, public administration reform, and social spending enhancement. Initiatives in these areas could together cost almost 11 percent of GDP annually by 2010 (Table 1). This would clearly be infeasible in the absence of other offsetting measures.

Table 1.

Ukraine: Annual Cost of Structural Reforms

(In percent of GDP)

Sources: Ukrainian authorities; and Fund staff estimates and projections.

Table 1.

Ukraine: Annual Cost of Structural Reforms

(In percent of GDP)

			2006	2007	2008	2009	2010
Total costs			4.23	6.31	7.75	9.19	10.64
	Tax reform		0.50	1.75	2.00	2.25	2.25
		VAT rate	0.50	0.50	0.50	0.50	0.50
		Payroll taxes	…	1.25	1.25	1.25	1.25
		Trade tax reform	…	…	0.25	0.50	0.50
	Infrastructure		2.50	2.50	2.50	2.50	2.75
		Transportation	1.50	1.50	1.50	1.50	1.50
		Communal services	1.00	1.00	1.00	1.00	1.00
		Social sector	…	…	…	…	0.25
	Social spending		0.25	0.50	0.75	1.00	1.25
		Education	0.05	0.10	0.15	0.20	0.25
		Health	0.20	0.40	0.60	0.80	1.00
	Civil service pay		0.98	1.56	2.50	3.44	4.39

Sources: Ukrainian authorities; and Fund staff estimates and projections.

View Table

Table 1.

Ukraine: Annual Cost of Structural Reforms

(In percent of GDP)

			2006	2007	2008	2009	2010
Total costs			4.23	6.31	7.75	9.19	10.64
	Tax reform		0.50	1.75	2.00	2.25	2.25
		VAT rate	0.50	0.50	0.50	0.50	0.50
		Payroll taxes	…	1.25	1.25	1.25	1.25
		Trade tax reform	…	…	0.25	0.50	0.50
	Infrastructure		2.50	2.50	2.50	2.50	2.75
		Transportation	1.50	1.50	1.50	1.50	1.50
		Communal services	1.00	1.00	1.00	1.00	1.00
		Social sector	…	…	…	…	0.25
	Social spending		0.25	0.50	0.75	1.00	1.25
		Education	0.05	0.10	0.15	0.20	0.25
		Health	0.20	0.40	0.60	0.80	1.00
	Civil service pay		0.98	1.56	2.50	3.44	4.39

Sources: Ukrainian authorities; and Fund staff estimates and projections.

99. Tax rate reductions, without a compensating effort to broaden the tax base (see below), could entail annual costs of 2¼ percent of GDP. The large informal sector—which some estimates place as high as 33 percent of GDP—is an indication that tax rates need to be cut. While progress has been made for income taxes, high VAT and payroll tax rates remain to be addressed. The baseline already incorporates the planned VAT rate cut from 20 to 17 percent in 2005, but the extra cut planned for 2006, to bring the rate to 15 percent, would cost ½ percent of GDP. Regarding payroll taxes, total rates range from 39-41 percent, and some consideration is being given to reducing them. Cutting by 5 percent to 35 percent would entail annual costs of about 1¼ percent of GDP.³³ Finally, the cost of trade tax reform must be borne in mind, particularly given Ukraine’s aim to join the WTO. As an indication of possible costs, if Ukraine were to cut its average effective tariff from 7 to 5 percent and eliminate half of its export duties, the annual loss to revenues would be about ½ percent of GDP. Payroll and trade tax reductions are assumed to be phased in after VAT rate reductions.

100. After a decade of underinvestment, Ukraine suffers from a significant infrastructure rehabilitation gap, which could demand an extra 2½ percent of GDP in annual spending. In the transportation area, 90 percent of the road network is in need of urgent repair, as are 40 percent of bridges. To maintain and expand the transportation network, the authorities estimate that they would need an extra 1½ percent of GDP per year for some time. In the communal services area, half of water supply systems are either fully depreciated, or in need of urgent repair, and the situation is probably similar for heating.³⁴ The public transportation fleet is also aging (the average bus age for most cities is 12-15 years). The authorities plan to spend an additional 1 percent of GDP per year between 2005-2010 to address these and other communal services deficiencies. While some of the transportation and communal services spending would be expected to tail off after 2010, investments in social infrastructure are likely to pick up. Indeed, there has been negligible investment in school upgrades since 1990, and almost no investment in schools in rural areas: the total need could amount to 2½ percent of GDP, beginning with very small amounts late in the decade.³⁵

101. Partially redressing large civil service pay inequities, without cutting employment (as discussed below), would cost the government over 4 percent of GDP per year. At 9¾ percent of GDP and 26 percent of total general government expenditures, the wage bill is not out of line with EU accession countries (an average of 9¾ percent of GDP and 23¾ percent of spending). However, civil servants in Ukraine are very poorly paid, receiving only 25 percent of what the private sector is thought to receive (including unreported income).³⁶ The authorities do plan a pay reform: they would reorient away from the 70-80 percent of pay which is now based on non-transparent and horizontally inequitable bonuses and allowances towards a system 90 percent comprised of basic pay. They also wish to decompress, although the present compression ratio, at 9.3, is not badly out-of-line with international standards.³⁷ Bringing average wages to about 50 percent of private sector levels by 2010 (still well shy of the 80 to 90 percent typical in more advanced economies), would raise the wage bill by 4¼ percent of GDP to about 14 percent of GDP.³⁸ With no offsetting action, this would leave Ukraine well out of line with regional comparators, and likely crowd out other spending.

102. Education and health care spending would need to rise by 1¼ percent of GDP in order to attain levels observed in more advanced transition economies. The authorities’ work is not well advanced in the health and education areas, but some convergence to higher income regional comparators is likely over time, as Ukraine’s own per capita income rises.

• Education spending amounts to about 4½ percent of GDP, below EU accession comparators (5¾ percent of GDP average). Informal payments are widespread, presenting barriers to access. Wage increases (matched with employment reductions—see below) would add ¼-½ percent of GDP to education spending, and capital projects, another ¼ percent of GDP. Another ¼ percent of GDP would help address operations and maintenance deficiencies; for instance, textbook provision and utilities (though more work is needed in this area to assess specific needs). This would place education spending in Ukraine broadly in line with accession countries.

• Health spending falls well short of levels in EU accession countries, at 3.7 percent GDP (versus 5.6 percent). Health statistics are poor in Ukraine, and show a rising incidence of disease, including AIDS and tuberculosis. As in the education sector, informal payments present a significant barrier to access. Higher wages (again matched by employment reductions—see below) would raise health spending by ½ percent of GDP.³⁹ Nonetheless, another 1 percent of GDP would be needed to approach comparators, money which would allow better equipment to be procured, cover operations and maintenance costs and help reimburse pharmaceutical expenses (in both cases, now largely borne by patients).

103. In addition to education and health costs, pensions and social privileges represent contingent social spending liabilities. If political pressures lead to increases in pension replacement ratios and legal pressures force a full funding of social privileges, costs could mount to well over 5 percent of GDP annually:

• At present and into the next decade , pensions do not present a significant problem for Ukraine. Pensioners suffer no greater rate of poverty than the population as a whole.⁴⁰ However, given current demographic trends, the relatively low income replacement ratio of 35 percent is set to fall over time to 30 percent or lower by 2040. If political pressures force an increase, costs would be significant in the short term. For every 5 percentage point increase in the income replacement ratio, the annual costs would be about 1¼ percent of GDP.⁴¹

• Budgeted social benefits and privileges amount to some 2 percent of GDP at present. This reflects the authorities restriction of privileges to individuals at or below the subsistence level of income (problems with the targeting of benefits nonetheless remain). However, the constitutional court has recently called this restriction into question. If the government is in fact forced to fund all social privileges, the latest estimate of additional annual cost (which accounts for the recent re-registration effort and elimination of double-counting) would be about 4¼ percent of GDP.

C. Identifying Resources to Cover the Cost of Fiscal-Structural Reforms

104. There are a variety of options open to the authorities to help finance the cost of fiscal structural reform (Table 2). Indeed, it will be important to adopt these measures prior to undertaking related structural reforms. For instance, tax base broadening should be linked to tax rate reductions, and employment reductions linked to higher civil service pay. However, the options discussed below do not fully cover costs, and leave an unfinanced gap, perhaps approaching 3¼ percent of GDP.

Table 2.

Ukraine: Resources for Structural Reforms

(In percent of GDP)

Sources: Ukrainian authorities; and Fund staff estimates and projections.

Table 2.

Ukraine: Resources for Structural Reforms

(In percent of GDP)

			2006	2007	2008	2009	2010
Total costs			4.23	6.31	7.75	9.19	10.64
Total identified resources			2.37	4.43	5.47	6.58	7.48
	Tax system		0.75	1.85	1.85	1.85	1.85
		Income tax to 15%	…	0.60	0.60	0.60	0.60
		Increase excises	…	0.50	0.50	0.50	0.50
		Base broadening	0.75	0.75	0.75	0.75	0.75
	Expenditure reallocations		1.17	1.93	2.77	3.68	4.38
		Subsidy cuts	0.25	0.50	0.75	1.00	1.00
		Other recurrent cuts	0.18	0.28	0.39	0.52	0.66
		Employment cuts	0.74	1.15	1.63	2.15	2.72
	Pension reform		0.45	0.65	0.85	1.05	1.25
		Raise retirement age	0.20	0.40	0.60	0.80	1.00
		Base broadening	0.25	0.25	0.25	0.25	0.25
Residual unfinanced gap			1.86	1.88	2.28	2.62	3.16
Memorandum items
Net impact on tax ratio			0.50	0.35	0.10	-0.15	0.15
Net impact on expenditure ratio			2.36	2.23	2.38	2.47	3.01
Net impact on recurrent spending			-0.14	-0.27	-0.12	-0.03	0.26
Net impact on the wage bill			0.24	0.41	0.87	1.29	1.67

Sources: Ukrainian authorities; and Fund staff estimates and projections.

View Table

Table 2.

Ukraine: Resources for Structural Reforms

(In percent of GDP)

			2006	2007	2008	2009	2010
Total costs			4.23	6.31	7.75	9.19	10.64
Total identified resources			2.37	4.43	5.47	6.58	7.48
	Tax system		0.75	1.85	1.85	1.85	1.85
		Income tax to 15%	…	0.60	0.60	0.60	0.60
		Increase excises	…	0.50	0.50	0.50	0.50
		Base broadening	0.75	0.75	0.75	0.75	0.75
	Expenditure reallocations		1.17	1.93	2.77	3.68	4.38
		Subsidy cuts	0.25	0.50	0.75	1.00	1.00
		Other recurrent cuts	0.18	0.28	0.39	0.52	0.66
		Employment cuts	0.74	1.15	1.63	2.15	2.72
	Pension reform		0.45	0.65	0.85	1.05	1.25
		Raise retirement age	0.20	0.40	0.60	0.80	1.00
		Base broadening	0.25	0.25	0.25	0.25	0.25
Residual unfinanced gap			1.86	1.88	2.28	2.62	3.16
Memorandum items
Net impact on tax ratio			0.50	0.35	0.10	-0.15	0.15
Net impact on expenditure ratio			2.36	2.23	2.38	2.47	3.01
Net impact on recurrent spending			-0.14	-0.27	-0.12	-0.03	0.26
Net impact on the wage bill			0.24	0.41	0.87	1.29	1.67

Sources: Ukrainian authorities; and Fund staff estimates and projections.

105. The VAT base can be broadened, and income and excise tax rates raised, to help pay for VAT and payroll tax rate reductions. With a revenue ratio of 36 percent of measured GDP, and a substantial informal sector, the burden that government places on the economy in Ukraine is not high. Even if the revenue ratio is maintained as actual and measured GDP converge, it would be slightly below the levels of eastern European comparators, and would fall well below them if 2½ percent of GDP in revenues were to be cut as discussed above.⁴²

• There is significant scope for upfront tax base broadening: ¾ percent of GDP in VAT preferences would remain even after permanent elimination of ½ percent of GDP in preferences planned for 2004.⁴³ Key areas where exemptions could be removed include, publishing, firms registered in Chernobyl affected areas (an exemption subject to wide abuse), and automobiles. Base broadening prospects are not confined to the VAT: ¼ percent of GDP in import and profit tax preferences could also be eliminated, targeting publishing and free economic zones.⁴⁴

• Regarding the income tax, so long as legislation is not reversed, the rate will rise to 15 percent in 2007, generating over ½ percent of GDP in additional annual revenues. Excise tax collections are low in Ukraine relative to eastern European comparators, at 2¾ percent of GDP (versus 3½ percent). This suggests some scope to raise rates (for instance, on automobiles, where some excises are now suspended). An extra ½ percent of GDP in additional annual revenues could be targeted.

106. Civil service employment can be significantly scaled back to help fund civil service pay increases. General government employment in Ukraine, at close to 9.1 percent of the population and 21.2 percent of employment, is well above the average in EU accession countries (7.4 percent and 16.4 percent). At an attrition rate of 4 percent per year, a typical figure, there is ample scope to both reduce and realign employment over time. If 3½ percent per year of this was devoted to reduction, ½ percent (amounting initially to some 22,000 persons) could be hired, while other employment needs could be covered by transferring employees from overstaffed functions.⁴⁵ The total employment reduction would amount to almost 20 percent by end-period, entailing annual savings of about 2¾ percent of GDP (at the higher wage rates suggested above).⁴⁶ While this would leave employment in the range of EU accession countries, attrition-based reductions in employment would need to continue 3-4 more years to bring the wage bill down towards the regional average.

107. New pension reforms can be enacted to offset the impact of any payroll tax cut on the pension system. Phasing in an increase in the retirement age for women from 55 to 60 years (over a period of 10 years) could lift the balance by over 1 percent per year by 2010. There is also scope to expand the pension contribution base to those taxpayers operating under the simplified and fixed agricultural tax regimes (which could raise 0.2 percent of GDP in new revenues). While not a public sector measure, full budget transfers to cover non funded pension types could lift pension fund revenues by ¼ percent of GDP annually (and thus the payroll tax cut would be fully financed at the level of the pension fund).

108. Recurrent spending reallocations could provide additional resources for education and health initiatives.Subsidies could be cut by 1 percent of GDP. At present they amount to about 1½ percent of GDP, including 1 percent to the coal sector and ½ percent of GDP to the agricultural sector. Agricultural subsidies may be politically difficult to cut, in view of present international practice, but the coal sector could be targeted, and a gradual phase out provided to allow the sector to adjust. In addition the large cuts in employment proposed above imply less of a need for operations and maintenance spending. If this support spending could be cut by just half as much, proportionately, as employment would be, some ½ percent of GDP in annual savings could be realized. A good budget process (discussed below) would help to identify where spending was no longer needed.

D. Options to Address the Remaining Resource Gap and Other Potential Costs

109. There are options to help address both the remaining resource gap and potential contingent spending liabilities, but ultimately some reforms will need to be scaled back. While there may be scope to finance some additional public investment, financing the whole gap of 3½ percent of GDP would not be feasible: this would quickly lead to sharp increases in the public debt ratio, to levels well above what would be considered appropriate for Ukraine. Other options to generate resources should be pursued, but are too uncertain in their timing and yield to be pre-programmed. Measures to contain contingent spending liabilities will also need to be implemented, lest these liabilities otherwise sidetrack the whole process.

110. Consideration could be given to financing some additional public investment. An additional 1 percent of GDP per year in public investment could in fact be financed between 2005-10 without increasing the debt ratio by end-period. Some institutional constraints would need to be overcome, as discussed below, and this approach would need to wind down (since the factors underlying the baseline debt decline are temporary). Linking financing to additional public investment would be attractive in one sense: it would emphasize the aim to preserve government net worth.

111. Tax administration and better management of state assets offer the prospect for significant gains, but international and Ukrainian experience indicates that these channels should not be relied upon ex-ante:

• There is substantial scope for tax administration improvements in Ukraine. Looking only at the VAT, the productivity ratio of only 0.3 compares poorly with the 0.37 average in more advanced central and eastern European countries.⁴⁷ While this may seem a trivial difference, it amounts to 1 percent of GDP per year in revenue collections.⁴⁸ Gains of a similar magnitude may be available through other taxes. However, international experience with tax administration reform is that the gains are uncertain in both magnitude and timing.

• There is also great scope for improving the management of public assets. Eliminating the estimated quasi-fiscal energy sector deficit of ½ percent of GDP in the electricity and heating sectors would allow for extra investment in this sector and for increased dividend payments. For the 2,600 state enterprises operating under the umbrella of ministries and central and local executive agencies,⁴⁹ data for 2003 show a decline in profitability by about ¼ percent of GDP to 1¼ percent of GDP total. Indeed, despite stronger than expected economic growth, 41 percent of enterprises did not meet their financial targets, and one-third were loss-making. The government is presently servicing some 2 percent of GDP of state enterprise debt. With many of the profitable state enterprises likely to be privatized between 2005-10, some improvement in the performance of remaining enterprises is necessary just to compensate lost dividend payments. If their performance were to be improved, they would also be more marketable, and privatization could be accelerated. Higher proceeds would be a useful way to finance one-off public investments, but Ukraine’s past problems with overestimation illustrates the need to receive them before programming their use.

112. With a difficult-to-bridge resource gap, it will also be important for the government to take measures to contain additional contingent spending liabilities. In this context, it would be possible to improve the income replacement offered by pensions by increasing the voluntary component of pensions savings (accelerating development of the second pension pillar). Further parametric changes (i.e., additional increases in retirement age) could also be considered. There should also be ample scope to better target social spending. Looking at social privileges, as noted in recent parliamentary deliberations, despite efforts to tie receipt to a below subsistence level income, the bottom income quintile gets only 8 percent of the privileges by value, while the top quintile gets 35 percent.

E. Other Constraints to Accelerating Fiscal-Structural Reforms

113. While the major constraint to fully adopting structural reform initiatives is a lack of resources, a number of other issues argue for some delay of initiatives. There are key institutional impediments to accelerating debt-financed investment, and many institutional changes would be needed to fully support recurrent spending reprioritizations. The discussion here identifies the issues; a full discussion of each is beyond the scope of this paper.

114. Pressing ahead with an expanded investment program, especially if it partly debt-financed, should await several institutional improvements:

• There is a need to improve project management—selection, appraisal, and implementation—and to buttress the process against political interference. Poor selection of projects will not deliver the desired return. Poor costing may undermine all projects, as budget allocations are spread too thin.

• Fiscal policy would need to be anchored in some fashion. Loosening targets to pay for one-off investment could lead to indiscipline on the recurrent side. In this context, establishing targets for the current balance could be useful, and a number of countries have taken this route. It would also be important to maintain strict controls over subnational borrowing, to ensure municipalities’ capacity to repay.

115. Recurrent policy reforms would also benefit from some delay to put in place supporting institutional reforms (many of these were noted in the recent Fiscal ROSC):

• A fully developed medium-term expenditure framework, called for in the budget code but not yet fully realized, would greatly assist in reprioritizing spending in a multi-year context. With forward estimates now in place, the challenge will be to force line agencies to justify and show, in their forward planning, continuing programs and their costs, alongside new programs and their costs. The multi-year ceilings identified then need to be enforced.

• A fully developed and reoriented audit function would help to identify the effectiveness of programs in achieving their stated goals, and whether there might be more effective means. Ineffective programs could become candidates for cutback.

• Adjustments need to be introduced to Ukraine’s intergovernmental framework. More objective criteria need to be developed for the allocation of capital grants, to assist lower levels of government in their budget planning.

• State enterprise governance must be improved. A system for timely and regular monitoring of activities needs to be introduced; accounting needs to be upgraded; a clear dividend policy established; and a system established for systematically reviewing the continued use of public assets. Commercial goals need to be clearly established, independent regulatory regimes strengthened (and introduced where necessary) and compensation mandated and granted for any non-market activities.

116. There are some clear sequencing issues that must be respected. Given the political economy of some expenditure reform, social spending reforms should be in place concurrent with, or before, subsidy (and thus enterprise restructuring) reforms take place. Without a better safety net, the pressures from interest groups for reversals would likely be intense.

F. Conclusions

117. While Ukraine has made enormous progress in restoring the health of public finances, huge fiscal challenges loom. Work remains to be done in defining the specifics of the full fiscal-structural reform agenda (especially in social spending areas). However, based on the authorities’ work, and comparisons with more advanced transition economies, addressing tax, infrastructure, civil service, pension and social spending deficiencies could cost the budget almost 11 percent of GDP annually by 2010. Not addressing them would subvert economic growth and curtail progress in reducing poverty.

118. This paper has provided some well-defined measures that could cover perhaps two-thirds of likely reform costs, and has discussed strategies to fill the remaining gap. There are well defined tax, public employment, subsidy and pension changes which can be taken to neautralize much of the impact of the reform measures above. There is also scope to finance some additional public investment, but an annual gap of some 2 percent of GDP would remain. Other measures, covering tax administration and the management of state assets could help to close much of the remaining gap, but involve significant institutional changes and cannot be relied on ex ante.

119. Beyond the need to fully develop reform plans and the existence of resource constraints, institutional considerations also argue for some delay in implementing structural reforms. Management capabilities need to be built up for an expanded investment program, and budget process reforms need to be introduced to facilitate planning, including the sequencing of measures. In this constrained environment, the political process, supported by a better medium-term budgetary analysis of costs, will need to sort out near-term reform priorities.