Resolving a Large Contingent Fiscal Liability: Eastern European Experiences
Author: Mr. Mark J Flanagan

Contributor Notes

Author’s E-Mail Address: mflanagan@imf.org
Publication Date:
01 Jun 2008
eISBN:
9781451870176
Language:
English
Keywords:
WP; financial market; monetary policy; current account; real interest rate
Download PDF (505.1 KB)

On occasion, a government may find itself confronted with a need to address a large contingent or off balance sheet fiscal liability. Implementing a settlement raises issues of fiscal sustainability and macroeconomic stability. This paper surveys the key design issues, and draws lessons from recent Eastern European experience. It then considers in more detail the particular case of Ukraine, and how it might approach its own large contingent liability-the so-called lost savings-which at end-2007 amounted to as much as 18 percent of GDP.

Abstract

On occasion, a government may find itself confronted with a need to address a large contingent or off balance sheet fiscal liability. Implementing a settlement raises issues of fiscal sustainability and macroeconomic stability. This paper surveys the key design issues, and draws lessons from recent Eastern European experience. It then considers in more detail the particular case of Ukraine, and how it might approach its own large contingent liability-the so-called lost savings-which at end-2007 amounted to as much as 18 percent of GDP.

I. Introduction

On occasion, a government may find itself confronted with a need to address a large contingent or off balance sheet fiscal liability. Usually, the contingent or off balance sheet debt in question is not being serviced. The experience in several European countries during the last decade provides some indication of how such claims may arise (Table 1). The debts have in some cases dated to periods of macroeconomic stabilization, when fast-decelerating nominal revenue growth, borrowing constraints, and rigid expenditures combined to create large build-ups of government expenditure arrears. In other cases, they have arisen from banking systems, when the government chose to absorb large deposit liabilities. Finally in other cases, they have arisen from damage and restitution claims, related to war or previous expropriation. The imperative to address the debt may arise from accumulating court orders for repayment, from a past guarantee (e.g. of deposits), or simply from political considerations.2

Table 1.

Recent European Cases of Large Off-Balance Sheet Liabilities

article image
Source: IMF Country Documents

The usual public finance and banking recapitalization approaches do not offer much guidance about how to devise a solution. The public finance literature emphasizes the need to disclose, regulate, and control contingent liabilities, to account for them in fiscal analysis, and to provision against their realization (see, e.g., Brixi and Schick, 2002). However, provisioning against a sudden payout would not be feasible when very large amounts are involved. For cases in which the liability arose in the banking system, it is tempting to think of settling the issue by a standard bank recapitalization exercise (see, e.g. Hoelscher, 2006). However, in these cases typically the liability has already been taken on by the government, and there is therefore no underlying bank solvency question (even if the banks in question are administrative agents for the liability).3

There are good economic reasons for a government to address these liabilities with a settlement, but also reasons for caution. The alternative to settlement—explicit and unilateral default—carries a heavy reputational penalty even while potentially leaving the underlying issue intact (to the extent legal actions against the government follow). A well-designed settlement—which eliminates uncertainty—may help to lower risk premia and debt service costs. It can also remove the risk to the budget from court ordered awards (which may lead to disruptive seizures of government assets). However, an overly generous or poorly structured settlement could instead undermine the fiscal policy plans, create debt service problems, and seriously undermine macroeconomic stability.

The diversity of recent Eastern European experience offers perspective about implementing a settlement. A settlement must foremost be designed with economic considerations in mind— fiscal sustainability and macro stability—and the countries in question have taken different measures to contain fiscal and macroeconomic impacts. There are also technical aspects to a settlement design (which can play a key role in determining the design’s overall fiscal and macro characteristics): countries have approached the administrative challenges in different ways; have staged settlements in different ways; and have used different repayment techniques (ranging from cash to netting to securitization).

Against the background of recent Eastern European experience, this paper looks at the issues in more depth, and then considers the case of Ukraine, now facing the problem. Section II considers the key economic considerations underlying a settlement design, and how to manage fiscal and macro risks. Macroeconomic stability issues are illustrated via the IMF’s GIMF model. Section III looks at the issue of technical design of a settlement, elaborating how this may determine the fiscal and macroeconomic characteristics of the settlement. Section IIISection IV shifts the focus to Ukraine’s efforts to resolve the so called lost savings problem, and offers specific design suggestions.

II. Design of a Settlement: Economic Considerations

A. Fiscal sustainability

A change in the level and/or structure of government debt may place fiscal sustainability at risk. Equation (1) sets out the general requirement for fiscal solvency (see Chalk and Hemming (2000)): the present discounted value of expected primary surpluses (PS) must be large enough to pay off the debt (B). The debt-to-GDP ratio will be falling (equation 2) when the primary surplus is large enough to offset movements due to differences between the real interest rate (R), and the growth rate (g) (with an adjustment for GDP deflator growth, (p)).4

(1)Bt=∑j=0∞R(t,t+j)−jPSt+j
(2)PSt>B−1[Rt−1,t−p(1+g)−g](1+g)(1+p)

  • From (1-2) it is easy to see that if the level of debt rises (because the government must pay off a contingent liability), expected primary surpluses may no longer be enough to keep the debt falling or stable (given the level of the real interest rate and GDP growth rate), let alone to ensure solvency. The primary surplus would need to rise at some point to ensure debt sustainability, but it may also be subject to some constraints (Box 1 discusses this issue in more detail).

  • Of course, (1) abstracts from the structure of the debt, whereas the literature has also emphasized a liquidity channel for debt crises. That is, excessive rollover requirements or excessive foreign exchange exposure (both public and private) can raise the probability of a debt crisis (see Detragiache and Spilimbergo (2001)). Thus if the structure of the debt shifts in the course of a settlement, this may be an independent cause for concern.

The sustainable level of primary surplus.

This requires a careful evaluation of tax and expenditure policy constraints and needs:

  • Tax policy. The total tax burden may be constrained by the mobility of tax bases, and by the potential for evasion, including because of weak tax administration. Since taxation distorts incentives, and in particular may reduce the supply and accumulation of factors of production, growth aspirations also put limits on tax rates.

  • Expenditure policy. This may be constrained by historical legacies (e.g. pensions), development aspirations (public investment in human and physical capital), and by equity-redistribution preferences (social transfers).

A medium-term budget framework is very helpful is fleshing out these constraints.

An iterative approach to assessing extra debt carrying capacity will thus generally be needed in designing a settlement. That is, a government must first determine the sustainable level of primary surplus and then assess how much debt can be added, given conservative growth and real interest rate assumptions. This determines the maximum net present value of a settlement, which in turn suggests possible combinations of face value, maturity and interest rate on any new debt to be issued, or the time path and interest rate applied for any cash settlement. The final step is to ensure that the chosen maturity structure of any new debt limits rollover risks. If the resulting settlement is viewed as too small, then the primary surplus must be reconsidered.

Eastern European experience lends some perspective to this decision process:

  • In the case studies, the primary surplus does seem to rise even before a settlement, with the difference between the pre- and post-settlement average about 1½ percent of GDP (Figure 1). In countries with relatively small amounts of claims to be repaid, this may be more than adequate to cover extra debt service costs. Other countries facing larger claims relied less on budget adjustments—in light of rigidities in taxes and spending—and more on sometimes significant (in NPV terms) write downs of the debt (Table 2). For instance, in Bosnia (2004) the government felt the existing primary surplus was the right assumption to underpin the exercise. It made conservative assumptions about growth and the real interest rate, and on this basis identified large necessary write-downs.

  • Countries were also cognizant of debt service requirements. Serbia provides a good example of how to handle this in the context of an upfront securitization. Bonds were carefully structured to limit debt service requirements to no more than 1 percent of GDP in any one year (with an exception related to privatization proceeds, discussed below) (Table 3).

Figure1.
Figure1.

Primary Surplus in Successful Episodes of Debt Regularization 1/

Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

Source: IMF WEO database1/ Bosnia (2004); Croatia (2006); Lithuania (1997); Macedonia (2000); Serbia (2001)
Table 2.

Settlement Structures for Large Contingent Fiscal Liabilities

article image
article image
Sources: IMF Country Reports

Size, in percent of GDP, as of the date specified. Amounts in most cases accumulated over several years.

Maximum and minimum annual payment

Table3.

Structure of the Serbian Debt Settlement

article image
Sources: National Bank of Serbia; and Fund staff estimates.

The total amount above EUR 530 is multiplied by the conversion factor to determine the amount of bonds issued for each maturity.

Initial settlement period GDP (2001)

Actual, through 2007; WEO projection, through 2013; extrapolation of WEO growth projections, through 2016.

B. Macroeconomic stability

A settlement may also place macroeconomic stability at risk. The macroeconomic impact would depend on several considerations:

  • The wealth effect of the debt shock. Fully Ricardian consumers would perceive the need to eventually pay higher taxes, and would thus save the temporary windfall to smooth their intertemporal consumption path. However, consumer behavior is likely to be non-Ricardian, due to different cohorts of agents with compressed planning periods (see Blanchard, 1985 and Weil, 1989); or because some proportion of consumers are liquidity constrained—unable to borrow as much as they need to smooth their consumption (see Gali et al, 2007).5 Thus a positive wealth effect and some demand stimulus is likely.

  • The policy response to rising demand. To the extent the government adjusts, taxing other individuals or reducing other spending, the impact on aggregate demand will be diminished, and potentially even offset. The central bank can also, with enough lead time, engineer adjustments to aggregate demand to offset the shock (crowding out investment and net exports with adjustments in interest and exchange rates). The monetary policy framework may prevent this, however: under a peg and free movement of capital, foreign resources can be drawn in and if not sterilized by the central bank, the excess liquidity may amplify the demand shock.

  • The cyclical position of the economy. If the economy is at or near capacity, a demand shock would lead to a relatively small increase in output, significant additional pressure on inflation (through non-traded goods prices), and deterioration of the current account. Nominal interest rates would rise, and the real exchange rate would as well. On the other hand, for an economy well below capacity, the demand shock might simply raise output.

Empirical examination of economic impacts is difficult. With so few cases to study, the diversity of settlement approaches, and so many shocks to control for, it is not possible empirically to isolate the impacts of settlements or aspects of their design on subsequent macroeconomic developments. And there has been insufficient variance in policy regimes (especially the exchange rate) to capture effects through this channel.

The IMF’s GIMF model can be used to illustrate how these key macroeconomic considerations may play out. The “Global Integrated Monetary Fiscal Model” is a new open-economy macro model with explicit microfoundations. Box 2 provides an overview of key features. See Kumhof and Laxton (2007) for a full description of the model. The model is first calibrated for a steady-state, with the impact of shocks considered against this baseline.

The Global Monetary and Fiscal Model 1/

Microeconomic foundations:

  • Optimizing forward-looking firms and consumers.

  • Nominal rigidities (prices (cascading); wages; plus pricing to market).

  • Real rigidities (habit persistence; investment and import adjustment costs)

Detailed Production structure:

  • Tradable and non-tradable sectors.

  • Raw materials, intermediate goods, and final goods.

  • Intra-industry trade across economies.

  • Endogenous labor and capital supply

Non-Ricardian features:

  • Multiple (4) distortionary taxes, plus productive investment.

  • Liquidity constrained agents without access to financial markets.

  • Life-cycle income patterns.

  • OLG agents with finite lifetimes and high subjective discount rates.

Monetary and fiscal policy reaction functions.

1/ See Kumhof and Laxton (2007) for full details.

For the purposes of this paper, the home country in the model is calibrated to resemble Ukraine, on its post-2002 balanced growth path. That is, net energy imports; a high share of trade and size of government; a low deficit and public debt, a moderately high level of net foreign liabilities, a moderate current account deficit, and a interest rate risk premium of 300-350 basis points. Moderate inflation is assumed under a fixed exchange rate, with fiscal policy targeting the headline deficit. Ratios are drawn from the two most recent years of data, with adjustments to capture the underlying current account balance and fiscal deficit (consistent with IMF 2008).6 Structural parameters are drawn from the literature on the Czech Republic (Laxton and Pesenti, 2003; Allard and Muñoz, 2008); and Western Europe/the U.S. (Bayoumi, Laxton and Pesenti, 2004; Everaert and Schule, 2006; Kumhof and Laxton, 2007). See Appendix I for details of the calibration.

The GIMF simulations highlight the following:

The approach taken in case study countries on the staging of the settlement—discussed in more detail below—is broadly consistent with GIMF findings. Large upfront settlements have been a minority and have generally been undertaken at a more favorable macroeconomic conjuncture (Table 4).

Table 4.

Debt Settlement Timing and Macroeconomic Conjuncture

article image
Source: IMF WEO data base

Median

The case studies, and other country experience, do highlight other approaches to reducing macro impacts and risks (beyond overall settlement design):

  • Creating incentives to save cash payouts. Lithuania created a new class of small-denomination government savings bonds, and offered above-market interest rates to claimants if they maintained their (now-unfrozen) bank account (IMF 1999a).

  • Drawing on broader country experience, penalties for selling newly issued securities can also serve as a direct disincentive to spending. Rediscount restrictions on securities issued would reduce the sale price, and would be effective unless there were close substitutes for “restitution” bonds in banks’ portfolios.

III. Design of a Settlement: Technical Considerations

A successful debt settlement must confront several technical issue of design, including how it can be administered at minimum cost and risk; how it is most effectively staged or sequenced; the technique through which claims will be resolved; and, as a special case of the latter, the use of public assets in a settlement. These design issues, and their role in determining the overall fiscal and macro characteristics of the settlement, are discussed in what follows. Table 5 summarizes findings from the case studies.

Table 5.

Settling Large Contingent Fiscal Claims: Issues Raised

article image
article image
article image
article image
Sources: IMF Country Reports

A. Administration

Implementing a debt settlement presents a government with a number of potential costs and risks. A settlement is often composed of a staggering number of individual claims, raising a potentially significant the administrative burden. Without appropriate controls, improper claims may be paid out, resulting in losses for the government and taxpayers. A settlement that is not final—at risk of an ongoing process of renegotiation—may have deeper costs. It benefits neither the government nor claimants to have ongoing uncertainty. At an extreme, the dangers are easy to see: a solution that is knife-edge in terms of fiscal or macroeconomic sustainability would break down if revisited.7

The experience of European countries shows several strategies for addressing administrative costs and fiscal and legal risks:

  • Stock-taking and verification of claims. Verification protects the government against fraudulent claims, which could otherwise give rise to large losses (given the typically large size of the settlement involved). For instance, in Bosnia verification reduced amounts outstanding by 6½ percent of GDP. How a government chooses to verify is important: a good check and balance is to involve, along with internal audit, an independent body; for instance bank auditors in the case of frozen deposit claims, or the supreme independent audit institution in the case of expenditure arrears claims. Where government capacity is insufficient, the involvement of private sector accounting and auditing firms can also be justified.

  • Prioritization of claims—but within limits. The stock-taking exercise can be used to identify characteristics of the claims—their age, size etc. It is usually the case that a majority of the claims represent a small amount of the total outstanding debt. Settling these small claims upfront and in cash may not be overly taxing in a fiscal or macroeconomic sense, but can vastly reduce administrative costs. Macedonia and Serbia followed this approach. Lithuania, Serbia, and Ukraine all at some point gave preference in settling claims to the elderly, disabled or to those trying to fund their education. This approach raises administrative costs, however, as a new level of verification is required, and it may create legal problems if courts do not accept some criteria for prioritizations (e.g. Ukraine, where the constitutional court rejected an age-of-claimant based criterion).

  • Transparency. Auditing and public disclosure of government payments and receipts helps prevent abuse. When these were not up to international standards, for instance during netting transactions in Russia, Ukraine, and Moldova in the late 1990s, widespread concerns developed about losses to the government (in particular, overvaluation of the private sector’s netted claim; see Commander, Dolinskaya and Mumssen (2000)). Better clarity about the fairness (horizontal equity) of the settlement may help build overall public support for the specific design of a scheme.

  • Voluntary settlement. The legal status of a large contingent claim is often ill-defined, leaving any settlement open to some legal risk. An accepted settlement reduces these risks, and in this context, almost all countries have required claimants to take action to receive their settlement. A time limit can also allow the books to be closed on difficult-to-trace claimants. Some countries (Croatia), have also negotiated with umbrella organizations for claimants to improve the odds that a settlement will be accepted.

  • Moratorium legislation. A settlement may take time to prepare, and in the interim successful legal challenges may lead to disruptions (e.g. seizure of government assets). Ex-post, lawsuits over debt haircuts may also prove successful, since the law may be unclear about the burden between the public and individual interest under a settlement. Again fiscal disruption may result. These concerns were especially prominent in Bosnia (IMF, 2005). The solution adopted there, and elsewhere (e.g. Ukraine), was moratorium legislation, which freezes a claim until budget resources are identified.

B. The staging of a settlement

There are arguments in favor of both upfront and spread out settlements, and the countries in question have used both approaches (Table 5). Given the sums involved, upfront payouts have been via securitization of the debt (Serbia, Macedonia, Bosnia (planned)). Spread out payments have more typically been in cash (Lithuania, Ukraine).

Spread out flexible settlements can have benefits relative to upfront settlements:

  • A phased approach does not lock the fiscal authority into what could be a fiscally unsustainable settlement. Under a staggered approach with some discretion (e.g. Lithuania), if assumptions about real growth, the real interest rate, and the sustainable primary surplus turn out to be incorrect, in principle the NPV of the settlement can be altered ex-post by shifting the time profile of remaining payouts. But the lack of closure is not itself without risk, as discussed below.

  • A phased and flexible settlement schedule can help in the management of macroeconomic impacts and shocks. Phasing reduces the fiscal stimulus in any one year, and thus reduces the pressure on monetary and fiscal policy to manage the demand impact. This could be a critical consideration in countries with weak capacity to implement policies. Looking at the case studies Lithuania (1997) was able to use phasing to good effect in managing macro shocks. When it faced an extremely challenging external environment in 2000-01, it was able to postpone cash payouts, which were targeted to be completed in 10 years, but not specified on an annual basis (Figure 7).

    Figure 7.
    Figure 7.

    Timing of Debt Restitution and Macroeconomic Factors

    Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

    Source: Country authorities, IMF WEO database, and Fund staff estimates

However some considerations argue for upfront settlements:

  • An upfront settlement rapidly achieves closure. For countries going this route, once the scheme has been designed, implementation has been rapid, and that has been the end of the issue.8 At the other extreme, Ukraine (1996) set no time limit for redeeming recognized liabilities, and no annual guidance. This helped avoid problems during 1999-2002, when Ukraine faced difficult external circumstances and severe fiscal financing constraints, but in 2003-04, payments fell below the macro-fiscal capacity of the government (Figure 7). Pressures have since mounted for an entirely new solution and at a much less attractive macroeconomic conjuncture (see Section IV below).

  • An upfront settlement can be leveraged toward immediate financial market development. In Serbia, claims were securitized by a series of bonds, varying in maturity from 1-16 years. Volumes were carefully chosen to provide liquidity along the yield curve (Table 3, above). However, the Serbian case does also contains a note of caution: the bonds issued were fx-denominated, and may have contributed to a creeping euroization of the economy, which is now complicating monetary policy.

In sum, neither approach dominates. Generally, an upfront settlement should be preferred when: (i) macro and fiscal risks are well contained; (ii) there are financial market development needs; and (iii) achieving closure (time consistency) is a pressing concern.

C. The settlement technique

The case study countries have pursued a variety of techniques in settling debt. Some have used cash payments (Croatia, Lithuania, Ukraine); some have securitized the debt (Bosnia, Macedonia, Serbia); some have netted the debt against amounts owed to the government (Moldova, Russia, Ukraine); and some have paid in-kind (Moldova).

The case studies highlight some benefits and pitfalls of settlement techniques (Table 5):

  • Cash repayment and securitization, either alone or in combination, offer a range of benefits to both the government and claimants. They offer the most flexibility, ease-of-administration, and transparency. Among all techniques, they offer the greatest improvement in claimant welfare, as consumers can achieve their optimal consumption bundle at minimal transaction cost (Ramos, 1998). Securitization may have the added benefit of encouraging financial market development (Table 3, above). Cases of successful settlement invariably have taken these approaches.

  • In-kind payments impose a range of costs, and should always be avoided. Country experience shows a high cost of administration, lack of transparency (including valuation problems), and significant fiscal side-effects, in the form of reduced cash revenue receipts for the budget. By encouraging a barter economy, in-kind settlements are thought to have been highly negative for growth (see IMF 1998). Finally, of all the settlement techniques, they improve claimant welfare by the least, by imposing a consumption bundle that may bear little relation to what people want, and one which can only be transformed into another bundle, or shifted into another period, at high transactions cost.

  • Mutual debt settlements (netting) carry heavy risks, and should only be undertaken under very specific circumstances. First, as noted above, netting operations do not lend themselves to transparency and disclosure, and in particular were associated in the CIS with rampant valuation problems (government expenditure arrears tended to be inflated by excessive prices and/or late payment penalties, while the government was willing to write down interest and penalties on tax arrears). Second, if the netting extends beyond bilateral debt cancellation to chains (e.g. involving state enterprises and claimants’ debt to them), the administrative cost becomes very high.9 Third, netting disintermediates the economy and distorts price signals thereby undermining macroeconomic performance. Finally, netting involves considerable moral hazard. Once introduced as a technique in settling debts in the CIS it quickly became entrenched with deepening macroeconomic and fiscal costs. Box 3 discusses this last point in more detail (see also IMF 1999b). In sum, experience suggests that netting is a highly risky strategy, and only if the government can credibly commit to a one-off transaction and can exert control over valuation abuse, should it play a role in a settlement.

Mutual Debt Settlements (Netting) in the CIS

A mutual debt settlement may appear to be a one-off phenomenon, but in reality it can create a problematic incentive for future periods. The government effectively teaches claimants that the way to obtain their remaining claim (if it has not all been repaid) is to not pay current obligations to the government (tax liabilities or payments to such state enterprises as public utilities). Figure B3.1 illustrates how this process was at work in Russia. Arrears did not decline over time, despite netting, and rebounded very rapidly after netting rounds. The situation did not right itself until the underlying problem—the liquidity strains preceding Russia’s debt default—disappeared. Figure B3.2 illustrates declines in compliance consequent on Ukraine’s public utility netting scheme.

Figure B3.1.
Figure B3.1.

Russia: Corporate Tax Arrears and Netting

Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

Source: Country authorities, and Fund staff estimates.
Figure B3.2.
Figure B3.2.

Ukraine: Utility Payment Compliance and Netting

Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

Source: Country authorities, and Fund staff estimates.

When netting becomes entrenched, fiscal and macroeconomic problems multiply. Fiscal sustainability comes into question and fiscal management is complicated as the budget loses the cash it needs to pay debt service, wages and pensions. If the netting operations are of sufficient size, the monetary and price system begin to collapse, with extremely negative consequences for structural adjustment and growth.

D. The use of public assets in a settlement

Many countries have linked their debt settlements to asset sales, either directly or indirectly (Table 2, above). Lithuanian repayments were linked to privatization proceeds, up to a cap;10 and a direct link was also made in Croatia. In Serbia and Macedonia, securities issued as part of the restitution exercise could be used at face value to buy public assets, including both land and companies.11

Political economy and financing considerations seem to lie behind this oft-observed link. In theory, there is no reason for such a link. In practice, linking to privatization may have helped build public support for privatization. Privatization remains controversial in several ex-Socialist economies, not least due to a perception that insiders have been the ultimate beneficiaries of past deals. An explicit earmark lends the appearance that citizens will benefit directly, and in proportion to their claims on the government (claims which, in the case of deposit savings, originally allowed creation of the public assets being sold). The government may also prefer to finance the transactions with asset sales rather than debt issuance on risk/return grounds. A state enterprise is an illiquid asset, often offering a low and unpredictable return in government hands (especially if the governance framework is weak); while a debt obligation imposes rigid servicing requirements (in effect reducing the government’s financial flexibility).

Experience suggests that there are macro-fiscal disadvantages to a privatization link which need to be addressed if it is made. First, as with any earmark, it reduces budgetary flexibility, committing resources to one type of spending, This may subvert higher priorities that unexpectedly arise at a later date. Second, the earmark may create a need for large offsetting policy actions to maintain macro stability. For instance, where privatization reflects FDI under a peg, spending of the proceeds is a direct liquidity injection into the economy (see Davis et al (2000)). It may be relatively simple to sterilize small amounts (either via government or central bank bond issuance), but for large inflows—and privatization has reached 5 percent of annual GDP in some of these countries—the sterilization burden becomes rather large. This was one motivation for Lithuania’s cap on the amount of privatization proceeds that could be used in any one year.

Experience also suggests that any link to government assets needs to be formulated very carefully. A government may not wish to accelerate privatization without a transparent well-governed process in place. More direct debt-asset exchanges have run into problems of valuation, and have been associated with losses to the government (e.g. the Russian loans-for-shares scheme). Finally, the exchange of debt against privatization vouchers could run into the problems observed with earlier voucher privatization schemes: too widespread dispersion of ownership can create problems for corporate governance (see Havrylyshyn and McGettigan (1999)).

IV. Application: Ukraine and the lost savings problem

The government of Ukraine expressed a desire in early 2008 to bring a long-standing contingent liability—the so called “lost savings”—to closure.12 The government has commenced a verification exercise, and set aside up to 2¼ percent of GDP in funds in the 2008 budget for initial payments in cash (including amounts financed by an earmark related to privatization proceeds in excess of the budget target). There would also be a triangular netting exercise involving communal service (local utility) payment arrears and local utility companies’ tax arrears to the government. It remains to be determined how the government will deal with the still large residual claims that will remain after this initial phase.

Evaluation

a. Technical design considerations

Ensuring full verification and undertaking some prioritization would help reduce administrative burdens. Given the passage of time, the inheritance of many claims (as some deposit holders died), and the change in Ukrainian identification documents since the early-post Soviet era, special care is needed in verifying claimants’ identity. In terms of paying claims, the initial focus should be on eliminating the large number of de minimus claims to reduce administrative burden (Figure 8).

Figure 8.
Figure 8.

Ukraine: Distribution of Lost Savings Claims

Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

Source: Country authorities.

The netting exercise should be reconsidered, in light of fiscal risks. An alternative would be to withhold payment for those in arrears to the government. If netting is to proceed, transparency would be crucial: netted claims should be audited, and the government should compensate utilities on-budget for their costs. Second, the government would have to somehow convince those participating that this is a one-off exercise and that new arrears will not be tolerated.13

The role for privatization in the settlement is less clear. The existing earmark against excess proceeds may, by accelerating privatization to high levels, create both a macro and fiscal risk (the latter to the extent transparency in the privatization process cannot quickly be improved). And while this form of financing may encourage faster private sector development (to the extent it speeds up privatization), it works against the possibility of using the settlement towards debt market development. If an earmark is to be used in 2009 and beyond, it would be wise to (i) cap annual spending through this channel, perhaps at no more than 2 percent of GDP; and (ii) eliminate the “excess” formulation of the earmark to keep the focus on designing a transparent, feasible and well-paced annual privatization plan.

b. Fiscal sustainability and macroeconomic stability

From a pure fiscal sustainability perspective, there is room to resolve the lost savings. Public debt is only 10 percent of GDP in Ukraine and falling. It is sustainable for debt shocks in the range of the existing liability, implying that from a fiscal standpoint, either an upfront securitization or spread out cash repayment could work (Figure 9). Of course, consistent with the Serbian experience, in the case of a securitization the feasibility of annual debt service requirements would need to be carefully considered (with private sector debt rollover also kept in mind).

Figure 9.
Figure 9.

Ukraine: Debt Shocks and Fiscal Sustainability

Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

Source: IMF staff estimates.

However, the existing plan—and moreso a full upfront settlement—raises considerable concerns about preserving macroeconomic stability. IMF staff estimates suggest that the Ukrainian economy is presently operating beyond its capacity, with the excess demand contributing to inflation and a rising current account deficit (Figure 10). At the same time, many of the lost savings accounts are thought to belong to liquidity constrained pensioners (given the 16-17 year age of the claims), and it could thus be expected that a good proportion of restitution will be spent. Finally, privatization proceeds are likely to come from abroad and, under Ukraine’s currency peg, this could accommodate the higher demand with money creation. The GIMF simulations above—albeit tailored to a Ukraine steady state—offer some indication of the size of the potential problem.

Figure 10.
Figure 10.

Ukraine: Current Macroeconomic Situation

Citation: IMF Working Papers 2008, 159; 10.5089/9781451870176.001.A001

Sources: Fund staff estimates.1/ See IMF (2008)1/ Estimated via CGER approach. See IMF (2007).

Consistent with the message from the case studies and GIMF model simulations, to limit macroeconomic concerns the government could consider:

  • Spreading out the settlement over several years. In light of already significant 2008 payouts, and high and rising inflation, further payouts in 2008 could be halted. But this should not be a replay of the 1996 law: a time limit, perhaps 5 years, is needed.

  • Moving to a countercyclical fiscal rule. Policy will then tend to offset the second round effects of the shock.

  • Using the room available under the planned gradual transition to a flexible exchange rate to tighten monetary conditions and lean against residual inflationary pressures.

A spread out repayment does not rule out using the settlement towards financial market development. Within a multi-year framework, the government could approach securitization opportunistically. It could continue to pay small amounts annually through the budget, and when the macroeconomic conjuncture is more favorable, it could pay off all remaining claims in one securitization transaction, modeled on the Serbian experience.

V. Conclusions

When a government finds itself confronted with a need to address a large contingent or off balance sheet fiscal liability, it faces several challenges. It must act in a way that preserves both fiscal sustainability and macro stability, and ensure that the settlement design also minimizes administrative burdens and legal risks.

Preserving fiscal sustainability requires attention to both the post-settlement level and structure of government debt. An appropriate level should be assessed on the basis of reasonable assumptions about the evolution of the government’s primary revenue and spending, and reasonable assumptions about the real interest rate and growth. NPV write downs may be needed. Debt structure should avoid bunched maturities.

Preserving macro stability requires attention to the cyclical position and structure of the economy; the size of the wealth effect from the settlement (which depends on, e.g., the share of liquidity constrained consumers), the design of the settlement (including its staging), and the fiscal and monetary policy response. GIMF simulations, calibrated in Ukraine (which has many features typical of Eastern European economies), show that impacts on inflation and the current account can range from alarming to moderate depending on these factors.

Managing administrative and legal risks is a matter of getting a number of technical details right. Eastern European experience highlights the importance of verifying and prioritizing claims, and of transparency. It also provides insight into non-cash settlement techniques (i.e. in-kind payments, securitization, tax or public enterprise receivable offsets, and asset swaps): securitization can promote financial market development, but other methods bring high risks. Indirectly using asset sales to finance a transaction may bring some benefits, however.

The review of Eastern European experience suggests some difficult issues that governments designing a settlement will need to focus attention on. When does a spread out settlement become well enough defined that markets are willing to lend against it, effectively making it upfront? How does a settlement affect the share of liquidity constrained agents, and thus consumption? A government must also assess whether using assets to finance the settlement (i.e. privatization) outweighs the financial market development benefits from securitization.

The Ukraine application highlights how macroeconomic factors can constrain possibilities for settlement even when there is no real fiscal sustainability concern. There is a clear need to spread out the restitution (which does not rule out securitization at a later date). Eastern European experience also suggests Ukraine should avoid netting, be careful not to over-accelerate privatization (to fund an earmark), and should carefully audit the settlement.

Appendix I. The GIMF Model Calibration

Table A.I.1.

Calibration of GIMF: Macroeconomic Ratios and Policies

article image
Sources: WEO, IFS and Fund staff estimates.

Ukraine.

Table A.I.2.

Calibration of GIMF: Structural Parameters

article image
Sources: Laxton and Pesenti (2003); Bayoumi, Laxton and Pesenti (2004); Everaert and Schule (2006); Kumhof and Laxton (2007); and Fund staff estimates.

Ukraine.

Appendix II. Ukraine: A Brief History of the Lost Savings

The so-called “lost savings” arose during 1992, when hyperinflation wiped out frozen Ukrainian household savings deposits. A 1996 Law revalued the deposits, and established state liability. However, the law at the same time clarified that savings were to be redeemed subject to availability of budget resources. No interest was to be paid, nor were any plans made for indexing the debt. Over time, there have been several initiatives to reach a more permanent solution. At the same time, with the debt unindexed and the nominal size of the economy growing strongly, the ratio of this debt to GDP has shrunk dramatically.

Table A.II.1.

Ukraine: Major developments with the lost savings

article image
Source: Ukrainian news media.

In percent of GDP

References

  • Aitken, B. (1999), “Falling Tax Compliance and the Rise of the Virtual Budget in Russia”, mimeo, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Allard, C. and S. Muñoz (2008), “Challenges to Monetary Policy in the Czech Republic: An Integrated Monetary and Fiscal Analysis”, IMF Working Paper 08/72, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Bayoumi, T., Laxton, D. and Pesenti, P. (2004), “Benefits and Spillovers of Greater Competition in Europe: A Macroeconomic Assessment”, NBER Working Paper 10416, (Cambridge MA: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation

  • Blanchard, O.J. (1985), “Debts, Deficits, and Finite Horizons”, Journal of Political Economy, no. 93, pps. 223-47.

  • Brixi, H. and Schick, A. eds. (2002), Government at Risk: Contingent Liabilities and Fiscal Risks, (Washington D.C.: World Bank).

  • Chalk, N. and Hemming R. (2000), “Assessing Fiscal Sustainability in Theory and Practice”, IMF Working Paper 00/81, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Commander, S., Dolinskaya, I. and Mumssen, J. (2000), “Determinants of Barter in Russia: An Empirical Analysis”, IMF Working Paper 00/155, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Davis, J., Ossowski, R., Richardson, T. and Barnett, S. (2000), “Fiscal and Macroeconomic Impact of Privatization” IMF Occasional Paper 194 (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Detragiache, E. and Spilimbergo, A. (2001), “Crises and Liquidity: Evidence and Interpretation”, IMF Working Paper 01/2, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Everaert, L. and Schule, W. (2006), “Structural Reforms in the Euro Area: Economic Impact and Role of Synchronization Across Markets and Countries”, IMF Working Paper 06/137, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Gali, J., López-Salido, J.D. and Vallés, J. (2007), “Understanding the effects of Government Spending on Consumption”, Journal of the European Economic Association, no. 5(1), pps. 227-70.

    • Search Google Scholar
    • Export Citation

  • Havrylyshyn, O., and McGettigan, D. (1999) “Privatization in Transition Countries: A Sampling of the Literature”, IMF Working Paper 99/6, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Hoelscher, D. ed. (2006), Bank Restructuring and Resolution, (Washington D.C.: International Monetary Fund).

  • IMF (1997), “Box 2: The Savings Restitution Plan”, in Lithuania: Staff Report for the 1997 Article IV Consultation and Fifth Review Under the Stand-By Arrangement, IMF Country Report 97/103, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (1998), “Barter in the Moldovan Economy”, in Republic of Moldova: Recent Economic Developments, IMF Country Report 98/58, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (1999a), “Box 2: The Savings Restitution Plan”, in Lithuania: Staff Report for the 1999 Article IV Consultation, IMF Country Report 99/73, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (1999b), “Non-Monetary Transactions and Arrears Accumulation”, in Russian Federation: Recent Economic Developments, IMF Country Report 99/100, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2000a), FYR Macedonia: Recent Economic Developments, IMF Country Report 00/72, (Washington D.C.: International Monetary Fund).

  • IMF (2000b), “Box 4: Budgetary Arrears and Offsets” in Russian Federation: Selected Issues, IMF Country Report 00/150, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2001), “Box 2: Frozen Foreign Currency Deposits” in Federal Republic of Yugoslavia: Request for a Stand-By Arrangement—Staff Report, IMF Country Report 01/93, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2004), “Domestic Claims on the Governments”. Chapter 4 in Bosnia and Herzegovina: Selected Issues (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2005), “Fiscal Sustainability and Budget Spending”, in Bosnia and Herzegovina: Selected Economic Issues, IMF Country Report 05/198,(Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2006a), “Box 2: Repaying the Pensioners Debt” in Republic of Croatia: Second Review Under the Stand-By Arrangement—Staff Report, IMF Country Report 06/128, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2006b), “Box 2: Impact of the ‘Pensioners Debt’ Repayment” in Republic of Croatia: Third Review Under the Stand-By Arrangement—Staff Report, IMF Country Report 06/346, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2006c), Bosnia and Herzegovina: 2006 Article IV Consultation-Staff Report, IMF Country Report 06/371, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • IMF (2008), Ukraine: 2006 Article IV Consultation-Staff Report, IMF Country Report 08/143, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Kumhof, M. and Laxton, D. (2007), “A Party without a Hangover? On the Effects of U.S. Government Deficits”, IMF Working Paper 07/202, (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Laxton, D. and Pesenti, P. (2003), “Monetary Rules for Small Open Emerging Economies”, Journal of Monetary Economics, no. 50(5), pps. 1109-52.

    • Search Google Scholar
    • Export Citation

  • Obstfeld, M. and K. Rogoff (1996), Foundations of International Macroeconomics, (Cambridge: MIT Press).

  • Ramos, Alberto (1998), “Government Expenditure Arrears: Securitization and Other Solutions”, IMF Working Paper 98/70 (Washington D.C.: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Ricciuti, R, (2003), “Assessing Ricardian Equivalence”, Journal of Economic Surveys, no. 17(1), pps. 55-78.

  • Weil, P. (1989), “Overlapping Families of Infinitely-Lived Agents”, Journal of Public Economics, no. 38, pps. 183-98.

1

I would like to thank Bob Ford, Albert Jaeger, Laurent Moulin, Balazs Horvath, Eric Mottu, Milan Cuc, Mark Griffiths and participants at the European Department SEE Network Seminar for helpful comments and suggestions. I would also like to acknowledge the research assistance provided by Ihor Shpak. Any remaining errors are my own.

2

Settling a large contingent debt would likely involve some redistribution of income—those paying to service the debt are likely to be different than those receiving repayment of the debt—so the politics can be complex.

3

In general, the choice between bank recapitalization and assumption of liabilities will depend on the size of the liabilities (and whether it is feasible in a fiscal sustainability sense to recapitalize the bank). Legal considerations may also come into play, for instance in cases where government guarantees are extended.

4

This abstracts from fx financing and currency valuation issues, and from financing via privatization proceeds.

5

The evidence on the importance of Ricardian equivalence in practice is inconclusive (see Ricciuti, 2003).

6

Ukraine is not in a steady state at present, as a credit boom and fiscal expansion have created a positive output gap, high and rising inflation, and a growing current account deficit. If Ukraine soon returns to a balanced growth path (at the underlying ratios used), the results that follow could be seen as having predictive content. Section IV below discusses the specific application to Ukraine in more detail.

7

Other dimensions of closure, including time consistency (building in enough barriers to prevent the issue from being politically revisited); and incentive compatibility (ensuring that the settlement, as designed, does not give rise to new claimant behaviors that may disadvantage the government in the future) are discussed below

8

The exception is Bosnia, where legal problems grounded the initial scheme. Arguably the same legal problems would have grounded an equivalent spread out scheme, since the question related to the haircut on claims

9

Money is often modeled as a matching technology. Arguably, it was the long experience of central planning in the former Soviet states that reduced costs of administrative matching and paved the way for the netting

10

To smooth out the time profile of restitution, the extra budgetary fund set up for this purpose borrowed from the government in anticipation of budgeted privatization proceeds.

11

The bonds carried a below market interest rate, implying a direct link between the amount of privatization and the NPV of the settlement.

12

Appendix II provides a brief history of the lost savings.

13

If the exercise fully pays out for lost savings claims this is less of a concern. If it does not, there is probably little that can be done to contain compliance impacts. In theory, strengthened collection powers for utility companies and stronger enforcement by the tax administration of its claims would help, but these are longstanding needs suggesting that a sudden assertion of change could be viewed with some skepticism.

Same Series

Other IMF Content

Other Publishers

Cover IMF Working Papers
Resolving a Large Contingent Fiscal Liability: Eastern European Experiences
Author: Mr. Mark J Flanagan
Volume/Issue:
Volume 2008: Issue 159
Publisher:
International Monetary Fund
ISBN:
9781451870176
ISSN:
1018-5941
Pages:
40
DOI:
https://doi.org/10.5089/9781451870176.001

Primary Surplus in Successful Episodes of Debt Regularization 1/

Up-front Debt Shock

Sensitivity Tests

Debt Shock with Counter-Cyclical Fiscal Policy

Spread-out Debt Shock

Debt Shock with All Offsets

(staggered; counter-cyclical fiscal response; and flexible exchange rate)

Timing of Debt Restitution and Macroeconomic Factors

Russia: Corporate Tax Arrears and Netting

Ukraine: Utility Payment Compliance and Netting

Ukraine: Distribution of Lost Savings Claims

Ukraine: Debt Shocks and Fiscal Sustainability

Ukraine: Current Macroeconomic Situation